Wikipedia talk:Manual of Style/Dates and numbers/Archive 79

Proposed clarification regarding comma usage in American dates
The article does not make clear that, in the American month-day-year usage, the year is set off with commas. One comma between the day and year, and one comma after the year (unless some other puncutation follows the year).

See Chicago Manual of Style, Section 6.46:

"In the month-day-year style of dates, the style most commonly used in the United States and hence now recommended by Chicago, commas are used both before and after the year. In the day-month-year system—sometimes awkward in regular text, though useful in material that requires many full dates—no commas are needed. Where month and year only are given, or a specific day (such as a holiday) with a year, neither system uses a comma."

This problem contributes to the widespread misunderstanding that a phrase such as "September 11, 2001 attacks" is appropriate. (The Wikipedia article on 9/11 is incorrectly punctuated in accord with that misunderstanding; it should be "September 11, 2001, attacks.) Similar mispunctuations are ubiquitous on Wikipedia.

Let's start the process of fixing that by correcting this Manual of Style first. Thanks! Lowell33 16:25, 10 August 2007 (UTC)


 * Thought I would add some examples of the correct punctuation of September 11, 2001, for illustration:


 * The 9/11 Commission Report, Preface at xv: "September 11, 2001, was a day of unprecedented shock and suffering in the history of the United States"; New York Times, June 24, 2006, A3: "majorities in Egypt, Indonesia, Jordan and Turkey . . . said, for example, that they did not believe that Arabs had carried out the Sept. 11, 2001, attacks in the United States"; New York Times, June 19, 2006, A11: "The United States said a small cell of Al Qaeda, made up of foreigners, had set up shop in Mogadishu after the Sept. 11, 2001, attacks and were being protected by court leaders"; New York Times, April 30, 2006, 44: "Mr. Deutch also began to require special approval for the use of unsavory characters as agency informants -- a policy suspended after the Sept. 11, 2001, attacks, when officers argued that only terrorists would know of plans for the next attack."


 * Lowell33 21:54, 10 August 2007 (UTC)


 * Having heard no objection, I added the proposed clarification. Lowell33 21:33, 13 August 2007 (UTC)
 * I removed the change, as no other editors have agreed to it. In many cases, silence implies consent, but silence in WP is not necessarily consensus. This style seems a bit stiff to me, but that's not my main objection. As dates without a comma after the year "are ubiquitous on Wikipedia", as you say, I am not anxious to open the floodgates. The Chicago Manual of Style may call for a comma, but that doesn't make Wikipedia's style wrong, so it doesn't need a "process of fixing", as WP can set its own style. I didn't respond earlier because I didn't want to lead the opposition; it seems like that's all I get to do lately, as just about every other day brings a proposal to turn Wikipedia on its head. Nothing personal, but let's not charge ahead on this until we hear from some other editors. Chris the speller 01:01, 14 August 2007 (UTC)
 * Maybe I could have explained the rationale better. It's not just because the Chicago Manual says so; it's because the year in the American day-month-year format is a parenthetical element, just like the "Illinois" in "Chicago, Illinois, is a nice town." See also Purdue University Online Style Guide ("July 22, 1959, was a momentous day in his life."). I understand that "WP can set its own style," but it should still follow basic rules of grammar, right?Lowell33 14:51, 14 August 2007 (UTC)
 * I don't like it. Tony 15:11, 14 August 2007 (UTC)
 * No offense, Tony, but that's because you've gotten so used to seeing it the wrong way that the correct way doesn't look right any more. I know it's a very common mistake, but it's a mistake nonetheless. See also the University of Wisconsin School of Law Style Guide ("The year should be set off by commas when a complete date is given: He always said that February 8, 1990, was the most important day of his life"); another guide  ("Classes begin Monday, Sept. 2, 2003, at the high school. [Note commas after the day of the week and the year.]"); Blue Book Guide to Punctuation, Rule 5a  ("Use a comma to separate the day of the month from the year and after the year. Example: Kathleen met her husband on December 5, 2003, in Mill Valley, California."); University of Texas Style Guide  ("When a phrase is used with a month, date and year, set both the date and year off with commas."); University of Minnesota Style Manual  ("Use commas around the year when it follows a specific date . . . . The committee agreed on December 12, 1979, as the next meeting date."); Rutgers Editorial Style Guide  ("If the day is included, the year is set off by commas before and after. . . . The events of April 18, 1775, have been celebrated in song and story."); Clemson University Editorial Style Guide  ("When writing a date, place a comma between the day and the year as well as after the year. January 23, 2003, was cold and snowy."). Lowell33 15:22, 14 August 2007 (UTC)


 * I'm not at all convinced that a comma after the year needs to be enforced, or even mentioned. In Lowell's example, the first post-year comma is required to mark the end of a nested phrase; the second is good because it separates to numerical items. The third and subsequent instances are within direct quotes, so cannot be touched (I wouldn't have used them, especially the last one, which is silly—bump, bump, hiccup, bump). Tony 01:25, 14 August 2007 (UTC)
 * I'm not at all sure what a "nested phrase" is or how the date in the sentence "September 11, 2001, was a day of unprecedented shock and suffering" is "nested" in any way. It's a noun, and the year is set off by commas because it is parenthetical. Regarding "the second," I think you're referring to my citation. Regarding the "third and subsequent" instances, I don't understand what you mean by "cannot be touched." I was just using them as examples. Finally, I understand that you "wouldn't have used them" and think they're "silly," but do you have any reasons for that?Lowell33 14:51, 14 August 2007 (UTC)
 * The "direct quotes" are examples used by those sources. Offsetting years (at least in American-style dates) with commas on both sides is correct per every English teacher I have ever had.  Omitting the comma after the year is incorrect.  Lowell33 is right. —  Aluvus   t / c  22:40, 14 August 2007 (UTC)


 * These sources are all American, and Lowell is defining "right" and "wrong" purely in terms of Amercian practice. Everywhere else, commas have been dropped from dates, starting by moving the day away from the year (17 January 1903). It's not an American project, but international. Tony 23:57, 14 August 2007 (UTC)
 * Tony, I agree completely that no commas are necessary in the day-month-year format, but I don't see why that would lead us to omit a necessary comma from the month-day-year American format. I also hope I didn't offend with talk of "correct" and "incorrect" punctuation. It's nothing personal, but, in my view, it affects Wikipedia's credibility to keep repeating this punctuation error everywhere. You won't find it in mainstream encyclopedias, American or not. See Encyclopedia Britannica Online ("The second of two commercial jetliners hijacked by al-Qaeda terrorists and crashed into the World Trade Center in New York City on September 11, 2001, approaches the building, while smoke billows from the crash of the first airliner."); Britannica Concise  ("Working in small groups, the hijackers boarded 4 domestic airliners in groups of 5 (a 20th participant was alleged) on Sept. 11, 2001, and took control of the planes soon after takeoff.").Lowell33 14:35, 15 August 2007 (UTC)
 * For American-style dates, American practice would be rather important to consider. Internationally, dates of that format have both commas.  Dates of other formats, such as the day-month-year format seen in Europe and elsewhere, are a separate case. —  Aluvus   t / c  23:00, 15 August 2007 (UTC)
 * We willingly accept both American and British English; we should use either as their speakers do; a hybrid style evolved from sterile doctrine is less comprehensible to either division of the common language. Septentrionalis PMAnderson 23:28, 15 August 2007 (UTC)
 * Thanks for input, Septentrionalis, but I didn't follow. What is the "hybrid style" that you're opposed to?Lowell33 02:23, 16 August 2007 (UTC)
 * Yes, well most of what Anderson writes is hard to decipher. I did understand his calling me "ignorant" and "illiterate", though. I fear that we're heading towards a nasty war. On the issue at hand, Lowell, can you advise whether it's now mandatory to insert a comma after all autoformatted dates? I don't see how the final comma can be included without its inclusion in other date-forms (that would be very unsatisfactory). Tony 03:19, 16 August 2007 (UTC)
 * I don't know enough about autoformatting to answer that question. I'm thinking it would be easiest to get a consensus on the correct punctuation first.Lowell33 03:46, 16 August 2007 (UTC)
 * Well, our loose-cannon friend Anderson went ahead and inserted his own sloppy version of your proposal; I've fixed it, but if it's the case that all autoformatted dates will be displayed with a comma where it's applied to just (original) US ones, I think that's a serious problem. I've checked it below, and this is the case. [Later: I'm removing the clause from the Manual until this problem is sorted out.] Of course, if I had my way, we'd ditch the autoformatting for a whole bunch of reasons. But that doesn't seem to be on the cards.

1 January, 1959, stands out as ...

Tony 04:33, 16 August 2007 (UTC)
 * Rather than wait for consensus, and in spite of the limitations of Wikimedia's formatting, Anderson reinserted the faulty instruction into the manual. I removed it again. Chris the speller 17:39, 16 August 2007 (UTC)
 * And again. The manual should not be changed until problems are identified and dealt with. Chris the speller 18:01, 16 August 2007 (UTC)


 * This may not be something that can be autoformatted. Another problem is that, where another punctuation mark follows the year, there should be no comma. For example, a birth/death date should be punctuated as "January 1, 1930--August 2, 1965." Note that there should be no comma after 1930 because there is a dash, and there should be no comma after 1965 because there is a period. Is autoformatting sophisticated enough to recognize situations like that? I doubt it, but like I said I'm not up to speed on autoformatting. Lowell33 14:52, 16 August 2007 (UTC)

At the top of this manual, it states "These guidelines have been developed through the consensus of many editors and should be followed in Wikipedia articles. Before making major changes, please raise them on the discussion page to ensure consensus"

In light of that, I would like to point out that it is wrong for one editor to change the guideline before there is consensus. The manual is not the place for any one editor to offer general suggestions, or to insert directions that differ from what a consensus of editors has previously built on that page. Let's determine what a consensus of editors think is the desired style, and then determine how close we should try to come to that style, given the limitations of the Wikimedia software. Until we do both, and figure out exactly where we're going and how to get there, let's not dispense instructions to other editors that could disrupt Wikipedia. Chris the speller 18:47, 16 August 2007 (UTC)
 * Thanks, Chris. I agree that the first step should be to reach a consensus on the "desired style" before figuring out how to implement it. I've explained the substantive issue about as well as I can in my posts above. I couldn't find any reliable source advocating a form like "September 11, 2001 attacks," and that form doesn't make sense grammatically. I reiterate my proposal to clarify in the guidlines that the year in the American day-month-year format should always be followed by a comma (unless it is followed by some other punctuation). Thanks! Lowell33 19:13, 16 August 2007 (UTC)
 * Let's not have a "desired style". What Chris proposes here is contrary to policy, which has long since decided that we should not prefer any national variety of English. Speaking of consensus, does anyone else support Chris's preference of the Automagic date conversions over the American final comma? Neither Tony nor I do. 19:17, 16 August 2007 (UTC)
 * I didn't understand your post. Just to clarify, my original post had nothing to do with preferring any "national variety of English" over any other. There is a correct way to punctuate the British day-month-year format and a correct way to punctuate the American month-day-year format. I am talking only about the correct way to punctuate the American month-day-year form, in which the year is parenthetical and should be set off by commas before and after (unless some other punctuation follows the year). Thanks Lowell33 19:51, 16 August 2007 (UTC)
 * Then I misunderstood yours. I agree that the style actually used in American English is the final comma. Chris removed the following paragraph, which states that as fact, and one of the elements to be considered; I am still unclear whether he disagrees, and if so, on what grounds:
 * American English, the month-day-year style of dating ("The attacks of September 11, 2001, were among the most serious.") treats the year as parenthetical; as in the example, the year is divided off from the rest of the sentence by commas. If the year is immediately followed by other puncutation ("on September 11, 2001; the world..."), the trailing comma is omitted.
 * I trust this advice, for those editors who prefer to write American, can be restored. Septentrionalis PMAnderson 20:00, 16 August 2007 (UTC)
 * We're on the same page, then, Anderson. What prompted me to propose this clarification is the fact that many articles contain phrases such as "the attacks of September 11, 2001 were among the most serious," including the title and body of the article on 9/11. This guide should explain why it should be "the attacks of September 11, 2001, were among the most serious," instead. Chris, what do you think? Thanks Lowell33 20:14, 16 August 2007 (UTC)

You seem to be forgetting that the manual of style does not say that "American dates" can be used. All dates containing day, month and year (in any order) should be linked, to allow the date style preference to work. The comma between day and year is regulated by the template translator, to adapt to the preference. The trailing comma is not modified. So if you include the comma, it will also appear in the day-month-year or year-month-day styles, in which it is incorrect. So it seems there is no solution yielding always correct results. Perhaps the choice that makes two out of three styles correct should be preferred. &minus;Woodstone 20:22, 16 August 2007 (UTC)
 * I would be fine not allowing "American dates" at all on Wikipedia, but I don't see that happening. If the American form is going to be used, editors should have a guide to how to use it correctly. As it stands, Wikipedia is reinforcing the commonly held misconception that a comma after the year in the month-day-year format is unnecessary. That's not right. Lowell33 20:43, 16 August 2007 (UTC)


 * You did not get my point. If an editor writes "On September 11, 2001, something happened", an editor with British preference setting will see "On 11 September 2001, something happened". Note the comma after the year, which is incorrect. So there is no guideline to make it appear correct in all cases. &minus;Woodstone 21:17, 16 August 2007 (UTC)
 * Okay, I get it. It seems like we have a consensus (however limited) that, in the American style, a comma is required after the year (unless some other punctuation follows the year). The problem is how to implement that. The first option is for an editor writing in the American style to type the second comma manually (i.e., without changing autoformatting). The problem with that option is that a reader with a British preference setting will see an out of place comma after the year. That won't work. Option two: we change autoformatting to put a second comma after the year in American style dates. The problem with that option is that it will result in double punctuation after the year in many instances (for example, if the date is at the end of a sentence). Is there any way to fix that problem? Lowell33 21:42, 16 August 2007 (UTC)
 * Thanks, Woodstone, for getting through to everybody (well, all but one, the one who thinks I'm pushing one national variety of English). Although I'm American, I would be happy to settle on whatever would make Wikipedia look consistent, and then let viewers specify how they want their dates to appear. I don't have a subscription to Chicago Manual of Style, but I've seen enough websites discussing it, and according to them, Chicago is starting to recommend day-month-year over month-day-year. I can go with that. Americans are not so stupid as to be unable to understand 4 July 1776. But even if every editor who ever read this guideline agreed, we would still need Wikimedia software to properly add commas for those who prefer to see American month-day-year formats. Until they do (and it's not hard, folks, it's not hard at all, whatever they tell you), the imperfect status quo that has gotten us through the last several years will just have to do for another few months or few years. Chris the speller 02:15, 17 August 2007 (UTC)

(Outdent) It appears that there is consensus that: (1) US usage should properly include the final comma, unless the item is immediatly followed by another punctuation mark, including an unspaced en dash (as in full date ranges); and (2) manually inserting the final comma incorrectly renders the formatting of the non-US date preferences, and therefore cannot be done. There seems to be a general feeling that it wouldn't be an easy task to fix the software in this respect, let alone the other three issues that I've listed below. We have three options:
 * 1) Do nothing, thus continuing the incorrect rendering of US date formatting (we've put up with this for years, without mention) and the other dysfunctions as listed below;
 * 2) maintain the status quo and mount a renewed push to have the autoformatting system upgraded (will take time and energy, and may not succeed);
 * 3) change the wording in MOSNUM/MOS to something more explicitly optional than the current "normally" WRT the use of the autoformatting function; and/or
 * 4) actively discourage editors from using it (the strategy I've been pursuing).

I'm sorry to say that the original good intentions of those who developed the autoformatting are turning out to be poorly served by the system. I'm quite happy to put up with a number of date formats on WP, unlinked, as long as they're consistent within each article. This would be the same situation as for the varieties of spelling that we tolerate. Tony 07:28, 17 August 2007 (UTC)
 * I support the third of these four options, and perhaps the fourth. I approve of the automagic system; but the second option cannot succeed fully unless our display software becomes a flawless English parser (for it must decide whether the comma after "on September 11, 2001," is required by the rest of the sentence or only by 2001). The choice is, fundamentally, between requiring those articles written in American to be ungrammatical, or accepting that the mechanical translation of dates into British usage will sometimes be ungrammatical. We can duck this by not autolinking American dates; but I would prefer to say that those who rely on mechanical translation will sometimes be disappointed. Septentrionalis PMAnderson 15:53, 17 August 2007 (UTC)
 * I support option 2, and oppose options 3 and 4. It is not progress to have non-US readers begin to see month-day-year dates, or to have a comma pop up where it once did not, and where it does not belong. The devil you know is better ... Chris the speller 16:26, 17 August 2007 (UTC)
 * In light of the technical problems, I agree with option 2. I would, however, like to explain in the manual of style why things are the way they are. It should explain that readers with American settings will often see a "hanging" year such as "January 1, 2007" not because it is correct, but because otherwise readers with British settings will see a superfluous comma after the year.Lowell33 16:51, 17 August 2007 (UTC)
 * The technical problem will not be fixed; even if it were routine programming, Bugzilla is slow; and detecting needed commas is one of those things which requires a very complex program, if it can be done at all.
 * Why is it preferable to require all American readers to see a grammar violation so that some British readers do not? If it is true that very few readers set these preferences, why are we catering to them?
 * But Lowell is correct that we should at least explain the problem.Septentrionalis PMAnderson 17:34, 17 August 2007 (UTC)

"summer months" and "winter months"
I'd like to suggest an addition to the section on dates, if it's important enough for the MOS (3042 hits on "summer months").


 * When mentioning the season, summer and winter are used instead of the longer the summer months and the winter months, unless the reference is to something that changes on the first of some month exactly.

&mdash;JerryFriedman 03:55, 11 August 2007 (UTC)
 * This is good advice, but would be inconvenient as a rule. "Summer months" is a reasonable way to explain two-season years, like the Viking calendar or Thucydides; let's not make it impossible. What context are you thinking of? Septentrionalis PMAnderson 04:04, 11 August 2007 (UTC)
 * There's a point about Northern/Southern Hemispheres, too. Better to give the actual months. Tony 05:14, 11 August 2007 (UTC)


 * I don't think it needs to be included here specifically. It is just good practice to omit needless words. —Centrx→talk &bull; 05:18, 11 August 2007 (UTC)

PMAnderson, I admit I can't see why "summer months" is better than "summer" for two-season years. Can you direct me to an example?

Tony, sorry to be unclear. This would go after the present paragraph on seasons, which makes the point about hemispheres.

For examples of the context I have in mind, I'll give the first few hits on "summer months":


 * Geography of Chile: 'The area receives considerable rainfall during the summer months in what is commonly known as the "Bolivian winter"'. I'm not going to fix this right now since it needs a little work, such as an explanation that in the Spanish-speaking tropics, the rainy season is called invierno (winter).


 * Koresand: "In the summer months it is possible to travel from Mandø to Koresand by a tractor-drawn wagon." Fixed.


 * Lush For Life: "During the summer months, Avalon still finds time to go back to New Orleans, his hometown, where he still owns a condo in the historical French Quarter." and "He currently lives in Tampa, Florida, and during the summer months he resides at his remote villa in Key West…" Fixed.


 * Margate City, New Jersey: "Margate is a popular Jersey Shore destination, and is highly popular during summer months." Changed to "especially during summer".

In all of these I think "summer" is better than naming the months, which would be too precise.

Centrx, I think you may be right. &mdash;JerryFriedman 12:52, 11 August 2007 (UTC)

If I have no idea where the place is, "summer" and "winter" have no meaning. Mention the months by name. The word "approximate" appears in my dictionary. It seems a useful word. 81.178.208.1 2007-08-11 20:48 UTC.


 * I've found it useful myself, but "summer" is a good deal more concise than "approximately June through August" in articles about a place in the northern hemisphere. If you were reading those articles instead of my excerpts, you'd know which hemisphere they referred to.  &mdash;JerryFriedman 21:10, 11 August 2007 (UTC)


 * We need to remember WP:CSB (aka Wikipedia is not American) - Wikipedia reaches a global audience, therefore notions of "summer" and "winter" will not be applicable or appreciated in all regions. Assuming a particular hemisphere viewpoint with season names becomes an WP:NPOV problem. Season name usage should be limited to specific cases e.g. quotes, material about seasons themselves, or where using season names is the lesser of evils. Even in those exceptions, the seasons should be deliberately identified by region for global clarity (e.g. "North American summer", "Australian winter months"). Note the WP:SEASON section of MOS:DATE suggests wording options other than month name ranges. These could be stated as "mid-2005", "early in the year", "second quarter", etc. Dl2000 22:31, 11 August 2007 (UTC)


 * Uh, there are summers in both the northern and southern hemispheres. Folks who live in the tropics and have only wet and dry season, and read English know generally what summer and winter mean.  If you're talking about roads that can only be travelled when frozen solid in the winter in Siberia, its not necesary to relocate Siberia's hemisphere, because this has alredy been done in the introduction.  KP Botany 23:17, 11 August 2007 (UTC)


 * Also, I agree with the MOS here: the seasons can be mentioned when "there is a need to do so", not merely as a lesser evil. For instance, "Apple trees bloom in spring and bear fruit in fall" would cover both hemispheres and be far better than anything that mentions months.
 * In regard to the examples above, such wording options as those suggested in MOS:DATE are not available. They work when talking about events&mdash;I understand the reason not to say, "Trade relations between the U.S. and South Africa worsened in the spring of 19xx"&mdash;but not when saying that a town is a summer resort or that a certain weather phenomenon occurs in summer in Chile.
 * Dl2000, I have no idea why you mentioned that Wikipedia is not American. I chose my examples essentially at random, two of the four were not American, and one was in the opposite hemisphere from the United States.  I hadn't edited Geography of Chile at the time I mentioned it, but I have now, and I left "summer" with the southern-hemisphere meaning.  &mdash;JerryFriedman 00:38, 13 August 2007 (UTC)


 * The examples quoted above would be covered by Don't use summer months or winter months when summer or winter conveys the same information. This is good advice, but why do we need it here? This is not a list of clichés; although we could make one (the link is not useful). I would support Eschew surplus words (with a citation to Strunk and White), if anyone feels they need a quote to employ on the stubborn. Septentrionalis PMAnderson 23:34, 12 August 2007 (UTC)


 * As I said when I proposed this change, it may not be appropriate for the MOS. I proposed it because I wasn't sure.
 * A real list of clichés to avoid in Wikipedia would help a lot, I think&mdash;if there could be a consensus. We could start with "A is located in B."  (I added "to avoid" because I think some phrases are clichés that shouldn't or can't be avoided.  For example, "In regard to", "summer resort", "weather phenomenon", and "essentially at random", all of which I used above.)  &mdash;JerryFriedman 00:38, 13 August 2007 (UTC)
 * It's coming: MOS urgently needs to expand its "Usage" section. Tony 23:53, 14 August 2007 (UTC)

Bizarre spaces in the decimal part of numbers?
At Featured_article_candidates, the nominator responded to my query about the spaces in the infobox of Uranus thus:


 * The space separators to the right of the decimal appear to come from ISO 31-0. It's been applied to all of the Solar System planet articles by an unknown editor. &mdash; RJH (talk) 16:34, 11 August 2007 (UTC)


 * Note that I don't have an issue with this spacing scheme; they make a long string of digits easier to read. &mdash; RJH (talk)

They look seriously bad, so I wonder whether action is required. I'm no expert on this. Tony 23:56, 11 August 2007 (UTC)


 * Actually, they've been around for at least several decades. Since it's inappropriate to use commas after the decimal point (or comma), a space is inserted after each group of three digits to enhance readability and reduce transcription errors. I didn't know that it had made it into an ISO, though. Askari Mark (Talk) 02:28, 12 August 2007 (UTC)
 * Thanks for that, Askari. They have a link to this section. Tony 05:42, 12 August 2007 (UTC)

Dates of birth and death duplication.
If the biography has an infobox, can we have the dates in there only instead of duplicating it after the name? -- Jeandré, 2007-08-12t11:58z
 * Nothing would please me more. Is there any support for this? Tony 12:17, 12 August 2007 (UTC)
 * Wrong forum; we should work it out on the talk page for WP:MOSBIO. Chris the speller 15:43, 12 August 2007 (UTC)
 * Linked to this from there. -- Jeandré, 2007-08-13t11:02z
 * On that talk page, I reiterated the question and asked for it to be discussed there, not on this talk page. Chris the speller 15:11, 13 August 2007 (UTC)

OT: rules.

 * Add the option; the fewer piddling rules the better. Septentrionalis PMAnderson 23:26, 12 August 2007 (UTC)
 * You seem to be anti-rule. This is not the place for you. Tony 23:40, 12 August 2007 (UTC)
 * No, I endorse the policy in the paragraph we restored today: if two methods are both acceptable, as both would be here, either can be used; don't edit solely to switch from one to the other. Do you disagree? Septentrionalis PMAnderson 23:51, 12 August 2007 (UTC)
 * As for not belonging here: an editor who cannot tell the difference between, on the one hand, a choice among methods to convey information (which is our purpose), and, on the other, those rules to which English does not allow exceptions, is hardly the ideal writer here. Septentrionalis PMAnderson 00:11, 13 August 2007 (UTC)


 * All too subtle for me; or is it that I just can't be bothered trying to make out the meaning? Tony 06:02, 13 August 2007 (UTC)
 * Presumably the latter. After all an editor who is too politically correct to accept him as common gender, and so cunning that he cannot change to him or her (or change number to them) can hardly be lacking in subtlety, can they? Septentrionalis PMAnderson 18:46, 13 August 2007 (UTC)
 * Your'e right, I am too aware of the need to write inclusive language to pass over the now-unacceptable assumption that the male is the default gender. Singular they is one of several solutions, and not one I favour. Tony 01:17, 14 August 2007 (UTC)
 * I did not propose singular they; had I done so, I would not have written "changing number". It would be no problem to change the sentence to speak of readers. Please read what is in front of you. Septentrionalis PMAnderson 02:52, 14 August 2007 (UTC)

Unsatisfactory edit
"Always convert dates to correspond with years starting on 1 January."

has gone to:

"Always be clear about what style an article uses; in general, this will mean starting the year at January 1, and saying that the year does begin then, unless that is clear from context."

which is unclear, inconsistent and misleading. It will have to be changed.

Thanks for raising it here first. Tony 23:39, 12 August 2007 (UTC)
 * Nonsense; I will tag any such modification as disputed. This is both clear and consistent as an objective.

On a wider matter, I see that there seems to be some confusion on how to write a working manual of style; this edit summary (The local stuff about this will have to go, then; I'll remove later today. )supposes that if we have a general rule or objective, we don't need to give examples for specific cases. Precisely the opposite is true; many editors will look for the clause on the matter they have in mind, and not bother to read the header. If it is important to remind editors what is not worth revert warring over, it is often worthwhile  to do it twice. Septentrionalis PMAnderson 23:48, 12 August 2007 (UTC)

Principles:
 * State the goals, which must be achieved.
 * Advise between the various methods to accomplish them
 * If a reminder is needed, put it in; if it is useful to put it in more than once, do so. Septentrionalis PMAnderson 00:14, 13 August 2007 (UTC)

OK, then every single option—there are many, many of them—will need to be tagged with the clause about consistency within an article, and not changing from one option to the other without good reason. Every single one. Are you going to do that? And before you lable my comment as "nonsense", look at your own English, which is ... little short of it.

So then, who is to judge when a reminder "is needed" or "is useful" to duplicate again and again. I say that it's not needed there. And to claim that your edits were not unclear, inconsistent and misleading is just poppycock. Want me to point out in detail why? Tony 06:00, 13 August 2007 (UTC)
 * No, thank you; you have already demonstrated your level of competence in English usage and your understanding of how to write guidelines. More is not necessary. Septentrionalis PMAnderson 18:49, 13 August 2007 (UTC)

Sprinkling "reminders", as you call them, locally through a MOS is not a good idea. The shorter and cleaner the text, the more likely its comprehension by users. Overriding principles need to be stated at the top, preferably once unless there is some critical need to do otherwise. I don't see that here. The current imbroglio at MOS talk on the very issue of converting BCE to CE and vice versa would not have been prevented by the insertion of a local "reminder".

I must ask that you not unilaterally make major edits to the text without posting them here first, because of the extent to which they need to be fixed. The adjustments should be done here; that's what a talk page is for. Please address my invisible inline queries on your latest attempt. You have a talent for redundancy and vagueness, and they're difficult to fix without your specialised knowledge. Um ... collaboration, I think it's called. Tony 01:14, 14 August 2007 (UTC)
 * It is unfortunate when confusing edits are introduced to the Manual of Style; it can happen at precisely the time an editor is referred to MOS, and they find confusing information. I couldn't decipher today's edits; glad they're sorted out now.  Sandy Georgia  (Talk) 01:27, 14 August 2007 (UTC)
 * Tony's "fixes" are illiterate, and demonstrate ignorance of common scholarly usage. Analysis follows.
 * ''Where a calendar other than the Gregorian is used, this must be clear to readers.
 * For the Middle Ages, when the Gregorian calendar was unknown? For the Roman Republic, where the invariable (and largely unavoidable) custom is to use the Roman calendar (which is neither Julian nor Gregorian, and for parts of which the correction to Gregorian is unknown)? For ancient Greece (normally, where feasible, converted to Julian? Septentrionalis PMAnderson 02:39, 14 August 2007 (UTC)
 * ''There are three other options.
 * There are at least four other options; see the paragraph below: there are a large number of variants which include both calendars. Making clear that the "both calendars" involved are Gregorian and the local calendar, whatever it may be, would be harmless; it rarely hurts to spell out an implication, however clear it originally was.
 * Removed It is less important which option an article uses (they all convey the same facts) than that the article communicate to the reader which one it does use.
 * Why? This should be consensus. Clarity is the goal; the MOS is a means. It should not be used as an power-trip . Septentrionalis PMAnderson 02:39, 14 August 2007 (UTC)


 * I have a client now for five hours, so I'll address these points later, including an illiterate and ignorant proposal for wording the "both/several" issue. In the meantime, might you propose wording here as an improvement of "''Where a calendar other than the Gregorian is used, this must be clear to readers." Please don't frame this as "an power trip" [sic]; it's just crafting and negotiation, with a reward for all at the end. Tony 03:11, 14 August 2007 (UTC)

Old Style New Style

 * See also Old Style and New Style dates, and on the talkpage Talk:Old Style and New Style dates

This paragraph appears to be unwisely put:
 * If it is important to preserve consistency with primary sources, a date may be given in the original style with its equivalent in the modern style; for example, Elizabeth I of England died on 1602-03-24 (Old Style)/1603-03-24 (New Style). Although this would correspond to 3 April, 1603 if fully converted into the Gregorian calendar, the month and day of a British event are normally not converted.

There are at least four possible styles here:
 * 24 March 1602 (Julian, as used in England)
 * 3 April 1603 (Gregorian)
 * 24 March 1603 (composite)
 * 24 March/3 April 1602/3 (and its variants)

The first, the contemporary usage, is Old Style; but either of the middle two can be called New Style. To say that the composite is normal is simply wrong; It is probably most common, but it's more likely to be a plurality than a majority, and it depends on the level of competence the author expects of the reader (and the date of the subject; New Year on January 1 was more common in the early eighteenth century than before). Serious books at least use both years for dates between January and March. (This would be clearer, btw, if the example were earlier than March...) Septentrionalis PMAnderson 00:34, 13 August 2007 (UTC)
 * George Washington's birthday (February 11, 1731 OS) might be a better example here, because it is in February. This has nothing to do with Anglo-Americanism; but since someone will think so, unless denied, I propose it here. —The preceding unsigned comment was added by Pmanderson (talk • contribs) 19:22, 13 August 2007  (UTC)
 * Using birthdays of people alive during the transition are probably not the best to use because clause VI. of Calendar (New Style) Act 1750, specifically regulates how birth dates should be presented for those born before "14th Day of September". --Philip Baird Shearer 17:54, 15 August 2007 (UTC)
 * This may be an Anglo-American difference then; Washington chose freely among the possibilities, in choosing to celebrate the Gregorian date. Septentrionalis PMAnderson 21:06, 15 August 2007 (UTC)

There are two different interpretations of what constitutes New Style. One is a change of the calendar date to 1 January as in this document. The other is a full conversion to Gregorian calendar as in this source.

Now for what I think is wrong with the page:
 * ''There are several methods for dealing with dates before the local introduction of the Gregorian calendar; in Europe, the earlier calendar involved will be Julian:
 * ''Give dates in the local calendar only &mdash; with an adjustment for the start of year as 1 January if necessary. This means that the dates will match the dates in the primary sources for that period. In Europe, there should be explicit indication that the dates are in the Julian calendar if this is necessary; for non-European calendars, it will usually be obvious they are not Gregorian.

I am sorry but it is not usual to give an explicit indication that the dates are in the Julian calendar. For example with the exception of the Battle of the Boyne, all events during the C17th in the UK are usually recorded as Julian with year adjustment. The should not be in the Gregorian calendar.
 * Clearly if this is necessary, which was intended to cover the case in which the reader will expect Julian, is not emphatic enough. I've made it explicit. Septentrionalis PMAnderson 21:35, 15 August 2007 (UTC)

The only problem comes when one is talking about events that take place in both Julian and Gregorian regions, for example in the article Glorious Revolution. This also throws up the problem of whether to refer to events as (Gregorian calendar) or New Style, because it is not clear if the date is New Style start of year adjustment or New Style Gregorian calendar adjustment. For example this sentence is wrong "William and Mary were offered the throne as joint rulers, an arrangement which they accepted. On February 13, 1689 (Old Style), February 23 (New Style) Mary II and William III jointly acceded to the throne of England." It should be February 13, 1688 (Old Style), not 1689. But then one has to decide if it is February 13, 1689 (New Style) or February 23 1689(New Style)


 * '' Convert the dates to the Gregorian calendar. This enables the correlation of dates in countries and at times in which different calendars are or were used. The conversion should be explicit unless this is obvious to readers; it is useful to check that the sources have not already converted them.

This is clearly wrong advise, all British dates should be Julian with the start of year adjustment until the conversion to the Gregorian calendar. What is needed here is follow the lead of reliable secondary sources.


 * '' Give dates in both the local and Gregorian calendars, for example, 'William Shakespeare died on 1616-04-23 (Old Style)/1616-05-03 (New Style).']

This advise is not the best. Again it should be follow the advise of reliable secondary sources.


 * ''It is less important which option an article uses (they all convey the same facts) than that the article communicate to the reader which one it does use

Not true Wikipedia should follow the lead of reliable secondary sources, or it ends up with dates that look nothing like those in other reliable sources.


 * ''It is reasonably common, and may be useful for consistency with primary sources, to give both years for dates between January and March. For example, Elizabeth I of England died, according to her English contemporaries, on 24 March, 1602, the eve of New Year; the equivalent Gregorian date was 3 April, 1603. Either may be used, or the common composite form 24 March, 1603; but 24 March, 3.

The layout of this is wrong it should follow the normal convention that would be 24 March 1602/03.

--Philip Baird Shearer 10:34, 14 August 2007 (UTC)
 * Much of this is well taken; I regret being so conservative with what I found when I got here. There are two overall points, on which I think reasonable editors can agree:


 * There are some articles on which we should use Julian (Charles II of England); some on which we should use Gregorian (Fronde); some on which we must indicate which (War of the Spanish Succession).
 * We should be use local dating when it is expected (the English articles here; the birth and death of Julius Caesar — his birth is of course neither Julian nor Gregorian.)
 * It does no harm to remind readers that they are dealing with a non-Gregorian calendar; especially in Europe, when we are using the Roman months. But it is useless rule-making to require every article to do so. Septentrionalis PMAnderson 18:25, 14 August 2007 (UTC)

I would not revise this section into a mandate in the other direction either. We should remember that there will always be boundary cases (much of the eighteenth century in English-speaking countries) where a mixed approach is desirable. We should not prohibit it in favor of Julian any more than in favor of Gregorian. Septentrionalis PMAnderson 18:38, 14 August 2007 (UTC)


 * My own thoughts on this is that:
 * The dating method used in a Wikipedia article should follow that used by reliable secondary sources. Unless otherwise stated in the article, Julian calendar years will start on 1 January. If there is a need to mention Old Style or New Style in an article (Like in the Glorious Revolution) then on the first usage, a footnote should be provided stating if the "New Style" refers to a start of year adjustment or a conversion to the Gregorian calendar.
 * --Philip Baird Shearer 10:24, 15 August 2007 (UTC)
 * Included a recommendation to follow reliable sources. I am not as convinced as PBS that actual usage is as uniform as all that (I've seen a lot of double-dating works of reference); but if it is, the recommendation will apply.. Septentrionalis PMAnderson 21:35, 15 August 2007 (UTC)

percentage vs percentage point
I propose to add the following last point to the section on percentages.


 * Percent or per cent are commonly used to indicate percentages in the body of an article. The symbol '% may be more common in scientific or technical articles, or in complex listings.
 * The symbol is unspaced (71%, not 71 %).
 * In tables and infoboxes, the symbol is used, not the spelled-out percent or per cent.
 * Ranges are preferably formatted with one rather than two percentage signifiers (22–28%, not 22%–28%).
 * Use percentage points, not percentages, to express a change in a percentage or the difference between two percentages (The agent raised the commission by five percentage points, from 10 to 15%. If the 10% commission had instead be raised by 5%, the new rate would have been 10.5%). Percentage point is always spelled out ; after the first occurrence, it may be abbreviated to point, unless this would be unclear , and should not be confused with basis point, which is a hundredth of a percentage point. Tony 14:19, 15 August 2007 (UTC)


 * Strongly oppose the last suggestion, which will produce unlimited confusion with basis point. A caution to be careful would be justified. The rest of these, which already exist, should be pruned; they  are (except perhaps for no spaces before %) more examples of the sort of thing MOS should not be deciding. Leave articles which succeed in communicating with the reader alone. Septentrionalis PMAnderson 20:58, 15 August 2007 (UTC)
 * Thanks to Anderson his/her point about possible confusion with basis point, although a more measured tone would be welcome; I've removed mention of the abbreviation to point and cautioned against confusion with basis point, as suggested.
 * I strongly oppose his/her comment about the other points, and doubt that it has support among others, here or at MOS. The logical extension is that MOS and MOSNUM would be "pruned" to a skeleton, and cohesion in the project sacrificed.
 * I'm tiring of continually having to fix Anderson's edits to the page, which are made in what appears to be a casually incorrect version of English—basic errors have been evidence of a slap-dash approach. Please make amends. Tony 00:11, 16 August 2007 (UTC)
 * You may stop doing so at any time.  Septentrionalis PMAnderson 02:25, 16 August 2007 (UTC)
 * That's the kind of attitude that will land this page in a mess. Again, I ask you to desist from blundering in with your error-prone language. I wonder why I bother to raise things here first, if you take it upon yourself to unilaterally make substantive changes to the page. Tony 03:11, 16 August 2007 (UTC)
 * No, the attitude that has landed this page in the mess it is in is provincial insistence on comprehensive-school shibboleths; as with the nonsense above about not having a comma after September 11, 2001. We should not insist on violations of idiom; and if we are going to play schoolmaster, we should get it right.  Septentrionalis PMAnderson 16:34, 16 August 2007 (UTC)
 * No idea what he's talking about; can anyone decipher it for me? Tony 06:32, 17 August 2007 (UTC)

This page fails to distinguish between three kinds of recommendation:
 * Rules, which no competent writer of English would violate (without extraordinary reason). Even here, it fails to recognize differences of importance; the only example of these in the section on percents is the absence of space in "71 %", and does that really matter? If it turns out to be the standard usage in Indian English, for example, we should permit it, as we use Mumbai; and would lose little by doing so.
 * Recommendations, or good ideas. The example here would be using % in tables, because it's shorter. This recommendation has no more force that that; if there comes a time when a table doesn't need to squeeze out those characters, we should use "per cent". Two examples:
 * If the headers of a table are "Absolute change", "Per cent",...; we should usually spell out Percent; especially if we do not lose a column by doing so.
 * A table of powers of the Federal Reserve Board may mention "percent" in passing; there is no reason to abbreviate.
 * Choices, unmotivated, or barely motivated, between good methods of accomplishing the same end. The canonical example here is the choice between footnotes and Harvard referencing; each has its good points, and some editors feel strongly about the choice between them. Here we should permit both; ideally, we would describe the advantages and disadvantages of each.
 * The lengthy paragraph about percentage points comes under this head. There is a trade-off between the clarity of points and the additional length and weight it adds to the sentence; it should not be mandated. A caution is the proper action here; if it is necessary to refer to this, add a sentence: in this case, it's worth adding "points". Septentrionalis PMAnderson 16:34, 16 August 2007 (UTC)


 * Point 1: no, standard in Indian English isn't necessarily a reason for blanket acceptance. A spaced percentage sign is harder to read, IMV, and there's benefit in forcing a standardisation. Almost all instances are unspaced, anyway.
 * There is very little benefit in forcing standardization for the sake of standardization, except for the pleasure of bullies. I do not regard this as a good. Septentrionalis PMAnderson 15:59, 17 August 2007 (UTC)
 * Point 2: Maybe, but % in a table header seems fine to me. I don't think it's worth making an exception for headers. This has been in the guidelines for quite a long time, I think. Why not abbreviate? The more white space in a table, the less cluttered and easier to read it is. I see no reason to change long-standing policy here.
 * I do not advocate making an express exception for headers; I advocate recognizing that this is a question of convenience, not of necessity. Septentrionalis PMAnderson 15:59, 17 August 2007 (UTC)
 * Point 4: The use of percentage points is absolutely essential. People write statements such as "the proportion of the population who are privately insured has risen by 10% in five years"; do they really mean from 40 to 44%? Usually they mean 10 percentage points. This distinction needs to be enforced, as occurs explicitly in quite a few house styles. Tony 06:41, 17 August 2007 (UTC)
 * No; clarity needs to be enforced; "percentage points" is one way, and often not the best way, to obtain clarity. Septentrionalis PMAnderson 15:55, 17 August 2007 (UTC)