Wikipedia talk:Manual of Style/Dates and numbers/Archive 150

This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 145

←

Archive 148

→

YYYY-MM-DD not good enough because of when day is missing in citations (YYYY-MM?)

Maybe there is a workaround? In citations w/YYYY-MM-DD and then say a stray "May 2014" is that the correct style? It seems inconsistent (that is not ok) and YYYY-MM gives a template error (actually I've had this issue backwards and "fixed" to "May 2014"). When introducing a stray citation it seems to violate WP:DATERET (I'm not sure if it applies to citations). Worst possibility is if people refrain to add a citation. Is this a good argument generally against YYYY-MM-DD in citations? Note: the issue doesn't strictly apply to archivedate etc. but up to now I thought those should be use the same format (but I see some people might even prefer two date formats is citations (and even three in article as a whole)..). YYYY-MM-DD seems to be more neutral than MDY (and DMY) but if date formats should be changed when forced as described, then why start with YYYY-MM-DD (it's not like "the script" makes changing this easy as it is)? Another issue (but not for old computer articles?!) are pre-1582 that might pop up..:

WP:CITESTYLE: "Although nearly any consistent style may be used, avoid all-numeric date formats other than YYYY-MM-DD, because of the ambiguity concerning which number is the month and which the day. For example, 2002-06-11 may be used, but not 11/06/2002. The YYYY-MM-DD format should in any case be limited to Gregorian calendar dates where the year is after 1582."

And: Should this say forbidden, not avoid? Note this is a separate issue from if "other than YYYY-MM-DD" should be dropped. comp.arch (talk) 19:50, 4 February 2015 (UTC)

The quoted text says the avoidance is "because of the ambiguity concerning which number is the month and which the day" – but that does not seem like a correct rationale for avoiding YYYY-MM. I suppose the problem with "2002-06" is the somewhat different problem of the ambiguity of whether it refers to "June 2002" or "the range of time between 2002 and 2006". WP:BADDATEFORMAT doesn't even seem to discuss YYYY-MM at all. I think some comment about it should be added (and may proceed to do so). —BarrelProof (talk) 20:27, 4 February 2015 (UTC)

Before you edit or discuss further, it may be worth reviewing last year's RFC, archived from this very Talk page. – Jonesey95 (talk) 20:29, 4 February 2015 (UTC)

Yes. I see that you yourself brought up the ambiguity between YYYY-MM and YYYY-YY (and YYYY–YY, depending on whether the reader and editor are paying strict attention to the difference between a dash and a hyphen), but I also see that it may be wise to seek a greater indication of consensus before taking action. —BarrelProof (talk) 20:49, 4 February 2015 (UTC)

Note, I'm not pushing for YYYY-MM date format (or even sometimes YYYY). It just seems to be a natural consequence of the YYYY-MM-DD citation style. Say for may "May 2003" what should be put into ditation templates? "2003-04" is just dangerously ambibious ("May 2014" wouldn't strictly be). "2003-04-00", would be bad "2003-04-??" even worse as the date is not missing, as in not known (well, kind of). "2003-04-NA" I don't really like (currently invalid). Since there is no good solution, what I'm saying should YYYY-MM-DD be out? It seems when a citation like "May 2003" pop up, some rule of the MOS needs to be broken. If that is the case the MOS is broken. Isn't there a rule in WP that says policies can't be inconsistent and then it needs to be fixed? comp.arch (talk) 09:18, 5 February 2015 (UTC)

The MOS is not quite inconsistent but their are some problems. As you picked up, an article with only 2011-12-31 style citations is fully compliant with the MOS. But if even a single citation date without a day is required then it has to be of the form December 2011 (mmm yyyy), which then conflicts with the MOS because it differs from the other citations. Solutions raised so far are:

Quietly mix 2011-12-31 and December 2011 styles and hope nobody raises a fuss.
Ban yyyy-mm-dd altogether.
Allow yyyy-mm-dd but as soon as the first citation without a day is added then convert the all the citations over to d mmm yyyy or mmm d, yyyy. Runs against the grain of WP:RETAIN.
Allow yyyy-mm. Some dates can be confused with year ranges (eg 2011-12) but most dates will be unambiguous (eg 2012-11). Will the world really end if some people can't figure the date?
Allow yyyy-mm in dates and change MOS to specify that year ranges are in the form yyyy-yyyy.
Write it as yyyy-mm-01, ie just assume the first day of the month. Not strictly true but does it really matter?

Previous discussions have not managed to agree on any of the above and no new suggestions have come forth. Stepho talk 13:56, 5 February 2015 (UTC)

My god, you're right! (6) is absolutely out. (1) might be OK, with the understanding that anyone who fusses will simply be banned for life. There really are places (e.g. tables) where YYYY-DD-MM is the best presentation, and now that you've brought it up, there may be times that YYYY-MM is the best presentation too. Here's a thought:

4X. Allow YYYY-MM, but only in e.g. tables or other places where a header or such makes it clear what's meant

or even

4Y. Both YYYY-MM-DD and YYYY-MM are allowed only in [etc etc like I just said for YYYY-MM]

EEng (talk) 15:29, 5 February 2015 (UTC)

Does SPELL09 require the "integers" to be exact? I don't think so

MOS:SPELL09 says: "Integers from zero to nine are spelled out in words."

Someone just changed an edit of mine from e.g. "... about three times as fast ..." to "... about 3 times as fast ...", explaining "these are not exact integers".

I reverted. I do not believe that SPELL09 requires the number expressed in the text to be exact. The number was written in the text as an integer, so it should be spelled out as a word. It does not matter that the actual measurement might have been, say, 3.1 times as fast, or that the source knew that "three times" was an approximation.

Who's right? Jeh (talk) 21:11, 3 February 2015 (UTC)

I'm the one that made the change. I do a lot of edits changing small integers written as a digit to spelling it out. However, these figures about the relative speed are not really integers, they are measurements or even approximations. So my opinion is that they should not be spelled out. Bubba73 ^{You talkin' to me?} 21:32, 3 February 2015 (UTC)

By the way, I was not specifically reverting your edits. I saw the "three" and thought it was wrong. Bubba73 ^{You talkin' to me?} 18:29, 5 February 2015 (UTC)

A Manual for Writers of Research Papers, Theses, and Dissertations by Turabian, section 2:23 says "in scientific material all measurements are expressed as figures. In nonscientific material numbers are sometimes spelled out and sometimes expressed as figures ... The Chicago style is to spell out all numbers through one hundred. Exact numbers over one hundred are written as figures." And "by exact numbers" she means integers.

Section 2:49: "In mathematical, statistical, technical, or scientific text, where physical quantities for distance, length, areas, volumes, pressures, and so on [i.e. measurements], all amounts should be expressed as figures, whether they are under or over one hundred." The CDC 6600 article is technical writing and ratio of speeds of the computers is a measurement.

In the APA style, Publication Manual of the APA, 3.42, "Numbers expressed as figures", part d, "numbers that represent statistical or mathematical functions, fractions or decimal quantities, percentages, ratios, and quartiles". And it gives the example "3 times as many". The relative speeds of the computers is a ratio.

So my conclusion from both of these sources is that since the relative speeds is a measurement and not an integer (not "exact"), it is a ratio, and the article is technical writing, digits should be used in this case. Bubba73 ^{You talkin' to me?} 22:24, 3 February 2015 (UTC)

So "about three times as fast" would be incorrect? I'm sorry, but I think that position is nonsensical. The numeral "3" (even without ".0" appended) suggests an accuracy that is not there; it is the very fact that the number is approximate that demands the less-technical form "three". Also: WP is written for the general audience. We do have articles that present technical information, but that does not make this "technical writing", and certainly not "scientific". Jeh (talk) 22:38, 3 February 2015 (UTC)

Jeh is right, certainly in the case where the number is "about". EEng (talk) 22:42, 3 February 2015 (UTC)

I think it is "about 3", not "about three", since it is not an integer - it is a ratio, as in the paragraphs above. Bubba73 ^{You talkin' to me?} 23:16, 3 February 2015 (UTC)

Huh? How does X being a ratio somehow mean X can't be an integer? If you're right about this then we better revise Molecule, which says that "A compound's empirical formula ... is the simplest integer ratio of the chemical elements that constitute it. is the simplest integer ratio of the chemical elements that constitute it." I guess Dalton was wrong. EEng (talk) 23:50, 3 February 2015 (UTC)

It is possible for a ratio to be an integer, but in this case it certainly isn't. Bubba73 ^{You talkin' to me?} 00:47, 4 February 2015 (UTC)

It is correct that "about three" is not an integer, but "about three" is not a number. The number in this case is "three" which is an integer, so according to our style, we would spell out that number as "three". I hope that helps clarify. SchreiberBike talk 00:07, 4 February 2015 (UTC)

OK. Bubba73 ^{You talkin' to me?} 00:47, 4 February 2015 (UTC)

Just curious, would you say that the 6600 is faster than Stretch by "a factor of 3" or a "factor of three"? Bubba73 ^{You talkin' to me?} 00:53, 4 February 2015 (UTC)

If I were saying it, it would sound the same either way. *grin* Seriously, because I know such things are estimates, and there are specific workloads where the factor would be less or more than three, I'd write "a factor of about three". Jeh (talk) 02:37, 4 February 2015 (UTC)

Search/replace

I've been really careful when search/replacing say " Gbps" to "<nobreak>Gbit/s" to exclude titles and quotes. Until I saw, MOS:QUOTE: "Formatting and other purely typographical elements of quoted text should be adapted to English Wikipedia's conventions [..] These are alterations which make no difference when the text is read aloud".

It's a hassle, sometimes, to have to read everything over (a script would be needed to change automatically (and for case insensitive search (in Chrome..)).

As space and no-break space read aloud the same, I thought I had been overtly conservative. I wander still if that change should be disallowed (for titles since its a different Unicode code point if you had to look titles up..). I'm more sure however that, for titles (but not quotes), changing "Gbps" to "Gbits/s" that read the same, would not be ok. comp.arch (talk) 22:05, 4 February 2015 (UTC)

If you really feel you need to change Gbps to Gbit/s in quoted material, then indicate the change via brackets: [Gbit/s] -- this is a change to the text which needs to be called out. Using nbsp where called for is entirely different, and can be done silently. EEng (talk) 23:29, 4 February 2015 (UTC)

Could do that but I will not :) That is probably an even more inferior solution - people might wander why the brackets and did the editor who put them in really do it right or misinterpret. It also has the "hassle" issue.. and then I might as well change back to Gbps. If changing titles is NOT allowed then maybe a note be added to MOS:QUOTE, just in case people the rule also applies to titles (they are kind of a "quote"). For quotes in "quotes" (e.g. not titles), actually it seems the rule allows substituting one abbrevation of a unit for another that means the same thing and reads the same out loud. comp.arch (talk) 09:07, 5 February 2015 (UTC)

I wouldn't assume that abbreviations and symbols are pronounced the same; there are all sorts of inconsistencies. To pick just one example, psi and lbf/in² are often pronounced differently and though in my experience the f is silent, that may simply betray my background. Certainly swapping one for the other would be a shift in register and/or speciality, akin to translating the writer's words into at least a different dialect. NebY (talk) 18:49, 5 February 2015 (UTC)

Exactly. EEng (talk) 18:59, 5 February 2015 (UTC)

British geography units of measurement

Should British geography articles have metric units first? MoS is not very clear about that, and articles dealing with that are often a complete mess.--Carnby (talk) 23:39, 5 February 2015 (UTC)

Please dear God, not again. EEng (talk) 05:16, 6 February 2015 (UTC)

From previous discussions it is obvious that Britain hasn't decided whether it will be Imperial or Metric and is continuing with a mix of units. So the first major contributor to a topic may choose the units and the rest should politely continue with the same units. Stepho talk 07:14, 6 February 2015 (UTC)

It's a most unsatisfactory situation that a few warriors have insisted on replicating on en.WP. Anyone who wants to fight them risks a nervous breakdown and/or site-banning. Tony (talk) 07:39, 6 February 2015 (UTC)

(edit conflict)Not exactly, Stepho. The Manual of Style does provide guidance at WP:METRIC. It is a minefield. As a result of repeated arguments here and elsewhere and conflicting edits to articles, sanctions now apply regarding units in UK articles as well as to editing the MoS in that regard and discussions here (see top of this page). Carnby, one of the main reasons these sanctions were agreed was that systematic attempts to prioritise one system or the other had caused long and bitter conflict. The archives of this page are full of arguments on the subject that went on for months, and there are similar arguments on other pages too. NebY (talk) 07:45, 6 February 2015 (UTC)

I have wondered about this: is GB not offically metric? Has there been some kind of unravelling? ~ J. Johnson (JJ) (talk) 23:14, 6 February 2015 (UTC)

The UK is mostly metric but not all things are one or the other, one of the most obvious is that vehicle speeds and related road signs are still in miles per hour and despite fuel being dispensed in litres we still use miles per gallon. Editors have argued that articles on British subjects should replicate this mix of metric/imperial, although the daft thing is they dont want one or the other but just the order (either metric/imperial or imperial/metric) in each article to reflect usage. The fact both system are used without a major problem and most readers can easily cope whichever order they are in seems to have passed them by. Perhaps they think only British users read British articles and would be confused which is also daft as the rest of the world doesnt have a problem reading for example French subjects with Metric first and American subjects with Imperial first (or whatever the Americans call it). These daft arguments from the different camps have wasted a lot of wikipedia time completely missing the point that all the reader is looking for is some consistency and they dont really care which is first as long as they can see both. MilborneOne (talk) 23:35, 6 February 2015 (UTC)

For all our sakes could we please not restart the debate on this guideline? It has been highly acrimonious in the past and the debates can last for months on end (hence EEng's comment).

If you're interested in more detail of the position in the UK, I don't mind answering on my talk page. But nothing good will come from continuing this discussion here. Kahastok talk 23:48, 6 February 2015 (UTC)

Perhaps it would help to include a link to the general sanctions that apply to units of measurement in UK articles in MOSNUM. Wikipedia:General sanctions/Units in the United Kingdom Michael Glass (talk) 02:26, 12 February 2015 (UTC)

K (or k) the computer kilo- prefix - for ancient computers.. not using bytes

0. If there is a better place to discuss this point me to this.. I started at Talk:Kilobyte#kilowords - "K" ambigious.

1. In general I add non-break space (and B for byte) in cases like "64 KB". However for eg. at NCR 315 I'm not sure what is best. Note the unit is missing, and it's not KB. Would you add the space "10-40 K memory" it looks like Kelvin (if you are dense and not reading context..) but "40K" looks like 40KB to me.. Even if the unit is the byte (and you just are not sure) or if you know it is not a byte what would you do?

2. In the above case, both "k" (1000) and "K" (1024) are used, I could guess that 1000 is meant but I'm not totally sure. If I "fix", nobody might see this and think I did it wrong. Would you leave it inconsistent or change per BOLD..? Or add some banner? The one I found isn't ideal.

4. Any other issue you see with this and the case at Talk:Kilobyte? Should something be added about obscure (exceptional) computers in the MOS? comp.arch (talk) 16:31, 15 January 2015 (UTC)

NCR 315 is unclear and I couldn't see a source cited, so I added some citation requests. Dondervogel 2 (talk) 17:22, 21 January 2015 (UTC)

If I'd stumble over something, where B is not an octet (8 bits), I'd convert unclear units to decimal SI-Units counting bits (lower case k for 1000 if needed, M for 1000,000, etc.). In theory B is the smallest addressable unit of memory, not necessarily an octet, but the last time this was relevant is decades ago. For octets I'd pick WP:COMPUNITS with binary KB, MB, etc. But upper case K is unclear, and BOLD is good. –Be..anyone (talk) 15:06, 29 January 2015 (UTC)

Any forward movement on this needs to start with WP:COMPUNITS. Good luck. EEng (talk) 15:20, 29 January 2015 (UTC)

The word for the smallest addressable unit of memory, particularly on machines where that is larger than what is usually used to store one character, is "word", not "byte". e.g. an HP 2100 can have up to 32K (and that's a binary "K") words of 16 bits each. This could be described as "64 KiB" but the bytes are not individually addressable. Historically, machines whose smallest addressable unit of memory was larger than eight bits did not generally use eight-bit characters. The HP 2100 is an exception. When you look at machines with word sizes above 16 bits the exceptions are very few. Jeh (talk) 15:34, 29 January 2015 (UTC)

I told you how byte used to be defined as of 1976, and why OSI standards used octet to avoid confusion, please add references if the definition was changed after 1990. –Be..anyone (talk) 23:16, 1 February 2015 (UTC)

Was this a reply to me? I'm not sure how it applies. I thought we were talking about machines that didn't use bytes.

Whatever - I do not think it is reasonable to quote memory sizes in bytes if the machine isn't byte-addressable and doesn't use 8-bit characters. I realize that "byte" has been used to mean other than 8 bits, but such uses are rare and not what most readers expect the word to mean.

If the machine has some sort of commonly-used character representation (six-bit characters were very common), then we could expand the memory size quotation accordingly. e.g. "32,768 36-bit words (196,608 six-bit characters)".

I do NOT think it reasonable to add "equivalent to 147,456 bytes" (based on taking the number of bits and dividing by eight). It's not equivalent in how the machines were used, because such machines were not used to store eight-bit characters.

Re. the NCR 315 article, the source (from NCR) spells out the numbers (e.g. 5000 slabs, 10,000 slabs, etc.). I changed the article to match. "kilo" takes no fewer letters than ",000" and the latter is more precise, and there were only three occurrences in the whole article, so it doesn't look as if it's crying for abbreviation. I think this is a reasonable approach for a lot of older computers, since the possible memory sizes were almost never large enough that using mega or mebi would be required. Many of the articles are pretty short too. (not that we can use mebi, of course) For a longer article we can write out the number the first time, followed by the abbreviation, e.g. "65,536 words (64 Kwords)", and use "Kwords" in the rest of the article. Jeh (talk) 00:57, 2 February 2015 (UTC)

I agree, and think something short to that effect should be appended to the "Bytes"-MOS-section along with forbidding K or k only (especially important for these old computer articles, even though you might find that in sources). I might BOLDly edit the MOS (unless someone beats me to it or disagrees), but am not a native English speaker and tend to complicate things to much.. comp.arch (talk) 17:28, 3 February 2015 (UTC)

You could BOLDly edit, or you could propose a change here. Jeh (talk) 21:04, 3 February 2015 (UTC)

I think forbidding K/k is going too far. Offhand, though, I think it might be OK to say something like K/k should be used only where the context makes clear [something something]. I might point out that even if we were to require saying Kbyte or Kword every single time, that just leaves us with the problem that words (and to a lesser extent, bytes) vary in size too. (And BTW, some architectures worked in "characters", IBM 1401 being my favorite by far -- and yes, I programmed one in the day.) EEng (talk) 13:47, 12 February 2015 (UTC)

kilos?

It's weird I never thought of this before, but would it ever be OK to say kilo or kilos for kilogram or kilograms? EEng (talk) 04:28, 24 February 2015 (UTC)

In German, yes, but not in the article namespace of dewiki, it's colloquial. –Be..anyone (talk) 06:14, 24 February 2015 (UTC)

Why have you thought of this now? Has a specific case arisen where it might improve the article but is currently being disputed? It's hardly something we'd want to encourage; in written English we have kg when we don't want to use the full word. NebY (talk) 07:40, 24 February 2015 (UTC)

Here [1] -- 4th bullet down -- some evil impulse made me want to change kilograms to kilos. I think it's certainly understandable to English speakers but I realized (being an American) I'm not sure whether it is, as someone here just said, too colloquial. EEng (talk) 11:22, 24 February 2015 (UTC)

To lose one kilogram may be regarded as a misfortune; to lose eight looks like carelessness. Dondervogel 2 (talk) 11:29, 24 February 2015 (UTC)

Can't you be Earnest for just one minute? It's important. EEng (talk) 11:36, 24 February 2015 (UTC)

But hearty congratulatons, nevertheless. Martinevans123 (talk) 11:43, 24 February 2015 (UTC)

All this banter is driving me wild. Robevans123 (talk) 11:52, 24 February 2015 (UTC)

Introducing 'kilo(s)' in the meaning 'kilogram' would be a really bad idea. Yes, in several languages 'kilo' when spoken in many contexts means 'kilogram', but this is not a reason for dispensing with the accuracy here. Lklundin (talk) 21:53, 24 February 2015 (UTC)

It's quite informal, as others have said. Not really encyclopedic prose. Similar to "klicks" or "kays" for "kilometres" and "mils" for "millimetres/millilitres". More typical of spoken than written language, like most slang. Moreover, it's potentially confusing to non-native English speakers, who are generally less familiar with English colloquialisms. I think in some languages, "kilo" can be short for "kilometre", for example. Archon 2488 (talk) 12:01, 24 February 2015 (UTC)

Well, I don't think it's at the level of hip militarism of "klicks", and "mils" has serious ambiguity problems (being applicable to 1000ths of an inch too). I don't think it counts as slang, as I've certainly seen it in serious newspaper and magazine articles, in particular in the context of weight loss, which is why it occurred to me in this particular case. ~~But anyway I'm convinced we shouldn't use it.~~[Changed my mind -- see below] EEng (talk) 12:30, 24 February 2015 (UTC)

Kilo is common parlance in the UK and Ireland, not slang. It is probably more commonly used, and probably more widely understood, than kilogram. The OED gives its primary use as a noun meaning kilogram, in use since 1870, and with the plural being kilos. I wouldn't hesitate to recommend that it be used in non-scientific articles in Wikipedia in place of the more awkward and clumsy kilogram or the meaningless kg. Chief archivist (talk) 21:49, 24 February 2015 (UTC)

Glad you've joined the conversation. I almost let all those purists above run over me. What you're saying is the impression I had. Let's hear some more thoughts from others too! EEng (talk) 22:23, 24 February 2015 (UTC)

Speaking of pure.... I tried typing "kilo of" into a Google toolbar to see what auto-complete might suggest. Coke, weed, crack and kratom. (Kratom?) Perhaps we need to make a street-cred exception to the MOS. NebY (talk) 22:30, 24 February 2015 (UTC)

Wouldn't Coke be in liters? EEng (talk) 22:39, 24 February 2015 (UTC)

No. Tradition demands quart bottles. RGloucester — ☎ 22:40, 24 February 2015 (UTC)

Yes, but only if it's in returnable containers delivered to your door step. Robevans123 (talk) 22:45, 24 February 2015 (UTC)

Yes, but only if delivered to one's doorstep by horse-drawn milk float in glass returnable bottles. RGloucester — ☎ 22:52, 24 February 2015 (UTC)

Now you're just courting controversy. NebY (talk) 22:49, 24 February 2015 (UTC)

In the UK, or parts thereof, terms such as "pikey", "chav", "jobby", "tadger", "spiv", "gaffer", "bloke", "scab", "old bill" and many more, are in common parlance. As a native English speaker from the UK, I am well aware of what they all mean. However, they are undoubtedly slang terms. They are not terms that one would expect to find in serious prose, or indeed in an encyclopedia. People forget the concept of register in language, perhaps unfairly dismissing it as unfashionable. You don't need to make an encyclopedia sound "street" in order to make it intelligible. Just stick with standard terminology, because on balance it's best for everyone.

I have no idea how cutting the word "kilogram" in half is supposed to make it more intelligible; "kilo" is simply a prefix, not a unit. Kilopascal? Kilometre? Kilogram? Kilofarad? You get the idea. Saying that SI symbols are "meaningless" is totally ridiculous, and I'll not entertain that sentiment any further. Archon 2488 (talk) 23:24, 24 February 2015 (UTC)

.... not to mention Ks, kryptonite, half a khat, an ounce of scratch and sniff, a Two Quid Deal, a pack of "Vera Lynns" and some lines of coke, and of course, just lately, Chinese molasses and Beijing burns!! Martinevans123 (talk) 00:07, 25 February 2015 (UTC)

Let me be clear that I don't think we're going to use kilo in Wikipedia, but nonetheless the discussion is interesting. Despite all this huffing about register and so on (as if the rest of us didn't know about that -- we just have a more catholic idea of what might be acceptable) it appears that kilo, meaning kilogram, is long-established usage, and it's not slang nor even informal -- and this usage is not restricted to English, either. [2][3][4][5][6][7][8][9][10][11][12] I'm beginning to think that in non-scientific contexts, it might be OK after all. EEng (talk) 03:22, 25 February 2015 (UTC)

I'm glad to see you "don't think we're going to use kilo in Wikipedia". We've had a good laugh, but you're surely aware how much robust opposition a serious proposal would meet - and we know from experience just how robust discussions here can be. Can we now drop this per WP:NOTFORUM? NebY (talk) 08:55, 25 February 2015 (UTC)

I'd suggest that in spoken English, in the UK, the word "kilogramme" is used very much less than "kilo". It's not heard very often at all, especially at the cash register. But I guess written English is different, and certainly in an encyclopedia. Martinevans123 (talk) 09:02, 25 February 2015 (UTC)

When I opened this thread it was primarily because I was startled to realize that, to my memory, the question had never come up here. I figured someone would point to a solid usage authority and that would be that. But instead, all we have so far is lot of unsupported tsk-tsking and snickering.

In addition to the links given earlier, it's turning out that Brittanica was using kilo as early as 150 years ago [13][14][15][16]. So I'm uncertain what it is you're laughing at, unless it be the irony of someone's complaint about "the usual inability to recognise or even consider the possible existence of one's own misconceptions and prejudices".

Having said all that, it may be that modern usage considers kilo informal [17]. Instead of huffing about NOTFORUM and using Google autocomplete as a source, you might have just pointed to something like that. EEng (talk) 13:24, 25 February 2015 (UTC)

Looking more closely at that reference, it seems that they are talking about an entirely different "kilo", which appears to be some sort of archaic measurement from Greek-speaking Europe or Asia Minor. It says that a "kilo" is equivalent to 60 pounds, so one must assume this is not the kilogram they are talking about. Bear in mind that the prefix "kilo" is derived from Greek in the first place. Archon 2488 (talk) 15:29, 25 February 2015 (UTC)

Ouch. Well, I guess that wraps this one up. Now you can laugh.[I keep giving up and then Chief Archivist shows up to restore my spirit.] EEng (talk) 16:26, 25 February 2015 (UTC)

"Pikey", "tadger", and the like are slang and not comparable to "kilo" which is a simple shortening of an unwieldy word. It is similar to using "taxi" instead of "taximeter cab", "lunch" instead of "luncheon", "bus" instead of "omnibus", "plane" instead of "aeroplane", "exam" instead of "examination", "fridge" instead of "refrigerator", "phone" instead of "telephone", "flu" instead of "influenza" and many others.

Let's not be so precious about "kilogram" (or even "kilogramme") and let's not prohibit the use of a word that is already in common use in the media and on many official, including government, websites. Chief archivist (talk) 21:13, 25 February 2015 (UTC)

Examples of media and official usage... [Bolding added by EEng]

UK government: Bound for the UK: 250 kilos of cocaine seized in Caribbean joint operation, Traveller found with 94 kilos of dried caterpillars at Gatwick Airport, [18], [19]
Irish government: [20], [21], [22].
BBC: [23], [24].
RTE: [25], [26].

Chief archivist (talk) 21:26, 25 February 2015 (UTC)

I hope C.A. doesn't mind my interpolating two particularly alluring titles in the above links. Those participating in the "joint" operation must have been thrilled to come up with cocaine instead. EEng (talk)

[Addition examples by Martinevans123:]

Goodness me, Chief, so many drugs! Let's not forget coins, greengrocers and explosives.... Martinevans123 (talk) 22:57, 25 February 2015 (UTC)

Seems to me a prima facie case is emerging for acceptance of kilo in high-quality sources. I'd like to hear what other editors think now (holding the laughter, please). EEng (talk) 23:54, 25 February 2015 (UTC)

What exactly is this going to achieve? Some sources will say "kilo", yes. Sources will differ on their conventions relating to quotation marks, but we pick one convention and stick with it. Why can't we do the same here? It makes sense to use the same word every time you use the unit for the sake of simplicity and consistency. If you find "kilogram" too long (I don't think we should accept someone's judgement that it's an "unwieldy" word; it's no worse than any other three-syllable word) you can use the symbol "kg".

This proposal appears to be to complicate a simple situation for no benefit. "Kilogram", "kilo" and even "kilogramme" (which is archaic and deprecated, especially in non-British English) mean the same thing, and I don't think WP gets any benefit from introducing a Byzantine set of rules about when to use which. That would require changing god-knows-how-many articles, for no benefit. Adding yet more options to the convert template to display "kilos" (or, god help us, "kilogrammes") as the output unit name – again for no benefit. Archon 2488 (talk) 11:40, 26 February 2015 (UTC)

I am 100% behind Archon 2488 on this. This is a style guide. Its self-stated purpose is to promote harmonisation, not Balkanisation. Dondervogel 2 (talk) 11:49, 26 February 2015 (UTC)

They used "ouncia" and "Litra-Libra" in fact: [27]. Martinevans123 (talk) 11:56, 26 February 2015 (UTC)

I support the use of "kilogrammes" in British English articles. Just as we use "centre" and "colour", we must protect the French etymological origins of this word. RGloucester — ☎ 14:54, 26 February 2015 (UTC)

"lol", RG, your arcane historical traditions will be death of this project. Martinevans123 (talk) 15:23, 26 February 2015 (UTC)

In fact, I think I grant this project more gravity than it is due. RGloucester — ☎ 18:03, 26 February 2015 (UTC)

Gravity. It isn't just a good idea. It's the law. EEng (talk) 22:31, 26 February 2015 (UTC)

... seven Oscars!! Martinevans123 (talk) 22:39, 26 February 2015 (UTC)

Arbor-treeish break

Here again is what got me thinking about this -- a "Did You Know" entry:

... that Chris Wu lost 8 kilograms (18 lb) in ten days when he was cast in The Third Wish?

Notice that it doesn't tell us kilograms of what -- oats? Gold? It's kilograms of bodyweight, of course. Now, in a scientific article we probably would say "bodyweight", but this is an actor's biography; everyone knows what's meant and no one over at DYK objected (nor, I would wager, did the thought even occur to anyone -- it certainly didn't occur to me). So my first question, before I go on, is whether anyone objects to an article talking about personal bodyweights in this way e.g. "After losing 5 kilograms, he qualified for welterweight matches", "He competed at 87 kilograms the first year, and at 82 kilograms the second" -- stuff like that. Thoughts so far? EEng (talk) 15:45, 26 February 2015 (UTC) P.S. For simpilicity I'm leaving out the conversions to lb that would normally be included as well.

I think it's evident in these cases that bodyweight is intended. Referring to "kilograms" or "kg" is clear and encyclopedic. Referring to "kilo" or "kilos" is informal and should not be used, even though it may be clearly understood by most English speakers (but perhaps not all those who are not native speakers). An exception would be when quoting someone (He said: "I lost 15 kilos [33 lb]!" but He claimed that he lost 15 kilograms (33 lb) or He claimed that he lost 15 kg (33 lb)). —sroc 💬 11:50, 27 February 2015 (UTC)

I've wavered back and forth and, given Chief Archivist's and Martinevans' links (above -- I've added bold) I'm really unsure whether there might not be a place, for example in a DYK hook thus:

... that Chris Wu lost 8 kilos (18 lb) in ten days when he was cast in The Third Wish?

Did you take a look at those links? EEng (talk) 20:29, 1 March 2015 (UTC)

sroc, I'm lonely! EEng (talk) 17:44, 4 March 2015 (UTC)

@EEng: There are places where you can meet people and circumvent loneliness. I have now seen those links with sources using "kilo"; I still wouldn't use "kilo" on Wikipedia generally, but I wouldn't be so concerned as to change it if someone else used it in the context of the DYK example. —sroc 💬 13:41, 5 March 2015 (UTC)

Thanks for the tips. Now will you please give me some guidance at #Resolution? EEng (talk) 14:23, 5 March 2015 (UTC)

For guidance of MOS see Wikipedia:Manual of Style#Do not use unwarranted abbreviations:

Avoid abbreviations when they might confuse the reader, interrupt the flow, or appear informal [emphasis added]. For example, do not use approx. for approximate or approximately, except in a technical passage where the term occurs many times or in an infobox or a data table to reduce width.

Cinderella157 (talk) 00:09, 6 March 2015 (UTC)

I don't see it as an abbreviation in the sense the approx. is -- more like autos instead of automobiles (which you might use, say, in an article on traffic flow where it would be awkward to say, over and over, automobiles/hour, automobiles/hour, automobiles/hour, automobiles/hour, automobiles/hour. My feeling, and this time I'm gonna stick with it, is that one might use kilo (under MOS' "common sense/exceptions apply" proviso) in unusual circumstances, like the DYK hook that prompted this whole thing (DYK being slightly more flexibly because of its come-hither mission), or maybe in some situation comparable to the autos/hr example just given, though offhand I can't think what that would be

I raised this point just out of friendly interest, and now, with the concurrence of my esteemed fellow editors, I'd like to close the discussion. I'm not proposing any change to MOS (and didn't expect to). EEng (talk) 01:28, 6 March 2015 (UTC)

convention of the primary units being the same as the cited value

At [28], The following amendment was made:

In historiographic or historiographic sections of other articles, it is appropriate for the primary unit of any particular measure reported to reflect the units of measure used for the quoted value as taken from the cited source. This convention maintains the integrity of the citation process and avoids errors by compounding conversions.

The reversion following, states:

... this is already covered ('Where the article's primary units differ from the units given in the source...", below)

The part being refered to more fully states:

Where the article's primary units differ from the units given in the source, the {{convert}} template's |order=flip flag can be used; this causes the original unit to be shown as secondary in the article, and the converted unit to be shown as primary.

This process does not make it apparent to the reader the original units reported by the cited source. A reader does not see the code. It does not accurately represent what is being quoted from the source. Accuracy in citation is the essential element of this amendment. I offer this amendment for the D (discussion) part of BRD. Cinderella157 (talk) 05:01, 27 February 2015 (UTC)

No – Sourced-based units have been rejected many times, for good reason. Regardless, the absurdity of a specific exception for "histographic articles" (what is a "histographic article", pray tell?) does not shine a good light on this proposal. Instruction creep, surely. If there is a problem with "histographic articles", it can be dealt with on the talk pages of the articles in question. RGloucester — ☎ 05:21, 27 February 2015 (UTC)

I see you've corrected it to "historiographic" articles. The same reasoning applies, however. No need for special guidelines. RGloucester — ☎ 06:01, 27 February 2015 (UTC)

historiographic: relating to the writing of history (thank you for making me aware of my misspelling). Histories are often written with a variety of units, depending on the timing of the event and the writing of the source. Representing a figure as a primary unit in the article when it is not a primary unit in a source, is to misquote the source. As for the premise, that it is instructional creep, it does not mandate style nor does it add additional restrictions or like to existing style, but permits a valid alternative which, incidentally does not appear to be in conflict with WP:MOS. I have already seen evidence of a failure to identify the source measurement has led to the introduction of errors. Cinderella157 (talk) 06:22, 27 February 2015 (UTC)

I know what "historiographic" means. However, you've not justified why this particular matter needs a special rule in the MoS. It doesn't. This is especially true if the proposed rule is redundant, as you yourself just said. RGloucester — ☎ 06:26, 27 February 2015 (UTC)

NO! This is well-settled. Source-based units have been rejected over and over. This is an encyclopedia, not a book for historians by historians. Jc3s5h (talk) 09:36, 27 February 2015 (UTC)

The OP starts by worrying about how to represent quantities taken a "cited" source, but then slips into worrying about misquoting the source. These are two different things. Practically the very first thing this page says is

In a direct quotation, always retain the source units. Any conversions can be supplied either in the quote itself (in square brackets, following the original measurement) or in a footnote.

EEng (talk) 10:33, 27 February 2015 (UTC)

I agree with what others have said above. Source-based units have been proposed in several different forms, all of which have been soundly rejected. A few seconds contemplating how such a proposal could play out – the thousands of unnecessary unit adjustments that would need to be performed for no actual benefit, the absurd over-complication of when to use which units (as if the UK units section were not enough of an overcomplicated mess as it is – I don't understand why the proposed new rule was put there, of all places – does it refer exclusively to UK sources? Why?) and the possibility of people picking sources that use the units they want to use in the article – should be enough to dissuade anyone from supporting it. Archon 2488 (talk) 12:42, 27 February 2015 (UTC)

Firstly, it does not mandate change. There is no complication at all. If a source is cited as quoting a figure, the figure is reported in the units in which it appeared. It has nothing specifically to do with the UK and cherry picking sources for particular units sounds somewhat far fetched. It comes down to a matter of accuracy in quoting and verifiability. I would note that figures are generally direct quotes, even if not part of a longer passage of quoted text or enclosed in quote marks. Cinderella157 (talk) 15:20, 27 February 2015 (UTC)

Sorry, I got confused because you put your proposed new text directly below the UK units section. I now see that you intended it to be separate.

However, I am not fundamentally any less confused. What do you mean by "cited"? If it's a direct quotation then we already have a rule that the original value goes first, as EEng said above. There is no need for a special "historiographical" rule, and as far as I can tell there is no MOS reason to treat historical articles differently from other articles. Merely finding a source that gives a measurement in unit X is not ipso facto a strong reason for an editor to put unit X first in the WP article, be it a history article or no. If we adopted this rule, as you seem to be suggesting, then it would entail large-scale changes for no real benefit, because in order to make historical articles comply with the MOS we'd need to go thru them all and check the sources for each figure, seeing whether the primary source reported a distance in yards, metres, or something else, etc. It would also potentially allow people to source-shop for sources that gave measurements in the units they wanted, such as (relating to an ongoing SPI) miles in Pakistan. These debates can be acrimonious enough, and we do not really need any more uncertainty about which units go where.

What you need to explain in order to reduce my (and, I gather, others') confusion is why your proposed rule improves on what exists. Archon 2488 (talk) 17:25, 27 February 2015 (UTC)

You say, "cherry picking sources for particular units sounds somewhat far fetched". Not as far-fetched as one might like, I'm afraid. Kahastok talk 20:17, 27 February 2015 (UTC)

No, for all the numerous reasons that have been brought up by others and in previous discussions. Same as we don't necessarily use the same date format or language (or dialect of English) as the source. As others note, direct quotations are already covered separately so respecting quotations is not a good reason to change. Kahastok talk 20:17, 27 February 2015 (UTC)

Clarify I have been asked to clarify and provide more detail. Firstly, to say in the proposal, "it is appropriate ... ", does not mandate wholesale change, it is merely to say this is an appropriate style and an existing style need not, and should not be changed. If this non-mandatory proposal is somehow seen to create a mandate, then it does so, not of itself, but by a fundamental misapplication of the MOS, which represents a serious failing in the MOS. This proposal is not analogous to dates or times formats, since the value of the date or time is unaffected by the format in which it is presented. Measures are affected by the conversion process. To say that editors would 'cherry pick' sources for units does sound far fetched. If it were to occur, I would be much more concerned about what else has been 'cherry picked'. All relevant, reliable sources identified should be reported and can be handled with reasonable objectivity but how this might be done, depends on the actual figures and a range of variable factors such as, whether the sources are reporting independent measurements of the same quantity, the precision indicated, the agreement and so on.

I will give a specific example from an article have been writing. In reference to a tropical grass, one source said that it "grows to over 6 ft" and another, that it "grows up to 2 m". Quoting both sources, I wrote, that it grows "to up to 2m (over 6 ft)". This clearly reflects the content and in the absence of any other clear distinction, the metric was given a slight preference. A zealous editor applied the convert template to this to give the following: "grow to 2 metres (6.6 ft)". Apart from the issue of implied precision, it is no longer consistent with both sources and favours one over the other. If there is 'cherry picking', then perhaps it is an attribute of the 'status quo' and not potentially confine to or specific to the proposal.

Current advice is somewhat ambiguous. It gives a preference for metric as the primary units, except for articles relating to either the US or the UK. It fails to deal with areas of overlap, compounded by the historical context. Imperial units have been (until about the 70s) the predominant system of measurement used by English speakers, and the US version making up much of the balance. In consequence, sources from before this time will predominantly report in US/imperial units. This proposal is not a position of 'metric martyrdom' but about a distortion of material through what is often a poor understanding of how to correctly apply conversions. By placing the cited units as the primary units for a particular measurement, the reader is better equipped to make an informed decision about the cited value. What are the correct primary units to use when an article (and sources) overlap the US, the UK and/or other English speaking countries and even non-English speaking countries? Contemporary matters will be written about in units which are contemporarily relevant to the national contexts that they pertain to. Contemporary matters have a different context from historical matters.

The proposal relates to historiographic articles or historiographic sections or articles. Why? Because sources are likely to report quantities in contemporaneous units - either of the time of the event or of publication of the source. I would see this as also being relevant to science and technology articles, reporting historical measurements where these may have been made under a lb.ft.sec system of measurement rather than metric. In science and technology, it is convention (outside WP at least) to report primary figures for dimensions and specifications of manufactured or constructed items, 'as constructed'. For other measurements, these are reported 'as determined'. The precision of the reported measurement will be affected by the precision of the conversion factors applied. Some comment has been made regarding the provisions that apply to direct quotation. I submit that (in most cases) the quoting of quantities in text are direct quotations of the source material, even though I am not quoting a passage of text. If I report the figure in primary units not the same as the source, then I am misrepresenting the source. If I apply a conversion to a cited figure, I am applying a POV as to the precision of the original figure. Using the source units for the primary figure allows the reader to readily identify the cited quantity and it makes the conversion applied transparent and verifiable.

In an article on Noah's Ark, it would be quite ridiculous not to give the primary dimensions in cubits. The following is a paraphrased sentence used in the lead of the Ark article: "... it will be 300 cubits long (137.16 m, 450 ft), 50 wide (22.86 m, 75 ft), and 30 high (13.716 m, 45 ft) ..." Cubits, and metres have been used as the primary units. This may be justifiable as a 'special consideration'. The question is then, when does this cease to be a consideration? I would also point out the spurious precision of the conversions. If it were determined appropriate to use metres as the primary unit, this article would be a mess of misinformation. It is saved from this because the units of the source have been used as the primary units. Obviously, this is a failure to comply with the principles of Uncertainty and rounding, but these principles are not well understood and not well followed, despite the MOS. Multiple conversions can compound conversion errors, leading to an 'error creep' and inaccuracy. I appreciate that the 'flip' parameter might be used, but the information it holds is concealed from the reader. The best way to avoid this problem is to not become a part of the problem by indicating to the reader which is the value to which the conversion has been applied. The reasonable assumption is that the parenthetically reported value is the conversion, since it is offered as a conversion of the primary units.

Another example from an article of my writing was was a figure of 5 ft (1.5 m). The source gave the value as "5 ft". The significant figures of the conversion should probably be 2 m, technically, but this would have been fallacious. It is reasonable to assume that the uncertainty is in the fractions of a foot and two significant figures in the conversion is not unreasonable. Again, the conversion template was applied by another editor on the metric value with the following result: 1.5 metres (4.9 ft). The resultant conversion implies a different level of precision and is an incorrect representation of what was stated in the reference.

I believe that the WP:MOS is sufficiently unspecific so as not to preclude the use of this convention. "The main unit in which a quantity is expressed should generally [emphasis added] be an SI unit ...". The WP:MOS also talks about a quantity (an amount measured or reported?) in the singular and not in the context of the unit of a 'dimension' or 'physical quantity'. Certainly, the general understanding of a quantity would be a reference to an amount measured or reported. In this respect, the MOS is ambiguous because its clarity (if this is what is meant) relies upon a specific meaning, which has not been defined, and not on the general meaning. If the MOS has precedence over sub-pages (as it does), then it is quite inappropriate for this page to mandate practice not consistent with the MOS. However, Wikipedia:Manual of Style/Dates and numbers#Unit choice and order states: "Quantities are typically expressed using an appropriate "primary unit", displayed first, followed by a conversion in parentheses, when appropriate, e.g. 200 kilometres (120 mi). For details on when to provide a conversion, see the section § Unit conversions. The choice of primary units depends on the circumstances ..." [emphasis added]. Neither the MOS nor Wikipedia:Manual of Style/Dates and numbers appear to mandate against the convention. Perhaps including the proposal is redundant, except that some are prepared to act upon the matter despite this. Perhaps it need not be included in the body of the text but rather as a footnote that clarifies meaning, scope and degree of force?

I would summarise in conclusion that the proposal is a convention with value and acknowledging its value does not, of itself, mandate change to articles where another style has been used. There are reasonable reasons of accuracy, fidelity and verifiability for acknowledging this as an acceptable style. 'Cherry picking' may exist but is not inherent to the proposal. Cinderella157 (talk) 03:18, 28 February 2015 (UTC)

I have not read Cinderella157's long essay, but I agree with her previous posts, stating that the original value should be respected. That is not the same as source-based units. Consider an article that has chosen kg as its primary unit, converting to lb as secondary unit. Where a source uses kg this presents no problem. But what about sources using lb? For those sources, it is appropriate to carry out the conversion in such a way that the original lb value is preserved, even though apparently converted from kg. If that is what she is saying, I agree. Dondervogel 2 (talk) 10:11, 28 February 2015 (UTC)

Yes - but I think MOS:CONVERSIONS and the point in the Unit choice and order section already make this clear. A measure like "4 ft 11 ins" can be converted with any precision up to 1.4986 metres. Similarly, 1.5 metres can be converted with any precision up to 4 feet 11 7⁄127 inches. Fact is that in many cases, 1.5 metres and 5 feet are close enough to be both perfectly accurate - but in some cases they won't be and it is not unreasonable to expect editors to use a certain degree of judgement in determining what the most accurate conversion is in their particular circumstances. As per the status quo.

But that isn't what the proposal actually says. It says that the "primary unit of any particular measure reported" should "reflect the units of measure used for the quoted value as taken from the cited source". Which means, source-based units. Change the source, and you change the order of the units. If one sentence is source to a document written using pounds and the next to a document written using kilograms, our text uses pounds in the first sentence and kilograms the next. For all the reasons given in this and previous discussions on this topic this is a bad idea. Kahastok talk 10:57, 28 February 2015 (UTC)

I see. The need for clarity make it is essential IMO for each article to use consistent (primary) units throughout, so source-based units in that sense can never work.

@Cinderella157: can you clarify between Kahastok's interpretation and mind, in slightly fewer words? Dondervogel 2 (talk) 11:19, 28 February 2015 (UTC)

The easiest thing is to offer an example:

The area was low-lying and featureless. Buna air strip is 5 ft (1.5 m) above sea level.[44] The elevation is only double this at Soputa, 7.5 mi (12 km) inland[44] and 85 m (279 ft) at Popondetta, 21 km (13 mi) inland.[45] Not surprisingly, the water table is reportedly shallow at about 3 ft (0.9 m).[46][47] This impacted upon the construction of weapons pits and defensive positions.[48]
Areas not waterlogged were either dense jungle or swathes of kunai grass.[49] Coconut plantations filled the wider areas of dry ground along the coastal strip but neglected, undergrowth had reclaimed the ground.[50] The dense kunai grass could grow to 2 m (over 6 ft) and the leaves were broad and sharp.[41]
Temperatures over the period of the battle ranged from an average daily low of 72°F (22°C) to 89°F (31°C) but with an humidity of 82%, this could be oppressive.[51] In the humid conditions, kunai grass trapped the heat and it was not uncommon for temperatures to reach 50ºC (122°F).[50]

I would note the reference to kunai grass, which is perfectly consistent with sources but would be difficult to rephrase and maintain the same consistency. Cinderella157 (talk) 20:46, 28 February 2015 (UTC)

The problem with that is the flipping between units and between unit systems. To avoid this it would need to read (first para only) either

The area was low-lying and featureless. Buna air strip is 5 ft (1.5 m) above sea level.[44] The elevation is only double this at Soputa, 7.5 mi (12 km) inland[44] and 279 ft (85 m) at Popondetta, 13 mi (21 km) inland.[45] Not surprisingly, the water table is reportedly shallow at about 3 ft (0.9 m).[46][47] This impacted upon the construction of weapons pits and defensive positions.[48]

or

The area was low-lying and featureless. Buna air strip is 1.5 m (5 ft) above sea level.[44] The elevation is only double this at Soputa, 12 km (7.5 mi) inland[44] and 85 m (279 ft) at Popondetta, 21 km (13 mi) inland.[45] Not surprisingly, the water table is reportedly shallow at about 0.9 m (3 ft).[46][47] This impacted upon the construction of weapons pits and defensive positions.[48]

Dondervogel 2 (talk) 21:09, 28 February 2015 (UTC)

No - for the reasons given by EEng's original revert (MOS - exceptions apply/use commonsense), and the other reasons given above (direct quotes covered already, and for the potential inconsistencies/arguments/conflicts that would arise from sources using different units). But the discussion did prompt some thoughts on historic units, identifying the source value in conversions, and the precision of converted values.

Using historic units

Examples that come to mind are Noah's Ark, as already mentioned, and various pyramids, all using cubits of one sort or another. Any attempt to change these from using cubits as the primary unit can presumably be countered by the exceptions apply/use commonsense guidance. However, articles using historic articles should be clear about which units are being used. Noah's Ark has some discussion on this, but it's buried in Origins section and it's not obvious that the 18 inch to a cubit conversion is the one being used in the article, or that the conversion factor is itself an estimated value.

Also, since the conversion factor used is in US Customary/Imperial units then perhaps the order of units should be cubits/US customary/metric so that "50 [cubits] wide (22.86 m, 75 ft)" becomes "50 [cubits] wide (75 ft, ≈23 m)". BTW 22.86 m is definitely over precise in this context. See notes on precision below.

Identifying the source value

Also mentioned in the discussion above were objections to using |order=flip with {{convert}}: ...as this causes the original unit to be shown as secondary in the article, and the converted unit to be shown as primary. This process does not make it apparent to the reader the original units reported by the cited source.

I believe this is a valid point and worthy of attention. A few possible solutions present themselves. Firstly, show the original units in bold - so that, for example, 1.5 metres (5 ft) becomes 1.5 metres (5 ft). Secondly, explicitly state which figure is in the source (possibly with some additional typographic change so that 1.5 metres (5 ft) becomes 1.5 metres (5 ft _(source)). And finally, use a footnote, for example, 1.5 metres (5 ft)^[a].
.
.
.

Notes

^ Figure in brackets is the value given in the source document.

I'm not sure if any of these are entirely suitable options (comments welcome). The use of bold is the cleanest, and does suggest the primacy of the source value, but is not as immediately obvious in meaning as the 2nd and 3rd options.

Precision of converted values

As discussed above, the precision of converted values can cause problems, and while MOS:CONVERSIONS gives some advice and also warns against False precision, the guidance to ...use a level of precision similar to that of the source quantity value... is in itself rather imprecise. It does seem to be sometimes interpreted as "show the same number of significant figures", so results in statements such as (from Cinderella157's notes above) The source gave the value as "5 ft". The significant figures of the conversion should probably be 2 m, technically, but this would have been fallacious., which is itself fallacious, but most likely based on the statement (from MOS:CONVERSIONS) Small numbers may need to be converted to a range where rounding would cause a significant distortion, so one mile (1–2 km), not one mile (2 km). I would contend that this statement is very misleading - it might be appropriate advice to apply to a statement such as "he walked about a mile (1–2 km)" but would quite wrong in statement such as "Roger Bannister ran the first sub four minute mile (1–2 km) in a race" where a value of 1.61 km would be much better (see also examples below on context of conversions). Also, the tendency to judge the appropriate accuracy of conversions based on the number of significant figures could lead to under precision based on the conversion factor. If the conversion factor is an integer (or possibly a factor based on an integer based fraction), then it would be a mistake to round a conversion from, for example, (imperial) tons to (US Customary) pounds: 47 tons is exactly 105,300 lbs - which doesn't need rounding to, say, 160,000 lbs, based on the number of significant figures.

I think MOS:CONVERSIONS should say something like ...use a level of precision appropriate to the accuracy of the source quantity value.... And False precision could give some advice on the dangers of being under-precise as well as guarding against over-precision. These difficulties often arise when converting US Customary/Imperial units to metric values, particularly because the older systems break smaller divisions into smaller (different) units, rather than giving a decimal value. A figure of 5 ft won't necessarily say that it really means 5 ft 0 inches (± some fraction of an inch) and so the precision of the source value can be difficult to gauge, and depends on the context, as shown in the examples below.

Examples of 5 ft conversions
Context	Notes	Implied accuracy	Range (in feet)	Range (in metres)	Suitable conversion
The room was about 5 ft wide and 10 ft long	Possibly given to the nearest foot (a rough "ball park" figure)	±0.5 feet	4.5 - 5.5	1.37 - 1.68	1.4 - 1.7 m or ≈1.5 m or approx. 1.5 m
Buna air strip is 5 ft above sea level (as mentioned above)	Given to the nearest foot? Or possibly 1/2 foot?	±0.25 feet	4.75 - 5.25	1.45 - 1.60	1.5 m
Freda Bloggs was 5 ft tall	Height usually given to the nearest 1/2 inch	±0.25 inch	4.980 - 5.021	1.518 - 1.530	1.52 m
The rail gauge was 5 ft	Usually given to nearest 1/4 inch	±0.125 inch	4.99 - 5.01	1.521 - 1.527	1.524 m

All this has reminded me of the old joke about precision used on building sites where bricklayers work to the nearest half brick, plasterers work to the nearest 1/2 inch, carpenters work to the nearest 1/8 inch, Electricians work to the nearest 1/16 inch, and labourers work to the nearest pub...

Robevans123 (talk) 17:41, 1 March 2015 (UTC)

I see no special need to highlight the unit that was used in the source in any way at all. All your proposals look a bit clumsy to me, and suggest we're worrying to much about it.

In cases where a unit is set via definition, or else part of a name, we put that first always. In all other cases there is some leeway for measurement difference, and it is basically always possible to convert the one the other to an appropriate degree of precision without loss of accuracy. Using the |order=flip notation gives notice to editors which unit is the original, which is useful in avoiding false precision (because if it's wrong it can be easily corrected). But chances are the reader doesn't care that much whether the source said the elevation was 5 feet or 1.5 metres, and if they do they can always read the source.

When it comes down to it, while I'm not ruling out supporting a change in our guidance on dealing with precision issues, anything we do is going to have to rely on editors applying thought to their particular situation. No guidance we give here can ever apply in every situation or even the majority of situations. The best we can do is advise editors to think about the precision appropriate to their situation and to guard against both overprecision and underprecision. Which, I believe, is what we already do. Kahastok talk 18:13, 1 March 2015 (UTC)

I am not certain about the proposals by Robevans123, certainly not bolding and the latter two, while having some virtue, may be cumbersome to implement. Being able to place a '*' (or similar) as a generic note would possibly work better and could be placed inside the brackets. The note would then read "value given in source document". This same note would then link to multiple uses in the text. This may be be Robevans123's intent but I am not certain of the the functionality to allow the same note to be referred to in text multiple times. Perhaps this may be able to be adapted as part of the functionality of the flip parameter.

Misrepresenting precision is to misquote a source. Any quote of a figure from a source is effectively a direct quotation, even if it appears in paraphrased text. Dismissing this matter appears somewhat inconsistent with the broader principles of fidelity, verifiability and reliability that apply to reporting of material in articles. Clearly, precision in conversions is poorly understood and is likely to continue, regardless of any efforts to instruct otherwise. If this is accepted, the other solution is to indicate (to the reader) what the source has actually said. The question then, is how to do so?

I said that technically, 5 ft should be converted to 2 m. The basis is that the figure is reported to one significant figure and, accordingly, the answer should reported to one significant figure, so 1.52400 m is rounded to 2 m. However, I used a similar reasoning to Robevans123 to assert it was more reasonable to report the conversion as 1.5 m. Cinderella157 (talk) 22:51, 1 March 2015 (UTC)

You say, "Any quote of a figure from a source is effectively a direct quotation, even if it appears in paraphrased text".

False.

If the source says that the Everest is 29029 feet high, then that does not mean either that 29029 feet is the original measurement, nor that saying it's 29029 feet is necessarily more accurate than saying it's 8848 metres. The fact that the source says 29029 feet does not mean that the mountain is precisely 8848.0392 metres high. We aren't quoting a source in that situation - we're citing it as a source for information and the information can be given equally accurately in either system.

The idea that "technically, 5 ft should be converted to 2 m" is nonsense. Technically, 5 feet should be converted to whatever the most appropriate conversion is in the circumstances.

That is very (very) unlikely to be 2 metres because 2 metres is more than 6 feet. The most appropriate conversion is far more likely to be 1.5 metres, but it could be anything down to precisely 1.524 metres. Depending on what it is you're measuring. There is no rule that we can give that can tell you exactly what it is - the only way of doing it is for the editor to figure out what's best in their particular situation. Kahastok talk 19:23, 2 March 2015 (UTC)

<applause> Exactly. For a measurement reported as a single digit, the idea of significant figures completely breaks down; good sense and good judgment (ideally including investigation of what the source says about the accuracy and precision of its own figures) are required. EEng (talk) 20:16, 2 March 2015 (UTC) Can I please have some help at #Resolution?

The conventions of significant figures do not break down for a single digit if the conventions of significant figures have been followed in reporting that single digit. The conventions report a number as including all certain digits and the first uncertain digit. To say, "technically, 5 ft should be converted to 2 m" is perfectly correct by the conventions of significant figures. To report a higher precision is to assume that a higher precision was actually intended in the initial measurement. It is to assume that the conventions of significant figures were not followed in the reporting of the initial measurement and I did make an observation in line with this. If a source says Everest is 29029 ft high, that is what it says - nothing more and nothing less. And if it says something happened at 9:48 pm on 23 January 1856, that is the time and date given by the source - not some other time or some other date. BTW 29029 ft is reported to five significant figures and a conversion to the same number of significant figures would be 88448.0 m. Cinderella157 (talk) 22:33, 2 March 2015 (UTC)

Oops - you've got an extra digit in there somehow - it should be 8,848.0 m. Robevans123 (talk) 23:15, 2 March 2015 (UTC)

I should really remember to take of my boxing gloves when typing.Cinderella157 (talk) 00:27, 3 March 2015 (UTC)

Tautological declaiming like "When a source says Everest is 29029 ft high, that is what it says" isn't helpful. A source discussing the geodesy of Everest would hem in such a statement with all kinds of complicated stuff about precision, accuracy, and sources of error, and (as Kahastok said so well) we as editors would have to figure out what we want readers to understand and how best to communicate that. No serious quantitative work relies simply on significant figures to carry the burden of representing uncertainty, though it's fine for exposition such as we do here. However, the significant-figures approach ("includ[e] all certain digits and the first uncertain digit") indeed breaks down when the only digit is the uncertain one, because (in the extreme example where the reported magnitude is 1) the implication is that the "true" value is perhaps 0 or 2, corresponding to errors of + or - 100% of the reported value i.e. the apparent error completely overwhelms the quantity being reported. (And that's assuming the uncertainty in this last -- only! -- digit is one unit, not more!).

Anyway, we've lost track of what we're discussing here. There's 0.0000% probability (note deliberate use of significant figures -- and I'm a frequentist so this is saying a lot!) that the present discussion will lead to any change along the lines you've proposed. That's not to say there isn't a place in MOS for guidance on how to handle precision/accuracy/error issues (even when there's no conversion of units involved) but, as have many discussions before, it illustrates the folly of trying to fashion such guidance in a content vacuum i.e. without actual issues in actual articles that need resolving. To trot out the tired old trope (here adapted to the present discussion) that MOS regulars have heard me intone over and over:

There's a further reason that disputes on multiple articles should be a gating requirement for adding anything to MOS: without actual situations to discuss, the debate devolves into the "Well, suppose an article says this..."–type of hypothesizing -- no examples of which, quite possibly, will ever occur in the real life of real editing. An analogy: the US Supreme Court (like the highest courts of many nations) refuses to rule on an issue until multiple lower courts have ruled on that issue and been unable to agree. This not only reduces the highest court's workload, but helps ensure that the issue has been "thoroughly ventilated", from many points of view and in the context of a variety of fact situations, by the time the highest court takes it up. I think the same thinking should apply to any consideration of adding a provision to MOS.

So, once you've run into this a number of times on actual articles, and duked it out with your fellow editors several times, and have a proposal for new MOS guidelines that will streamline such discussion in the future, then come back here. Until then this is waste-of-time hypothesizing. EEng (talk) 00:13, 3 March 2015 (UTC)

Declaiming a statement as tautology does not declaim the logic of such a statement. Acknowledging what a reference actually states is a matter of fidelity. It does not, of itself "hem in such a statement with all kinds of complicated stuff about precision, accuracy, and sources of error". If anything, it uncomplicates matters. The reported figure is identified free of any bias, error, perception of misperception that might arise in making a conversion. Yes, reporting '1' alone where it is an uncertain digit does, without other information or assumption, imply a potential uncertainty of ± 100% but reporting '9' implies an uncertainy of ± 10% (rounded from 11%). More importantly though, it is reporting that the observation appeared closer to one, than either zero or two. I do believe that a number of real examples have been raised and, based on a variation of the proposals of Robevans123, there exists a possible neat solution that is satisfactory to both positions. Cinderella157 (talk) 01:13, 3 March 2015 (UTC)

Huh? You'd better look up declaim in a dictionary.
(Double-)huh? I didn't say that "acknowledging what a reference actually states" is what "hems in such a statement [etc.]"; rather, I said that a source discussing (in any serious way) the height of Everest would hem in its statements with error data.
Your interpretation of the final digit is nonsense. A final (or only) digit of 1 does not mean that the true value of that digit is closer to 1 than to 0 or to 2 (partly because in general no one knows what the "true" value is -- you seem to be confusing uncertainty with rounding). It means that 1 is the best summary (in any of several senses of "best") of a (possibly wide) distribution of possible values. And this talk of all the uncertainty being in the last digit is nonsense as well: continuing your favorite Everest example, this high-quality source gives its height as "29,035 feet, with an error margin of plus or minus 6.5 feet".
Could you please learn to indent your posts like everyone else?

EEng (talk) 04:28, 3 March 2015 (UTC)

Please define how 'everyone else' indents their posts and where WP gives guidance on this matter? 'Declaim' has several meanings and I believe that my previous usage is quite consistent with the one I intended. I apologise if I misunderstood what was actually being hemmed (and where); however, I believe that the point I was making in consequence remains valid. An observer making an observation reports an uncertain digit on the basis that they estimate it to be closer to the reported digit than to the other possibilities. This was the context of my statement and my statement, I believe, was correct in that context. If you imply by 'best summary' a averaging and a statistical analysis of precision, that is quite something else. Herein lies the distinction between the actual value and a reported value. I would note that the Everest example was not originally mine. In the absence of other information, the conventions of significant figures indicate that the uncertainty is in the last reported significant digit. To say that the uncertainty is in the last digit does not mean that the range of the uncertainty cannot change the second last digit - such as 26.0. A figure of plus or minus 6.5 feet would appear to suggest an estimate derived from a statistical treatment. This is outside the scope of employing significant figures as a basic means of representing and reporting precision - which I believe to be the scope of this discussion as it relates to significant figures. The point remains that there are real examples and a reasonable solution to deal with the issue that has been raised. Cinderella157 (talk) 06:24, 3 March 2015 (UTC)

WP:INDENT
You're mixing up declaim with (perhaps) decry -- perhaps you mean declaim against.
You keep insisting stuff like "An observer reports an uncertain digit on the basis that they estimate it to be closer to the reported digit than to the other possibilities" and that just isn't right -- it's one of many ways a particular string of digits might be chosen for reporting, out of many other possible ways. Nor do more-than-single-unit uncertainties arise only when "an estimate [is] derived from a statistical treatment": if the text says The victim estimated he ran 1 to 2 miles before reaching the road there's no statistics involved, yet we still need to figure out how to communicate that in km. Anyway, I repeat that this should be discussed at MOS only after several illustrative dilemmas have been hashed out on actual articles, by actual editors who actually care about those articles. That's the best way to ventilate the various issues before trying to fashion project-wide guidelines.

EEng (talk) 23:23, 3 March 2015 (UTC)

You say: "'An observer reports an uncertain digit on the basis that they estimate it to be closer to the reported digit than to the other possibilities' and that just isn't right". It is a perfectly correct statement. I did not say it was the only way. You quite correctly point out that such an observation could also be reported as a range. Nor did I say that double digit uncertainties could only arise from a statistical treatment - merely that the example suggested the likelihood of this being the case. Where an error analysis has been conducted, it is quite common to report the error to one more decimal place than the last significant figure of the associated measurement. An error analysis is not confined to a statistical analysis. The example: The victim estimated he ran 1 to 2 miles before reaching the road is a clear illustration of when it would be a nonsense this using anything other than miles as the primary units.

Regarding the use of declaim, see wiktionary:declaim (first definition): "To object to something vociferously; to rail against in speech."

The original question was in regard to misperceptions created when an 'apparent' conversion was actually the value given in a cited reference. Robevans123 has made a number of suggestions. Placing a note inside the parentheses of a conversion, as I had originally thought, is problematic: firstly because it is not supported by the convert template but also, it might be confusing when the unit consists of superscript 'powers' (eg m²). Using notes generated by {{refn}} (or similar), if used multiple times, would produce multiple notes of the same text: "Figure in brackets is the value given in the source document." Unless there is a variation that overcomes this, a manually generated note (eg <sup>§</sup>) and manually inserted note text is likely the most effective way of overcoming the problem. I believe that this discussion and this outcome has been useful. Furthermore, annotation (of the MOS) is an effective way of clarifying and improving the MOS - though it is clear that some would rail against any improvement, particularly if it were not the result of an adversarial process. If this perceived 'railing' is because this matter has been 'hashed over' repeatedly, then perhaps this itself needs to be clarified and made more apparent. It is one thing to be open to change and continuous improvement, even if proposals are ultimately dismissed, it is quite another to be dismissive and resistive of any change. Cinderella157 (talk) 01:00, 7 March 2015 (UTC)

Still you're not indenting -- again I've done it for you -- and if that keeps up people will begin to think you won't put even basic effort into learning how things are done here.
Over wiktionary I'll take [29], which considers declaim, in the sense of protest or criticize, to be intransitive -- you need to add against for that sense.
I don't see the need to close whatever communication gap remains between us, because I'm going to insist (for myself, at least) that you come back here when you can cite multiple issues, in multiple articles, which would have gone better (in terms of time saved, or outcome improved) had some change to MOS been in place.

EEng (talk) 01:59, 7 March 2015 (UTC)

Edit - height uncertainty example

Re edit [31], there now appears to be a redundant example and the examples don't include a conversion. Cinderella157 (talk) 20:41, 8 March 2015 (UTC)

Oops -- I overlooked that an example on that point was already present, which is funny since I'm pretty sure I'm the one who put it there. Not every example needs to include conversions, even though in actual article writing conversions would usually be included. The job of an example is to illustrate the point at hand, not to illustrate every other part of the guidelines at the same time, and this page can ill-afford further bulk. EEng (talk) 00:27, 9 March 2015 (UTC)

Representing controversial population statistics

What is the proper way to represent a controversial city population statistic if the total city population count in RSs includes areas not often considered part of the city? Sir William Matthew Flinders Petrie | ^{Say Shalom!} 18 Adar 5775 15:51, 9 March 2015 (UTC)

Anyone? Bueller? Sir William Matthew Flinders Petrie | ^{Say Shalom!} 28 Adar 5775 01:57, 19 March 2015 (UTC)

Sorry to sound like a broken record (see elsewhere on this page), but this question should be threshed out with your esteemed fellow editors on the talk page of the article concerned. This isn't a MOS matter unless you think something should be added to MOS to handle this situation -- and I would counsel against that unless and until it's become a recurring problem in multiple articles. EEng (talk) 03:45, 19 March 2015 (UTC)

It's more like we were wondering how this question should be handled according to the MOS as we couldn't find anything ourselves. I asked at the Help Desk originally and was directed here. This is a unique situation as no other city had this status, but the original representation of the numbers had parentheses and we couldn't figure out what purpose they served and reckoned there might be something about it in the MOS.... On a side note, don't you think it might be a good idea to put a sign at the top saying this page is for discussions of changes to the MOS only, not referencing? Someone coming here, particularly a person who was directed here, wouldn't know otherwise until you told them, which seems dreadfully inefficient. Sir William Matthew Flinders Petrie | ^{Say Shalom!} 28 Adar 5775 03:51, 19 March 2015 (UTC)

Sorry if I seemed off-putting. I followed your link to Help Desk, and unless I missed something there's nothing there pointing to the actual article so we can tell what you're actually talking about. And that's the problem with these kinds of inquiries -- they're abstract, lacking all the little details that might point to one solution or another. So I counsel you (and your fellow editors on the article) to have faith in your own judgment and work out a solution that seems good to you, for the situation at hand. From the looks of things this doesn't seem to be a point of contention so revel in that! EEng (talk) 04:08, 19 March 2015 (UTC)

UK metric first

I know there was a big fuss about this, and WP:UNIT states that for non-scientific UK articles the primary unit for height is feet/inches. Can someone indicate how much of a "guideline" that is. Also, MOSNUM does not seem to specify whether templates such as {{convert}} and {{height}} should be used. My opinion is that a template is highly desirable to avoid nonsense when editors subsequently edit one number but not the other (if a template is not used). I'm asking due to a difference of opinion with MetricStronk (talk · contribs), for example, diff1 and diff2. Johnuniq (talk) 01:39, 16 March 2015 (UTC)

After a quick review of MetricStronk edits, this appears to be a single purpose account to change primary measurements to metric. The town of Cairns, California was 11 miles northeast of Lincoln, it is now 17.7 km.[32] This edit to KFNW (AM) is vandalism.[33]- SWTPC6800 (talk) 16:39, 16 March 2015 (UTC)

First, with this editor, if he's systematically changing units on UK-related articles without consensus since this warning, then you can now call on the general sanctions at WP:AN.

Second, the rule is that personal height should be feet-first, in the absence of any good reason in specific circumstances to do anything else. At the same time, the general sanctions as written mean that you can't necessarily rely on this guideline for changes to the order, without also getting a local consensus for the specific changes. It's a little contradictory, but it's what was agreed at the time and it's what we'll have to cope with for the foreseeable. The fact that the guideline is here, though, should guide the consensus toward it in the absence of a good reason in specific circumstances.

On the US-based stuff, act as though he changed the language to use British English. It's the same thing.

I would urge editors not to restart the whole discussion as to what the rule should be, that will last months and go nowhere. We have a rule, we have a corollary in the general sanctions. Wasting everyone's time repeating the same discussions over and over again helps nobody. Kahastok talk 18:58, 16 March 2015 (UTC)

WP:UNIT could point to the templates a little more clearly. It already says "For details on when to provide a conversion, see the section § Unit conversions." and that section does mention {{convert}}. Might WP:UNIT say "on when and how to" and MOS:CONVERSIONS mention {{height}}? Many editors may be unaware {{height}} exists or just appreciate the reminder. NebY (talk) 19:28, 16 March 2015 (UTC)

This is maybe a dumb question, but why are there two separate templates for this? I'm not aware of what functionality the height template provides that the convert template does not. Archon 2488 (talk) 23:23, 16 March 2015 (UTC)

I don't know the history of the templates, but I'd imagine some editors might well find {{height}} more straightforward to use, with less parameters to wonder about and documentation focused on that one use. NebY (talk) 19:47, 18 March 2015 (UTC)

Like Archon 2488, I question the need for {{height}}. Proliferating templates just makes maintenance more difficult. There wasn't a discussion leading to a consensus to include it in the MOS page, so I've removed it for the present. Peter coxhead (talk) 11:50, 21 March 2015 (UTC)

{{height}} has already proliferated. There are over 98,000 uses on en.wikipedia. That's not surprising; it's a clearly titled template with a straightforward set of options and it's useful. Far from just making maintenance more difficult, it probably makes pages easier to maintain because it's a little more transparent than the one we do mention, {{convert}}. NebY (talk) 13:08, 21 March 2015 (UTC)

@Peter coxhead: do you still question the need for {{height}} to the extent that you believe it should not be mentioned and would revert me if I re-instated that mention? @Archon 2488: do you think it should not be mentioned? NebY (talk) 14:43, 25 March 2015 (UTC)

I wouldn't go so far as to say the height template should not be mentioned. Indeed, it might already have proliferated so far that it would be hard to deprecate it, and if editors will encounter it in article-space it seems hard to justify removing it from the MoS. But it still seems to me a redundant template. Archon 2488 (talk) 14:56, 25 March 2015 (UTC)

Proliferation isn't evidence for a consensus in a reasoned discussion. I don't think its use should be encouraged. Editors encounter lots of things in article space that are not encouraged in the MOS (and indeed many that are deprecated). If you look at many of the uses, {{convert}} is used in the same article (e.g. for weight in the case of sportspeople). Should we have another template "weight", another "speed" (for use in articles about vehicles), and so on? Each could be claimed to make it easier for editors, but the end result is ultimately not helpful. So I still believe there must be a wider discussion, with a consensus, as is normal for changes to the MOS. Peter coxhead (talk) 16:41, 25 March 2015 (UTC)

I've opened a separate section which might gather more attention; it's only a small suggestion, not worth an RFC. I haven't tried to summarise your case - I hope that's OK. NebY (talk) 20:36, 25 March 2015 (UTC)

Back to the question. Using {{convert}} properly feeding it feet and inches you get this displayed '1.85 metres (6 ft 1 in)'. That has the advantage of keeping the sourced number and avoiding reconversion errors. Vegaswikian (talk) 23:40, 16 March 2015 (UTC)

MOS:DATERANGE for decades

Hiya, what's the appropriate formatting for this: From Gainesville to Boston: 1970s–1980s? MOS:DATERANGE doesn't seem to address this type of range (and if it does, I'm blind and apologetic)—Based on the guideline for normal date ranges 1970–80, my guess is that the proper formatting would be 1970s–80s (endash in the middle). Es correct? Thanks, Cyphoidbomb (talk) 16:01, 25 March 2015 (UTC)

Everyone else is too busy arguing about feet and meters to bother with you. You're right DATERANGE doesn't speak to this specifically. However (here comes the broken record) I always recommend that new issues be hammered out on the talk page of the article involved, and that it be discussed here only after it's become a time-wasting problem on multiple articles. The one thing I can say for sure is that the little dash-hyphen thing in the middle should be an {{endash}}. EEng (talk) 01:47, 26 March 2015 (UTC)

Why are ordinal dates unacceptable?

Standard formal written English uses ordinal dates, yet our poor Manual of Style proscribes them. Why? They make prose smoother, and more natural. They avoid the stilted news-style journalistic writing that "20 January" et al. forces upon us. Encyclopaedias ought be written to the standard expected of an esteemed publication. There are many instances where "20 January" is necessary and proper, but likewise, ordinal dates are also useful for writing good prose. Application should be based on context. I see no reason why a standard English-language usage is blanket forbidden. It is rather absurd, come to think of it. I propose we remove the proscription, encouraging application as suits a particular context. RGloucester — ☎ 06:43, 17 February 2015 (UTC)

"He was born on 20 January 1940" will be read aloud as "he was born on the 20th of January 1940" by most English speakers. So what's the case for writing "20th January" rather than "the 20th of January"? None. "20 January 2015" or "January 20, 2015" are simply two conventions for writing dates in text, independent of how they are read. I don't see that we need more conventions. Peter coxhead (talk) 10:18, 17 February 2015 (UTC)

Hart's Rules (Oxford) specifies cardinal numbers, e.g. "16 August 1960". Some British journalists (The Economist?) may use ordinal numbers (and may use the American order and/or different punctuation), but I would regard such journalistic usage as less formal. I agree with Peter coxhead: it is a convention of written English; the definite article and the ordinal suffix are added when reading. --Boson (talk) 13:53, 17 February 2015 (UTC)

Agree. Similarly, we wouldn't write each of the eighteen children received a 1/18th share. We'd write 1/18 but the reader would speak out one-eighteenth. I can vaguely imagine a construction like rent was due on the 3rd of each month, but (a) hard to see why you couldn't write third instead, and (b) that's not really a date anyway. If the OP has a particular situation in mind I'd like to hear it. EEng (talk) 14:07, 17 February 2015 (UTC)

This is entirely incorrect, and goes against centuries of good English usage. We write out the ordinals. "20 July" is read as "twenty July", not as "20th of July". I'm not asking for more conventions, merely removing the proscription on standard English usage, at once. Formal English usage demands the use of ordinal dates in prose to assist in writing proper and pure-sounding English. RGloucester — ☎ 15:50, 17 February 2015 (UTC)

Let's suppose, for the sake of argument, that some readers read 20 July as "twenty July" -- and perhaps others read it as "twentieth of July" or "the twentieth of July" or "July twentieth" or ... So what? It's understood no matter what. Outside writs of replevin and royal decrees and the letters of Lord Beaconsfield I haven't seen anyone actually write out July 20th in decades. Do we have to fuss about absolutely everything? Once again I ask: do you have an actual article editing situation in which this has come up, or are we just arguing about how many angels can dance on the head of a pin? EEng (talk) 16:49, 17 February 2015 (UTC)

Yes, I do. That's why I'm here. The problem lies in the cadence of the prose. There are many instances where using an ordinal would smooth out a sentence to give it a finished lustre. Without them, one is left with sentences that sound unprofessional and un-English. One must be able to use the tools given to one by the English language. To proscribe standard English usage is a travesty, as is to prescribe choppy prose. I do not know of your origins, but in Britain, ordinal dates remain the standard when writing English formally. RGloucester — ☎ 17:43, 17 February 2015 (UTC)

I'm an American, and I'm all for luster, or even lustre. Can you please cut to the chase and tell us in what specific article-editing situation this has arisen – particular text, in an actual article, which you think would be improved via use of st, nd, rd or th? EEng (talk) 17:53, 17 February 2015 (UTC)

We don't usually base our MOS on what was considered standard several centuries ago, or standard by outliers like you. But if you have modern guides that recommend ordinals in dates in some contexts, please do share them for our consideration. Dicklyon (talk) 16:22, 17 February 2015 (UTC)

The modern is unacceptable. Only the traditional will suffice. RGloucester — ☎ 16:39, 17 February 2015 (UTC)

Uh, oh. Trouble ahead. EEng (talk) 17:24, 17 February 2015 (UTC)

Indeed. The English Wikipedia is not just for usage by the English. Rather, in an act of inverse imperialism a great many countries have descended upon the English language and now claim it as their common language. Thus, the century long English monopoly on the English language have ended. At least with regards to Wikipedia. Terribly sorry, old chap. Lklundin (talk) 21:05, 17 February 2015 (UTC)

Really? I'm a native speaker, but when I parse your example, "He was born on 20 January 1940", I would definitely read and speak it as "he was born on twenty January nineteen-forty". Transforming it to "the twentieth of January 1940" feels weird. Not exactly wrong, but perhaps overtly formal. Curiously, if you had given me "He was born on January 20, 1940", I would have instinctively said "January twentieth nineteen-forty." Anyway, I certainly can't agree with the contention above that the definite article and ordinal suffix are automatically added when reading. I simply don't think that is true of the dialect of American English I grew up with. Dragons flight (talk) 16:17, 17 February 2015 (UTC)

Just to clarify, I'm perfectly fine with using cardinal numbers when writing dates. I am merely disagreeing with the specific contention that they would somehow be implicitly converted to ordinal numbers by the reader. Dragons flight (talk) 16:39, 17 February 2015 (UTC)

Good. As I said in a post above, I don't think it matters how the reader "reads it out", as long as it's understood, and within the radius of normal convention. EEng (talk) 17:26, 17 February 2015 (UTC)

Precisely. Writing 20 January 2015 allows it to be read in any dialect. Peter coxhead (talk) 18:47, 17 February 2015 (UTC)

By that logic, we might as well start writing in dialect, with "ain't", "thou", and "outwith". Sure, the same information will be conveyed. However, will it be conveyed in a cadence befitting an encylopaedia? No. We must have brilliant prose. RGloucester — ☎ 19:18, 17 February 2015 (UTC)

No, because ain't is outside the bounds of acceptable serious writing; 20 June and June 20 are inside bounds. It was in anticipation of your raising such a strawman that I was careful to say, above, "I don't think it matters how the reader "reads it out", as long as it's understood, and within the radius of normal convention." EEng (talk) 20:02, 17 February 2015 (UTC)

Guess what, the normal convention is to use ordinals, and has been for centuries. RGloucester — ☎ 20:50, 17 February 2015 (UTC)

And so we arrive at the nub of the problem (or your problem, anyway): you think good usage is a single point instead of (as it is in most cases) a radius of options. EEng (talk) 21:10, 17 February 2015 (UTC)

I don't know what a "radius" is. However, if by "radius" you mean a platter with assorted cakes, then I agree. I agree that we should have many options available at our disposal, and I believe that there is no good reason for the MoS to exclude a commonly used standard variety for no apparent reason. I never said I would mandate ordinals, nor would I even use ordinals most of the time. I'd use them when they add a lustre to a sentence, or seem appropriate. RGloucester — ☎ 21:14, 17 February 2015 (UTC)

The logic is precisely the opposite. We write in a standardized style, determined by the MOS, so that readers are not distracted by dialect differences (so far as this is possible, given the differences between the ENGVARs allowed here). I can read "born on 20 January 2015" as "born on the 20th of January 2015"; an American can read it as "born on 20 January 2015". Peter coxhead (talk) 19:45, 17 February 2015 (UTC)

This is a question of register, not of information conveyed. I fear there are too many rationalists hiding in these woods. RGloucester — ☎ 19:47, 17 February 2015 (UTC)

Snore. I'm asking you in a new subsection below to give us an actual example of what you're talking about. Otherwise this is going nowhere. EEng (talk) 20:02, 17 February 2015 (UTC)

One exception that comes to mind is phrases like "He died on the 5th". If a day is given without explicitly including the month (presumably because the month is implied by context), then I'm pretty sure the ordinal form is standard. I don't think the present MOS really address that circumstance. Dragons flight (talk) 18:11, 17 February 2015 (UTC)

Yes, that's very much along the lines of the rent was due on the 3rd/third of each month example I gave earlier. I think that's OK, and note that the guideline as written doesn't forbid it -- what is says is to write e.g. June 9 or 9 June but not June 9th or 9th of June. I'd also be OK, if the context was right, with Originally rescheduled from June 3 to June 5, then again moved to the 8th/eighth, the meeting was finally held on the 10th/tenth (though I think I'd prefer eighth/tenth to 8th/10th). EEng (talk) 20:02, 17 February 2015 (UTC)

As an American who learned to read and write in England, and also lived in Canada, I concur with Dragons_flight here; I read that in my head, and would read it aloud, as "twenty January", and it seems perfectly fine to me (though not an American usage). RGloucester's preference for sounding it out in modified form as "the twentieth of January" is certainly insufficient reason to change MOS to favor ordinals. — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼ 22:46, 5 March 2015 (UTC)

The reason is historical. In the dim and distant past every date was wikilinked, because firstly we had a lot less pages and it did not seem unreasonable, secondly there was some vague idea about "metadata" i.e. semantic mark-up, and thirdly it allowed preference formatting. This would have been more clunky with ordinals, especially as the presentation layer could make them ordinals (at least in theory).

I have also had a few qualms about "20 July" vs "the 20th of July" recently, but not enough to bring the matter up myself. All the best: Rich Farmbrough, 19:39, 17 February 2015 (UTC).

Since Hart's (Oxford) and Butcher's (Cambridge) specify the use of cardinal numbers in dates, I find it hard to accept that formal (British) English usage demands the use of ordinal dates". --Boson (talk) 22:23, 17 February 2015 (UTC)

It does not demand their usage, but it does demand that they be allowed to remain in the toolbox of the proseman. RGloucester — ☎ 01:21, 18 February 2015 (UTC)

The proseman??? How Shakespearean thou art. EEng (talk) 03:17, 18 February 2015 (UTC)

No. I fight battles on many fronts. RGloucester — ☎ 03:23, 18 February 2015 (UTC)

It's disgusting, and above all, retarded. | **robot voice** - "PLEASE INPUT CURRENT DATE" | **human voice** - "20 February" | **robot voice** "DATE INPUT ACCEPTED, THANKS HUMAN" *bleep bloop bleep bleep* --Andiar.rohnds (talk) 17:02, 20 February 2015 (UTC)

Angels on the head of a pin

I'm asking RG, out here in the open, to give us an actual editing situation in which this issue has arisen, so we can stop wasting our time in the abstract. EEng (talk) 20:02, 17 February 2015 (UTC)

I refuse. I will not submit to rational analysis. We must have brilliant prose. RGloucester — ☎ 20:50, 17 February 2015 (UTC)

Then you're just stirring the pot for no reason. EEng (talk) 21:04, 17 February 2015 (UTC)

No, I've been forced to write plain dates when an ordinal would make a more pleasant sentence for far too long. These rationalists who privilege data above beauty and tradition must be pushed out. RGloucester — ☎ 21:12, 17 February 2015 (UTC)

@RGloucester: I suggest you start working on the Irratiopedia and leave Wikipedia to us rationalists. Peter coxhead (talk) 22:30, 17 February 2015 (UTC)

Warning! Unhide only in an environment where bursting out in uncontrollable laughter will not lead to embarrassment or danger of physical harm. Do not drive or operate machinery until you know how the hidden content affects you.
Perhaps we need an "Irascipedia"! --Boson (talk) 22:37, 17 February 2015 (UTC)

"I'd use [ordinals] when they add a lustre to a sentence, or seem appropriate." Subjective notions would only lead to edit-warring over what seems appropriate in individual cases. We need, if anything, fewer standardised options, not more. "We must have brilliant prose." This is an encyclopedia for objective recording of facts; brilliant prose is for authors and poets. "I will not submit to rational analysis." Then there is no benefit in arguing with you logically. —sroc 💬 02:41, 18 February 2015 (UTC)

We already have "subjective notions" with MDY and DMY and the like. No reason not to include another perfectly acceptable "subjective notion". If you'd like to settle on one particular date format, I'd be all for that. Until then, however, there is no reason not to accept ordinals. RGloucester — ☎ 02:58, 18 February 2015 (UTC)

When you have an actual example of where you'd like to do this, discuss it with your fellow editors on the talkpage of the article in question, citing MOS' oft-forgotten "use common sense/subject to occasional exceptions" provision. If after that you're still aggrieved, then bring your example back here. For now you're demanding a tool for which you refuse to give even a single use case, other than your lustre thing (or, as we Americans would put it, luster -- a word that seems a bit naughty now that it look at it). EEng (talk) 03:17, 18 February 2015 (UTC)

That would be an amenable solution. However, no one has yet justified the proscription, and hence it isn't. I shan't talk about "single use cases" on the grounds that I don't want people following me to places where they don't belong, as often happens in these cases. RGloucester — ☎ 03:23, 18 February 2015 (UTC)

"We already have 'subjective notions' with MDY and DMY and the like." We can have objective arguments of whether MDY or DMY should be used in individual cases based on the considerations enumerated in MOS:DATETIES. If you can justify using ordinal rather than cardinal numbers in dates based on something less vague than "lustre", you might have more than a snowball's chance of forming consensus for your proposed change. —sroc 💬 04:34, 18 February 2015 (UTC)

I shall hang on a tree before that happens, I fear. Enjoy your poorly sounding dates. RGloucester — ☎ 04:38, 18 February 2015 (UTC)

RG, are you all right tonight? In all seriousness, let's all take a breather. EEng (talk) 05:30, 18 February 2015 (UTC)

"Standard formal written english uses ordinal dates, yet our poor Manual of Style proscribes them. Why?". TBH, I don't understand the queston. Isn't the MoS matching formal written english? GoodDay (talk) 17:17, 20 February 2015 (UTC)

Contrary to what some editors keep insisting, there is good deal of variety and flexibility in formal English. In many cases MOS reflects (explicitly or implicitly) that flexibility, while in others it prescribes one, or a few, of standard alternatives. EEng (talk) 20:10, 20 February 2015 (UTC)

And then, as quickly as they had come, they were gone. EEng (talk) 04:26, 24 February 2015 (UTC)

Unlike these tasty ~~ordinal~~ traditional dates, with a tree for RG to hang on. Martinevans123 (talk) 23:03, 24 February 2015 (UTC)

Superscript Ordinals

[Moved here from a new thread opened elsewhere on the page -- somewhat related]

I need some things EXPLAINED and good arguments coupled with those explantions. This dialogue took place and you're the people apparently who can clear it up... User_talk:Magioladitis#Yobot_Query

David-King (talk) 17:32, 4 March 2015 (UTC)

It's just "house style" -- arbitrary choices made long ago. I know such rules can be grating when one is used to some other style, but for better or for worse it is what it is. You're entitled to begin a discussion about changing it -- presumably you'd like MOS to allow superscript ordinals (I note your particular concern was about ordinals arising outside the context of dates, unlike the discussion earlier in this thread) -- but you'd have to have very, very strong reasons that readers will be better served by such a change, not just that you think it looks better etc. Trust me on this; to see what you'd be getting yourself into, click here -- if you dare! EEng (talk) 17:57, 4 March 2015 (UTC)

The most likely reason is that any text that is super-scripted can mess up line spacing. The various style guides around are mostly based on many years of experience of print publishing (Chicago and Oxford) although the Microsoft MOS has a section on web content. Sometimes the reasons for any guidance in a style guide are explained, but often they are not. Other considerations, apart from line spacing, may include readability, accessibility, and whether a feature is available in a wide range of fonts. Robevans123 (talk) 18:33, 4 March 2015 (UTC)

I have no care for ordinals in dates. There are many formats for dates and why one should be picked over another is simply arbitrary. I would simply be arguing that we should superscript ordinals when referring to a position in a chronology or a position in a race for example. As I've stated before, Microsoft Office programs' default is to superscript ordinals. I really can't imagine there will be any issues of readability and accessibilty or such like or even any issue of font compliance. David-King (talk) 19:24, 4 March 2015 (UTC)

What is the perceived benefit of using a superscript? Dondervogel 2 (talk) 19:49, 4 March 2015 (UTC)

What's the perceived benefit of the alternative? David-King (talk) 20:00, 4 March 2015 (UTC)

Aside from the considerations Robevans123 already mentioned (which are real, whether you can imagine them or not) the benefit of the current guideline is that it's the guideline we've got, and absent some quantum of indication that there's a benefit to changing it, time that would be spent rehashing it should be invested instead in working on articles. So the ball really is in your court, D-K. I hope this doesn't come across as dismissive, but if you follow the link above you'll see that this point, like every other goddam thing in MOS has been discussed, and discussed, and discussed to death, for no apparent reason except that different people are stuck on different arbitrary conventions. EEng (talk) 21:50, 4 March 2015 (UTC)

So what, you're saying we don't superscript ordinals because of accessibility and readability issues as well as because of font difficulties? I've never heard such spurious sanctimony emanated from an apparent expert in all my born days. Thus far, you have all come across as dismissive and contrary to what I said I wanted, you have only given me personal assertions which have not been evidenced by anything substantial. It's insulting. David-King (talk) 19:34, 5 March 2015 (UTC)

Spurious sanctimony emanated from an apparent expert? Gracious! I've never heard such unconscious consonance and all-out alliteration in all my born days. And who said anything about being an expert?

To repeat: the rule is (likely) largely arbitrary and (quite possibly) partly motivated by technical and accessibility considerations, and (almost certainly) no one still hanging around knows specifically. But (also again) dropping in and expecting everyone to reopen the question with you, when you pointedly refusing to give even a hint of what you think the benefit would be, is inappropriate. Follow my link above to the past discussions of this issue, and if after you think you can shed light that hasn't been shed before, by all means let the rest of us know. EEng (talk) 22:22, 5 March 2015 (UTC)

Actually, you're not repeating yourself. You're making yourself clearer to compensate for your lack of coherence in your previous responses. The benefit is that it looks more professional and indeed in schools, pupils are taught to adhere to the standard which is to superscript ordinals. There isn't a great amount I can 'shed light' on as it is a relatively peripheral and poxy issue in the first place so let's not make this something it isn't. Reopening it is no big deal especially given that it obviously wasn't sufficiently discussed originally. David-King (talk) 00:24, 7 March 2015 (UTC)

As you'll soon learn if you hang around here for any length of time, those confusing something Miss Snodgrass taught them when they were 12 for "the" standard to which all "pupils" (how quaint!) are taught to adhere are setting themselves up to be made fools of. In the present case, I note that Chicago and AP, at least, say no to superscripts, as does, apparently, the practice in American legal work [34][35]. And this very thoughtful short discussion [36] sends them up this way:

Microsoft Word and its ilk reintroduced this fetish from the Victorian era and made it the default, so any ordinal number (1st, 2nd, etc) gets the ordinal indicator as a superscript instead of the normal 1st, 2nd, etc. ... the use of the superscript form seemed to have disappeared around the 1940s and 50s — until its corpse was reanimated by Microsoft.

So just how are you going to turn superscripting into "the" universal and professional standard? EEng (talk) 04:39, 7 March 2015 (UTC)

Chicago and AP don't set the international standard; perhaps they do the American standard but not the international standard. Funnily enough, Microsoft stipulations are more internationally widespread due to its wide use and so if anyone sets the universal standard, it's Microsoft. It's not quaint at all; the standard taught in most schools worldwide is that ordinals be in superscript. The world doesn't revolve around America you know. There is China for example who own the US debt. David-King (talk) 22:39, 7 March 2015 (UTC)

<rolls eyes> Other than your personal experience of world educational systems, can you point to anything actually recommending superscripting? You'll pardon us for not accepting the word of someone who doesn't know a hyphen from a dash, and thinks that "Professors of English always say it is preferable to keep hyphens out of prose" [37].

EEng (talk) 00:08, 8 March 2015 (UTC) Martinevans123, I'm guessing you will have something incisive to add at this juncture of the narrative.

Followup: Oh, look! Here's what the Oxford Style Guide [38] says:

Spell out words for ‘first’, ‘second’ and so on up to and including ‘tenth’; use numbers and ‘st’/ ‘nd’/ ‘rd’/ ‘th’ for larger ordinal numbers. Don’t use superscript (to prevent problems with line spacing).

Don't worry‍—‌maybe Cambridge wants superscripts. Why don't you check and get back to us? If not, try China.

EEng (talk) 01:14, 8 March 2015 (UTC)

Enjoy this marvellous educational video. It has a super script. Martinevans123 (talk) 09:31, 8 March 2015 (UTC)

No superscripts, I note for the record. EEng (talk) 13:24, 8 March 2015 (UTC) "It has a super script." You crack me up.

Don't impose experiences upon me; I never said I had "personal experience of world educational systems". I was just stating something I'd read. Back in the day, when Britain was a colonial power, people were taught British English overseas and I read, although I can't find exactly that which I did read at the present time, that students were taught that ordinals be in superscript. As generations have passed, standards have slipped and what is taught in schools has changed by consequence.

Actually, I do know the difference between a hyphen and a dash and as a training lawyer at a Russell Group institution, being adept in using all punctuation is quintessential. However, my A* in my English Language A-level speaks for itself really. To have got anything less would have not justified the £14K spent per year on my secondary education at a school which prides itself on academic excellence and the upholding of its magnificent academic record but there you go. I however didn't care for the correct use of the pertinent jargon in that instance because I didn't realise that I was being tested. I should think twice now in future I suppose. I however do concede that I made a mistake and that I was sloppy for which I apologise.

Before you start extrapolating, I should tell you that context is important and that which you quote me verbatim is hyperbole. I merely wanted to potently enforce my point. I did rounds with a survey asking English professors at my university whether it was acceptable to use dashes in prose. The consensus was I found that it is acceptable to use them but If YOU CAN phrase your sentence without them, THEN DO SO.

Finally, you come with some PROOF instead of blind assertions evidenced by nothing!!!! As Oxford, AP and Chicago do indeed say there are issues with line spacing when putting ordinals in superscript, I will now accept that point albeit reluctantly. In my mind, they do look more professional and were computers more advanced, issues of line spacing wouldn't prevail. Of course there will be a source which supports the superscripting of ordinals. It wouldn't be done full stop if there weren't any sources which supported it. David-King (talk) 00:37, 14 March 2015 (UTC)

To summarize:

A lot of MOS is arbitrary, as is true of all style guides
My link way back gives the discussions that led to the no-superscript choice
Authoritative usage guides (American and British) deprecate superscripting
Nothing has been offered supporting the idea that superscripting is more "professional" or is what "is taught in schools", despite your ALLCAPS, your bolding, and your expensive and therefore impressive credentials.

EEng (talk) 10:33, 14 March 2015 (UTC)

As EEng notes, a choice has to be made one way or the other, and maintained for consistency. Some people like superscripting; others don't (it seems to me old-fashioned now). No good reason has been offered to change the existing MOS. Peter coxhead (talk) 11:11, 14 March 2015 (UTC)

(ec) @David-King It seems to me that, while there might be arguments for and against superscripts, there is no knock-out punch either way. The important thing to achieve the harmonisation that is MOS's main goal is to choose one of them, however arbitrary that choice might seem. MOS made that choice eons ago. Without a good reason to change them, a good enough reason to keep things the way they are is that changing them would result in instability. Dondervogel 2 (talk) 11:24, 14 March 2015 (UTC)

You just can't debate well can you EEng; take another look at my reply and please tell me how ALLCAPS, BOLDING and the citation of my credentials was used to enforce any of my points on superscripting ordinals. You invoked an issue you had with my misuse of jargon in an entirely separate discussion and I responded to that specific point by citing my personal education history as well as using BOLDING and ALLCAPS. Not anywhere did I use them to enforce any of my assertions on superscripting ordinals. Dondervogel 2 is right; there is no knock-out punch either way. However, I will accept maintaining the status quo is an easier option and by consequence the preferable option. British and American usage guides don't deprecate ordinals in superscript; they just don't decree them ideal because of font and line spacing issues which we here on Wikipedia may or may not run into. David-King (talk) 21:00, 18 March 2015 (UTC)

Whatever helps you sleep at night. EEng (talk) 22:58, 18 March 2015 (UTC)

Oh believe you me, I have far far more important things keeping ME awake at night. I'll be back though... David-King (talk) 13:23, 21 March 2015 (UTC)

If you say so. EEng (talk) 18:13, 27 March 2015 (UTC)

Are metric units compulsory when describing events from 18th/19th century Australia?

A user left a threat of sanctions on my homepage, stating that MOS does not allow me to change measurement units back to the more contemporary units after he has changed them to metric. The change concerned an 1830 event in Australia.

Is this really the case, as there are plenty of good articles which do not seem to comply with this bizarre sounding restriction?

Looking at the use of the complainant's account, "Archon 2488", it appears to exists solely and for no other reason than to metricate as many Wikipedia articles as possible. It has been used to systematically visit huge numbers of articles solely to undermine the work of the creative editors and to change the measurement units to metric, regardless of context. — Preceding unsigned comment added by Hugh8 (talk • contribs) 16:56, 25 March 2015 (UTC)

I am amused by this blatant bad-faith attack. Since anyone who has been around the block more than once knows perfectly well who we are dealing with here, I'll not address him directly; I'll just leave a few comments for others, to provide a bit more context.

It is not a "bizarre-sounding restriction" to require articles about a country's history to use the units in normal use in that country. The article in question won't be read by Australians from 200 years ago, for obvious reasons, so unless there are good article-specific reasons for preferring some other system of measurement (such as nominal values, which are already covered) metric should go first by default. To my knowledge, nobody has seriously disputed this.

The "threat of sanctions" was a reminder that revert-warring, especially when you are aggressively reverting to a MOS-unfavoured style for no good reason, is likely to result in sanctions. Archon 2488 (talk) 20:43, 25 March 2015 (UTC)

I think it is bizarre to to require articles to reflect any characteristics of their subject. Should an article on Napolean be in French? Or use the calendar of the Revolution? I don't think so. Or should metric conversons in U.S. topics use different conversion templates depending on whether they are primarily before or after the adoption of the survey foot in 1959? (Perhaps, but would still be a bit bizarre.)

What I think is proper is that articles present information in a manner most suitable for the audience. That most of the world's population - and even most of the English-speaking population - is most familiar with the metric system is, I think, a strong argument for universally listing metric values first. That we tolerate variance in cases where editors have strong preferences where there is a "national" (also cultural?) basis is acceptable. But the basis for this (absent any special considerations) should be the usage of the current audience, not of the historical context of the topic. ~ J. Johnson (JJ) (talk) 22:23, 25 March 2015 (UTC)

I agree up to a point – as a compromise we say that USA-related articles should use American English per ENGVAR, and so on. Likewise it would be natural to follow some conventions of the article subject when that is what reputable sources do; in the case of the French Republican Calendar, for example, we would speak of the Coup of 18 Brumaire VIII. But of course this is not an unlimited justification for using older conventions in general historical settings, such as insisting on giving all measurements in articles on Roman history in Roman units of measurement purely for the sake of historical authenticity. WP is a modern document and should follow modern conventions. Archon 2488 (talk) 01:10, 26 March 2015 (UTC)

The absence of a definitive "yes" answer by now indicates to me that metric units are not compulsory in these circumstances.

Is it acceptable that a virtually single-purpose account such as "Archon 2488" is using such threats against those who challenge his attempts to mass-metricate Wikipedia, especially as he does not describe such edits correctly in the summary and he even flags them as minor changes! Account "Johnuniq" characterised "Archon 2488" as an "SPA" (single purpose account) and described him succinctly on the talkpage of the "Thundersnow" article with this statement: "... someone being disruptive by changing the style of existing articles to SI with edits marked as minor". Hugh8 (talk) 20:48, 26 March 2015 (UTC)

Comments from two editors on a talk page don't "indicate" anything about MOS consensus. They are certainly not carte blanche for you to follow other editors around, hitting the revert button.

FTR if you read the relevant discussion you'll see that "account" Johnuniq (dude, not everybody here is a sockpuppeteer – they are usually best thought of as people rather than "accounts") was referring to someone completely different; a SPA that was active briefly a few weeks ago (which he calls "1"). His SPA #2, by the way, is our friend here. Archon 2488 (talk) 20:57, 26 March 2015 (UTC)

To clarify what I wrote at Talk:Thundersnow#Units of measurement, two SPAs have recently started pushing opposing views regarding units. They are (1) MetricStronk (talk · contribs) and (2) Hugh8 (talk · contribs). Regulars here know that Archon 2488 is an established and very helpful editor. I currently don't have a view on the issue of which unit goes first, and ask for views on the example in the second para at Halls Head, Western Australia—should it say:

received a land grant of some 6,715 hectares (16,594 acres) or
received a land grant of some 16,594 acres (6,715 ha)

Let's clear that up and leave discussion of editors for the WP:SPI page. Johnuniq (talk) 22:33, 26 March 2015 (UTC)

Sorry if I misunderstood you, but the only apparent difference between the actions of MetricStronk (talk · contribs) and Archon 2488 (talk · contribs) is that where the former is blatant and explicit with his metrication of articles, the latter is secretive and underhand. Although the latter has certainly been doing it longer, his editing history nevertheless shows him to be an SPA and fits your definition as "someone being disruptive by changing the style of existing articles to SI with edits marked as minor". Indeed his most recent edit (to "Edinburgh of the Seven Seas" at 22:02, 26 March 2015) exemplifies this. How can you defend this action whilst condemning that of the former? It sounds to me like double standards. Hugh8 (talk) 07:19, 27 March 2015 (UTC)

For Australia-related articles, the rules are very clear: metrics main, imperial in parentheses. It doesn't matter whether the topic involves the current day, 1970 (before the adoption of metric units in that country), or 1870. Tony (talk) 08:03, 27 March 2015 (UTC)

I believe Tony is correct, and that should be the end of it. EEng (talk) 17:59, 27 March 2015 (UTC)

Yes. And simple edits in accordance with a clear rule are properly deemd minor. ~ J. Johnson (JJ) (talk) 21:32, 27 March 2015 (UTC)

...save for the standard exceptions, which apply on every article but are particularly likely to come up when describing nineteenth-century Australia. If the land grant was defined as being precisely 16,594 acres - and I rather doubt it - then it would in general be appropriate for acres to go first.

And sockpuppet or not, Hugh8 does have a point that the edit made by Archon to Edinburgh of the Seven Seas both broke the MOS and had the potential to trigger the general sanctions. Kahastok talk 15:41, 28 March 2015 (UTC)

Thank you for mentioning that one edit without context and for feeding into a certain lunatic sockpuppeteer's obsession with discussing editors he doesn't like. It's nice to see that out of god-knows-how-many small fixes and stylistic adjustments you pick up on the one that could paint me in a bad light and present it – where plenty of others can see – in complete isolation (and the fact that a sockpuppet did exactly the same thing before you is a weak excuse). Of course, it's not at all possible that somebody wouldn't intuitively guess that an island in a different part of the world, with a political connection to the UK that most people frankly couldn't explain to you, was implicitly included in the general sanctions, for reasons relating to some argument that a different group of editors had about a different island years ago. This gives me the impression that you're just waiting to "catch me out", which is frankly a bit oppressive and passive-aggressive.

And for what it's worth, neither the General Sanctions nor the MOS section about the UK actually mention overseas territories or crown dependencies (or whatever other fossils of the empire might still exist). So in a lovely Kafkaesque fashion, it's apparently entirely possible to break a rule for reasons that are not explained. Archon 2488 (talk) 19:06, 28 March 2015 (UTC)

If you have personal problem with another editor, take it to ANI. Screwing up on a single edit would be nothing (haven't we all?), but obsessing about it just makes it more significant. Time to put this thread out of its misery. ~ J. Johnson (JJ) (talk) 20:30, 28 March 2015 (UTC)

Please, please will you all just stop this so this thread can be archived? This page is for discussion of changes to MOS. EEng (talk) 20:00, 28 March 2015 (UTC)

As I see no clear agreement that metric is compulsory in such circumstances, I'll assume therefore that it is not. In which case, I suggest "Archon 2488" seeks consensus before making them, or at least clearly summarises the edits as "metrication" or similar, and certainly stops marking them as "minor", to increase their visibility. Can we at least all agree on that? Hugh8 (talk) 00:13, 29 March 2015 (UTC)

No. Talk:MOSDATE is for discussion of changes to MOSDATE, not to micromanage edit summaries. EEng (talk) 00:42, 29 March 2015 (UTC)

Well, the problem is that both of these editors are invoking various guidelines and this discussion in support of their strongly colored interpretations of what should be done. Since these two can't work it out I am wondering if we could assemble an ad hoc "Committee of Supervision" to look at whatever article all this arose from, settle matters how ever they deem best (generally and any exceptions), and lay down that this is the way it is going to be, end of discussion. ~ J. Johnson (JJ) (talk) 19:34, 29 March 2015 (UTC)

I realize I may seem like a bit of a monomaniac on this point, but I think it's highly undesirable for Talk:MOSDATES (or the Talk of any part of MOS) to take on that kind of refereeing role. The reason, again, is that the immediate goal of such a discussion will be some compromise (or whatever) to solve the dispute at hand, but that conclusion -- really fashioned for a particular situation and particular personalities -- will inevitably take on a kind of precedent value for future disputes on the same topic. In other words, it will to some extent have the effect of an actual change to the text of MOS, even though the discussants didn't think of themselves as doing that, and didn't mean for that to happen. If people seeing this thread want to go over to the Talk of the article(s) at issue and help out there, that's fine, but not here. This may seem like a distinction without a difference, but I really think this page should be for, and only for, discussions of changes to MOS. EEng (talk) 20:00, 29 March 2015 (UTC)

You, J. Johnson and others may be interested in the RfC at Wikipedia:Village pump (proposals)#Proposal: creation of "style noticeboard". NebY (talk) 20:30, 29 March 2015 (UTC)

EEng: I understand. My idea is that an ad hoc but distinct group of editors could pull this matter out of this forum. Leaving it to what ever editors happen to drift in to the article(s) does not seem to be working. The only precedent I would want to set is the example that if editors persist in arguing article issues here they might get boomeranged with an arbitrary decision. ~ J. Johnson (JJ) (talk) 21:21, 29 March 2015 (UTC)

What's the article in question? EEng (talk) 15:17, 1 April 2015 (UTC)

Hugh8 is now shopping his complaint to ANI. Wikipedia:Administrators'_noticeboard/Incidents#Paranoia.2C_a weird coincidence.2C_or something underhand going on.3F EEng (talk) 00:18, 4 April 2015 (UTC)

I don't understand the question. Nothing is "compulsory". The question is just how to best serve the reader, and part of that is answered via the consensus represented in the MOS. If there's a dispute about what's best, after reviewing the general guidelines, let's address the specifics, not talk about what's "compulsory". It seems to me that adding metric is generally a good thing; if there's a question about which comes first, point us at some alternatives to look at (e.g. disputed diffs perhaps). Dicklyon (talk) 00:35, 4 April 2015 (UTC)

There is no question. You must understand, Mr Lyon, that this Hugh so and so is likely part of a longterm disinformation and abuse campaign. Better to ignore it. RGloucester — ☎ 00:36, 4 April 2015 (UTC)

Mention Template:Height in MOS:CONVERSIONS?

Should we mention {{height}} as well as {{convert}} in WP:MOSNUM#Unit conversions thus (insertion in italics): "Conversion templates can be used to convert and format many common units, including {{convert}} and {{height}}, which include non-breaking spaces"? {{height}} has existed since 2006 and there are over 98,000 uses of it on en.wikipedia. It's very similar to {{convert}} but with less options and more focused documentation, so it may be easier to use for some editors who might be put off by the power of {{convert}}. There's a brief discussion above in the second half of WT:MOSNUM#UK metric first (don't worry, this part isn't about metric/imperial), where you'll see that I did once go ahead and insert it and Peter coxhead reverted, so we're now in the discussion stage. NebY (talk) 20:31, 25 March 2015 (UTC)

I would say not. {{Height}} is redundant; it would best be deprecated. Once we had a plethora of conversion templates but as {{convert}} became more and more functional most of these became redundant to it and were subsequently deleted. The only reason {{height}} survived was that it could do something that {{convert}} couldn't: produce fractional output. Now, though, this limitation is gone (thanks to Johnuniq) so there is really no reason to keep {{height}} around. Having a number of templates doing the same thing just makes it all the more difficult for users to fathom what's going on, let's not encourage it. Jimp 01:13, 29 March 2015 (UTC)

MOS:DATEUNIFY allows DMY and MDY in the same article?

Wikipedia:Manual of Style/Dates and numbers § Consistency (MOS:DATEUNIFY) says:

Dates in article body text should all use the same format: She fell ill on 25 June 2005 and died on 28 June, but not She fell ill on 25 June 2005 and died on June 28.
Publication dates in an article's citations should all use the same format, which may be

the format used in the article body text,
any other format from the "Acceptable date formats" table, or
the format expected in the citation style being used (however, all-numeric date formats other than yyyy-mm-dd must still be avoided).

For example, a single article might contain one, but only one, of:

Jones, J. (20 September 2008)

Jones, J. (September 20, 2008)

Jones, J. (20 Sep 2008)

(among other possibilities).

Access and archive dates in an article's citations should all use the same format, which may be:

the format used for publication dates in the article,
the format expected in the citation style adopted in the article, or
yyyy-mm-dd

For example, a single article might contain one, but only one, of:

Jones, J. (20 Sep 2008) ... Retrieved 5 February 2009.

Jones, J. (20 Sep 2008) ... Retrieved February 5, 2009.

Jones, J. (20 Sep 2008) ... Retrieved 5 Feb 2009.

Jones, J. (20 Sep 2008) ... Retrieved 2009-02-05.

(among other possibilities). When a citation style does not expect differing date formats, it is permissible to normalize publication dates to the article body text date format, and/or access/archive dates to either, with date consistency being preferred.

The highlighted text implies that the dates can be in DMY format in the body and MDY format in the publication and access/archive dates, or vice versa (as discussed at Talk:Microsoft Office 2013 § MOS:DATEFORMAT. This seems to be a mistake. It is understandable that publication and access/archive dates may take an abbreviated form that would not be allowed in prose, but the date elements should be in the same order (unless using YYYY-MM-DD). For example:

DMY dates:
- Body: She fell ill on 25 June 2005 and died on 28 June
- Reference: Jones, J. (20 Sep 2008) ... Retrieved 5 Feb 2009
MDY dates:
- Body: She fell ill on June 25, 2005, and died on June 28
- Reference: Jones, J. (Sep 20, 2008) ... Retrieved Feb 5, 2009
Mixed use (should not be allowed):
- Body: She fell ill on 25 June 2005 and died on 28 June
- Reference: Jones, J. (Sep 20, 2008) ... Retrieved Feb 5, 2009

To this end, I propose amending the MOS:DATERET as follows:

amend the words 'any other format from the "Acceptable date formats" table' to read 'an abbreviated format from the "Acceptable date formats" table, provided the day and month elements are in the same order'
delete the example 'Jones, J. (20 Sep 2008) ... Retrieved February 5, 2009.' or perhaps even use it to illustrate a bad example.

—sroc 💬 13:56, 28 January 2015 (UTC)

On closer inspection, I'm not sure why this example is permitted either:

Jones, J. (20 Sep 2008) ... Retrieved 5 February 2009.

In this case, the access date is not in the same format used for the publication date, nor is it yyyy-mm-dd. If it is "the format expected in the citation style adopted in the article" then this should be made clear, as it is impossible to tell from the information given what citation style is adopted and why a particular date format is required. In any case, I query whether the various examples are even necessary if the rules are made clear enough. —sroc 💬 14:14, 28 January 2015 (UTC)

Oppose Just because you don't understand behind the rationale behind something, it doesn't mean that the said thing suffers from spelling mistakes. It only means you don't understand why. This manual of style's aim is to cultivate diversity in distinct areas of one article and facilitate different global styles to be employed without any arbitrary favoritism toward one. That's foundation's way of work: Neutral towards all arbitrary styles. Fleet Command (talk) 14:59, 28 January 2015 (UTC)

What I would like to know, does any one (paper) source mix styles? I didn't get at first that some people prefer having DMY and MDY in the same article. Is there a better way for cultural diversity and/or consensus on this? How hard would it be to make a user preference setting work? comp.arch (talk) 15:48, 28 January 2015 (UTC)

The existing user preference setting can only customise dates that are stored in the database as a Unix date, such as those in your contributions, or in a page history. It cannot do anything at all about dates stored as text, which is what we're dealing with here. At one time, it was possible to enter dates into the Wikitext using a format like [[28 January]] [[2015]] and dates linked like that would be formatted according to user preferences: but the feature was removed over six years ago. I strongly suspect that it will not be reintroduced. --Redrose64 (talk) 17:24, 28 January 2015 (UTC)

@Comp.arch: Automatically converting dates is also quite hard because MDY dates usually require a comma after them depending on context, and it is not simple for computers to work out whether the comma should be inserted/removed in a particular context (e.g., a sentence might require a comma anyway even if the date is converted to DMY to suit the reader's preference). [See User talk:Ohconfucius/script/MOSNUM dates § Commas after MDY dates.] —sroc 💬 18:16, 28 January 2015 (UTC)

Ok, but most cases are easy? Seems to me a template around DMY dates would work, that does nothing if you do want DMY (or no changes) but converts and adds a comma for MDY. Note you would only use it for the exceptional cases. Maybe they are (almost) never required with some heuristics that add a comma without a template, if "On", "In" or "Since" (or in lower case) is in front. Most readers wouldn't be bothered by MDY only or DMY only in articles and the MOS could then require either choice but for readers that are fuzzy, they would enable the conversion. It seems to me at least changing an article to DMY shouldn't bother anybody then? comp.arch (talk) 14:24, 29 January 2015 (UTC)

@FleetCommand: Firstly, chastising me on not understanding something is not nearly explaining why. Secondly, we're not dealing with spelling mistakes, so I'm not sure why you'd bring that up. Thirdly, the MOS does not aim "to cultivate diversity" but rather to promote consistency of style across Wikipedia. Now, can you point to any reason why a section headed "Consistency" which states that "Dates in article body text should all use the same format" should allow contradictory DMY and MDY formats within a single article? —sroc 💬 18:10, 28 January 2015 (UTC)

Oh, God I am not "chastising" you. Why is everyone dramatizing everything out of proportion? You made a mistake, I pointed out. But of course, there is nothing else in your message to which I reply because your "thirdly" says that you disagree-period, and your last sentence with question mark is a trick question built on the assumption that I agree with your "thirdly". So, to summarize: You don't even want to listen to opposition. Fine. I have nothing more to say. My "Oppose" stands. Fleet Command (talk) 13:43, 29 January 2015 (UTC)

@FleetCommand: To address the third point, which remains a point of contention, what is your basis for your claim that: "This manual of style's aim is to cultivate diversity in distinct areas of one article"? In any case, I note that you have not elucidated on any reason why both DMY and MDY dates should be permitted within individual articles (aside from honouring citation styles) which would seem to contradict the underlying principle of MOS:DATEUNIFY. —sroc 💬 13:26, 30 January 2015 (UTC)

@Sroc: A person convert 87% of dates in one article from DMY to MDY, the opposing person does not receive fair treatment, someone offers a compromise based of the current reading of MOS and to defuse that compromise, you start this topic. You want elucidation: I have one word for you: WP:GAME! As for my own belief, you shouldn't have asked it because you don't like it but here it is: I believe all of Wikipedia must adhere to one unified date style and let everyone customize its own view. Fleet Command (talk) 02:20, 4 February 2015 (UTC)

@FleetCommand: Please. The conversion was to change the dates in the refs to match those in the article, in the expressed belief that an article about Microsoft Office had strong national ties to the U.S. Given that GF belief, the change was not improper. (It's been argued that Microsoft does not have strong national ties to the U.S., a position I find bizarre, but that's not the point here.)

And there is nothing wrong with coming here to ask about, or even challenge, perhaps with the result of changing, a point of MOS that's been brought up on an article talk page. That's how and why a great many things in MOS get changed.

And you didn't answer sroc's question. What is the basis for your claim that "This manual of style's aim is to cultivate diversity in distinct areas of one article"? (emphasis added - jeh) Was this "aim" ever discussed and agreed upon? I always thought that the point of having a house style was to show consistency. And is there any evidence that allowing date formats in different parts of the article (specifically body vs. refs) to differ actually cultivates or shows support for diversity? Or is it just someone's belief that it does? (I'm asking these questions in GF.) Jeh (talk) 22:48, 4 February 2015 (UTC)

Straw man! Straw man! Straw man!

"The conversion was to change the dates in the refs to match those in the article" Why? The refs were more numerous. Their style was the prevalent one. With or without sroc's proposal, changing 87% of dates in one article to another format is a violation the MOS that exists today, the MOS that existed in 2012 and the MOS that existed in 2006.

"It's been argued that Microsoft does not [~snip~]". Why it was argued in the first place? Microsoft wasn't the subject of dispute at all. Office was.

"What is the basis for your claim that [~snip~]". A manifesto of some sort from Jimbo Wales. But that's besides the point: When someone asks my personal opinion, what I give is my personal opinion. You don't have to agree with it; you just have to stick to MOS. And that's why Comp.arch and I had this discussion in my talk page, not article talk page.

Fleet Command (talk) 20:45, 8 February 2015 (UTC)

Did the "manifesto of some sort from Jimbo Wales" include the words "in distinct areas of one article"? (Please provide a link so we all can see what you're talking about.)

Re. "opinion": your claim "This manual of style's aim is to cultivate diversity in distinct areas of one article and facilitate different global styles to be employed without any arbitrary favoritism toward one" appears over your signature on this talk page, not your user talk page. No one here had asked for your opinion, personal or otherwise, and nowhere here do I see anything that disclaims your statement as "just my opinion". Taken as written on this page, it appears to be a claim of fact by you. Indeed, your phrasing ("just because you don't understand the rationale behind something") implies that this is the rationale for allowing different date styles in article body vs. refs. I see that you are now stating that it is just your opinion. Well OK then. Jeh (talk) 01:04, 9 February 2015 (UTC)

@Jeh: New policies and policy changes stem from people's own personal reasoning and nothing else. (Otherwise, the first founding policies are never created.) Also "my talk" stands the opposite of "Talk:Office 2013", because in the latter, the policy must be used. This talk page is excluded from my original clause because of the sentence subject "Comp.arch and I". Are these things so difficult for you to understand? Quite frankly, it seems you are just trying to get back at me for that small mistake that I pointed out a in Talk:Office 2013. I'd gladly have dug out the manifesto thing from Meta if I didn't think your next reply is some other form of denial. Fleet Command (talk) 20:34, 9 February 2015 (UTC)

@FleetCommand: I'm not saying you have no right to your opinion or personal reasoning or that these have no relevance to the discussion. But on this page, you stated "This manual of style's aim..." as if it was "the rationale" for the current guideline that permits mixed date styles. If that had been the case, then any attempt to change the support for mixed date styles would have to address that point. You now have stated that no, it's just your opinion that that's the rationale. Don't you think that it is useful to have established that perspective? Now, I would like to read the referred-to manifesto, if it isn't too much trouble for you to find. (I've tried searches - too many hits to go through.) And I frankly don't remember what "small mistake you pointed out" at Talk:Office 2013. Jeh (talk) 21:40, 9 February 2015 (UTC)

I've searched the MOSNUM archives for the word "diversity", which I feel should be reasonably specific to this issue. I find no support for your claims. There are only 18 page hits, with one or two hits per page. The closest I find to anything that comes close to supporting your position is this acknowledgement that diversity is one of WP's goals, so different styles are accommodated, here: "Wikipedia rightly strives for both consistency and diversity in usage. In some cases this means making a choice between competing usages; in other cases, the differing usages are accepted. So, the differences between British and American spelling is accommodated, while, sensibly, the rules state that individual articles should be internally consistent." (emphasis added - jeh) But (note the last part) I find no support in any of the archives here for differing styles within an article, nor any claim that this would promote an impression of diversity.

So, once again: Citation is Needed for your claim. I find nothing in MOSNUM's archives that arrived at the conclusion that "This manual of style's aim is to cultivate diversity in distinct areas of one article" (i.e. allowing one style in one "area" and another style in another), nor any that simply assumed that "aim" as a "given". In fact, I don't find that the concept has been mentioned previously at all! Might it be somewhere other than MOSNUM's archives? Jeh (talk) 23:23, 4 February 2015 (UTC)

Out of order, refuse to discuss. This is not a well advertised RFC. Also, the discussion should be in Wikipedia talk: Citing sources because the "Manual of Style" and its subsidiary pages do not control citaitons. There should also be a notice in Wikipedia talk: Manual of Style. Jc3s5h (talk) 16:53, 28 January 2015 (UTC)

@Jc3s5h: I didn't set this up as an RFC because I didn't think it was controversial. Not everything needs an RFC and some changes are made boldly, but I thought I would flag it here first to test the water. Anyway, it is best practice to discuss proposals before launching an RFC in order to settle appropriate wording. Now, do you have anything helpful to add on the issue or not? —sroc 💬 18:10, 28 January 2015 (UTC)

For the record, a stable version from 2013 stated:

Dates in article body text should all have the same format:

Correct: Julia ate a poisoned apple on 25 June 2005. She died three days later on 28 June.
Incorrect: Julia ate a poisoned apple on 25 June 2005. She died three days later on June 28.

Publication dates in article references should all have the same format. Although nearly any consistent style may be used, avoid all-numeric date formats other than YYYY-MM-DD.

For example, in the same article, write

Jones, J. (20 Sep 2008)
Smith, J. (Sep 2002)

but not

Jones, J. (20 Sep 2008)
Smith, J. (September 2002)

Access and archive dates in references should all have the same format – either the format used for publication dates, or YYYY-MM-DD.

For example, in the same article, write

Jones, J. (20 Sep 2008) ... Retrieved 5 Feb 2009.

or

Jones, J. (20 Sep 2008) ... Retrieved 2009-02-05.

but not

Jones, J. (20 Sep 2008) ... Retrieved 5 February 2009.
Jones, J. (20 Sep 2008) ... Retrieved Feb 5, 2009.
Jones, J. (20 Sep 2008) ... Retrieved February 5, 2009.

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

Date formats in quotations or titles should not be changed, even if this causes inconsistent formats in the same article.
It is acceptable to change other date formats in the same article to provide consistency, so long as those changes would otherwise be acceptable.

I believe that the words "nearly any consistent style may be used" suggested that the format for publication dates should be consistent with dates in the article (i.e., don't mix DMY and MDY) but this nuance was lost in various edits in 2014. I was a culprit in implementing this edit based on this discussion between User:EEng, User:Jc3s5h and myself – but I don't think that we realised that these edits could be used to support the idea of mixing DMY and MDY dates within the same article. This this edit by User:SMcCandlish further muddied the waters by changing some bad examples to good examples, perhaps misunderstanding why they were originally put there. With these and other edits, I think this section has gradually eroded to a point where it has departed from what was originally intended without any discussion on mixing DMY and MDY dates as far as I am aware, so it's time this was addressed. —sroc 💬 18:10, 28 January 2015 (UTC)

I think sroc is probably right that this section has unintentionally eroded, as he says. Certainly I, for one, didn't scrutinize all its changes the same way I did changes elsewhere on this page, because (a) its provisions are so contorted, the result of angry compromises years ago; (b) too many people are inexplicably dug in with one or another position; (c) I don't care sufficiently to make me deal with (a) and (b). So other than this comment, I think I'll stay out of this one. EEng (talk) 00:41, 29 January 2015 (UTC) An exception to what I just said: if somehow it comes out that an abortion like 2015, January 28 (see Jc3s5h's comment below) is acceptable, then I'll be reaching for my pistol. EEng (talk) 00:41, 29 January 2015 (UTC)

In the sense of preparing an RFC, I will make the following points:

If two guidelines contradict each other, the contradictory provisions in both are null and void. If WP:CITE and WP:MOSNUM disagree on dates in citations, there is no guidance at all on dates in citations.
WP:CITE says any consistent citation style may be used; that includes dates within citations, except all-numeric formats other than YYYY-MM-DD are forbidden.
There is good reason to be flexible with dates in citations, because editors using some tools such as Zotero to format dates for some citation style other than CS1 or the Citation template will want to be able to use the output from the tool without having to fix the dates.
There is at least one citation style, APA, which uses mdy in the body of articles but, for example, "2015, January 28" in the publication date field, so it will be easy to find a citation in the "Works cited" list when several publications by the same author are cited; the "Works cited" list is sorted alphabetically by author, and in case of ties, chronologically by citation date.
I am not aware of any recognized citation style that uses mdy in the article body and dmy in the citations (or vice-versa). The problem is, Wikipedia is the only publication I know of that uses MOS for the article body and "anything goes" for the citations. So if WP:Ties says to use mdy for the article body, but the citation style chosen uses dmy, then those two styles will be mixed. MLA seems to be an example of a style that uses dmy for dates. Jc3s5h (talk) 18:33, 28 January 2015 (UTC)

The (only) mixed style with "2015, January 28" is an interesting case, but that isn't an allowed format. YYYY-MM-DD is but neither would order so it seems mixed format wouldn't emulate any (paper) source. I thought the latter style was some compromise or for cultural diversity, but that has issues if the date is missing. comp.arch (talk) 14:22, 29 January 2015 UTC

For the horrible history of YYYY-MM-DD see Wikipedia:Manual of Style/Dates and numbers/Three proposals for change to MOSNUM, Wikipedia:Manual of Style/Dates and numbers/Date Linking RFC, and Wikipedia:Manual of Style/Dates and numbers/Three proposals for change to MOSNUM. Your statement "The (only) mixed style with '2015, January 28' is an interesting case, but that isn't an allowed format" is not correct, the guideline states "Publication dates in an article's citations should all use the same format, which may be...the format expected in the citation style being used...." Jc3s5h (talk) 15:46, 29 January 2015 (UTC)

Interesting.. but that APA style requires citations to be in order(?) Well, I haven't seen this used in Wikipedia, but I assume then as always citations must all be in the same format, and they must all be "manual" as WP would not order. I wander if you can change citation style if you do it for all of the article.. comp.arch (talk) 15:33, 30 January 2015 (UTC)

@Jc3s5h: Your argument is invalid: there can be no conflict with WP:CITE because MOS:DATEUNIFY specifically allows the use of any date format expected by a particular citation style. The issue here is that, even where there is no mandate for a particular date format in a particular citation style, it is still permissible for editors to choose any date format from the MOS:DATEFORMAT table. Allowing DMY dates in the body and MDY dates in references (or vice versa) without any mandate for a particular citation style contradicts the very principle of MOS:DATEUNIFY (i.e., keep the date formats within a given article consistent). —sroc 💬 13:26, 30 January 2015 (UTC)

The practical problem with the guideline is that it allows the use of the date format called for by the chosen citation style, but WP:CITEVAR allows any citation style, including an ad hoc style invented for a particular article. So if an article has a consistent style, the date format is allowed (unless it's something line 1/31/2015). I once proposed at WT:CITE that only recognized style guides like APA or MLA be used, but that was not adopted. Jc3s5h (talk) 15:18, 31 January 2015 (UTC)

What's the problem? (I'm just responding to the OP, not having read all the intervening material.) The accessdate field is not visible to the casual reader, so who cares what order it's in, or indeed if "20 September 2015" and "September 20, 2015" are both used in the article, for accessdates? Not me. The only firm rule should be to forbid ambiguous formats, such as "10-12-2014". As long a person can read it and understand it, it's fine. Herostratus (talk) 20:44, 29 January 2015 (UTC)

@Herostratus: The whole premise of "Consistency" (MOS:DATEUNIFY) is (spoiler alert) to promote consistency in date formats within individual articles. This is undermined if it permits using DMY in the body and MDY in references, or vice versa. —sroc 💬 13:26, 30 January 2015 (UTC)

Concur with sroc: It's pointless, confusing and a source of continual strife to permit multiple date formats the same article. Use one, end of issue. WP:ENGVAR already permits the format to be flexible to suit the English dialect begin used in the article, and that is sufficient. — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼ 05:59, 31 January 2015 (UTC)
Concur with sroc: Seeing mdy in one place in an article and dmy in another is like encountering spelling or grammar errors—or encountering speedbumps while driving: I can still get to where I'm going, but I'm distracted from my main task. This is why WP:ENGVAR states that although national styles are allowed, even encouraged, to differ from one article to another, that they should be consistent within each article. I have seen no good reason for making an exception of date styles in the references. Besides creating inconsistencies that, if noticed, call attention to themselves and therefore distract the reader, this exception increases editing workload (since editors have to "switch gears" between text and refs) and leads to editing disputes: editors who encounter these conflicting date styles often feel that the right thing is to change either the text or the references to make everything within an article consistent. The ensuing reverts, etc., help no one.

I wish to re-emphasize: My position is not at all contradictory to supporting different national styles in different articles. I simply think that this odd, counterintuitive exception that allows for conflicting date styles within an article - I'm not impressed by the "different parts of the article" claim - should be removed.

FleetCommand writes: "This manual of style's aim is to cultivate diversity in distinct areas of one article and facilitate different global styles to be employed without any arbitrary favoritism toward one. That's foundation's way of work: Neutral towards all arbitrary styles." My opinion is that this aim is perfectly well achieved by support of different national styles in different articles. If it isn't, then why stop at date styles in refs vs. text? Why do we ask that styles (including spelling, grammar, etc.) be consistent within an article? Why don't we allow "color" in one sentence and "colour" in the next? Wouldn't that do even more to show that Wikipedia cultivates diversity? No, it would just look amateurish, as if we are even more sorely in need of copyeditors than we already are. Consistent style within an article is the right thing. The exception that allows mdy in text and dmy in refs, or vice-versa, should be removed. Jeh (talk) 11:54, 31 January 2015 (UTC)

(Edit - added:) Also, WP is not "neutral to all arbitrary styles". There's a whole column full of "unacceptable" styles in the table (WP:BADDATEFORMAT). Jeh (talk) 12:59, 31 January 2015 (UTC)

Concur with sroc: we should return to the version from 2013. Changing from this did not result from discussion and consensus. Peter coxhead (talk) 11:59, 31 January 2015 (UTC)

I have to say that I prefer some aspects of the presentation (like the table "unacceptable date formats" table) in the current version to all of the text in that version. What I am supporting is to just change the current version to state that date style (and the rest of ENGVAR, for that matter) should be consistent between body text and refs. (Of course, the current allowance for yyyy-mm-dd in refs for brevity should be retained. References are obviously a terse format and it makes little sense to require a "prose-like" date like "February 23, 2012" in them.) One thing at a time - more likely to make progress that way. Jeh (talk) 12:59, 31 January 2015 (UTC)

So Jeh, do you plan to remove "the format expected in the citation style being used (however, all-numeric date formats other than yyyy-mm-dd must still be avoided)"? If you do WP:CITE and WP:MOSNUM will contradict each other, so they cancel out and there is no guidance. Jc3s5h (talk) 14:34, 31 January 2015 (UTC)

My impression is that adherence to any "citation style" gets pretty short shrift on WP and I really wasn't thinking about that. But it seems to me that simply dumping the second major bullet under "Consistency" and changing what was the third one to "Publication, access, and archive dates in an article's citations" will do. However I am not really interested in being distracted here with noodling over the exact wording changes that would be required. Jeh (talk) 23:35, 31 January 2015 (UTC)

Concur with sroc In fact I strongly agree. Why have different formats in the citations. In fact why can't we just get rid of YYYY-MM-DD date formats altogether. They prove nothing. No English speaking country uses them. And they only cause problems among editors.--JOJ ^Hutton 15:03, 31 January 2015 (UTC)

@Jojhutton: YYYY-MM-DD dates are a completely different issue. They are used by some English-speaking countries, e.g. official Canadian forms require them. There's no consensus for getting rid of them, but there does seem to be for avoiding MM DD, YYYY and DD MM YYYY in the same article, even between text and citations. Peter coxhead (talk) 17:15, 31 January 2015 (UTC)

I still say that we should get rid of them. I think they are pointless and have no possible use within any citation.--JOJ ^Hutton 17:31, 31 January 2015 (UTC)

YYYY-MM-DD is no more or no less pointful or useful than any other date format within a citation. You can say WP:I DON'T LIKE IT, but that's no reason to get rid of it. Others do like this format; it's brief and it puts the information in order of importance. Peter coxhead (talk) 17:39, 31 January 2015 (UTC)

You are right, I don't like it. I don't like anything that serves no purpose other to appease the wants and needs of a few editors. Most of us have moved on from this outdated date format in the English language world. It only survives due to the determination of a few editors who continue to stand by it for unknown reasons. I know its unpopular among some editors for me to hold this opinion, but facts are facts. There is no reason for the YMD date format any longer.--JOJ ^Hutton 17:48, 31 January 2015 (UTC)

I think you should have written "my opinion is my opinion", not "facts are facts". "Outdated?" Are you aware that it's an ISO standard? "No reason"? yyyy-mm-dd is brief, unambiguous, takes the same amount of width in a table regardless of the date value, sorts in date order by simple numeric sort (edit-added: i.e. easily sorted by eyeball), etc. All good things in citations. Do you want WP page histories (example: [39]) to use dmy or mdy too? Hey, maybe times there should likewise be written out as e.g. "15 minutes past 4 o'clock". No? Yeah, I didn't think so. yyyy-mm-dd is nothing more than the principles of the common numeric time format (all numbers, with most significant parts on the left) extended to cover dates. If you're fine with HH:MM but not yyyy-mm-dd, I think you're being inconsistent. Anyway, allowing use of yyyy-mm-dd is not what got this started. If you want to get rid of yyyy-mm-dd completely, that should be a separate proposal. Jeh (talk) 23:35, 31 January 2015 (UTC)

Well whatever we believe individually, it is "My Opinion" that there is no need to use YMD over the more highly accepted formats of DMY and MDY. And I would propose to get rid of yyyy-mm-dd altogether if I thought it would have any chance of passing. But, Again My Opinion, too many people still want to use it and will defend its use to the last, even when there is no logical reason to keep it around.--JOJ ^Hutton 23:45, 31 January 2015 (UTC)

@Jeh: You say "Do you want WP page histories ... to use dmy or mdy too?" - anybody who wants these formats for dates in page histories (also user contribs, watchlists, etc.) can set them very easily. For me, they are already in dmy, and that is because I have the third selection enabled at Preferences → Appearance → Date format, so I'm guessing that you have the fifth selection enabled there. The dates and times in page histories are stored as Unix times, and can be converted by software into almost anything that resembles a date, but the MediaWiki software restricts us to a choice of four (not five: the first of the five options, "No preference", is the same as the third, on English Wikipedia at least). --Redrose64 (talk) 23:51, 31 January 2015 (UTC)

@Redrose64: I'll be darned! I must have made that setting so long ago I forgot about it. Thanks. (I still think the comparison to "prose-style" times is apt!) Jeh (talk) 00:29, 1 February 2015 (UTC)

FWIW, JOJHutton, I prefer YYYY-MM-DD for citation formatting because it is clean, simple and easily understood by everyone. The use of the YYYY-MM-DD format does not cause problems among editors. Pedantic editors that needlessly try to force changes to "February 4, 2015" instead of "2015-02-04" are the ones that cause problems. Reso lute 23:57, 4 February 2015 (UTC)

I always thought it's a nice touch (when I remember) to use words (February 2, 1982 or Feb. 2, 1982) for the date in the citation proper, and YYYY-MM-DD for the access date, if any. Access dates are sort of modern and technical, and so is YYYY-MM-DD, so it fits. And that way the eye immediately distinguishes them. EEng (talk) 00:52, 10 February 2015 (UTC)

Section break

Given the raft of "Concur with sroc" responses above, I made the change which was promptly reverted (along with another subsequent edit) by FleetCommand. Despite being in the clear minority, FleetCommand has not been clear in explaining their opposition, to wit:

"This manual of style's aim is to cultivate diversity in distinct areas of one article and facilitate different global styles to be employed without any arbitrary favoritism toward one." Despite repeated requests, FleetCommand has not provided any evidence to support this claim. In any event, the proposed change do not seek to favour any particular style but to promote consistency, which is, after all, the point of a section headed "Consistency" which explicitly states that dates within an article "should all use the same format".
"A person convert 87% of dates in one article from DMY to MDY, the opposing person does not receive fair treatment, someone offers a compromise based of the current reading of MOS and to defuse that compromise, you start this topic. You want elucidation: I have one word for you: WP:GAME!" I do not advocate for either DMY or MDY, but for consistency. I attempted, in good faith, to fix what seems to me (and others) to be a glaring error in the current wording of MOS. FleetCommand has apparently assumed the worst of me instead of focussing on the issue at hand, namely, closing a loophole in MOS which is being use to support illogical arguments elsewhere.
"Why [change the dates in refs to match those in the article]? The refs were more numerous. Their style was the prevalent one. With or without sroc's proposal, changing 87% of dates in one article to another format is a violation the MOS..." Actually, MOS says: "When a citation style does not expect differing date formats, it is permissible to normalize publication dates to the article body text date format, and/or access/archive dates to either, with date consistency being preferred."
"As for my own belief, you shouldn't have asked it because you don't like it but here it is: I believe all of Wikipedia must adhere to one unified date style and let everyone customize its own view." Feel free to propose it, but know that it is fraught with difficulty and was removed as a feature many years ago. In the meantime, it's no justification to oppose a perfectly good change to MOS or to revert it when it is supported by the majority. —sroc 💬 03:22, 10 February 2015 (UTC)

Hey, sroc, cut out the fo-cussing. It's rude. EEng (talk) 05:33, 10 February 2015 (UTC)

Sorry, EEng, I genuinely don't know what you mean. I assume the hyphen in "fo-cussing" is a clever play on words but it's gone over my head. —sroc 💬 12:28, 10 February 2015 (UTC)

You wrote of "focussing" on something. See... focussing ... fo-cussing ... cussing? A small joke perhaps, but mine own. EEng (talk) 12:33, 10 February 2015 (UTC)

So you weren't really reprimanding me as I'd originally thought? I didn't think I was doing any faux cussing, let alone any genuine cussing. —sroc 💬 16:29, 10 February 2015 (UTC)

No, I was faux-reprimanding you. EEng (talk) 16:43, 10 February 2015 (UTC)

Yes, I see the fauxny side now. —sroc 💬 16:51, 10 February 2015 (UTC)

@Sroc:

That's not my reason for revert. The reason for the revert was the ill-publicized nature of the discussion that makes it not binding. But let's have a non-bureaucratic compromise: Let's ask a non-involved admin to close this discussion and explicitly spell out the compromise. That way, the closure is binding. I am sure he or she will close it in your favor. Method of contacting this admin is unimportant: Talk page, IRC, IM, email or even the crowded Request for Closure noticeboard; all are okay to me.
Very well. Let's accept that you are acting in good faith and are attempting to eliminate a glaring error, shall we?
Double-standard alert! If your proposal is to be accepted, the "permissible to normalize" part of the sentence that you quoted may only and only be taken to mean: (a) contracted MDY dates become expanded MDY dates, (b) contracted DMY dates become expanded DMY dates, and (c) YYYY-MM-DD dates become expanded DMY or MDY dates depending on what the article's prevalent date format is. The quotation must not be taken to endorse changing the prevalent date format of the article, which WP:DATESRET opposes.
Duly noted.

Fleet Command (talk) 11:24, 13 February 2015 (UTC)

Comment - Sorry that I am late to the party, having just found this discussion while reviewing my watch-listed MOS pages. Late though I may be, I feel the need to express my opposition to any form of MOS sanction, permission, approval or prescriptive mandate in favor of multiple date formats within the same article, and especially within individual footnotes. This is a long-time pet peeve of mine, and the careless or intentional insertion of ISO formatted dates for linked footnote retrieval dates is an ongoing aggravation to those many editors who work to maintain stylistic consistency within our articles. My primary objection? The use of multiple date formats in the same article makes Wikipedia look like the Gang Who Couldn't Shoot Straight. It's one thing to vary date formats from article to article on the basis of national ENGVAR usage, or common use within a given subject area, but to insert machine language date formats (e.g., 2014-01-31) side by side with American MDY (e.g., January 31, 2014) or British/Commonwealth/military DMY dates (e.g., 31 January 2014) in the same article -- and especially within the same footnote -- looks amateurishly inconsistent to our readers. Let's please put an end to this. There is no valid reason not to be internally consistent in our use of date formats within any given article (and individual footnotes), and, contrary to assertions above, that is not inconsistent with the existing language of WP:DATESRET. Dirtlawyer1 (talk) 13:34, 13 February 2015 (UTC)

@FleetCommand: Thanks for your reply. Regarding the first point, as I stated in my reply to Jc3s5h above, I didn't think this was so controversial as to require more widespread publicity, but feel free to request third-party closure. Regarding the third point, I always understood WP:DATESRET to be about the selection of DMY or MDY format in individual articles (and they are the only acceptable date formats for general use) to avoid reversions back and forth based purely on individuals' preference, so the "permissible to normalize" comment does not conflict with this rationale as I see it. —sroc 💬 13:41, 13 February 2015 (UTC)

As far as citations go, it doesn't matter what WP:DATESRET calls for; WP:CITEVAR calls for a citation style that has been established for a particular article to be retained. I think that should be modified a bit; I would only retain recognized styles such as CS1, CS2, MLA, APA, Chicago, etc. Also, I would do away with YYYY-MM-DD in citations. But in a number of very long RFCs in the past my preferences did not win out. Jc3s5h (talk) 15:04, 13 February 2015 (UTC)

@Dirtlawyer1: Okay, I disambiguated my "it" by replacing it with "the quotation" and changing my sentence to negative. Does it address your concern? Fleet Command (talk) 08:51, 15 February 2015 (UTC)

@Sorc: I can't contact any admin. Let's just say they hate me. (And let's just say you wouldn't blame them if you knew why.) Also, there is a form of drama I've seen before: I do a change, and cite the MOS; someone says "oh, look, this change is made on such and such date and the editor who made it twice said he wasn't looking for consensus", and they reject authenticity of the change. Fleet Command (talk) 09:00, 15 February 2015 (UTC)

Resolution

Absent further dissent, would anyone care to implement the revision? EEng? —sroc 💬 09:49, 27 February 2015 (UTC)

So my reward for having the sense to stay completely out of this is to be dragged in as the fall guy at the very end? What have I ever done to you??? But OK. Honestly I've paid no attention to this at all I'll need a bit to review. And I'll do it only under one condition: you have to join the discussion at #Arbor-treeish_break. EEng (talk) 10:24, 27 February 2015 (UTC)

Done. I was simply hoping someone would do the sensible thing and I instantly thought of you! —sroc 💬 11:57, 27 February 2015 (UTC)

I'm now about 1/2way through the debate. There was a call at some point for an admin to close -- you do all realize I'm not an admin, right? -- though you may be forgiven for mistaking me for one, given my noble bearing. I hope that doesn't impair my qualification for the important role I've been assigned here. EEng (talk) 19:40, 27 February 2015 (UTC)

(A) OK, am I wrong in concluding that what's been settled on is a reinstall of this edit [40] (and BTW I support doing so)?
(B) If so, I have one worry, to wit that one of the "prongs" for Publication Dates --

an abbreviated format from the "Acceptable date formats" table, provided the day and month elements are in the same order

-- is ambiguous. Do we mean --

(B1) ...provided the day and month elements are in the same order in all publication dates (but not necessarily as in dates in the article body)

or

(B2) ...provided the day and month elements are in the same order as in dates in the article body

--? I'm guessing it's B2.

EEng (talk) 19:15, 1 March 2015 (UTC)

BUMP???? HELLO??? EEng (talk) 17:35, 4 March 2015 (UTC)

Sorry, I didn't see this earlier. (A) Yes, that's right. (B) Yes, (B2) is the intended interpretation. The point is that if the article body uses DMY dates, the publication dates should too (whether in full or abbreviated form), and likewise with MDY dates. We don't want a mix of DMY dates in the body and MDY dates in references, or vice versa. —sroc 💬 15:18, 5 March 2015 (UTC)

I concur with sroc. — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼ 22:39, 5 March 2015 (UTC)

Looking at the edit advocated by EEng, I noticed a phrase that was introduced by User:SMcCandlish on 25 October 2014 in a flurry of other edits, and didn't receive any discussion:

When a citation style does not expect differing date formats, it is permissible to normalize publication dates to the article body text date format, and/or access/archive dates to either, with date consistency being preferred.

I am concerned this could be interpreted to mean that, for example, if the body and publication dates are in the MDY format, the access dates are in the YYYY-MM-DD format, and the archive dates are in the YYYY-MM-DD format, it is permissible to change all the access and archive dates to the MDY format, "date consistency being preferred". I have no objection to chucking YYYY-MM-DD entirely, but I don't believe this passage has genuine consensus. Jc3s5h (talk) 15:48, 5 March 2015 (UTC)

This is most how our policy and guideline pages are authored – someone notes a best practice already in play, and records it. Four months is certainly long enough for objections to have been raised, especially given that this is one of the most-watchlisted projectpages on the whole system, and subject to a greater degree of nit-picking scrutiny even than other MOS pages. Even you don't appear to be objecting, but simply noting that someone added it, and expressing a concern that doesn't seem to raise an actual issue. Does anyone really dispute the idea that we prefer a single date format in articles over two conflicting ones, and absolutely over three of them? If you agree we don't have any use for YYYY-MM-DD dates, then you appear to be making an argument for people intentionally using DMY format in the references section of an MDY article. Why would anyone do that? Just because they're American and like them that way and want some corner in which to force that format on the rest of the world, no matter what? [I write this as an American most familiar with MDY, by the way.] Does anyone seriously expect to see the date format reverse itself confusingly just because we're looking in the "References" or "Notes" section instead of some other section? And could they possibly expect to see access/archive dates differ from publication dates, in turn differing from dates in main article text? Of course not. It logically, necessarily follows that a) if it's permissible to normalize publication dates to the same format as the dates in the main text, and b) it's desirable to have only one date format in the references section, then c) in the course of normalizing publication dates to the date format in article, the access/archive dates would be given the same format as well. The only other alternative would be to force publication dates to have the different (e.g. YYYY-MM-DD) format that someone decided to use for access dates, on the supposition that random editor preference with regard to access/archive dates is the #1 concern of date formats on Wikipedia, consistency of dates within the refs section secondary, and consistency within the article doesn't even rate. That's completely backward. — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼ 22:39, 5 March 2015 (UTC)

I'm completely serious. Some editors are in love with the YYYY-MM-DD, and will raise a ruckus if anyone changes access dates to DMY even though the body of the article and the publication dates use DMY. If I remember correctly, one article I tried to change to all-DMY was Anno Domini, but I got reverted. Why don't you go over there and change the dates, citing the passage you put in MOSNUM, and let us know if your change stuck (that is, if you don't get blocked). Jc3s5h (talk) 23:38, 5 March 2015 (UTC)

Nothing at blocking policy would appear to apply to such a situation, so I'm having a hard time taking this seriously even if you are "completely serious". I'm sorry you got reverted, at one article, by someone who doesn't pay attention to MOS. [shrug] I normalize dates in the refs section to the format used in the article on a fairly regular basis, without controversy, as part of other generally WP:ENGVAR-related cleanup (e.g. fixing later-added "-ize"/"-ise" spellings to match the rest of the text, and so forth). The only time I can recall ever being reverted on such a date change, it was by someone who was reflexively revertwarring against virtually every edit I ever made to any of the articles that "belong" to "his" wikiproject; it was personal antagonism, not a principled position about date formatting. I certainly don't encounter ISO date reversion on a regular basis, and if you watch WT:MOSNUM, you know that the format is falling increasingly into disfavo[u]r here. I think you ran into a fluke situation. :-) — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼ 12:53, 9 March 2015 (UTC)

I'll just sit quietly while y'all work this out. sroc? EEng (talk) 01:20, 6 March 2015 (UTC)

<bump again> sroc??? EEng (talk) 12:08, 21 March 2015 (UTC)

@EEng: Sorry, I don't recall getting the previous ping. As far as I'm concerned, the issue at hand about (A) and (B2) has been settled and should be implemented; will you do this? The issue regarding YYYY-MM-DD is altogether separate and I don't have a view on it; it certainly need not delay finalising the original issue, I don't think. —sroc 💬 13:51, 21 March 2015 (UTC)

OK, done. Please check me. EEng (talk) 01:24, 26 March 2015 (UTC)

At this late time, I would make an observation. The citation tool as part of the editing tool inserts dates (eg access dates) in a default DMY format. References are not part of the 'readable text'. Date formats used in references should be consistent within the references/citation sections as should dates within the body of text. However, I see no advantage or benefit in suggesting that the formats used in both the body of text and in references should be identical - particularly if the inconsistency arises from using the WP tool for inserting references. There is something inherently wrong in creating or having a 'rule' with which WP systems are intrinsically incompatible. Cinderella157 (talk) 21:44, 29 March 2015 (UTC)

Exception for "quotations or titles"

I just wanted to explain a couple of recent edits for the record.

The following text was added at the beginning of Wikipedia:Manual of Style/Dates and numbers § Dates, months and years by J. Johnson on 13 May 2014:

These requirements do not apply to dates in quotations or titles. Special rules apply to citations; see Wikipedia:Citing sources#Citation style.

Firstly, this addition was redundant because:
- the first part was already covered under § Quotations, titles, etc., which says:
  Quotations, titles of books and articles, and similar "imported" text should be faithfully reproduced, even if they employ formats or units inconsistent with these guidelines or with other formats in the same article. If necessary, clarify via [bracketed interpolation], article text, or footnotes.
  It has said this for some time. There is also an embedded link to that section in the following table heading:
  Unacceptable date formats (except in external titles and quotes)
- the second part was already covered under § Consistency, which states that the publication, access and archive dates may follow the citation style being used.
Secondly, and more worryingly, the apparent exemption for "titles" could be construed as meaning that date format guidelines do not apply to Wikipedia article titles (e.g., it would be acceptable to use 1982/83 South Pacific cyclone season as an article title rather than 1982–83 South Pacific cyclone season). I do not believe that there is consensus for this and understand that the exemption is meant to apply to external titles (e.g., titles of books, referenced journal articles, etc.) but not to Wikipedia article titles. This interpretation is consistent with the sections quoted in the first bullet point above.

I have therefore deleted the redundant and potentially confusing wording. —sroc 💬 14:24, 29 March 2015 (UTC)

You have therefore increased ambiguity and likely confusion. My somewhat vague recollection is that the part you took out at MOS:DATEFORMAT was added because someone arguing for a specific format in a citation had gone straight to that section, and to § Consistency (aka MOS:DATEUNIFY). Note that the latter does not clearly state that while dates in citations may follow the style of the article, they are not required to do so, and that special rules may apply. If you feel that there was some ambiguity regarding titles feel free to augment that part, but taking out the whole line makes it less clear that "unification" is not compulsory on citations. (Which leads to arguments on interpretaton, such as the one in which you are currently involved.) I suggest you revert. ~ J. Johnson (JJ) (talk) 20:52, 29 March 2015 (UTC)

What is ambiguous about DATEUNIFY?

Publication dates in an article's citations should all use the same format, which may be:

the format used in the article body text,
an abbreviated format from the "Acceptable date formats" table, provided the day and month elements are in the same order as in dates in the article body, or
the format expected in the citation style being used (however, all-numeric date formats other than yyyy-mm-dd must still be avoided).

...

Access and archive dates in an article's citations should all use the same format, which may be:

the format used for publication dates in the article,
the format expected in the citation style adopted in the article, or
yyyy-mm-dd

...

If this is somehow ambiguous, then the solution is to amend this section to resolve the ambiguity, not to add a vague note to DATEFORMAT.

In any case, the line "These requirements do not apply to dates in quotations or titles" is inaccurate and should be omitted. —sroc 💬 07:28, 30 March 2015 (UTC)

On the contrary. The MOS clearly and unambiguously states (as you just quoted, above) that quotations "should be faithfully reproduced". I see no exceptions for quoted dates. That you can find some ambiguity in "title" (oh, this could refer to Wikipedia titles!) in no way invalidates the requirement for faithfully reproducing "imported" text, including titles of books and articles. If you find this alleged ambiguity too great to endure then I would quote your own suggestion: the solution is to amend. ~ J. Johnson (JJ) (talk) 20:56, 30 March 2015 (UTC)

I don't follow. MOSNUM already says "Quotations, titles of books and articles, and similar 'imported' text should be faithfully reproduced, even if they employ formats or units inconsistent with these guidelines ..." and this applies to all aspects of date and number formatting, so why do we need to reiterate this under DATEFORMAT specifically? We don't repeat it at MOSTIME or NUMERAL or UNIT or any other section of the guideline. Once is enough. —sroc 💬 01:23, 31 March 2015 (UTC)

Re: "My somewhat vague recollection is that the part you took out at MOS:DATEFORMAT was added because someone arguing for a specific format in a citation had gone straight to that section ..." The solution in that case would be to point the editor to "Quotations, titles, etc." which provides the exceptions to MOSNUM as a whole, not to amend the specific section that an editor was misguidedly following. —sroc 💬 01:27, 31 March 2015 (UTC)

In the first place, you are confusing two cases: dates in quotations or titles, and dates in citations. Second, what you deleted did (at least in regard of citations) exactly what you suggest: it did not describe the special handling, it simply alerted the reader to the existence of exceptions, and pointed the reader to the appropriate section. The lack of such a link for quotations was readily amendable, and did not warrant removing that material. ~ J. Johnson (JJ) (talk) 23:18, 1 April 2015 (UTC)

Military date format in biographical articles

This discussion is moved here from Talk:Audie Murphy#Date format (again). Apologies if this means re-hashing things again. In my opinion, a single sentence in the guidelines – "articles on the modern US military use day before month, in accordance with military usage" – seems to be widely over-interpreted. If you look at wikt:military#Noun, it doesn't mention persons. In particular, the word "modern" indicates to me that it is not about persons. What is a modern person?

So what do we do with persons, whether or not they (like Murphy) were more than soldiers? Was this single sentence really supposed to override WP:STRONGNAT when it comes to biographical articles on American people. I think not.

HandsomeFella (talk) 14:54, 18 March 2015 (UTC)

There must be a fundamental misunderstanding of this thing with the "military dates", which applies to "modern U.S. military". So Murphy's a "modern U.S. military"?

This rule applies to U.S. military organizations, battles, history, ships, aircraft, equipment, decorations, etc, but of course it does not apply to people. Millions of Americans have some sort of connection to the military. Some have been officers, some have just been recruits. Many have gained notability and have an article here. Should the fact that there is some sort of connection to the military, if ever so slight, override the standard mdy date format? Of course not. I think this single statement in the guidelines is widely over-interpreted.

In contrast, Dwight D. Eisenhower does not appear to be a "modern U.S. military", despite the fact that he's one of very few five-star generals in the US Army, and despite that his military career is what made him notable, and made it possible for him to become president.

To me it's pretty obvious that the guideline does not apply to people.

HandsomeFella (talk) 08:58, 17 March 2015 (UTC)

Sorry I was under the same impression and was also basing the date format change on Dwight D. Eisenhower, Charles Pelot Summerall, and Robert E. Lee it seems like this person had a well established notable life outside of the military, so I wasn't sure where the line is drawn. It does seem though that MM/DD/YYYY is more widely used, perhaps clarification is needed. Valoem ^talk ^contrib 19:35, 17 March 2015 (UTC)

Well, read threads above. And this was hammered out on the WP Military History, in various reviews and their talk page. I can't tell you how tired I am of this. You can't have two styles in an article. This is what the reviewers and others at the Military History project decided it should be. You are more than welcome to go over to their talk page and post all about it. — Maile (talk) 19:45, 17 March 2015 (UTC)

Agreed, just want to make sure there is a single format, sorry about the confusion I hope it was understandable. Valoem ^talk ^contrib 19:58, 17 March 2015 (UTC)

I'll tell you where the line is drawn. "Modern US military" does apply to people, and by "modern", the 20th and 21st centuries are meant per WP:STRONGNAT. Any attempt to change the format of any article without strong consensus violates WP:DATERET and will result in a trip to WP:ANI. ArbCom held in the Infoxes case that gnomes can be blocked or banned for attempting to override the article creators. Hawkeye7 (talk) 19:45, 17 March 2015 (UTC)

Okay, take a look at Wesley Clark and John McCain. They were born in 1944 and 1936 respectively. Are they politicians or are they military persons? Are they "modern"? Well, their articles are in the mdy format, and rightly so in my opinion.

Maile, consider this: if your interpretation is correct, howcome you end up in so many discussions that "[you] can't tell [us] how tied [you are] of this"? Does this possibly indicate that your interpretation is wrong? Just think about it. I mean, we have a host of articles with the dmy format, and a host of articles with the mdy format (such as those on Clark and McCain). Is this fortunate? Most of the former articles (hopefully) have some kind of military connection, but so do many articles of the latter category.

You don't say about people that they're "modern". Are you "modern", Maile?

HandsomeFella (talk) 21:54, 17 March 2015 (UTC)

The term "modern" here refers to modern history. It's a technical term. Hawkeye7 (talk) 01:08, 18 March 2015 (UTC)

End of copied discussion. Pinging @Valoem:, @Maile66:, @Hawkeye7:.

@Hawkeye7: care to elaborate on that post?

HandsomeFella (talk) 14:54, 18 March 2015 (UTC)

This talk thread is challenging the date style vetted by members of the WikiProject Military history in the FAC process. There is a thread on the Audie Murphy talk page above this one that says:

U.S Army manual on dates, capitalization, etc.

In regards to some recent well-meaning edits and reverts, please see the U.S. Army CMH Style Guide 2011.pdf.

Regarding dates, section 6.1 specifically says "Use the military day-month-year dating system (without punctuation)."
Regarding capitalization of ranks, section 1.47 specifically says "Lowercase military titles when standing alone, or when following a name."

Hope this information has been helpful. — Maile (talk) 22:15, 14 October 2014 (UTC)

One of the examples given in the U. S. Army manual as proper date usage is:

"On 1 August 2003 General Schoomaker succeeded General Shinseki as chief of staff."

The Army applies it to people. — Maile (talk) 19:35, 18 March 2015 (UTC)

We're not the Army. Needless to say, the Army manual applies in Army paperwork, not (necessarily) here. And what about the other people who have served notably in the military, like Clark and McCain above, or John Kerry, John F. Kennedy, George W. Bush, James Stewart and Elvis Presley, just to mention a few? (Stewart was even a brigadier-general.)

Doesn't the Army manual apply for them? (Answer: yes, in Army paperwork it probably did, but not here.)

HandsomeFella (talk) 23:09, 18 March 2015 (UTC)

Is someone proposing a change to MOS/Dates and Numbers? If not, this discussion doesn't belong here. If this is a chronic problem in multiple articles, then come back here with a proposed change to MOS which would clear up that problem. EEng (talk) 00:58, 19 March 2015 (UTC)

I think that this discussion is an attempt to gain guidance and find out how WP:STRONGNAT applies to those with military ties. IMHO, those who have a military connection during a period when the DMY format is used (by the service they have the military connection with), if it is already common practice at that article to use DMY, than the DMY format should be used through out. Otherwise, keeping with that persons nationality common date format should suffice.--RightCowLeftCoast (talk) 07:02, 19 March 2015 (UTC)

I understand, but Talk:MOS isn't a court of appeals for interpretation, because what happens is that you get a lot of people with only a passing interest and familiarity drifting to a conclusion which then takes on a kind of precedent value. Again I say: work it out on the Talk of the particular article involved, where (presumably) editors are steeped in all the facts and circumstances that might come to bear on the issue. If this has been a recurring problem on many articles, then bring that here, with links to all those discussions, with a proposal for how a change to MOS can save time and trouble. EEng (talk) 14:25, 19 March 2015 (UTC)

EEng...You are absolutely correct. On all of it. At the top where this thread is copied from the AM talk page, it had been suggested to take this issue to the Military History project, where the Audie Murphy series had been through the reviews. A little background. Neither Hawkeye7 nor I decided on DMY. It happened in the various review scrutinies. The DMY was questioned later, and it was dealt with on the MH project talk page. And there it stayed until Audie Murphy was March 17 TFA. On that date, the above editor made his one and only edit to the Audie Murphy article, and it was not about the dates. His talk page posts started that date. He got his answer, as you can see above. Not what he wanted. Putting it here seems like gaming. — Maile (talk) 23:40, 19 March 2015 (UTC)

Of course I'm right on everything. You're just realizing that now? Jeesh! Anyway, we don't need to speculate on anyone's motives -- important thing is the discussion belongs back on the article Talk. EEng (talk) 00:44, 20 March 2015 (UTC)

What can I say - when you're good, you're really good. — Maile (talk) 13:19, 20 March 2015 (UTC)

@Maile66: accusing me of gaming for starting a discussion here, which struck me as a wider and more appropriate forum (I didn't think of WP:MILHIST) than the AM talkpage, is totally unfair and uncalled for, and fairly close to a personal attack. When you have re-read WP:AGF, I hope you will retract it. HandsomeFella (talk) 12:48, 20 March 2015 (UTC)

I think Maile now understands that you are just looking for clarity. We don't need to go further down the road re motives. EEng (talk) 15:01, 20 March 2015 (UTC)

I don't care if the subject individual was raised in a military family and spent their entire adult life in the military; the article about them is not "on the modern US military". Presumably the article describes aspects of their life that had nothing to do with the military, including childhood, marriage, offspring, and retirement. George Patton himself had interests outside his military career. The guideline is not "over-interpreted", it is misinterpreted. ―Mandruss ☎ 07:35, 19 March 2015 (UTC)

If someone's biography was being written for a particular audience (say, perhaps, a military community, or even a military publication) use of "military" style dates would seem appropriate. But Wikipedia is written for a general audience, so should use a general style of dates. But here's the rub: is the issue here really about military style, or just MDY vs. DMY? In the present case, that Murphy was in the military does not meet the "military usage" exception of MOS#Choice of format, and seems irrelevant. The real issue seems to be MDY vs. DMY. And while the former is more traditional in the U.S., I note that the latter has become acceptable and even common in recent years, and presents no challenges of novelty. If an article is consistent, either format should be acceptable. ~ J. Johnson (JJ) (talk) 18:22, 19 March 2015 (UTC)

But what is often found is that those articles written by those who follow DMY are often changed to MDY, overriding local consensus, claiming WP:STRONGNAT. If that is the case that is in opposition to MOS#Choice of format.--RightCowLeftCoast (talk) 21:58, 19 March 2015 (UTC)

I agree. This is but one of several matters where some editors have a monomanic focus on blindly applying some alleged "rule", quite regardless of any exceptions or local consensus, which merges into WP:IDHT. Seems like we ought to have a catchy tag for this. ~ J. Johnson (JJ) (talk) 00:02, 21 March 2015 (UTC)

Clarification requested

Just for sake of clarity: could someone please explain, as simple as possible, how the guideline works? It apparently prescribes that the article on Audie Murphy should use dmy dates, while that on James Stewart should use mdy dates. Both men had notable military careers. Both men had notable acting careers. They both lived in the 20th century. Where's the difference?

Thank you. HandsomeFella (talk) 07:13, 22 March 2015 (UTC)

Don't change the existing date format just because you think you can. The chances are it will be percieved as contentious. Cinderella157 (talk) 09:54, 22 March 2015 (UTC)

I'm not changing the dates – I'm discussing the matter. HandsomeFella (talk) 10:10, 22 March 2015 (UTC)

I was not suggesting you were. I was offering a clarification, as requested. Cinderella157 (talk) 10:45, 22 March 2015 (UTC)

To be honest, telling me not to change the date format wasn't really much of a clarification of the guideline, was it? HandsomeFella (talk) 10:51, 22 March 2015 (UTC)

For the record, the relevant provision is Wikipedia:Manual of Style/Dates and numbers § Strong national ties to a topic (MOS:DATETIES):

Sometimes the customary format differs from the usual national one: for example, articles on the modern US military use day before month, in accordance with military usage.

This obviously refers to articles about US military topics in modern history. This issue is whether this should also apply to articles about people from the US military. I would think that, logically, this would apply to people who are primarily notable for being in the US military (whose articles would be mostly about that part of their life—e.g., Wesley Clark, Richard A. Cody, Martin Dempsey) but not to people who are primarily notable for other reasons who happen to have been in the military during their lives (e.g., George W. Bush, Elvis Presley, James Stewart). I would have thought that was obvious, but if that's not the case, then perhaps a clarification is in order in the MOS to avoid confusion and inadvertent "corrections" by well-meaning editors.

I'm not sure when DMY format because commonplace in the US military to know whether this should apply to older cases (e.g., probably Audie Murphy, served 1942–1966, but presumably not Robert E. Lee, served 1829–1865), so if this is a cause for confusion, perhaps "modern" should be clarified as well. —sroc 💬 11:24, 22 March 2015 (UTC)

@HandsomeFella: By way of addressing your query, look at the difference in the opening of each article:

Audie Leon Murphy (20 June 1925 – 28 May 1971) was one of the most decorated American combat soldiers of World War II, receiving every military combat award for valor available from the U.S. Army, as well as French and Belgian awards for heroism. At the age of 19, Murphy received the Medal of Honor after single-handedly holding off an entire company of German soldiers for an hour at the Colmar Pocket in France in January 1945, then leading a successful counterattack while wounded and out of ammunition.
James Maitland "Jimmy" Stewart (May 20, 1908 – July 2, 1997) was an American film and stage actor, known for his distinctive drawl voice and down-to-earth persona. Over the course of his career, he starred in many films widely considered classics. He was known for portraying the average American middle class man, with everyday life struggles.

The former is primarily described for his military career, the latter is primarily described as an actor. —sroc 💬 11:35, 22 March 2015 (UTC)

To further clarify my previous, both DMY and MDY are acceptable formats. From the previous discussion, MDY is not exclusively used in the US (excluding US military usage). Provided an article is consistent in date format usage, there is no good reason to change the date format used - I perceive it is just asking for an argument and might be perceived as Gaming the system. Cinderella157 (talk) 11:44, 22 March 2015 (UTC)

But the guideline says "articles on the modern US military use day before month"; it doesn't say "may use" or "sometimes use", nor does it say "may use either DMY or MDY format provided the same format is used consistently within each article". If an article on modern US military matters uses MDY, MOS:DATETIES would call for a change. MOS:DATERET would not apply, as it says:

If an article has evolved using predominantly one format, the whole article should conform to it, unless there are reasons for changing it based on strong national ties to the topic or consensus on the article's talk page.

—sroc 💬 11:51, 22 March 2015 (UTC)

Cinderella157, please do not make personal attacks. It must be possible to question and discuss fuzzy guidlines without being accused of gaming the system. If you have no good arguments, please abstain from commenting, or at least do not make personal attacks. Sroc, thanks for your thorough and valuable input. I will reply later. HandsomeFella (talk) 12:01, 22 March 2015 (UTC)

Wise counsel from sroc as usual, and everyone cool the accusations and warnings. There's no certainty to be found here on this page (some things on WP "just are") and the discussion should continue back on the talk page of the article in question, if at all. EEng (talk) 12:06, 22 March 2015 (UTC) sroc, I'll be attending to the DATEUNIFY matter later today.

I do not perceive that any 'personal attack' was implied nor was it my intent to do so. I was simply making an observation and answering a question that was asked. From the above discussion, it is apparent that it is increasingly less justifiable to assert MDY on the basis of 'strong national ties' and, even asserting this as a justification (if it were a universal national standard) might be perceived as sporting for an argument, particularly if it were not discussed first. Given that both MDY and DMY are acceptable formats for WP, if the article is consistent, I perceive no reasonable reason to change from one to the other (myself). However, I reiterate my former statement: I perceive that to do so is just asking for an argument and might be perceived as Gaming the system. Cinderella157 (talk) 12:27, 22 March 2015 (UTC)

Will you please stop saying that? It serves no purpose. I see no evidence that HF is acting in anything but good faith. Let us please have an end to this thread now. EEng (talk) 12:49, 22 March 2015 (UTC)

Did I say he was? No! Did I imply he was? No! I simply stated that this is how it 'might' be perceived. I also stated why. A personal attack (as some here are prone to) definitely serves no good purpose. An explanation, which explains how such actions might be perceived does serve a good purpose. I did not attribute these intents to HF and nor was it my intent to do so. I made a plain statement of how such action 'might' be perceived and I offered same in 'good faith' as a response to the question asked. To be very specific, I stated: "I do not perceive that any 'personal attack' was implied nor was it my intent to do so." Is it not reasonable advice that has been offered? 13:15, 22 March 2015 (UTC)

The only "advice" offered by you here so far has been telling me not to change the dates – which I didn't – and claiming that the discussion might be perceived as gaming tactics. Valuable advise indeed. If you have nothing more, please stay off the thread. HandsomeFella (talk) 13:27, 22 March 2015 (UTC)

End this thread now, back to article development End this thread now, back to article development End this thread now, back to article development End this thread now, back to article development End this thread now, back to article development End this thread now, back to article development End this thread now, back to article development EEng (talk) 13:29, 22 March 2015 (UTC)

I'm sorry if I misunderstood your request for clarification on 'how the guideline works'. I have provided a pragmatic response, whereas, you have wanted to make sense of the guideline. My mistake!! In reality, there are probably diminishing reasons to claim 'strong national ties' for using MDY over DMY. There is little reason for choosing one over the other (for the US, though elsewhere, DMY is more prevalent) and to some extent, the choice rests with the first to make an edit using dates - unless there is a discussed consensus to change. The question to ask though, is why change? For non-US articles, it is clearer to change to DMY, because this is more universal. But the underlying question should be, why change at all? The guideline is perhaps potentially contentious. It could be made clearer with perhaps stronger advice not to change. But there are those that would resist any attempt to clarify or improve the guideline without at least ten rounds bare knuckles in the minor leagues (article talk pages) before even suggesting a change here. Good luck! Cinderella157 (talk) 14:16, 22 March 2015 (UTC)

Okay, there was a misunderstanding. No harm done. I may have used a somewhat ambiguous wording. HandsomeFella (talk) 08:23, 23 March 2015 (UTC)

@Cinderella157: I haven't interpreted any personal attacks, but your claim that following MOS may be perceived as "gaming the system" is unjustified. The objective facts are:

MOS:DATEFORMAT states that both DMY and MDY dates are acceptable in general.
MOS:DATETIES states:
- "Articles on topics with strong ties to a particular English-speaking country should generally use the more common date format for that nation. For the United States, this is month before day; for most others, it is day before month. Articles related to Canada may use either format consistently." Thus, if an article is about an American person or organisation, MDY date would be used (and any that use DMY should be changed). I have not seen any evidence that both MDY and DMY dates are commonplace and acceptable in the US in general use;
- "Sometimes the customary format differs from the usual national one: for example, articles on the modern US military use day before month, in accordance with military usage." Thus, as an exception to the usual MDY format for articles with strong ties to the US, articles relating to the US military should use DMY instead (and any that use MDY should be changed).
MOS:DATERET states that the existing date format should not be changed "unless there is reason to change it based on strong national ties to the topic or consensus on the article's talk page." Thus, if DATETIES applies, DATERET doesn't.

To say "both DMY and MDY are acceptable formats" is an over-simplification which ignores DATETIES. The only issue here is clarifying the cases in which DATETIES applies. Your argument seems to be don't bother changing dates, it only leads to arguments (sorry if this is an over-simplification); this may be nice advice for keeping the peace, but it conveniently ignores the application of the MOS as it is currently written. If you believe that DATETIES ought to be removed to avoid debates over date changes or refined to restrict the cases in which it applies, then you can propose such a change and seek consenus. Until then, the guideline applies and simply saying "it is just asking for an argument" is unhelpful. —sroc 💬 15:30, 22 March 2015 (UTC)

To conclude that "if DATETIES applies, DATERET doesn't" is not only oversimplistic itself, it is quite arbitrary, and arises only from your rather skewed interpretations of MOS:DATETIES and MOS:DATERET. That "both DMY and MDY dates are acceptable" (per MOS:DATEFORMAT) is quite correct in itself, and there has been no showing of actual problems with either usage. Your insistence that these guidelines must be applied strictly (with emphasis on the lack of a permissive "may") where there is no need or benefit of such strictness, even where there may be local consensus to use one form or another, carries a distinct whiff of wikilayering. ~ J. Johnson (JJ) (talk) 00:16, 25 March 2015 (UTC)

@J. Johnson: Please explain. If DATETIES ("strong national ties") applies, then this makes the choice between DMY or MDY; in that case, they are not "both acceptable". DATETIES is an exception to DATERET ("unless there is reason to change it based on strong national ties"). There may be occasional exceptions to disregard these guidelines (Wikipedia:Ignore all rules), but Cinderella157 is seemingly refusing to accept this: claiming that DATERET overrides DATETIES ignores the express exception within the wording of DATERET itself. —sroc 💬 14:11, 25 March 2015 (UTC)

That you cannot demonstrate any actual problem or harm or any other basis for preferring MDY vs. DMY dates other than "the express exception within the wording of DATRET" - which is to say, insisting on a strict (and questionable) reading of the "letter" of policy, contrary to local consensus - certainly looks like wikilawyering. To the extent you are doing this deliberately to override a local consensus, and thus "avoid the spirit of consensus", amounts to (as others have noted) WP:gaming the system, and "is strictly forbidden". It is also becoming tendentious. A question that begins to loom is why you are "seemingly refusing to accept this".

I think the best good-faith construction that can be made of your comments is that some of our guidelines are not fully consistent, nor even accurate. However, it does not therefore follow that strict enforcement is mandatory. And as it does not appear that this thread is going to make any further progress I will echo EEng's that it be ended. ~ J. Johnson (JJ) (talk) 19:53, 25 March 2015 (UTC)

What is the "local consensus" you refer to? If there is a local consensus relating to a specific article or WikiProject, then I am unaware of it and my comments are directed to it.

My position remains clear:

In general, both DMY and MDY date formats are acceptable (MOS:DATEFORMAT), provided the same format is used within each article (MOS:DATEUNIFY);
Where a topic contains strong national ties to a particular English-speaking country, that will direct whether to use DMY or MDY format for that article (MOS:DATETIES);
Don't switch arbitrarily between DMY and MDY formats unless it is to apply DATETIES or local consensus (MOS:DATERET)—thus, DATETIES overrides DATERET.

If you think that editors changing dates to comply with DATETIES (or local consensus) is a violation of DATERET, then you have misunderstood DATERET. DATERET says: "If an article has evolved using predominantly one format, the whole article should conform to it, unless there are reasons for changing it based on strong national ties to the topic or consensus on the article's talk page." If you somehow interpret this to mean that the date format should not be changed even if DATETIES indicates that a different format should be used to reflect the topic's strong national ties, then: (a) I do not understand how you could possibly interpret these words to mean the opposite of what they say; and (b) DATETIES would cease to have any effect on any article where an editor has used the wrong date format. Perhaps I have misunderstood your position, in which case you could enlighten me? —sroc 💬 18:27, 26 March 2015 (UTC)

By "local consensus" I mean "consensus on the article's talk page", per WP:DATERET. (Why did you not catch either mention of that?) As you seem to be not listening, and your comments only reiterating what you just said, I will only reaffirm my last comment. ~ J. Johnson (JJ) (talk) 21:52, 27 March 2015 (UTC)

I know what "local consensus" means. Are you saying that there is local consensus regarding the Audie Murphy article to disregard MOS on this issue? (You have not said so explicitly here.) If so, then fine. The above discussion has been based on the principle of MOS (disregarding local consensus) and Cinderella157's accusations that applying DATETIES is "gaming the system" and your insistence that "both DMY and MDY dates are acceptable" in spite of DATETIES. —sroc 💬 06:05, 28 March 2015 (UTC)

sroc, The statements I have made are:

Don't change the existing date format just because you think you can. The chances are it will be perceived as contentious.

and;

... [if] there is no good reason to change the date format used - I perceive it is just asking for an argument and might be perceived as Gaming the system.

I also stated:

... the underlying question should be, why change at all?

I did not make an 'accusation' that applying DATETIES was gaming the system. I did state that it might be 'perceived' as gaming the system. I have advised caution in applying DATETIES and given reasonable reasons for doing so. I perceive the last as a departure from neutral language which does not accurately represent the statements that were actually made by me. Cinderella157 (talk) 07:39, 28 March 2015 (UTC)

@Cinderella157: I apologise for mis-stating that you considered that applying DATETIES was gaming the system, rather than that it may be perceived as such.

I think we are agreed that the DMY date format should be retained for the Audie Murphy article; as he is primarily notable for military service, DATETIES indicates that DMY dates should be used, which is already the status quo. If the article had used MDY dates, I would think this should be changed to DMY dates unless there was a local consensus otherwise (e.g., if editors considered that he was equally notable for his acting career and that it was better to leave it alone per DATERET).

Importantly, I would hope that you would agree that, if the article of another recent or current US military figure used MDY dates, then it should likewise be changed to DMY (absent any local consensus otherwise); this was the thrust of my argument above. —sroc 💬 13:59, 28 March 2015 (UTC)

@sroc, for myself, I would agree that it would be prudent to follow my own advice, particularly as I do not perceive that such an edit adds any value to an article in any way whatsoever. I would not take such a step as I believe it would be a mark of rudeness to do so. I agree that the MOS should be applied with common sense even if it does not appear to be written with same at times. I agree that this is a reasonable interpretation of what the MOS indicates. I have not disputed this. On the otherhand, I do not agree that the action it indicates is necessarily reasonable. Cinderella157 (talk) 14:40, 28 March 2015 (UTC)

Agree. I would go further, and state that notability of military service does not warrant "military" form and usage, in dates or otherwise. DMY/MDY preference should follow that customary with the readership, not the subject. Unless the main editors on an article agree otherwise. ~ J. Johnson (JJ) (talk) 23:31, 1 April 2015 (UTC)

Gbit/s or Gb/s?

A difference in interpretation has arisen concerning the advice on the correct symbol for gigabit per second. Feedback is requested at GDDR5. Dondervogel 2 (talk) 17:07, 10 April 2015 (UTC)

I urge my esteemed fellow editors to comment here

Wikipedia:Village pump (proposals)#Proposal: creation of "style noticeboard" EEng (talk) 15:09, 1 April 2015 (UTC) Thanks to Neby for bringing this to our attention elsewhere on this page -- important enough that I thought it should get more prominence.

Proposed style noticeboard

There is currently a discussion at the village pump about creating a noticeboard (similar to the RSN, ORN and NPOVN) for people with questions about how to implement Wikipedia's style policies. The proponents say that one centralized board would be easier for editors to find than many talk pages, and opponents say that it might be a venue for forum-shopping and drama. Participation is welcome. Darkfrog24 (talk) 18:59, 14 April 2015 (UTC)

[1] Figure in brackets is the value given in the source document.

[a]