Wikipedia talk:Manual of Style (dates and numbers)/Archive 51

From Wikipedia, the free encyclopedia
Jump to: navigation, search
Archive 45 Archive 49 Archive 50 Archive 51 Archive 52 Archive 53 Archive 55

Proposal: Mention about spelling out numbers up to 100

While we currently have the manual say spelling out numbers up to 10, I propose that we at least in one sentence explain the standard formal English usage of spelling out numbers up to 100. I was asked by an editor "why spell out "fifty" and not "five hundred and sixty-two" and responded that it was formal English practice because it was simpler and less complex verbiage for numbers up to 100. We could also do this implicitly by separating numbers above 100 from those between 10 and 100, because we really don't want people spelling out (as they "may") 14583, however this would be still more complex because it should still be acceptable to spell out "thirty million". A suggestion above from the MLA handbook about spelling numbers out that can be put into one or two words may be the way, but this does seem unusual and it may be difficult to word it properly? —CentrxTalk 23:54, 14 June 2006 (UTC)

I agree. Jimp 02:49, 10 July 2006 (UTC)


In the examples, why does GBP have a space before the digits but the other currencies have no space? -- SGBailey 12:19, 17 June 2006 (UTC)

Not only GBP, but also CNR has a space. This gives away the key: the $ sign is customarily written attached to the numbers. This is retained if the $ is preceded by a precision which dolar is meant. The all-letter abbreviations like BGP and CNR are followed by a space. −Woodstone 14:52, 17 June 2006 (UTC)

This is described but not shown as an example ... so can we add this to the Good list? $123 USD -- Robocoder (talk | contribs) 00:07, 12 July 2006 (UTC)

No. "$123 USD" is against the guideline. It should be, in the case of articles about the United States, "$123"; in the case of international articles, "US$123", and in the case of technical/currency-specific articles (or wherever else it seems necessary), "USD 123". Neonumbers 04:26, 12 July 2006 (UTC)

Unit format tool

For those interested in making unit formats consistent, please feel free to use my 'unitformatter.js' tool.

To get the tool, copy the entire contents of User:Bobblewik/monobook.js to your own monobook. Then follow the instructions in your monobook to clear the cache (i.e. press Ctrl-Shift-R in Firefox, or Ctrl-F5 in IE) before it will work. To make the tool work, simply click on the 'units' tab in edit mode.

Regards. bobblewik 12:57, 18 June 2006 (UTC)

Clarification. After you activate the "edit this page" tab, the tool places a few tabs to the right at the very top of the page. When you enter either of those tabs, the tool makes a number of substitutions and shows them in "differences" mode. It is then up to you to cancel, save or continue editing. −Woodstone 17:14, 18 June 2006 (UTC)

What is the big deal with non-breaking spaces?

The guidance says: Use   for the space (25 kg) to ensure that it does not break lines..

However, this guidance appears to be misinterpreted. Somebody recently said: the Manual of Style states that non-breaking spaces should be used with units. I don't see changing "20mm" to "20 mm" with a normal space as an improvement, since neither are in accordance with the manual of style.

I have always thought that the nbsp is vastly over-sold. Can we do something about this misinterpretation? bobblewik 19:26, 21 June 2006 (UTC)

Firstly, the SI rules prescribe a space, without specifying if it should be non-breaking. Secondly, since an average line contains say 120 characters, and a unit is mostly 2 characters, the chance that non-breaking has any impact is about 2%. So the tool gives at least a 98% improvement in compliance. −Woodstone 21:06, 21 June 2006 (UTC)
Could you please clarify the above? What do you mean by "98% improvement in compliance" and how do your derive it? I presume you mean that non-breaking HTML entitites should always be used. I think they are cumbersome. --Cedderstk 09:21, 22 June 2006 (UTC)

The tool changes 20mm to 20 mm. So a non-standard format is changed into a standard SI format in all cases. However when the space happens to be at the end of a line, it is not according to the WP style guide. This happens when the end of the line is reached within the two characters "mm". So about 2/120 chance (assuming a linesize of 120). My personal preference would be to use a normal space and adapt the manual of style accordingly. −Woodstone 11:01, 22 June 2006 (UTC)

Why? What is the problem with nonbreaking spaces? Did one pick on you in high school or something? ;) Chuck 12:54, 22 June 2006 (UTC)
It really isn't that hard, and adding nbsp improves readability. Also, we can't always assume a linesize of 120 (since screen size varies), and for some articles (such as ones on tropical storms), there are alot of units used, increasing the likelihood of a problem. --Spangineer[es] (háblame) 15:06, 23 June 2006 (UTC)
While I'm totally behind following SI style recommendations, I agree that using   is very cumbersome. I see the utility of having a non-breaking space between quantities and units, but I don't think it's worth burdening the author with having to type a bunch of character entities. Someone suggested before that the non-breaking space might be better handled by the Wikimedia software, and I tend to agree. It wouldn't be terribly difficult to add the capability to add   characters appropriately where known units are detected. -- uberpenguin @ 2006-06-23 15:17Z
That would be great, but the developers are overworked, so it might not get done right away. However, remember that Wikipedia is a work in progress—the number of authors who actually care about nbsp might be rather negligible, and that's OK. The idea of this style guide isn't so that we can delete people's badly formatted entries, but so that we can agree on the "ideal" style. If no one cares enough to implemenet it, fine, but I think there are enough sticklers like me roaming WP:FAC to at least keep our best articles looking great. 100% compliance with all guidelines isn't mandatory, it's the goal, and it's not a problem if some people focus on a few particular ones and ignore the rest. If enough people care, it'll all get done. --Spangineer[es] (háblame) 15:24, 23 June 2006 (UTC)
I'd hack together the code for this myself if I knew anything about Wikimedia's development cycle and thought that it might actually make it into Wikipedia eventually. -- uberpenguin @ 2006-06-23 15:32Z
There is no recommendation outside Wikipedia to use nonbreaking spaces in all cases. This is a Wikipedia recommendation, not an SI style recommendation. Gene Nygaard 14:29, 2 August 2006 (UTC)

I have changed it so that the recommendation is less demanding (must/should), now it is a subordinate part of the item about using a space, and says "Preferably". —Centrxtalk • 06:36, 24 June 2006 (UTC)

Sorry for posting to an aging discussion, but I am the user quoted above; I said I don't see changing 20mm to 20 mm (with a normal space) to be a worthwhile improvement. I think that the context is important here: this was in a discussion about Bobblewik's unit formatter monobook tool. I feel that if you're going to use a tool that automatically formats units, that tool should format them with nbsp. Doing so results in a slightly higher false positive rate. I think it's silly to say that it's not worth it to put up with that and you'd rather just auto-add regular spaces.
I don't think that there's anything wrong with writing articles using regular spaces between units. It's a lot more effort to insert nbsp, and it's more important to spend your time writing a better article than worrying about the nitpicky details of unit formatting. However, I think if you're going to make it your mission to correct unit formatting, you should correct it all the way, to use nbsp. TomTheHand 17:27, 30 June 2006 (UTC)
Several people have asked me to add that to the scope of my tool. I understand the frustration that they have. I have not spent the time and effort to do what they want and to eliminate the additional false positives. Remember though that I am moving Wikipedia towards the solution you want, not away from it.
If the community wants to jointly develop these two 'units' and 'dates' tools, I would be delighted to share the code with the community and advise where I can. It is all 'open source'. Indeed, there is actually nothing to stop people copying it right now. bobblewik 17:55, 30 June 2006 (UTC)
You've definitely been a huge help. It's just that when I read the above quote I thought I sounded pretty irrational, and I wanted to step in and explain the way I feel in detail. TomTheHand 18:00, 30 June 2006 (UTC)
OK. My mantra is that there is no such thing as user error. All good-faith editors are rational almost by definition. I was attributing a deficiency to the guidance, not you. I apologise if it could be interpreted the other way round. bobblewik 18:20, 30 June 2006 (UTC)
In general, it should be regular spaces. The NIST style guide, for example, allows breaks between numbers and units.
IMHO, however, there should not be a break within the number itself, nor within the units. So compound units should be separated with either a middot (which is nonbreaking) or a nonbreaking space. Furthermore, this is one of the strongest arguments against using spaces for thousands separators, because not using nonbreaking spaces in that context is bad, and using them really clutters up the editing page. Gene Nygaard 14:34, 2 August 2006 (UTC)


The manual currently states "The format of the numeric and percentage terms should match. Thus pair 7 with % and seven with percent." Yet, I found that when using Bobblewik's new tool numerous articles pair a numeral together with percent, as "45 percent". This seems reasonable to me, and it is acceptable in the Chicago Manual of Style. It is also very common elsewhere on the Internet and in printed text. What is the justification for requiring that the word percent only be together with a spelled out number? Is this a possible mis-application of the principle that numerals always be paired with numbers in a list or in the same context (see Wikipedia:Manual of Style (dates and numbers)#Numbers in words? —Centrxtalk • 22:34, 21 June 2006 (UTC)

Personally, I thing "45 percent" looks bad. if you're spelling it out, then spell it out. If you're using numerals, then use symbols. This is not particularly different than for currency. We don't use "50 dollars"; instead it's "$50". Chuck 12:57, 22 June 2006 (UTC)
The manual of style does not state that you must always pair numerals with the dollar sign, in "$50" rather than "50 dollars". It has an example which could be interpreted that way, but that would be over-reaching: the manual just ignores the matter for currency. If "45 percent" looks bad, why doesn't "45 people"? If it is because the number is so small, then that is a different matter not specific to percentages or currencies, whether numbers ought to be spelled out. Why is "percent" or "dollar" any different from the thousands of other words or countable nouns? —Centrxtalk • 00:05, 23 June 2006 (UTC)
From the article: "use US$100 or one hundred United States dollars'"
Your example using "people" is a little silly because there is no symbol for people. There is a symbol for percent. To answer your general question, "percent" and "dollar" are different because there are broadly recognized symbols that are generally paired with numeral representations. For example, it is sloppy to write "100 million dollars". It's either "$100 million" or "one hundred million dollars". This is the same reason you would not write "forty-five%". By your logic, this would be a perfectly reasonable alternative. Chuck 02:35, 23 June 2006 (UTC)
I know what the MOS says. That is an example. The examples are not exhaustive, a mere absence of an example does not mean a style is not acceptable; there are, throughout the MOS, perfectly acceptable formats for which there is not an example. There is no statement that pairing "100" with "dollars" is not acceptable.
The example of "people" was to elucidate how exactly you thought it was ugly when numerals are paired with words everywhere else.

Numerals paired with words are common and appropriate style. If you open up the New York Times, if you look at several books, you will find that it is rather normal. If you want to recommend against written English usage, you need to provide some better justification than simply saying it is "sloppy" or "ugly".

This does not entail using "forty-five%". It is just using symbols instead of words, for no reason, in what is supposed to be formal prose. The default in all prose is to use the spelled out word, except when doing so would be substantially less clear to the reader or where space is precious. In such exceptions, the word in question is changed to make it more clear. This is the reason for writing "1,393,040" rather than "One million three hundred and ninety-three thousand and forty", but it is not a reason for writing "%" yet forgoing writing the more complex "forty-five" as "45". "Percent" is a word, and using it just like any other word is perfectly acceptable. —Centrxtalk • 07:41, 23 June 2006 (UTC)
Ah, but "percent" is not just a word. To me writing "45 percent" is the equivalent of writing "45 per hundred", which would be wrong per the MoS. You would either write "45 per 100" or "forty-five per hundred". Since you'd be using the numerical symbol for "forty-five" (i.e "45"), you should use the numerical symbol for "percent" (i.e. "%"). Chuck 16:53, 23 June 2006 (UTC)

Percent is a special case of ratio, not any old "25/41". For rounded, commensurate ratios, such as "per hundred", "per thousand", or "per million", it is normal to to find numerals alongside them. For percent, I found in the OED that even when the ratio is more explicit, as "per Cent.", numerals were still used, like:

  • 1568 T. GRESHAM Let. 29 Aug. in H. Ellis Orig. Lett. Eng. Hist. (1827) 2nd Ser. II. 314 Th' interest of xij. per cent by the yeare.
  • a1687 W. PETTY Polit. Arithm. (1691) vi. 99 The Interest thereof was within this fifty years, at 10l. per Cent. forty years ago, at 8l. and now at 6l.
  • 1804 G. ROSE Diaries (1860) II. 136 The Funds rose 1 per cent. on the news.
  • 1843 J. A. SMITH Product. Farming 153 The ash of the turnip bulb contains 16½ per cent. of soda.
  • 1870 Nature 23 June 140/1 Neglect of it may result in an error up to 190 per cent.
  • 1973 Times 21 Dec. 14/6 This trouble's cut my social life by about 35 per cent.
  • 1990 Daily Mail 10 Mar. 10/2 A boy who was set on fire because he would not buy crack spent his 12th birthday yesterday recovering from 55 per cent burns.

These are in fact, more common (7 of 10), in these quotations, which are carefully chosen to be representative and canonical examples by the OED.

So, pairing the numeral with the word is common both now and in the literature, and is appropriate style in formal English. Why should the MOS recommend against it? —Centrxtalk • 06:31, 24 June 2006 (UTC)

I have now searched through all the archives, and it appears that this style recommendation, edit [1], was not made according to consensus and was made against valid arguments. The discussion that preceded the change, one month before it, [2], has three regular style editors stating positions that do not support the change, and three supporting it, one tentatively, one the implementing editor and another editor whose only contribution to the discussion was their vote of support. This is hardly consensus, some of the last comments in the discussion were about leaving it up to individual editor decision. Long after, [3], we find two regular style editors not involved with the other discussion recommending "percent" as standard style. Other slightly relevant discussions: [4]: Not especially relevant, but recommending against too much instruction creep. [5]: Discussion unrelated, but has U.S. Government Printing Office recommending acceptable "15 to 25 percent". Not found in this particulary discussion, but note also that this is Chicago style and AP style, among others. —Centrxtalk • 22:18, 27 June 2006 (UTC)

Centrx, what wording do you propose? bobblewik 22:58, 27 June 2006 (UTC)

Questioning space before unit

I would like to question the insistence on a space between a quantity and a unit, especially where the measurement is used as an adjective. I know that such spacing is in accordance with the SI standard, yet the style guide does permit separating thousands by commas, which is clearly not SI-compliant. In everyday usage, it would be far more common to see "35mm film" rather than "35 mm film". There is probably a good reason for it too: "75cl bottles" avoids any possible ambiguity. I would suggest restricting this guideline to scientific contexts. -Cedderstk 09:18, 22 June 2006 (UTC)

"35mm film", "70mm film", etc. have become names rather than descriptions of the film size (IMHO), so spacing is not relevant. (It's not a question of adjectival vs. nominal usage.) — Arthur Rubin | (talk) 16:42, 22 June 2006 (UTC)
The official SI standard states:
Even when the value of a quantity is used as an adjective, a space is left between the numerical value and the unit symbol. Only when the name of the unit is spelled out would the ordinary rules of grammar apply, so that in English a hyphen would be used to separate the number from the unit.
It gives as example "a 35-millimetre film". −Woodstone 20:46, 22 June 2006 (UTC)
Sorry, but whether this is SI standard or not, it's stupid. English has an inherent "multiple adjective" issue in its grammar. This is why English grammar insists on the hyphenation to disambiguate between multiple potential semantics.
For example, "large ship repairs" could be "large (ship repairs)" - a large amount of standard ship repair - or it could mean "(large ship) repairs" - repairs which are made to large ships.
The use of the hyphen clarifies, disambiguates and makes the semantics easier to read - which, after all, is the purpose of a style guide, surely? - because "large-ship repairs" leaves you in no doubt that we're repairing large ships.
This adjectival sense should, I feel, be maintained or it opens up a whole array of ambiguity and will prove to make articles more difficult, not easier, to read.
For example, are you really saying "8 256 Kbit 9 bit parity RAM chips" is easier to read than "8 256-Kbit 9-bit-parity RAM chips"? Okay, that example is contrived (I really should spell out "eight" at the start, if in the body of an article, after all) but ambiguities like this could arise naturally. If a noun is loaded with too many adjectives before it without hyphenation, it can produce more ambiguous - hence, more difficult to read - statements than not doing so.
In nominal use, yes, SI standards should dominate. But, in adjectival use, English grammar should override because English grammar does have ambiguities with adjectives inherently contained within the grammar of the language. The hyphen removes these ambiguities. It makes it easier to read.
Therefore, simply, is the objective here to make things easier to read or not?
Because following the SI "unit" standard on nouns loaded with multiple adjectives is going to cause more headaches for readers, not less. It is not going to ease or improve reading at all. Quite the contrary.
It makes something wholly unambiguous like "8-ball potting" and turns it into something ambiguous like "8 ball potting". It takes something unambiguous like "100 32-bit OSes" and produces confusions like "100 32 bit OS" (as the style guide also tells us not to pluralise the units: Arguably, "OS" is a "unit" - depending on how you want to read it - and now these conventions have produced far more unreadable nonsense).
As someone rightly suggested, English grammar has its conventions for good reason. It evolved them over centuries. By pragmatic precedent. Across whole populations of speakers. It is arrogant and dangerous and counterproductive to assume blind obedience to a standard made up overnight by a handful of people upon well-intentioned but unpragmatic "theory" alone would prove superior.
This is pedanticism above common sense and it will prove more ambiguous and far harder to read than adopting the good sense to see that evolved pragmatic precedent over centuries survives by its natural "fitness" for its use, not because of enforced blind organisational obedience.
It is my opinion that, in this matter, we choose the vastly poorer option and that we will wrongly create more ambiguity, confusion and unreadable articles by following this ungrammatical - and, apparently, vastly unthinking - "standard". It is clearly the wrong choice for our desired wish of making articles more readable. PetrochemicalPete 07:06, 19 July 2006 (UTC)
I agree that this can be an issue. I would say that in cases where there is ambiguity, an editor should just apply judgement and throw in the hyphen (of course there remains the problem that another editor will sooner or later quote this guideline, and set the sentence back to an ambiguous form). Michael Z. 2006-07-19 07:36 Z

Any examples of adjectival use that are not in the body of an article? If it is in the body of an article, then it should not use the abbreviation anyway and a hyphen is appropriate. Wouldn't most infobox items not need it? For example, specifying 35 mm film might be an item in a camera infobox, yet the heading of that item would be "Film", so we would have "Film: 35 mm", not "Film: 35 mm film", same for the bottles example if that is used anywhere. —Centrxtalk • 23:19, 22 June 2006 (UTC)

Articles about firearms or armoured fighting vehicles (AFVs) may cite weapon calibres a dozen or more times. Writing them out every time may become tedious and confusing for the reader. In books about AFVs, it's quite conventional to use the unspaced "00mm" form, for example anything written by Steven Zaloga, the guidebooks published by Jane's, The Economist's style guide (search the page for "calibre"), etc. This form is used successfully in the Wikipedia featured article T-34, where millimetre measurements appear 53 times. Michael Z. 2006-07-19 07:36 Z
I'm sorry, but does anyone else find it ridiculous to suggest picking and choosing which parts of SI convention to adopt as style guidelines? Either we should adopt SI or not, and SI has thus far proven invaluable to keeping units consistant on Wikipedia. If we suggest that the unit space is "optional" or "up to editor discression" it will invariably lead to arguments about which is better on many articles' talk pages. That's counterproductive. I don't think the rule in question is "stupid" in the slightest, so here's at least one dissenting opinion. In my editing scientific and computing related articles I have never seen conformation to SI standards cause any sort of problem except with the personal preferences of the random editor.
The given T-34 article inconsistantly uses a space between numbers and unit abbreviations, which in my opinion is far worse than not using the space at all. I don't see at all how changing the units in T-34 to conform to SI convention would be confusing or detract from the article's quality and flow. Even with the mass of millimeter measurements, non-breaking spaces can be used to prevent a unit from wrapping to a next line if that were a problem. -- uberpenguin @ 2006-07-19 14:02Z

Not including the space before a unit symbol also screws up search engine searches. Most search engines can search for a specific phrase if desired; most can't find the mm in pages containing only "85mm" and "101mm" and the like, nor the number in "0.45359237kg". Gene Nygaard 14:26, 2 August 2006 (UTC)

Abbreviated month names

I reverted a link to Nov 16 in the DJ Screw article. Is this style always ok, or only in a quote? I can't find the answer in the MOS. Tim Ivorson 2006-06-22

In the body of an article, month names should always be spelled out, except, as you say, in quotes. It may be appropriate to abbreviate in an infobox or other place where space is short, but the infobox in question on DJ Screw does not appear to be such a situation. The MOS only implicitly indicates this, by examples of fully spelled months. This is normal English writing, but perhaps it should be explicitly stated in the MOS. —Centrxtalk • 00:10, 23 June 2006 (UTC)
America's Army has two good examples where abbreviated months are appropriate. —Centrxtalk • 00:12, 23 June 2006 (UTC)
Thanks for the info, Centrx. I think it'd be helpful to make the MOS explicit, but I think I should leave that to people who know what they're doing. Tim Ivorson 2006-06-23

Month-date ranges

I found nothing in the current article about what to do with yearless month-date ranges like "August 16–20" (i.e., "16–20 August"). There are many occurrences in Wikipedia articles of this type of date range, but there seems to be no policy statement on whether or how to format them, even if to say there is no policy. Should we:

  • leave them unformatted (breaking consistency with other, formatted dates)?
  • format part of them (e.g., August 16–20), which won't look right for half the users whichever way they're formatted?
  • format the whole thing (e.g., August 1620), again looking bad for half our editors?
  • some other system not obvious to me?

At the very least, our policy page should acknowledge the problem. I didn't see anything mentioned in Wikipedia talk:Naming conventions (calendar dates). I didn't sift through the forty-eight archives of this talk page, but if there's some discussion buried in them, its resolution (or lack thereof) should long ago have been made a part of the policy page. Thanks for any assistance. ~ Jeff Q (talk) 07:41, 25 June 2006 (UTC)

Since it seems that formatting like "August 16–20" would with date preference applied show "16 August–20", which is not proper, in such cases I do not think the dates should be linked. A possible solution is to write it out as "from August 16 to August 20", which preserves date preferences and is otherwise perfectly acceptable. —Centrxtalk • 02:32, 26 June 2006 (UTC)
A horrible example is: [[November 12]]–[[15]], [[1942]] in Naval Battle of Guadalcanal (at time of writing). I propose that we add the guidance:
  • Date ranges: Generally do not link..
bobblewik 20:11, 28 June 2006 (UTC)
This isn't guidance already? I'm surprised, since it's what I've always been doing or changing ranged dates to. It's so obvious I'd just always assumed it was in WP:DATE. TheGrappler 01:17, 1 July 2006 (UTC)

I strongly object to this proposal. Are you people trying to get your way simply by tiring everyone else out and hoping that you'll slip this back into policy when no one is looking - again? Rebecca 02:31, 1 July 2006 (UTC)

What do you propose instead? If you are referring to date linking, this "proposal" is not about date linking in general; it is a very specific case that, as far as I can tell, was not discussed at all in the recent date linking discussion in Archives 42–46,48. —Centrxtalk • 07:01, 1 July 2006 (UTC)
November 12-November 15, November 12-15 and November 12-15 are all fine by me per se, but I do think there's no reason to break the date preferences setting. —Preceding unsigned comment added by Rebecca (talkcontribs) [6]
Ok for me too. Linking has the additional advantage that it's clear what is meant. -- User:Docu
I agree that there is nothing wrong with [[November 12]]-[[November 15]]. But that format is rare.
Many articles have broken dates because some editors don't understand the difference between 'hyperlink' and 'preference mechanism'. Editors do things like: [[November 12]]–[[15]], [[1942]]. It is probably worse than all the other errors because the page looks broken when you see:
  • 12 November-15.
Set your date preference to the non-US format and look at the the first sentence of: Naval Battle of Guadalcanal.
The proposal addresses that common problem. bobblewik 10:36, 1 July 2006 (UTC)
Furthermore, the example given by Rebecca presents exactly this problem.
Quote: ...[[November 12]]-[[November 15|15]] ... are all fine by me
That is broken and becomes 12 November-15. bobblewik 10:46, 1 July 2006 (UTC)

Date ranges & linking: in general no difference from other dates, I'd say. The only remark is that editors should be sensible to the implications, and make an appropriate choice:

  • unlinked:
    • May 12-15 gives de facto precedence to US notation, so could only be used w.r.t. topics that are clearly under a variety of English that uses this notation as a standard (See WP:MoS#National varieties of English)
    • 12-15 May gives de facto precedence to European notation, so could only be used w.r.t. topics that are clearly under a variety of English that uses this notation as a standard (See WP:MoS#National varieties of English)
  • linked:
    • Month + day at both sides of the hyphen/dash/"to", without pipes would give a consistent layout whatever the transformation by preferences setting of the user: [[May 12]]–[[May 15]] or [[May 12]] to [[May 15]], etc. (but yeah, this is not so "fluent", and probably not very good style)
    • Both with pipe, resulting in US notation (same remark as above for US notation, "unlinked"): [[12 May|May 12]]-[[15 May|15]] or [[May 12|May 12]]-[[May 15|15]]
    • Both with pipe, resulting in European notation (same remark as above for European notation, "unlinked"): [[12 May|12]]-[[15 May|15 May]] or [[May 12|12]]-[[May 15|15 May]]
  • half-linked as above, e.g.
    • [[12 May|May 12]]-15 (US notation)
    • 12-[[May 15|15 May]] (European notation), but only linking the second date doesn't look too well I suppose. Although I could imagine something like "After a short campaign (12-15 May) the elections were held" wouldn't be too silly. --Francis Schonken 12:00, 1 July 2006 (UTC)
Do you have an alternative wording for the proposal? bobblewik 12:31, 1 July 2006 (UTC)

The proposal is to change nothing. Something that is evident from existing guidance, should not be used as an excuse to change prior agreed upon provisions in the guidelines. In other words: there is no problem (apart from maybe the problem that you're on a date de-linking crusade again, as remarked above by Rebecca, and on your user talk page). There's no reason to change linking/delinking of dates guidance, and I go from the assumption you know that. The so-called linking of date ranges "problem" you elaborated as a result of Jeff's question above is an artificial construct, ignoring current guidance. I think I demonstrated that. --Francis Schonken 12:50, 1 July 2006 (UTC)

Why not add "Incorrect: February 14–15 and 15–14 February" to Wikipedia:Manual of Style (dates and numbers)#Incorrect date formats? —Centrxtalk • 22:39, 1 July 2006 (UTC)

Proposal for section Numbers in words

It is common English style to have numbers up to 100 spelled out, and under the MOS, acceptable. However, the current version lumps numbers 10 through 100 together with numbers 100 and up. While it is in formal use appropriate to have "eighty-seven" spelled out, it is rare to have "five hundred and sixty-seven" and totally inappropriate to have "one million, four hundred and fifty-six thousand, seven hundred...", which the MOS read literally currently accepts. Therefore, numbers 10 through 100 must be distinguished from those above 100. Here is probably the simplest solution:

  • Numbers between ten (10) and one hundred (100) may be written as numerals or spelled out in words. Within each article, style should be consistent.
  • Numbers above one hundred (100) must be written as numerals, except at the beginning of a sentence.

Note that one problem with this is it forbids "five hundred" or "ten million", which with a little wording could be corrected.

There are two other possibilities. First, we can explicitly recommend that all numbers between 0 and 100 be spelled out as words in the body/prose/text of an article. Second, per suggestion above, at Numbers from zero to ten which states it is MLA style, we can recommend that numbers that can be written out in two words or less (e.g. "eighteen", "ninety-nine", "five hundred", "five million", but "101", "499", "5,100,000" or "5.1 million"). This is a convenient solution, and corrects the problem with the above, simple solution. See also Wikipedia talk:Manual of Style (dates and numbers)#Numbers from zero to ten and Proposal: Mention about spelling out numbers up to 100. —Centrxtalk • 22:58, 27 June 2006 (UTC)

I like your second possibility ("written out if in two words or less"). It sums it up nicely, I never thought of it that way. Neonumbers 06:03, 6 July 2006 (UTC)
  • Support "if in no more than two hyphenated words". Regards, David Kernow 09:13, 7 July 2006 (UTC)

Done, see [7]. —Centrxtalk • 20:59, 9 July 2006 (UTC)

Proposal for section Percentages

The MOS currently recommends that numerals should always be paired with the percent symbol (%) and spelled-out numbers paired with the word "percent". This forbids the style of "55 percent", which is common on Wikipedia and is standard style in English prose, recommended by the Chicago MOS and AP style guide, common in the canonical quotations of the OED, s.v., and found in books[8][9], journals[10], and styleguides[11] everywhere. See also Wikipedia talk:Manual of Style (dates and numbers)#Percentages, [12], and [13].

Note, however, that while it is common and standard for "percent" to be written out in 'humanistic' copy, we find in above-mentioned discussions and in the Chicago MOS that in scientific or statistical use, the symbol % is more common or appropriate. However, such recommendations are for scientific journals or books, not for general-purpose encyclopedias that are covering scientific matters among many others. Note, for example, that the Encyclopedia Britannica uses "percent" uniformly (e.g.: "They consist of (by dry weight) roughly 50 percent to 75 percent crude protein, 4 percent to 18 percent fats,...". Therefore, the Percentages section ought to be changed to:

  • Use the word percent or per cent for a percentage in the body of an article. Examples: "twelve percent", "12 percent".
  • Use the symbol % with a numeral for a percentage in a table, infobox, or the like. Example: "12%"

Centrxtalk • 23:27, 27 June 2006 (UTC)

I agree with Centrx's proposal with two modifications. First, the use of numerals versus numbers should be based on the Numeral guideline in the MOS (zero to ten, 11+) and we should recommend one spelling of percent, namely "percent". Rillian 12:30, 28 June 2006 (UTC)
Apparently, "per cent" is common in Britain and "percent" in America, so I was just following Wikipedia:Manual of Style (spelling)#Preferred variants in allowing both. —Centrxtalk • 22:19, 28 June 2006 (UTC)
For numeral use, the example is in the range in which it is acceptable to have either "12" or "twelve". Or do you mean that the section should explicitly state that percentages should follow the Numeral guideline? —Centrxtalk • 23:49, 3 July 2006 (UTC)
Some pro-symbol references:
Use the sign % instead of per cent (The Economist)
% in headlines and copy (The Guardian)
Symbols for numbers and units versus spelled-out names of numbers and units (NIST)
Unlike most publications, Wikipedia is international. Therefore we should take this into account when choosing between a language-specific words and a language independent symbols.
The original complaint from Centrx was that '55 percent' is forbidden. I like the current guidance but I am sympathetic to the desire for deregulation. However, the proposed new wording is not liberalisation, it is inversion. What was forbidden would be mandatory and what was mandatory would be forbidden. Could we agree on deregulation instead? bobblewik 18:54, 28 June 2006 (UTC)
Nothing is "forbidden" by the MoS - Rich Farmbrough 10:44 5 July 2006 (GMT).
Yes, this is not the most accurate word, but it is understandable within the advisory context of the MoS, and the fact remains that one is liable to find some editors changing article style to conform to the MoS, sometimes by script/semi-bot, and find that changing it back is met with calls to the authority of the MoS. While the MoS is not exactly binding in the free flow of the wiki, its substantial effect can be beyond mere recommendation. —Centrxtalk • 02:19, 6 July 2006 (UTC)
Deregulation would not require much of a proposal and could be implemented by removing the section. Note, however, that following the standard in spelling out units in text[14], rather than using symbols, would indicate that "percent" should be spelled out as well. Note also that "percent" spelled out is far more common on Wikipedia, and the former policy was against common practice. Also, the current proposal is not exactly an inversion, formerly any editor who wanted could use "fifty-five percent".
The NIST recommendation is irrelevant, it is for scientific and technical papers, which Wikipedia is not. Regarding the Economist and the Guardian, note that the BBC styleguide[15] and The Times of London styleguide[16] both have "per cent" written out (In the U.S. also, aside from the AP styleguide it is found in the NY Times, the Wash. Post, and Time Magazine), and furthermore that Wikipedia need not consider issues of space like newspapers and magazines, and that books, non-technical scholarly journals, and encyclopedias overwhelmingly use the word spelled out, while it is still much more common for newspapers and magazines anyway (see also Google News).
I see no reason why the Manual of Style should not state that "percent" is written out when:
  • It is far more common style on Wikipedia.
  • It is uniformly common and standard style in professional works, books, encyclopedias, etc.
  • It is far more common style in magazines and newspapers
  • Only scientific and technical journals and books, which Wikipedia is not, and a minority of newspapers and magazines, which Wikipedia is not like, use % instead of percent
Centrxtalk • 22:19, 28 June 2006 (UTC)
Whichever way is chosen, it should be required (based on SI specification) to have a space between the number and the percent sign. The example should read:
  • Use the symbol % with a numeral for a percentage in a table, infobox, or the like. Example: "12 %"
Woodstone 19:06, 28 June 2006 (UTC)
This is very rare in general practice, and does not really apply to many uses of percent, which is not an SI unit. The MOS should not push such a minor usage against the vastly more common use on Wikipedia and elsewhere. Also, this is a separate issue for which there was a lot of debate in the archives, and so to keep this discussion clean we could have both in the example, though it might be best to just resolve it now. —Centrxtalk • 22:19, 28 June 2006 (UTC)
  • Prefer "percent" per Encyclopaedia Britannica, but probably best left optional...?  Regards, David Kernow 09:15, 7 July 2006 (UTC)

No space with % symbol

I have added the sentence "Note there is no space between the numeral and the symbol %." to the percentage section. This is based on Chicago Manual of Style, 14th ed., p. 299, section 8.18. My reason for adding it is that I have a hard time remembering this convention and have to look it up frequently. --Gerry Ashton 16:42, 30 July 2006 (UTC)

Dates (years BC)

The entry mentions, "Articles for the year 500 BC and earlier should be redirected to the relevant decade. Articles for the year 1700 BC and earlier should be redirected to the relevant century. Articles for the year 4000 BC and earlier should be redirected to the relevant millennium."

Shouldn't this read, "Articles for the year 500 BC and later should be redirected to the relevant decade. Articles between the year 1700 BC and 500 BC should be redirected to the relevant century. Articles prior to 4000 BC should be redirected to the relevant millennium." The current wording is ambiguous and includes overlap. —The preceding unsigned comment was added by Dr1819 (talkcontribs) 19:42, 30 June 2006 (UTC)

It appears to mean:
  • After 500BC: Link to year.
  • 1700BC–500BC: Link to decade.
  • 4000BC–1700BC: Link to century.
  • Before 4000BC: Link to millennium.
Centrxtalk • 07:09, 1 July 2006 (UTC)

It's the page itself that redirects, e.g. 502 BC (which is 500 years before/prior/earlier than year 1) redirects to 500s BC. One should still link to 502 BC. -- User:Docu

Different calendars

In the section on different calendars it says to use the Julian calendar for dates before 1582-10-15.

Dates before the adoption of the Gregorian calendar on 1582-10-15 should usually be given in the Julian calendar. The Julian day and month should not be converted to the Gregorian calendar, but the start of the Julian year should be taken to be 1 January (see below for more details).

I'm not totally clear on what this means. For example in citing history it may not be clear whether a date is Julian or Gregorian. On the other hand in using simulations (such as astronomical software) the dates are often presented as Gregorian dates no matter how far back the simulation goes. In addition, the Julian calendar is ambiguous for the starting year (see Julian calendar), so I assume this means use a Julian calendar with the AD/BC epoch. However, it's not entirely clear. Perhaps should simply say Gregorian calendar as the default unless there's some clear reason to use the Julian calendar. Is this common in the field of history to explicitly use Julian calendar dates before the first adoption of the Gregorian calendar? My assumption when I read Julius Cesear adopted the Julian calendar in 46 BC is that this is a Gregorian date: that this is 2051 years ago (with no 0 year in either calendar). Can anyone clarify this? Thanks in advance. --Cplot 06:31, 2 July 2006 (UTC)

After posing the question above I did find a useful tool that perhaps could be linked to in the manual of style to help make conversions. It's a | calendar conversion calculator for many different calendars (including Gregorian and Julian). It doesn't really answer my question, but I thought others would find it useful. It may also help in thinking through this issue. --Cplot 06:51, 2 July 2006 (UTC)

The Julian calendar doesn't say anything about numbering of years. In Julius Caesar's Rome years weren't even numbered, a year was known by the names of the two consuls of that year. The AUC ("Ab Urbe Condita") year number wasn't used as a means to number years in civil life. Livy wrote a historical work named "Ab Urbe Condita", but didn't "number" years in that historical work. Varro listed consuls, and only much later a numbering of AUC years was derived from a combination of historico-mythological works about the founding of Rome and the known lists of consuls. Numbering of years from the birth of Christ (as far as known at the time) originated in the early middle ages, so was completely unknown by Caesar, and, on the other hand, completely established by the time of pope Gregorius - so he didn't "change" that numbering of years, he only modified the start of months, and how many days months would have; these months were basicly still the same twelve months as in Caesar's day.

Another difference was at which month a year started. In old Rome a year started when the two new consuls came to power. Which was May in the oldest known times, later became the first day of March, and had more or less standardised on the first day of January by Julius Caesar's time (in fact since AUC 601 = 153 BC, see blue vertical line on this page: [17]). Also here "Julian Calendar" doesn't say anything about whether a year starts 1st of January, or 1st of March (or whatever other month). Gregorius did away with the differences (that apparently still lived on till his day), and in his calendar years always started first of January. For some countries that didn't yet move to Gregorian calendar at the time, a year still started at first of March (or whatever other day usual in the local calendar). That is the "old style" of some countries. So, moving to Gregorian calendar had two aspects: (1) having the year start at the first day of January, (2) adopting Gregorius' system of leap years, number of days for the month of February, etc. The first of these changes could be done independently of moving to the full Gregorian calendar. Anyway, for "old style" calendars, dates in the month of January and February could have a one year difference with calendars that started a new year on the first of January.

To answer your question "Is this common in the field of history to explicitly use Julian calendar dates before the first adoption of the Gregorian calendar?", the answer is yes, but you have to know that Julian calendar, by itself, says nothing about numbering of years. Doing otherwise (i.e. not using the Julian calendar for dates before the Gregorian reform) is called using a "proleptic" Gregorian calendar, which is thoroughly uncommon in historiography. As for the numbering of years, in current (Occidental) historiography the numbering of years starts at 1 AD/CE (and counting back from 1 BC/BCE for older dates). And this counting of years is influenced by which day of which month is seen as the first day of a year (which, for instance, appears in old style/new style differences for days in the two first, or last, months of a year). --Francis Schonken 09:08, 2 July 2006 (UTC)

Thanks for the reply that clear things up for me. I think some abridged version of this explanation could be included in the manual of style. Perhaps I take a stab at it. --Cplot 17:51, 2 July 2006 (UTC)

Inclusion of conversions

Under Wikipedia:Manual of Style (dates and numbers)#Units of measurement as of the time of this edit, the first point states that “Conversions should generally be included and not be removed.”  I hope that I’m not missing anything, but this seems to be a loophole to me.  I could add an astronomical amount of unnecessary unit conversions from the list of strange units of measurement but, according to policy, “Conversions should generally be included and not be removed.”  I don’t believe that the removal of these inessential conversions should be against policy.  Would anybody agree with me that these conversions should be limited to standard systems of measurement, such as metric and imperial conversions only? —Rofl 09:22, 7 July 2006 (UTC)

Makes sense. I suspect there is an intent to avoid "lawyering up" with too much specificity, leaving it to common sense and the WP principle (pillar #5) of not getting bogged down in rules. The general idea, I think, is to permit the use of both SI and the most common and relevant non-SI measurement (usually imperial or U.S., of the most comparable magnitude) for each quantity. ~ Jeff Q (talk) 09:32, 7 July 2006 (UTC)
Yeah, I see what you're getting at. (I was concerned about this when we redid that section, but no-one else seemed to consider it relevant). Jeff's right, the general idea is to allow conversions where necessary; but then you get disputes over what's necessary and stuff like that, I guess... the loophole needs to be fixed.
I think conversions should be limited to:
  • all units typically used for that context (e.g. in astronomy, light years or AU)
  • metric units
  • U.S. units (or whatever they call them.
But knowing the nature of this, I bet there'll be some exception somewhere. It's just a tricky call... it's not so much related to pillar five (or at least, not in my view... we just try to give good, solid, reasonable guidance.) Neonumbers 12:15, 7 July 2006 (UTC)
There will be certain exceptions like in the Jersey article, but for the most part I agree with Rofl. I think that common sense will prevail if someone tries to input a conversion in some odd unit like cubits to describe the CN tower's height. Someone will revert it. —MJCdetroit 19:20, 7 July 2006 (UTC)

How about "Conversions useful to many English readers should generally be included and not be removed."? —Centrxtalk • 03:19, 9 July 2006 (UTC)

Discussion on the interpretation of Eras section.

Hello. There is an ongoing discussion on the interpretation of the Eras section of this MOS in Talk:Montanism. The passages used for the two different positions are:

  1. Both the BCE/CE era names and the BC/AD era names are acceptable [..] When either of two styles are acceptable it is inappropriate for a Wikipedia editor to change from one style to another unless there is some substantial reason for the change. For example, with respect to English spelling as opposed to American spelling it would be acceptable to change from American spelling to English spelling if the article concerned an English subject. Revert warring over optional styles is unacceptable; if the article is colour rather than color, it would be wrong to switch simply to change styles as both are acceptable.
  2. Normally you should use plain numbers for years in the Anno Domini/Common Era

The matter of the dispute is:

  • in an article in which all of the years and centuries are AD/CE, is it possible/mandatory/prohibited to add "AD" to the first year/century nominated?

Please, help us to settle this matter.--Panairjdde 23:54, 8 July 2006 (UTC)

Hi! As a general rule, you should ensure the passage is comprehensible and consistent (this general principle lies behind all guidelines in the MOS). From what I gather, there is no dispute over whether AD or CE is better for this article. Therefore, the judgement is whether or not it is necessary to write "AD" for early years and centuries. (Correct me if I'm wrong.) I admit here that guidance for this time period might not be clear. This is my interpretation of the guideline:
Where you are refering to a century, e.g. "3rd century", it is not necessary (i.e. redundant) to use "AD" because the 1st, 2nd and 3rd centuries are, unless followed by "BC", normally taken to be refering to AD. Where you are refering to a year (number) that is in the first century, it would probably pay to use "AD". Where you are refering to a three-digit year, if the sentence implies that the number is a year or date, then it is probably not necessary to use "AD". (btw, where you see terms like "not necessary" or "probably", read them as a straight "shouldn't" or "should".) Neonumbers 10:54, 10 July 2006 (UTC)

Which midnight?

Someone added to the guidline:

So for consistency, always use 12 midnight to indicate the start of a given day: whenever the day is relevant. For example: “The winner crossed the finish line at precisely 12 midnight 17 March 2001.” indicates the beginning of the day 17 March (the end of the day 16 March).

Is this really the common way of doing it? A more usual occurrence would be a due date, like in:

  • submissions have to be handed in before midnight July 31

Would that not commonly be interpreted as midnight at the end of that day? −Woodstone 19:59, 9 July 2006 (UTC)

Yes, I don't think this is most common at all. Midnight implies the night of the given date, not the morning of it. For clarity, it might even be best to explicitly state "the night of" or "the morning of". I have reverted the changes of that line. —Centrxtalk • 20:44, 9 July 2006 (UTC)
  • submissions can be handed from midnight 1 June
I don't know, I see midnight as being just as much the begining of one day as it is the end of another, i.e. belonging equally to both; but I'd more usually view it as being between days, i.e. belonging to neither. Yes, it's best to be explicite. This is an advantage of the 24 hour clock: a day begins at 0:00 and ends at 24:00. On the other hand, if you adopt the convention that midnight is 12:00 a.m., then it would make sense to "always use 12 midnight to indicate the start of a given day". However I wouldn't recomend this be adopted here either for the sake of clarity.--Jimp 17:00, 19 July 2006 (UTC)

The "time" guideline is weird. It doesn't even follow what the referenced articles of 12-hour clock and 24-hour clock say. 00:00 refers to the start of a day. 24:00 refers to the close of a day.

Anyway, the best way to avoid ambiguity is not to define either way and ask all others to follow. Readers will still confuse anyway since readers will not read your style guidelines before they read any article in wikipedia. Specify instead of assuming!!

Worse approach: assume whether "between 23 and 50" includes or excludes the numbers of 23 & 50. Better approach: clear the ambiguity up: editors are encouraged to specify it: "between 23 and 50" (inclusive) or "between 23 and 50" (exclusive).

Finally I attach another explanation from one time website about Noon and Midnight:
A better explanation on Noon and Midnight

AM and PM - What is Noon and Midnight?

AM and PM start immediately after Midnight and Noon (Midday) respectively. This means that 00:00 AM or 00:00 PM (or 12:00 AM and 12:00 PM) have no meaning. Every day starts precisely at midnight and AM starts immediately after that point in time e.g. 00:00:01 AM (see also leap seconds) To avoid confusion timetables, when scheduling around midnight, prefer to use either 23:59 or 00:01 to avoid confusion as to which day is being referred to.

It is after Noon that PM starts e.g. 00:00:01 PM (12:00:01) --

Best regards. --Wai Wai (talk) 15:24, 6 August 2006 (UTC)

Proposal for section Wikipedia:Manual of Style (dates and numbers)#Eras

To recommend the use of AD when it would clarify the date, such as for small dates, like "16 AD" that could be confused for numbers without the AD, and for rather ancient histories in the first few centures AD, I recommend that following revision for the relevant item:

  • Normally, use plain numbers for years in the Anno Domini/Common Era. When there is the possibility of confusion, such as in an article that covers events of both BC and AD or in the early years of AD, use AD or CE for the date. Example: "35 BC–18 AD", "105 AD".

Centrxtalk • 21:26, 9 July 2006 (UTC)

Totally against. If you have an article with all the years in AD/CE era, why you should disambiguate adding AD/CE? And if you oblige to put BC/BCE after BC/BCE years/centuries, where is confusion?
Furthemore, what does "in the early years of AD" mean? I've seen AD used for years in the 10th century, who should decide if that is early or late years?
Furthemore, in a text like "In 15, Germanicus was proclaimed Imperator", how could you possibly confuse year 15 with number 15? Or, for the matter, how could a reader be puzzled if 15 here means AD/CE 15 or 15 BC/BCE?--Panairjdde 21:40, 9 July 2006 (UTC)
Putting AD with early years is to distinguish that it is a year, and not just one of the many common uses of low numbers. This is common practice on Wikipedia (at least before your unilateral edits) and in English, and would regardless warrant a "may" directive, permitting it under the styleguide. There is nothing currently that forbids the use of AD, and there ought not be in these cases.
Whether a year is early or not is not mandated by this proposal, but left up to editor decision on particular articles. The example "105 AD" was carefully chosen to include all two-digit numbers and to suggest low three-digit numbers but not strongly exemplify it.
One major example of where there would be confusion between the year and a number is with persons, where the number could instead indicate the person's age. Example: "At the age of 16, he joined the Roman legions. By 25, he was a centurion." This is only one of the more egregiously confusing examples; even without the first sentence, it could still be confusing. —Centrxtalk • 05:39, 10 July 2006 (UTC)
I would support such a provision for years in the first century only. However, I would not object to this for low three-digit numbers, that is, those less than 150. I would be sceptical for three-digit years later than that.
For two-digit years, while it can be inferred that it is a year, it can take a short while (a few seconds) for a reader to figure it out. For three-digit years, in my view, it is typically immediately or almost immediately obvious.
If we can form some agreement on it, we should have a more objective borderline — not necessarily a black-and-white one, just a more objective one. I agree with the idea in principle — just don't take it too far. Neonumbers 11:04, 10 July 2006 (UTC)
Centrx, since all of your proposal is based on distinguishability, would you be so kind and give me an example of an article in which a year could be confused with a number?
As regards your edits, in which way in the sentence "In 15, Germanicus was proclaimed Imperator" "15" could be understood as the number 15 as opposed to year 15?
Requiring not to introduce AD for low number years in articles dealing with all AD years would show respect for the reader (you don't suppose she's stupid), increase readability (AD is all capitals), reduce controversies (no AD/CE = no AD/CE controversy).--Panairjdde 14:53, 10 July 2006 (UTC)
The problem can be present in any article with BC years or low AD years. I gave an exceptional example where there would be confusion for even a reader explicitly told that no era name means it is AD. The confusion can still be present even in normal uses of low years. Even for the average reader, this can be true, and there are indeed children, non-native speakers, and stupid people reading the article. —Centrxtalk • 03:13, 11 July 2006 (UTC)
It is not necessarily confusion with numbers that is the issue (though it certainly could be), it's just that kind of, "(confused frown) What? Err... Oh..." that lasts about two or three seconds when even an intelligent reader reads something like "In 15, ...". They'd figure it out, of course, but good writers don't make their readers momentarily ponder over things like that (they have better things to think about). Writing something like, "In the year 15, ..." would also help eliminate confusion, though some people (not me) might not like that stylistically.
It is most certainly not supposing a reader is stupid to include "AD" before any year — obvious or not — it shows respect to aid the reader where the literal meaning is not immediately obvious. It just gets in the way (redundant) where you use "AD" for three- or four-digit years. Two-digit years aren't so nice. Neonumbers 04:38, 11 July 2006 (UTC)
  • It's funny, but I was going to propose that it be normal to specify that a number is a calendar date with the era, for any years that aren't four digits. Anyone will automatically and easily recognise that 1287 is a year. But for 987 AD, I am just more comfortable seeing 987 AD (or in the preferred style, [[AD 987) - not just 987. Of course, 987 isn't that jarring, but it's true that the smaller the number gets, the more customary it is to use the AD, at least in the first instance, for native English writers. How about 5 digits? Again, since it's not 4, I would want to see AD to make it glaringly obvious that it is a year and in what calendar. This is hypothetical, since we have about 8000 years to go into the future before we get a 5 digit AD date, and there are probably no instances of this on the wikipedia. But the fact that it isn't the normal four digits, means I would rather see 12487 AD, in place of just "12487"... Get what I'm trying to say? Now maybe we should all go for a nice cup of tea and a sitdown...! ፈቃደ (ውይይት) 14:57, 10 July 2006 (UTC)
  • I feel changing the policy is not the right step to take. I think the right step to take is to make era systems a user editable preference. This would avoid any future edit wars. In lieu of that, I feel the current guideline is just fine. As long as the numerical year is wikilinked, it avoids confusion. AD/CE is only necessary in a range of dates that starts in BC/BCE. All positive numbers go to the corresponding year article. I'm willing to allow the slight change that someone will be confused by seeing 17 or 120s to avoid edit wars over AD vs. CE or even worse, a compromise that results in both (AD/CE) which has just happened over at Paul of Tarsus, and happened at Jesus long ago.--Andrew c 21:52, 27 July 2006 (UTC)
This means that whether a number is wikilinked or not is supposed to convey information. I have problems with that. First and foremost, because when I read text I am not looking for this vehicle of communication. Second, because the color contrast between regular and wikilink text is not that strong, and may be indistinguishable to some forms of colorblindness. Third, because not all years are wikilinked. I'm undecided on policy rewording, as I feel naming convention ambiguity handles this, and a user pref is a better way to resolve it. I tend to want to see AD or CE on positive up to 150 or so, at least once per section or paragraph. Gimmetrow 22:26, 27 July 2006 (UTC)
There was a moratorium on the massive AD / CE edit wars that has mostly held place for something like a year, so it hasn't been a sitewide problem for both to co-exist as equally acceptable and both the be left alone in articles - that is, until now, when one user decided that they are henceforth equally unacceptable no matter what anyone else says, and he even began stalking my recent contribs removing all usage of AD for early AD dates, even where customary and after a BC span. The policy already gives due weight to the ArbCom decision about equal styles as it is - it just needs clarifying more, for those who defy common sense and the spirit of the law. ፈቃደ (ውይይት) 22:27, 27 July 2006 (UTC)
Keep in mind that this is not so much a policy. If a strong majority of Wikipedia articles use AD for low years, then that is the style and it should be reflected in the styleguide. Era preferences are very unlikely to happen, you would have to get a developer to sign on for that, and even if it were to happen, it would not be effective. Such preferences could not account for cases where articles have both BC and AD dates, in which case the AD notation is more appropriate, and are we to then also have a preference so the user can specify at what year the ADs start getting dropped? The solution which reflects common practice and which can make it clearer for the reader, is to state that usually low AD years have the AD notation, and that in general it is left up to the editors of an individual article to decide whether there could be confusion warranting AD notation, for their article. AD notation may not be strictly necessary for a computer parsing with a rulebook, but it does help a reader understand exactly what is referred to. Note also that not all years are linked, and regardless it should be able to stand on its own in print. —Centrxtalk • 23:04, 27 July 2006 (UTC)
Also, note that your interpration could allow such strange formations as January 1, 1, which are not appropriate. —Centrxtalk • 00:32, 28 July 2006 (UTC)