Wikipedia talk:Manual of Style/Dates and numbers/Archive 78

New text implemented at MOSNUM, not yet at MOS-central
Please raise unresolved issues here. I know that SMcCandlish will want to present further arguments for dots after imperial units. Question marks over some of the templates (dates of birth and death). I haven't yet inserted anything about four by four, etc (dimensions vs. mathematical operations). Tony 01:57, 30 July 2007 (UTC)
 * Dots are a minor detail that we can beat each other up about at a later date. I wouldn't let that hold back implement this draft. &mdash;MJCdetroit 02:04, 30 July 2007 (UTC)
 * Thanks, MJC. Yes indeed; I'm hoping that when everyone reads through it in full structure, they'll pick up or think of other issues (hopefully minor ones!). Tony 02:30, 30 July 2007 (UTC)
 * I'll live. I don't really have any "further arguments" to present. Seems pretty clear to me. There isn't any real-world evidence in support of no-dotting non-metric/scientific units, except in scientific writing. Not sure what else to say.  The MOSNUM can continue treating WP like it was Nature, but I think what will happen is that most editors who normally use periods after such units will continue to do so, and that the recommendation in MOSNUM will not really have much effect, other than leading to editwarring, as MOS sticklers revert it, and others re-revert it because to them it is obviously grammatically incorrect, and so on, round and round. Seen it enough times already... —  SMcCandlish  &#91;talk&#93; &#91;cont&#93;  ‹(-¿-)› 06:27, 30 July 2007 (UTC)
 * PS: The main reason to permit (not require) period after non-met./sci. units is simply strife and timewaste reduction in my book, for the same reasons that we permit US or UK (or whatever) English in articles, and so on. The headaches and hard feelings are not worth the over-standardization. (This is my reply to an earlier thread up there pushing for standardization about all else, as well).  Anyway, I don't have anything further to say on the draft at issue.  I'm happy with some of it, not with other parts, and so it goes. —  SMcCandlish  &#91;talk&#93; &#91;cont&#93;  ‹(-¿-)› 06:45, 30 July 2007 (UTC)
 * Well, the article Inch gives in as the abbreviation right at the top. What more could you want? —Preceding unsigned comment added by Tony1 (talk • contribs) 14:53, July 30, 2007
 * Speaking of our very own articles, you might want to visit Circa, which mentions encyclopedia articles, but does not follow the style of this guideline. Chris the speller 15:03, 30 July 2007 (UTC)
 * I suppose there's no chance of resolving whatever was in dispute about the Geographical coordinates section, is there? Pity to have that fierce dispute sign hanging over our shining new version ... Tony 05:26, 30 July 2007 (UTC)
 * Just an odd thought, could the word tonne for the metric ton be encouraged over the term metric ton please. There are some articles out there that have just got ton written. CR7 11:55, 30 July 2007 (UTC)
 * Sorry, but I think that part of the MOSNUM is well written. The word tonne is completely foreign to most American readers.  Most (even the well educated ones) could think that tonne is just the British way of spelling ton.  The MOSNUM is written in such a way to let those readers know that tonne is actually a metric ton; the term most a Americans would be familiar with. Besides, the MOSNUM already covers not using the word ton by itself anyway. &mdash;MJCdetroit 12:43, 30 July 2007 (UTC)

Doublecheck on comma in date format
Does anyone know where/how to doublecheck on this: Once I came across a discussion that the comma is never needed, because the auto formatting inserts it automatically. I don't know where to doublecheck that, but it fits with my experience, whether logged in or out (user preferences). Sandy Georgia (Talk) 15:19, 30 July 2007 (UTC)
 * either  (US editors) or   (others) will be rendered as either February 17, 1958 or 17 February 1958, according to a registered user's set preferences.
 * See the discussion above, "Omitting commas in Wiki-formatted dates", and there's an even earlier discussion in Archive D4. Chris the speller 15:47, 30 July 2007 (UTC)
 * Thanks; this talk page got a bit out of hand :-) So, the software does insert the commas, but tempest in a teapot.  Sandy Georgia  (Talk) 16:05, 30 July 2007 (UTC)
 * I think the long and short of it is that yes, the software inserts the comma if you don't, but that we're not about to burden people with a ban on its manual insertion. All for the good, too. Tony 13:56, 31 July 2007 (UTC)

"To"/"versus" hyphen, and "through"/"up to" en-dash
This stuff's been changing so much I'm not sure if this is even in there in any form. I think it should be specified that when a dash is a stand-in for "to" or "versus", as in "McDougal lost to Yamamoto, 3-12", it is a hyphen (or what some call a "hypen-minus" for some reason), in contradistinction to the en-dash used in "fl. 205–162 BCE", where is stands for "through" or "up to". En-dashes should emphatically not be used for sports statistics and the like, or the result will be misleading in many cases, implying a range rather than a 1:1 (or 1-1, but not 1–1 :-) comparison. —  SMcCandlish  &#91;talk&#93; &#91;cont&#93; ‹(-¿-)› 18:42, 30 July 2007 (UTC)
 * See WP:DASH. — Aluvus   t / c  23:35, 30 July 2007 (UTC)
 * No, it emphatically should be an en dash; ranges are only one of that symbol's functions. Representing relationships and opposition are others. Sports scores should be presented with them, and not hyphens: 3–12, not the bunched-up 3-12. Tony 00:04, 31 July 2007 (UTC)
 * I'll concede on this one, but only because Chicago 15th ed. recommends it. I strenuously disagree with the logic of it, but I'm outgunned on this one.  I do have to note, however, that I've been editing sports articles here for a year and a half, and have never even once seen sports scores use an en-dash in this way.  Whether Chicago and MOSDASH advise it or not, it is not common usage, and getting people to use it is likely to prove essentially impossible.  I'll try to remember to do it myself, but I doubt this will have any effect. —  SMcCandlish  &#91;talk&#93; &#91;cont&#93;  ‹(-¿-)› 23:26, 31 July 2007 (UTC)
 * Endashes in sports scores has been the norm at WP:FAC for many months, and there's been no problem with acceptance. Sandy Georgia  (Talk) 23:55, 31 July 2007 (UTC)
 * Yup, I've encountered absolutely no resistance when insisting on it. People soon see that it's much more readable. Tony 00:48, 1 August 2007 (UTC)
 * I didn't meany anyone would fight it, they simply won't do it. I've already seen ample evidence of that, if WP:MOSDASH has actually been around a while and saying what it says. —  SMcCandlish  &#91;talk&#93; &#91;cont&#93;  ‹(-¿-)› 09:59, 2 August 2007 (UTC)

outdated.js — script for reformatting date links
While we wait for a server-side update that makes date auto-formatting and date linking independent, I've put together a script that will separate date links from the rest and allow you to specify your own style for them. With this script, "January 1, 2008" and "100 AD" [for example–ed.] will render as black text in articles, or any other format you can apply with CSS.

See User talk:Outriggr/outdated.js for instructions. – Outriggr § 08:00, 31 July 2007 (UTC)


 * Great idea! Thanks. Maybe you could make it understand non-North American date order too. Stephen Turner (Talk) 09:15, 31 July 2007 (UTC)


 * Thanks. It will—the dates mentioned above are simply examples. – Outriggr § 09:20, 31 July 2007 (UTC)
 * Outriggr, if this works, you're a genius! Tony 13:54, 31 July 2007 (UTC) PS Heartily recommended functionality, but I won't take it on yet, coz I need to identify overlinking in FACs and FAR/Cs and insist on remedial action. Tony 14:06, 31 July 2007 (UTC)
 * That's the dilemma, but someone's got to use the thing! With the script, you can reduce the visual impact of date links while still seeing that there are date links, if you so choose. Tony, do you know of any other talk pages related to the date linking problem where I could mention this tool? – Outriggr § 22:48, 31 July 2007 (UTC)
 * Yes, FAC, FAR, and some of the major, active Wikiprojects (don't know how to identify them, but they're there). Bill it as a new functionality for those who are sick of blue spattering all over WP's text from overlinking. I suppose you've seen my change-colour solution (Alvurus's, acutally). Tony 00:47, 1 August 2007 (UTC)

Too many large changes?
I'm concerned that there are too many large changes going on at the moment without sufficient discussion. The page may have needed some tidying up, but this is more like a complete rewrite. I can't keep up with everything that's been done, and I'm convinced that changes in meaning have been introduced without being properly discussed here first. Does anyone else share my concern, or am I just not putting in enough work to follow it? Stephen Turner (Talk) 09:13, 31 July 2007 (UTC)

I've posted a reply on Stephen's talk page. Tony 11:41, 31 July 2007 (UTC)
 * Your reply is not at Stephen's talk page yet, but no, I don't share the concerns. Sandy Georgia  (Talk) 11:47, 31 July 2007 (UTC)


 * OK, I guess it's just me then. I wasn't paying close enough attention. Apologies. Stephen Turner (Talk) 09:51, 1 August 2007 (UTC)


 * My confusion was because I thought you were talking about putting a summary of this page on a central MoS page, rather than changing this one. So I skipped over the discussion. Stephen Turner (Talk) 10:26, 1 August 2007 (UTC)
 * It would have been good to have had your input at the time. Pity. Any serious issues? Tony 10:33, 1 August 2007 (UTC)


 * Well, one. I strongly disagree with the removal of the requirement to link full dates. I know you've disagreed with that requirement for some time, but I'm pretty sure there's a consensus for retaining it. I certainly don't think you should change it based on a conversation between about three people. Stephen Turner (Talk) 14:09, 1 August 2007 (UTC)


 * I don't think many of the 88 people who signed the plea for the autoformatting and linking functions to be separated, or at least for a parallel system of autoformatting without linking (thus retro-compatible), would agree with mandatory full-date linking. Since our well-organised and (IMV) highly reasonable request hit a brick wall, I changed my tune on autoformatting. If the system isn't going to be fixed, I can't support a guideline that forces people to use it. When the technical glitch is fixed, I'll willingly support a "mandatory" clause.
 * It's even worse that the autoformatting doesn't allow you to render date ranges without typing out (and reading) the full date twice. That is crude and, frankly, ridiculous (October 11–18, 1977 won't work). It won't allow slashes (the night of January 15/16). While the latter is a minor issue, the former is quite an irritant.
 * And after all that, the autoformatting works only for the vanishingly small proportion of readers who are registered and logged in and who have chosen a date preference. For all others, the original is displayed, so out there in the real world of WP use, almost everyone has learnt to accept the variety of standard date formattings—much as they accept the major varieties of spelling.
 * In the light of these shortcomings, forcing people to use it seems like a navel-gazing exercise, since it works for so few and gives WPian editors a false view of what everyone sees.
 * The new wording doesn't mean WPians won't use the autoformatting—indeed most people will continue to do so.
 * By analogy, technical issues have caused other functionalities to be downgraded in status. We've recently thrown out a proposal to recommend (although not compell) the use of a new template that produces a non-breaking thin space. It would have been just excellent, but silly IE 6 ruined everything by not rendering it properly. This shortcoming sank it completely. Tony 14:32, 1 August 2007 (UTC)
 * I share some of Stephen's concern about a lot of changes, at least to the point of requesting a summary of changes; an experienced editor who cares enough to try to follow these guidelines should not be required to grind through all of this to see if habits need to be changed. Can you put such a summary on this talk page, and consider mentioning it on the project page? I hate to request more work of you after such a Herculean task, but you can always decline. Chris the speller 14:54, 1 August 2007 (UTC)
 * I also agree with Stephen about linking full dates; I don't see a consensus to abandon, either. I don't know where you obtained the statistics to characterize the group who uses date formatting as "vanishingly small", but they at least care how the dates appear, and we should accommodate them. We have had visits from people telling us not to use suspenders (commas) because we have belts, and now you say belts are not required. Pants will fall down. Chris the speller 15:09, 1 August 2007 (UTC)


 * Tony, I understand your point of view, and I respect it; I just don't think it's the view of the majority. And I disagree with your belief that that "[not] many of the 88 people who signed the plea for the autoformatting and linking functions to be separated would agree with mandatory full-date linking". I think you're letting your own point of view influence you here. Speaking for myself, I've been an advocate of separating them for a long time, but I believe it's important to continue linking full dates until they're separated. And my belief is that that's the view of the overwhelming majority of people who've thought about the issue.


 * I feel that you've sneaked in a major change to the MoS which affects the majority of pages, under the cover of a large copy edit. No, I take that back &mdash; "sneaked in" could imply a deliberate deception, which I'm not implying. I don't doubt your good intentions. I just think that the discussion got lost in the large amount of discussion about making a summary for MoS central; only involved three other people, two of whom disagreed with your proposal; only lasted five days; and doesn't reflect a consensus.


 * Stephen Turner (Talk) 17:06, 1 August 2007 (UTC)


 * User:Chris the speller notes in another thread (but in reference to this topic): "Stephen Turner and I question the removal of the requirement to link full dates. As of a few days ago, the guideline was clear that all full dates should be linked. My suggestion is to link them all until there is consensus to drop the requirement." I concur with this. I don't feel strongly that it should be done this way or that way, and do feel strongly that the developers need to fix the underlying problem, but in the interim, I think things on this score should stay as they were. —  SMcCandlish  &#91;talk&#93; &#91;cont&#93;  ‹(-¿-)› 04:05, 2 August 2007 (UTC)


 * The current wording opens thus:

"The WikiMedia software formats full dates, and days and months, according to the date preferences chosen by registered users. This function is activated by inserting double square-brackets, as for linking;..."

Would you tolerate this opening, until the functionality is fixed?

"Full dates, and days and months, are normally autoformatted, by inserting double square-brackets, as for linking. This instructs the WikiMedia software to format the item according to the date preferences chosen by registered users."

Tony 06:19, 2 August 2007 (UTC)

PS List of all of the substantive changes? Phew, you're right, I'm nearly burnt out from the process. I'll try.


 * I've only just noticed this discussion, but I also want to place on record that I strongly object to dropping mandatory linking of full dates. -- Arwel (talk) 06:36, 2 August 2007 (UTC)
 * Just to clarify myself, by "full dates" I mean "day+month+year" and "day+month", but not "month+year". -- Arwel (talk) 11:40, 2 August 2007 (UTC)
 * The revision works for me. Again, in the interim; this really, really needs to get fixed on the developer side so that the problem goes away, as it is a very noxious problem. —  SMcCandlish  &#91;talk&#93; &#91;cont&#93;  ‹(-¿-)› 06:38, 2 August 2007 (UTC)

Viewing this from a European stand point makes the argument for only linking full dates look silly, if the policy is mandatory linking of full day/month/year dates but not partial (day/month) dates then that could lead to European users who have bothered to set their date preferences being confronted with articles where all the linked full dates are in their chosen format but partials are in the format they were entered eg  On the 13 October 1980 the debate started and it finished on October 15. I gather from other discussions on this page that there is/was a request for alterations so that the Wiki software would format but not link dates. Until that happens the MoS for dates should either state that all dates, both partial and full, should be linked or it should state that the same line for English/American English in articles should be followed i.e British date format for British articles and American format for American. As an average user I just want a clear indication about what the accepted format is. - X201 09:33, 2 August 2007 (UTC)
 * That's pretty good reasoning, too. —  SMcCandlish  &#91;talk&#93; &#91;cont&#93; ‹(-¿-)› 09:59, 2 August 2007 (UTC)


 * As I understand it, day/month/year and day/month items are at issue for autoformatting, and months/years, months alone, and years alone are the "partial" dates that can't be autoformatted. Isn't the proposed revision clear on this point? The only issue now is acceptance of the word "normally" in place of "almost always"; I am strongly against forcing people to use a dysfunctional system. "Normally" seems encouragement enough. Tony 09:57, 2 August 2007 (UTC)


 * Sorry, X201, I wasn't clear. Anything with a day and month must be linked. The page was explicit about this until recently.
 * Tony, I'm sorry to have to say this, but I feel you're still imposing your own views on this page, and I'm very close to reverting it. I just can't decide whether to revert the section on linking, or the whole section on dates. You have made some improvements, but a lot of good stuff has been over-abbreviated or omitted too, for example the explanation about why ISO 8601 dates shouldn't be used in prose.
 * Stephen Turner (Talk) 10:12, 2 August 2007 (UTC)
 * You revert and I'll be seriously at war with you: there will be no peace on this page. I've put a HUGE amount of work into this—more than I wanted to—and I won't sit idly by while you crassly undo it without yourself gaining consensus. You absented yourself from continual debate here about the overhaul—for whatever reason—so don't think you can just walk back in after the event and revert. As far as I'm concerned, there was consensus for the changes. I flagged, I planned publicly, I managed compromises, and most of all, I spent many hours fixing up what was, frankly, shitty wording and bad formatting. So pay the courtesy of discussing or butt out completely. I'm angry to be faced with this after all of my work. And take your mock "sorry" and put it where it belongs. Tony 11:16, 2 August 2007 (UTC)
 * I would not support any wholesale revert; this was a long and consensual process. If there's a problem about date linking, pls discuss that separately from the rest of the work (although it's been well covered in the past).  Sandy Georgia  (Talk) 13:55, 2 August 2007 (UTC)
 * I too would be against any wholesale revert of the recent changes. &mdash;MJCdetroit 17:07, 2 August 2007 (UTC)

I apologise, I was out of order in threatening to revert. Stephen Turner (Talk) 20:01, 3 August 2007 (UTC)
 * No problem; I apologise for what was now appears to have been an overly strong reply. I'm keen to debate your concerns, and if I have time, will post a list of substantive changes in the recent overhaul to assist that process. Tony 23:48, 3 August 2007 (UTC)

Autolinking dates?
Concerning WP:MOSDATE - is the "2007-07-31" format no longer valid anymore? How come it doesn't include an example of that, while the rest of the MOS uses that format as examples of correct format? Hong Qi Gong (Talk - Contribs) 14:37, 31 July 2007 (UTC)
 * The "Dates" section says: "ISO 8601 dates (1976-05-12) are uncommon in English prose, and are generally not used in Wikipedia. However, they may be useful in long lists and tables for conciseness and ease of comparison." The thrust of this point is unchanged in the overhauled policy. Tony 14:50, 31 July 2007 (UTC)

Implementing the MoS
I'm adding this in hope it might start a productive discussion -- or at least not be ignored, as much of what I write seems to be -- but after the MoS reworked & all who have contributed are happy with it, how does it get implemented?

If we allow the usual kinds of people to make these changes -- with or without bot help -- there will be an uproar. If you forgot the AD/CE wars, then have a peek at the battling over "Fair use images" war -- letting the MoS become a "license to edit without fear & win conflicts" will only serve to embitter many long-term productive Wikipedians & drive many other productive ones away. Yet if it is not somehow implemented, is there any purpose to all of this work?

There must be some kind of soft, gradual process to implement this MoS that doesn't unnecessarily ruffle feathers. Education is one method: writing a regular column to the Signpost would be one way to do this. Engaging people is another: perhaps asking on Talk pages or at WikiProjects if the proposed changes could be made. This runs the risk that the more people who know about this work, the more likely much of what has been hammered out will go back to the furnace & a new consensus made.

But some form of a gradual, soft strategy should be adopted, otherwise I suspect that there will be months of even more hostility than usual on Wikipedia. -- llywrch 21:58, 31 July 2007 (UTC)


 * Very good advice. Those who participate in implementation are, of course, skilled diplomats, yes? Bad idea to continue with mass changes against significant local resistance. Bring it up here or at MOS-central. Tony 00:44, 1 August 2007 (UTC)

Old Style and New Style dates
See Talk:Old Style and New Style dates --Philip Baird Shearer 23:59, 31 July 2007 (UTC)

Now implemented at MOS-central
I've copied over much of Sections 1–6. In the process, I realised that the stupidly named "Long short big small" at MOS-central belongs within the summary, and here (see vagueness), as does some of MOS-central's "Scientific style". All that remains of the old Scientific Style section at central is a short section on chemistry at the bottom of central.

I'm now adding an invisible note at the top of MOSNUM about the need to coordinate changes with MOS. In the next week, I'll change as many surface details as possible in the examples here, so that readers are forced to process the same guidelines anew; that will be good for both their understanding and the separate identity of this submanual. Tony 01:46, 1 August 2007 (UTC)

Autoformatting and linking
The Autoformatting and linking section states how linking works and what it is for, it states what you should not link but it doesn't explicitly state that you should either link every date on a page or that you should only link certain dates. I've read through the discussion archives and they have left me non-the-wiser with some users saying link everything and some complaining about over linking. What is the MoS standpoint on linking dates? - X201 18:19, 1 August 2007 (UTC)
 * We're talking about full dates, then? And by "link", I presume you mean "autoformat"? The wording indicates that you don't have to use the autoformatting facitility: people aren't forced to do this. Most people probably still will. Do you have a specific suggestion for additional text? Tony 00:59, 2 August 2007 (UTC)
 * In the above section "Too many large changes?", Stephen Turner and I question the removal of the requirement to link full dates. As of a few days ago, the guideline was clear that all full dates should be linked. My suggestion is to link them all until there is consensus to drop the requirement. Chris the speller 02:41, 2 August 2007 (UTC)

I've replied in the "Too many large changes?" thread, but to clear a few things up... I meant autoformating. I was referring to both full (day+month+year) dates as well as partial (day+month) dates but I failed to explain that my main concern was where partial and full dates appear in the same sentence, with full dates linked/autoformatted but partial dates not linked/formatted. - X201 09:41, 2 August 2007 (UTC)

BC/AD and BCE/CE, again
The following long standing material was removed from the article. This perspective was based on several long and acrimonious discussions and at least one RFA, and many editors have strong opinions on the use of both systems. I have already noted some recent confusion on changing dating systems on some articles. With this permanently removed, I would expect the issue to return with a vengeance. Opinions on replacing this in the policy or clarifying the policy in another manner? Best wishes. WBardwin 19:46, 1 August 2007 (UTC)
 * 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 British spelling as opposed to American spelling it would be acceptable to change from American spelling to British spelling if the article concerned a British 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. See also Requests for arbitration/Jguk.

I believe the default position is that if there is any conflict over usage of a term, spelling, etc it automatically goes back to the earliest non-stub. Unless there is a good reason, such as use of American English in a British article. John Smith&#39;s 21:44, 1 August 2007 (UTC)

This was not included because it applies generically in WP articles where a valid choice is first made. MOS-central says this. Tony 00:57, 2 August 2007 (UTC)
 * And we should say so as often as convenient. Stopping the stupid date wars is more important than removing redundancies; the assumption that every editor will have absorbed the entire MOS is unrealistic.. I will restore if no one else has done so.

Rewording of the Era section
Does anyone have a problem with rewording the section on BC/AD and BCE/CE to this:


 * Either CE and BCE or AD and BC can be used—spaced, undotted and upper-case—to specify the Gregorian era. Be consistent within the article. AD appears before or after a year (AD 1066, 1066 AD); the other abbreviations appear after (1066 CE, 3700 BCE, 3700 BC). The absence of such an abbreviation indicates the default, CE or AD.
 * When either of the 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.
 * Year ranges, like all ranges, are separated by an en dash (do not use a hyphen or slash: 2005–08, not 2005-08 or 2005/08). A closing CE/AD year is normally written with two digits (1881–86) unless it is in a different century from that of the opening year (1881–1986). The full closing year is acceptable, but abbreviating it to a single digit (1881–6) or three digits (1881–886) is not. A closing BCE or BC year is given in full (2590–2550 BCE). While one era signifier at the end of a date range still requires an unspaced en dash (12–5 BC), a spaced en dash is required when a signifier is used after the opening and closing years (5 BC – 29 AD).
 * A slash may be used to indicate regular defined yearly periods that do not coincide with calendar years (the financial year 1993/4).
 * Abbreviations indicating long periods of time ago—such as BP (before present), Ma and mya (million years ago), and Ga (billion years ago)—are spelled out on first occurrence.
 * To indicate about, c. and ca. are preferred to circa or a question mark, and are spaced (c. 1291).
 * Decades contain no apostrophe (the 1980s, not the 1980’s); the two-digit form is used only where the century is clear (the ’80s or the 80s).


 * Centuries and milliennia. There was no year 0. Thus, the first century AD/CE was 1–100 AD/CE, the 17th century AD/CE was 1601–1700 AD/CE, and the second millienium AD/CE was 1001–2000; the first century BC/BCE was 100–1 BC/BCE; the 17th century BC/BCE was 1700–1601 BC/BCE, and the second millennium BC/BCE was 2000–1001 BC/BCE. Note: Writers should choose one system or the other, but not both. It is shown in this explanation with both to express that the MOS does not favor one method over the other.

I took out the superfluous part about avoiding the Christian associations of BC/AD. All terms are linked within the section if an editor needs to know more about what the terms mean. I included the inappropriate to change from one style to the other...part because without that clause, an editor with a personal agenda WILL go through and start changing the era style in as many articles as they can. I changed the Centuries and milliennia part to include both methods; instead of just CE/BCE. I also included a note to pick one method or the other, but not both.

As it is written now, it tends to encourage writers to use CE/BCE and does not prohibit the changing from one method to another for personal/unclear reasons. Therefore, I tried to reword it as to be as neutral as possible and making it clear that the MOSNUM/MOS does not favor one over the other. &mdash;MJCdetroit 16:59, 3 August 2007 (UTC)

Sorry that I was unaware and did not participate in earlier discussions. I am particularly missing two things from the live version. As noted above, a statement discouraging changing between era notations should be restored. Also, I miss the part that says Normally you should use plain numbers for years in the Anno Domini/Common Era, but when events span the start of the Anno Domini/Common Era, use AD or CE for the date at the end of the range.-Andrew c [talk] 18:08, 3 August 2007 (UTC)
 * I fully agree. We have had enough disruption from idiots "correcting" dating styles in both directions. It's a good thing, by the way, that this text does not attempt to prescribe the placement of AD with respect to the number. The encouragement to be clear about the transition from BCE/BC to AD/CE is also needed. Septentrionalis PMAnderson 08:20, 10 August 2007 (UTC)

Problems: (1) "When either of the 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." As I said above, MOS clearly states in its first lead paragraph the very same thing, so that specific options do not need to cover this principle. To do so casts doubt on the global application of the principle. It should be removed.

(2) The same applies to this: "Note: Writers should choose one system or the other, but not both. It is shown in this explanation with both to express that the MOS does not favor one method over the other." And let's not tell them to "note" this: they should note everything in MOS. The last sentence needs to be reworded. Similarly, someone has inserted "Remember that in ..." in MOS-central's mirror points concerning "Centuries and milliennia" (did I misspell it?). Tony 09:34, 10 August 2007 (UTC)
 * I respectfully disagree with your opinion of (1). The third paragraph of theMOS eludes to the American/British styles of English and leaves a "grey area" for things as this.  There are more passionate feelings involved with BC/AD vs BCE/CE than whether something is spelled program or programme.  Therefore, it doesn't hurt to reiterate the fact that it is inappropriate to change the established style of an article for personal reasons.  I placed that sentence in there because I've seen the mass changing of styles before; even when there was a similar sentence in the MOSNUM.  This extra sentence makes it clear that's it is not ok to change the eras because you want to and it gives editors more "ammunition" to revert such PPOV edits.


 * As for (2), it looks as if it was rewritten since I introduced it anyway and I don't have any objections to its present revision. &mdash;MJCdetroit 17:33, 10 August 2007 (UTC)


 * Please tell me what is unclear about this statement, glaring out from the lead of MOS:


 * "In June 2005, the Arbitration Committee ruled that, when either of two styles is acceptable, it is inappropriate for an editor to change an article from one style to another unless there is a substantial reason to do so ... Edit warring over optional styles is unacceptable". What more could one want? The variant spellings are given explicitly as an example of this principle, not, as you're assuming, as the only application of it. It weakens the global force of this statement to restate it locally. No one should have concern that they lack the backing of MOS to revert wholesale changes from one option to another without substantive reason. There are so many options permitted by MOS/NUM; are we to restate this basic principle for all? Should we, then, restate the "doesn't apply to direct quotations" for every single point, too?

I draw your attention to another statement in the lead; this embodies two further principles that should not be restated for specific points: "An overriding principle is that style and formatting should be applied consistently throughout an article, unless there is a good reason to do otherwise, except in direct quotations, where the original text is generally preserved." Tony 05:25, 11 August 2007 (UTC)


 * I disagree. Adding an additional statement in the MOSNUM regarding eras doesn't weaken the global statement.  It reiterates it with a specific example of something inappropriate that is done very often regardless of what the MOS says.  This only strengthens the global statement and makes it very clear that changing era styles is inappropriate.  Remember, not everyone who edits Wikipedia has read the MOS and its submanuals top to bottom; I know I haven't.  Therefore, if we make it easier for the beginning editor to find an answer to the very common problem of switching eras, then I think we've improved the MOS/MOSNUM. &mdash;MJCdetroit 16:14, 11 August 2007 (UTC)
 * As above. Septentrionalis PMAnderson 23:16, 12 August 2007 (UTC)

Under Longer periods, the sentence "Either CE and BCE or AD and BC can be used—spaced, undotted and upper-case—to specify the Gregorian era." should not state Gregorian because Julian dates are preferred before 15 October 1582. "Gregorian" should be deleted, leaving "to specify the era." — Joe Kress 05:02, 17 August 2007 (UTC)

In the discussion of AD/BC, BCE/CE, and, in particular, c., we need to clarify that they are not italicised (except where the whole date is italicised). The examples, because they are examples, are usually in italic. I've just had a minor edit war over "c." (Kirkstall Abbey) in which the other editor says "Checked MOSNUM - italicised c. appears acceptable. In fact, irrespective of Wikipedia, it is proper to italicise any term borrowed from another language or any abbreviation derived from such a term". PamD 23:19, 27 August 2007 (UTC)

MYA or mya
MOSNUM says MYA, but the article is mya (unit); I've always thought it was mya. Which is it? Sandy Georgia (Talk) 23:14, 1 August 2007 (UTC)
 * To further complicate the question, Talk:mya (unit) says the international standard is Ma, not mya or MYA. Sandy Georgia  (Talk) 23:31, 1 August 2007 (UTC)
 * (edit conflict) I have always seen Ma (which stands for Mega annum, with the "ago" being implicit) in academic literature, but find mya to be more intuitive. --JayHenry 23:37, 1 August 2007 (UTC)


 * I always use "Ma" (or ka or Ga) when writing for the technical literature. To the extent it does appear, I think "mya" (lowercase) is more common than uppercase.  Dragons flight 00:04, 2 August 2007 (UTC)


 * And the Ma article says mya is deprecated. Sandy Georgia  (Talk) 00:28, 2 August 2007 (UTC)


 * A quick search of google found this page from the University of California.  Which isn't diffinitive, but does use mya.  --Rocksanddirt 00:40, 2 August 2007 (UTC)


 * It sounds like MYA is out. But there's a proposed merge of mya (unit) to Ma.  Does anyone know how to track down SI on this?  Sandy Georgia  (Talk) 00:43, 2 August 2007 (UTC)


 * The official SI standard of time is, of course, the second. Dragons flight 00:50, 2 August 2007 (UTC)


 * I much prefer Ma if it still matters at this point. Mya is kind of amateurish (trying not to sound snobby and failing). Sheep81 01:46, 2 August 2007 (UTC)
 * However, when it is used, I think I always see "mya" uncapitalized. Sheep81 01:49, 2 August 2007 (UTC)

This says Ma is not SI but is in common use. Like DragonsFlight, to me Ma and Ga or ka is totally standard in geologic usage and is increasingly common in more popular literature, when the phrase "million years ago" is not written out. I have felt (with no hard data) that "mya" was something of a Wikipedian construct. Cheers Geologyguy 01:49, 2 August 2007 (UTC)
 * I first picked up mya and "million years ago" from a book back in my youth (1980s). What it has going for it is the meaning is easier to grasp when written out.  In practical terms, I've never committed to one abbreviation on Wikipedia because of this very issue, preferring just to write out "million years ago". J. Spencer 02:18, 2 August 2007 (UTC)

Do any of you, then, have an opinion about what our Manual of Style should say? Do we say no MYA? Either Ma or mya? Or do we stay silent? Sandy Georgia (Talk) 02:33, 2 August 2007 (UTC)


 * Well, we should at least eliminate MYA from the MOS, since there is no consensus whatsoever for that one. My issue with Ma or Ga is that the casual reader may miss what it means, whereas mya can be guessed at rather easily.  Never thought I'd participate in this type of discussion!   Orange Marlin  Talk• Contributions 04:07, 2 August 2007 (UTC)
 * If the academic style is going towards MA that's fine with me (though i like mya). --Rocksanddirt 04:43, 2 August 2007 (UTC)
 * But it's Ma, not MA, right? Sandy Georgia  (Talk) 04:59, 2 August 2007 (UTC)


 * FWIW, I think I would count as a "casual reader", and I've never heard of any of these terms, nor do I think I could pick up the meaning of any of them purely from context. "Mya" is easier to grasp after someone tells you what it means, but I can't imagine just guessing its meaning. —  Aluvus   t / c  05:07, 2 August 2007 (UTC)


 * The current wording is:

"In articles on prehistoric topics, BP (before present) and MYA (million years ago) are spelled out on first occurrence."

As long as a standard form of abbreviation is used consistently in an article, and it is spelled out on first occurrence, I can't see the problem in being more inclusive. The current guideline, in any case, doesn't say that BP and MYA are the only forms permissable. So why not this simple addition of three words (which I've italicised here for clarity)?

"In articles on prehistoric topics, abbreviations such as mya (million years ago), Ma (Mega annum) and BP (before present) and are spelled out on first occurrence."

Tony 05:27, 2 August 2007 (UTC)


 * It implies that MYA is the common usage, when in fact mya and Ma are, while MYA doesn't appear endorsed by anyone (see OrangeMarlin's current FAC on Cretaceous–Tertiary extinction event). Sandy Georgia  (Talk) 05:33, 2 August 2007 (UTC)
 * Oops, I didn't read the foregoing discussion properly. How's that? Do we need BP? Both existing abbreviations, BTW, were unchanged in the recent overhaul: I didn't know any better. Tony 05:56, 2 August 2007 (UTC)
 * I'd just never abbreviated it nor seen any abbreviations before using WP, where I saw mya alot and now Ma. I don't care for any of them too much as I feel none is well known to the common reader, but my opinion is not too strong on this one. cheers, Casliber (talk · contribs) 07:24, 2 August 2007 (UTC)
 * I'm going to stick with mya in the article, since I think it's mostly clear. I'd be concerned if I had to spell it out in this article, since it's used so many times, it might make it look pretty trite.  I still think MOSNUM needs a revision to remove MYA and move to either Ma or mya.   Orange Marlin  Talk• Contributions 08:49, 2 August 2007 (UTC)
 * I'm a geosciences student and I only ever see Ma, and occasionally Ga in my course texts (and one of them was published this year.) saying that why not just write it as "X million years ago" as that is the least ambiguous way of all... no need for any interpretation then! --Vertilly 08:57, 2 August 2007 (UTC)
 * Heavily concur. These abbreviations are too specialist for use in a general-public encyclopedia, unless explained in context, e.g. "2.5 million years ago (Ma) ... Then, 1.4 Ma, ..." Abbreviation okay if both explained and linked to earlier. —  SMcCandlish  &#91;talk&#93; &#91;cont&#93; ‹(-¿-)› 09:09, 2 August 2007 (UTC)
 * MYA is not used in geological literature, it is always Ma, as it is Ga and Ka (Mega, Giga and Kilo Annum). In the Dutch wikipedia Ka, Ma, and Ga are used and explained in a lemma of their own, and if appropriate in the text where this is used. So I wonder why MYA has been here anyway. I assume that MYA is used by people who are ignorant of the geological literature. --Tom Meijer 09:23, 2 August 2007 (UTC)
 * Not entirely a geological issue at all. Biology/zoology and anthropology use some form of this as well; I'd be curious what the current lit uses (I'm used to stuff from over a decade ago in my university days, so I can't really say for sure). —  SMcCandlish  &#91;talk&#93; &#91;cont&#93; ‹(-¿-)› 09:59, 2 August 2007 (UTC)


 * (Outdent) So the proposed revision looks to satisfy these concerns. I've added Ga and Ma and removed BP; all are now linked here. Shall I implement it?

"In articles on prehistoric topics, abbreviations such as Ma and mya (million years ago) and Ga (billion years ago) are spelled out on first occurrence." Tony 09:31, 2 August 2007 (UTC)


 * What about ka/Ka? I've seen it spelled both ways in this discussion, but can't think of a reason that it would be "ka, Ma, Ga", given "KB, MB, GB", not "kB, MB, GB". Otherwise, no issues from my corner. Take that back, one issue: I'd change "Ga (billion years ago)" to "Ga (giga-annum or billion years ago)" —  SMcCandlish  &#91;talk&#93; &#91;cont&#93; ‹(-¿-)› 09:59, 2 August 2007 (UTC)


 * BP is absolutely standard for time periods less than a few thousand years, as comes up in paleoclimatology and carbon dating, and shouldn't be removed from the list of acceptable options. Dragons flight 10:04, 2 August 2007 (UTC)
 * OK, but does this have to be a comprehensive list? Why not just mention one or two? It does, after all, say "such as"; it's proscribing nothing, is it? I was probably unwise to add more to my proposal. This whole point is about "prehistoric topics", which doesn't include less than a few thousand years ago. Now it's becoming chaotic. Tony 11:19, 2 August 2007 (UTC)

I think the proposed revision covers all examples, while removing the incorrect MYA. Sandy Georgia (Talk) 12:43, 2 August 2007 (UTC)
 * I miss apparently the point of this entire discussion. Ga, Ma, Ka, BP and AD, all are standard abbreviations and normally used in literature (MYA is not), as far as I know not only geological (although, maybe not surprising, I consider geology as the most important user here). Shouldn't these conceptions be explained in an encyclopedia anyway? BP is before 1950 AD (so not before now) and is the common used expression in Carbon Fourteen dates, even if this is 30.000 year (I agree, this hardly makes sense, but it is an international standard), AD is after year zero from the Christian calender, etc. I don't understand why you shouldn't use these abbreviations that are already standard. As far as I am aware, the standard is also Ma, Ga, and Ka, and not MA, ma, mA, etc. There maybe not a proper 'reason' for this, it is just the standard (and that is apparently the reason)--Tom Meijer 12:46, 2 August 2007 (UTC)
 * I agree entirely with Tom Meijer. Ga, Ma, Ka are both widely used in geology and archaeology and are the official units/designations as approved by relevant professional societies, eg IUGS. Babakathy 12:25, 3 August 2007 (UTC)
 * You're a bit late. Tony 13:06, 3 August 2007 (UTC)

Changes and strategy
I've been bold and changed the autoformatting phrase from the current wording, which clearly had little support, to the wording that appears above (which did receive support from SMcCandlish, and is, is suspect, much closer to what Stephen et al. are comfortable with). Can we please continue discussion of this here, rather than threaten reverts? I'm keen that we keep the current text and work to address concerns one-by-one. Stephen has raised a problem with the ISO point. Let's gain consensus here. I've implemented the suggested changes to the "prehistoric abbreviations" point.

Another issue is that Crissov is unhappy with the amount of detail that has been transferred across to MOS. (See talk page there.) There's no reason that we can't negotiate the removal of a few points from that text, while retaining them here. I have a few in mind. Tony 00:47, 3 August 2007 (UTC)

List of substantive changes to MOSNUM in recent overhaul

 * I've done a fairly quick job, and hope that it's accurate; please edit it you find omissions or commissions.


 * Please add notes or comments in square brackets to the list, as I have in a few places, and sign. Important issues probably need to be raised below on this page.


 * It's categorized broadly by section, in terms of "Removed", "Added" and "Changed", but the distinctions between them are blurred in places.


 * I discovered that two mini-sections on numbers had been omitted (Natural and Non-base-10 numbers). I've copy-edited them and added here (but not at MOS).


 * "Magnitude prefixes" and the still-messy "Geographical coordinates" need to be filled in in the list.

Tony 01:05, 4 August 2007 (UTC)


 * Very nice; it helps me, especially as a sanity check, is bound to help editors who have not been following closely for the last few weeks. Check my slight tweak. Thanks, Tony! Chris the speller 01:50, 4 August 2007 (UTC)


 * I restored two items to MOSNUM (but not MOS) that were excised but which I believe are crucial (deprecate "17th Century", and "1700s" is a decade not a century in WP terminology). —  SMcCandlish  &#91;talk&#93; &#91;cont&#93; ‹(-¿-)› 19:32, 5 August 2007 (UTC)
 * I'm most concerned that many people take the 1700s to mean a century. Why is WP different? Tony 23:58, 14 August 2007 (UTC)

"The '60s" change proposed
I propose changing this:
 * Decades contain no apostrophe (the 1980s, not the 1980’s); the two-digit form is used only where the century is clear (the ’80s or the 80s).

to this:
 * Decades contain no apostrophe (the 1980s, not the 1980’s), and are not abbreviated except in contexts where the abbreviated form, with a leading apostrophe, is itself a culturally significant term, as in  ' 60s counterculture and the Roaring '20s, but not  ' 60s music or neo-conservatism of the '80s.
 * Decades contain no apostrophe (the 1980s, not the 1980’s), and are not abbreviated except in contexts where the abbreviated form is itself a culturally significant term, as in 60s counterculture and the Roaring 20s, but not 60s music or neo-conservatism of the 80s. (A leading apostrophe, as in  ' 60s, is superflous, as it would be in  ' phone, and can interfere with quotation and bold/italic formatting.)
 * &#91;proposal revised per Tony1's objection below. —  SMcCandlish  &#91;talk&#93; &#91;cont&#93; ‹(-¿-)› 01:34, 6 August 2007 (UTC)&#93;

The reasoning I hope is obviously, but to spell it out anyway: Abbreviation like the latter two examples is simply sloppy and unencyclopedic. Also, there is no reason to permit leaving off the leading apostrophe in the rare cases that '60s is appropriate; "60s" is an age range, not a date range. —  SMcCandlish  &#91;talk&#93; &#91;cont&#93; ‹(-¿-)› 19:25, 5 August 2007 (UTC)
 * Do you use a leading apostrophe for 'phone, still? Or 'plane? It's a particular problem when it starts a quote, too. The whole idea is that if the context is clear, it's very recognisable (preferably after a decade has already been spelled out in full. I hate the look of the leading apostrophe; again, it's not about sloppiness or laziness, but a smooth read for the public. I'm away until Saturday. Tony 00:58, 6 August 2007 (UTC)
 * You've convinced me completely, actually. Revising proposal accordingly. —  SMcCandlish  &#91;talk&#93; &#91;cont&#93;  ‹(-¿-)› 01:34, 6 August 2007 (UTC)
 * Actually, the examples obscure proper usage. They should preferably not express the decades as numbers, but rather as spelled-out words:  the Sixties counterculture or the Roaring Twenties.  Likewise, the counter-examples of 'phone and 'plane are obtuse; these are proper nouns in of themselves.  The reason for the leading apostrophe in abbreviated decade identities is that they are missing information, as are other forms of contraction – and contractions properly require an apostrophe.  Yes, a rendering of ’60s is ugly (especially in Arial font), but so is 60s (which normally is used to refer to an age range, not a calendar decade).  Since Wikipedia eschews the use of contractions (outside of quotations), we should discourage such use with decades as well – or punctuate them accordingly. Askari Mark (Talk) 22:46, 10 August 2007 (UTC)

I agree with the proposal, as amended. Using two digits for years, instead of four, cost the world many billions of dollars (that's 1,000,000,000s) in software corrections as the '90s wound down and we approached 2000. Oops, I mean the 90s. Oops, oops, I mean the 1990s. ;-) Chris the speller 01:03, 11 August 2007 (UTC)

Fix misuse of italic emphasis
This document is misusing italics rather massively. Italics are a form of emphasis. Example text should be given in "quotes", rather than emphasized. Aside from simply being weird style, it makes it very difficult to actually emphasize anything without going bold (which is generally too heavy-handed), since everything italicized for emphasis just looks like more example text. —  SMcCandlish  &#91;talk&#93; &#91;cont&#93; ‹(-¿-)› 19:28, 5 August 2007 (UTC)
 * See "Words as words" in MOS. If you want to change it, propose at MOS talk. The only way I could see of interpreting that rule was to italicise nominal groups (nouns and noun phrases), but to treat larger chunks as quotes. I pleaded for advice at MOS talk several times, to no avail. Needs to be addressed globally, among all MOSs. Tony 01:01, 6 August 2007 (UTC)


 * Done. —  SMcCandlish  &#91;talk&#93; &#91;cont&#93; ‹(-¿-)› 01:47, 6 August 2007 (UTC)