Wikipedia talk:Manual of Style (dates and numbers)/Archive 85

From Wikipedia, the free encyclopedia
Jump to: navigation, search

Date formatting

It seems to me that the date section was more clear on using spaces around the ndash in the date. I think it used to state that if the dates on either side of the ndash did not use a space, then use no spaces around the ndash; if the dates have spaces, then space the ndash. Thus 1910–1917 and February 11, 1910 – 22 March 1917. If this is correct, then the current text is not so obvious. --Gadget850 ( Ed) 11:49, 9 September 2007 (UTC)

This is precisely the sort of thing which we should not bother to standardize; for Heaven's sake, why? Septentrionalis PMAnderson 17:38, 9 September 2007 (UTC)
Yes, that was the standard I remember. It is standardised to ensure consistency; it is consistent so that this is a professional work. For some editors it is a stylistic issue, for others, just consistency. Think about it, if you saw 1910–1917 in one place, 1924-1926 in another, then 1932 — 1939 and then 1945 – 1952, it would look scrappy. This isn't something people should stress over, unless they're like me and they like details—remember the guide is for articles, not editors. As long as no-one deliberately makes edits which go against it. Neonumbers 01:28, 10 September 2007 (UTC)
It's precisly the sort of thing MOS should state: readability is at issue. No space where there are internal spaces in the items looks horrible and is confusing. Hyphensa are more readable as en dashes. The spacing point was and is still made in MOS. Tony 02:07, 10 September 2007 (UTC)
Not to me; I find the readability exactly equal. Neonumber's point seems to me to prove that this is another instance of a frequent case; we should have, but not stress over, consistency within an article. Septentrionalis PMAnderson 02:59, 10 September 2007 (UTC)
Problem here (with your general poin) is articles share navboxen, tmeplates, infoboxen, and bits get moved betweeen them. Sowhile having articles consistent internally is a (very) good start, it's not really enough. Rich Farmbrough, 19:21 12 September 2007 (GMT).
PMA, if they are "exactly equal" to you, why are you bothering to go on about it? Its been said here before that you frequently seem to be raising and continuing arguments just for the hell of it. This is a good example of a post that makes people think that. Funny thing is, I remember you being way more reasonable during the WP:ATT debates; why the combativeness here over things you appear to consider totally trivial yourself? — SMcCandlish [talk] [cont] ‹(-¿-)› 15:39, 17 September 2007 (UTC)
Because forbidding spaced emdashes has in practice, under some reviewers, been made into a FAC criterion - despite Tony's note above that it should not be so read. One solution would be to remove the mention of MOS as a criterion there, and let this page exist as what it was designed to be, a page of advice. I want to keep these trivialities out of FAC, and away from bullies; if some editor is not sure what to do, and comes here to check, fine. Septentrionalis PMAnderson 19:57, 26 September 2007 (UTC)

Actually, I found the ref I was looking for at WP:DASH:

Spacing: All disjunctive en dashes are unspaced, except when there is a space within either or both of the items (the New York – Sydney flight, the New Zealand – South Africa grand final, 3 July 1888 – 18 August 1940).

--Gadget850 ( Ed) 19:42, 12 September 2007 (UTC)

Capitalisation of SI units?

It is implied by the present wording (Conversions section) that the SI unit of absolute temperature (kelvin) should be written with a capital "K". SI units are written as lower case when written out in full, and the kelvin is no exception. The statement is incorrect. The reason why Celsius is upper case is because the celsius is not an SI unit (it is the degree Celsius). Thunderbird2 17:43, 9 September 2007 (UTC)

Quite right; but we should phrase to discourage, not forbid, the quite common surviving usage of "degrees Kelvin"; that is merely old-fashioned. Septentrionalis PMAnderson 18:04, 9 September 2007 (UTC)
Yes the MoS never "forbids". It is a style guide - and we have never expected editors to meet it (althoguh many do), that is for sub-editors. Rich Farmbrough, 19:17 12 September 2007 (GMT).
Let it be strongly discouraged then. On a related note, some discouragement of the practice of capitalising imperial (as in imperial system) might be worthwhile ... quite a common error. Jɪmp 03:52, 14 September 2007 (UTC)
A far more important issue, one more worthy of our attention than the debatable one you raise (it does refer to a specific, identifiable "Empire"; we don't often use that term in connection with the units of the Roman Empire or any other empire), is the fact that "imperial" is often used in these contexts when it should not be used, because its context includes units not acceptable in the British Empire in the post-1824 era. Gene Nygaard 14:13, 28 October 2007 (UTC)

The kelvin is a unit that is mostly used only in a scientific context, and as such it is quite wrong to call it a degree (and has been for 40 years). So "strongly discourage" sounds about right. Thunderbird2 13:29, 17 September 2007 (UTC)

Kelvin is an absolute scale. So it is quite wrong to say "degrees Kelvin". It is twenty Kelvin, or four-hundred Kelvin, etc. 81.178.90.168 (talk) 2007-09-20 00:30 UTC
They are not all lower case at all. They are especially upper-case when they were named after a real person: Watt, Newton, Kelvin, etc. 81.178.90.168 (talk) 2007-09-20 00:28 UTC

Yes, it is incorrect (in modern usage) to use degrees Kelvin. It is also incorrect to use the person's name (Kelvin) when talking about the unit (kelvin). Similarly the SI units of power (watt) and force (newton) are also lower case. Thunderbird2 05:46, 21 September 2007 (UTC)

Very useful summary: http://lamar.colostate.edu/~hillger/faq.html (talk) 2007-09-21 11:25 UTC —Preceding unsigned comment added by 81.178.210.249 (talk)
Their abbreviations are uppercase when named after a person but not the words themselves—that's just how the metric system works. Jɪmp 00:18, 25 September 2007 (UTC)
From the norm, examples of units starting with an upercase: base SI units: A (intensity), K (temperature), derived coherent SI units: Hz , N, Pa, J, W, C, V, F, S, Wb, T, H, °C, Bq, Gy, Sv,... pom 18:44, 25 September 2007 (UTC)

Use of place of birth/death

Greetings,

Another user is going around deleting the places of birth and death from the first lines of many bio entries, such as below:

- Lloyd Brown (Lutie, Missouri, October 7, 1901Charlotte Hall, Maryland, March 29, 2007)

He claims that including the place of birth and death is a 'violation' of WP: DATE. However, I have not seen a policy forbidding its use. Further, I note that the Encylopedia Britannica uses the above format, which allows for quick skimming. It doesn't make sense to bury the birth/death date info in the main part of the article. The main part of the heading...name, birthplace, birthdate, deathdate, deathplace...allows the reader to quickly establish the parameters (time, area/nation, etc). Thus, my question is of two parts: one, is this against WP:DATE policy and two, if it is, what can we do to change it? Ryoung122 03:10, 14 September 2007 (UTC)

The first question is easy: the MoS currently says: Locations of birth and death are given subsequently rather than being entangled with the dates. Until July 30, it used to be more strongly worded:
  • Locations should be included in the biography portion of the body article. For example, "(12 February 1809 in Shrewsbury, Shropshire, England–19 April 1882 in Downe, Kent, England)" should be separated to "(12 February 180919 April 1882) … He was born in Shrewsbury, Shropshire, England … and died in Downe, Kent, England".
The second question is very good. I never completely agreed with this style guideline (not a rule) either. There are many many articles that are in violation; so, there are obviously other editors who feel the same way as well. Neier 08:01, 14 September 2007 (UTC)
There was never a consensus to soften the wording, but a general overhaul of this guideline resulted in the compression of some instructions. The locations should be separated and placed in the body of the article. Many editors object to adding unimportant details to the opening; look at the talk page for WP:MOSBIO, and this is a recurrent theme. WP is not the Encyclopedia Britannica, has the right to set its own style, and has done so. This seems to surprise some editors who want to copy from the other encyclopedia. Saying that an editor is "going around deleting" sounds like an attempt to paint him or her as a loose cannon. The action was not a deletion, but a move of the information to the body of the article, where it belonged. Chris the speller 00:39, 17 September 2007 (UTC)
It really needs to be noted that not even close to everyone agrees with this, that birth/death place do not belong in the lead. There are thousands and thousands of bio articles that do put the places in the "(born, died)" lead segment. I prefer it there myself, but my point is that there isn't anything close to actual consensus on that at all. I don't believe this is a matter of sloppiness or "just not thinking about it", either, but rather the product of hundreds, probably thousands of editors' conscious decisions (otherwise I wouldn't bring the matter up). Basically, many of us do not find them to be extraneous or excessive details at all. While birth/death place are less "defining" characteristics than birth/death date, they are nevertheless still defining. Especially when dealing with common names in broad fields (e.g. football (soccer) in which there are many confusable instances of different people with the same given and family names. — SMcCandlish [talk] [cont] ‹(-¿-)› 12:58, 17 September 2007 (UTC)
Am I right in thinking that there is a consensus to link the dates (day/month and year), if known, in biographies? thanks. Itsmejudith 16:29, 20 September 2007 (UTC)
Yes for complete dates (July 14, 1968 or 14 July 1968). Partial dates (1999, or July 1832) are disputed. And I'm not certain about longer-than-a-year spans (1960s, 18th century, etc.), but logic would seem to dictate per WP:CONTEXT that they not be linked unless it is genuinely likely to be useful to the reader to do so, as with any other linking being done (e.g. we do not link like this, "The plane crashed around noon, possibly as late as 30:30 pm, during a lightning storm that produced a dark, overcast sky in which the sun was barely visible...") — SMcCandlish [talk] [cont] ‹(-¿-)› 03:54, 21 September 2007 (UTC)
Very helpful, many thanks. Itsmejudith 15:42, 21 September 2007 (UTC)


Change to precision in "Decimal points" section

The text before today's changes was better. There are times when 41.0 actually means 41.0000000 and can be rounded. But there may also be occasions when it should not be rounded. Isn't that what was intended by the old text? Thunderbird2 14:38, 17 September 2007 (UTC)

Oh dear. I've just read my post again, and it doesn't make sense at all! Concerning this edit, what I meant to say was that there are times when it is appropriate to add more significant figures after a conversion than were used before. For example, 180 degrees is not 3.1 radians, but (approximately) 3.14159 radians. I think that's what the older text was trying to say. Thunderbird2 19:18, 19 September 2007 (UTC)

All dates should be linked

'Wikipedia has articles on days of the year, years, decades, centuries and millennia. Link to one of these pages only if it is likely to deepen readers' understanding of a topic' - I do not think this is a good idea. We should link all dates, partial and full, for enabling the user to browse the almanac section of Wikipedia (ie its collection of articles on years, centuries, etc). It is a great way to enable users to find other events that happened around the same time mentioned in the text. NerdyNSK 03:15, 19 September 2007 (UTC)

Needless to add, I am a prolific user of Wikipedia's date links and the articles on years, and I use it as a means to browse the wiki to find articles on new topics etc. Many articles do not link all of the dates they contain, and as a user this creates difficulty in my Wikipedia browsing. I believe there must also exist other users who have similar views. NerdyNSK 03:19, 19 September 2007 (UTC)
This is an age-old issue that has been partially resolved by moving away from the absolute insistence on the linking of every full date, and the active discouragement to the linking of simple years. Decouple the autoformatting and linking functionalities, and I'm all for autoformatting every full date. But we've tried in a very big way, more than once, to be met with a shrug by the developer allocated to the matter. On simple years, I'm sorry, but those who invest time in writing year articles cannot expect that every year be turned bright blue throughout the project. Year pages might be interesting to the minute proportion of readers who have loads of time to follow rabbit tracks to information that has no particular bearing on the subject they've looked up. But most readers want to read the article at hand and be directed to useful, focused, vaguely relevant links, undiluted by year links. Just about every year link I've followed plunges into totally irrelevant information (so and so broke an Olympic record in such and such) in an article on the history of diesel engines. Hello? Tony 03:33, 19 September 2007 (UTC)
I know about the discouragement of linking simple years; but, when was there a move away from the linking of full dates that allow the date preferences to work? A lot of people have asked for the decoupling; but, I don't recall a concensus that it was acceptable to ignore the date prefs linkage just because of that. Neier 06:17, 20 September 2007 (UTC)
But, for users like me, having links that lead to irrelevant information is a feature, not a bug. I want to see links to pages that have little or no relevance to the article I'm reading, because I want to be able to discover new information that otherwise I would probably not be able or willing to find it. I have learnt a lot just because some good editors once put a link to a page that is not directly relevant to the article. If all links lead only to other articles that are in the same domain as the one I'm reading, then, as a reader, I feel locked in whatever articles I'm mostly interested in at the moment. The random article link isn't good for various reasons, and the other browsing features (portals, lists) are not as useful as blue links, at least not for the way I am used to surf. Many times I have to copy and paste words from a WP article into the searchbox just because someone thought that no one would ever want to visit article X while reading article Y because in their view they were unrelated. The total solution to the problem, of course, would be a simple software modification to enable links with different importance scores applied on them. 'irrelevant' links could be displayed with the same colour as the article's text by using a simple CSS class. Normal links could be left blue. And very important links could be bold. We could even do this now without software modifications, but software support for importance or relevance scores would make the process much faster. I talked about links in general here, but of course the same apply to dates, as links to dates is the last resort for people who surf in the way I do. Also note that the more links we have, the more we encourage editors to edit new articles. No link = fewer readers = fewer editors. NerdyNSK 04:02, 19 September 2007 (UTC)
Not everyone agrees with you; indeed, it is my perception that fewer and fewer do, with a sea change beginning some time around February 2007, and steadily growing since then, against linking bare years (and decades, centuries, etc.) without very clear reasons for doing so. The issue is one of context. Some examples (some real, some made up) and my takes on them (others might disagree somewhat, but I'd bet money that a majority would agree with most of them):
At Civil rights: "Civil rights activism peaked in the 1960s...", because the 1960s as an "age", a generation, a euhemerized period, is deeply linked with the topic of civil rights activism in Western culture, both ways (mutual interrelation and effect)
At Jim-Bob Dupree: "Growing up in the 1960s, race car driver Dupree...", because the '60s and a random sportsperson are not inextricably mingled topics.
At Fluffmosis III: "The king died some time near the end of the 4th century BC, and was entombed at...", because lots and lots and lots of stuff happened in that time period, and a long list of such things is not relevant to readers of this article, and one particular person, even the High King of Numismidia, cannot have had a particularly enormous effect on the period.
At Sesclo: "The Sesclo culture, ca. 68504800 BC...", because the culture straddled the entire span, and was very significant (and using the millennium dates even if specific year articles exist for 6850 and 4800 BC, because they are radiocarbon dating approximations, and as such have no demonstrable relevance to those year articles, or vice versa, at all).
At Willie Mosconi: "In 1933, Mosconi competed in the Billiard Congress of America (BCA) World Championship tournament. He nearly won the title but lost ... [big snip] ..." into the mid-1950s..." A link to the year as a ding an sich will not provide much if anything of use to a reader of this passage. Same goes for the decade in the later segment. And nothing Mosconi did during those years had a marked effect upon them.
At Nazi party: "Rising to increased power in the late 1930s...", because Nazism had a major world impact on that time period, and the world impacted back on Nazism.
Again at Willie Mosconi: "After suffering a stroke in 1956 ... but recovered fully, to win the 1957 BCA World Championship"; again, the 1956 year and Mosconi are not interrelated in any way other than coincidence. However a World Championship victory is significant, and might be worth comparing to related and competing events in the same year as at the 1957 in sports article that the "1957" points to (or something more specific, such as a 1957 in pool, if available), and the title should also be listed in 1957 in sports, such that the two articles, on this particular point of international significance, are in fact inter-related. But not 1957, again because that calendrical article does not help the Willie Mosconi reader, and Mosconi's big win here did not have any serious effect upon the world in that year (though it was certainly notable specifically in the sporting context).
I could go on, but I think the point is probably pretty clear by now. If there is no particular context or relevance, that runs both ways, then don't link a bare year/decade/century/millennium date (and especially not a day, like Wednesday! A link to that day article woudl be appropriate at something like Woden, but not as part of a date in a random article.) Such linking is a readability problem and offers little value to the reader.
NB: Anyone should of course feel free to borrow any of the above and rework it if the MOS should say something more specific on this topic than it presently does (I think it should, but I've been involved in enough MOS disputation this week to last me for several months, so I think I may back out for a while after those particular debates calm down).
SMcCandlish [talk] [cont] ‹(-¿-)› 09:51, 19 September 2007 (UTC)
Nerdy, there's nothing stopping you from typing just four numbers into the search box if you fancy checking out a year. Takes three seconds, and leaves the page cleaner. Tony (talk) 14:05, 19 September 2007 (UTC)
I think Nerdy NSK's comments (particularly "I want to see links to pages that have little or no relevance to the article I'm reading") go against WP:CONTEXT, a WP style guideline. If you feel that guideline is wrong, you may want to discuss it there. I, however, also disagree. If you want to learn things that are irrelevant to what you're looking at, you can look up a random date or use the random article function. Rigadoun (talk) 20:31, 19 September 2007 (UTC)
There's also, of course, WP:BTW... NerdyNSK —Preceding signed but undated comment was added at 02:04, 20 September 2007 (UTC)
Hmmm, yes, it's an embarrassment to the project, that one: have you read it? (1) The register is way out of whack with WP's guidelines—comes down to vague musings, suggestions, statements of philosophy, and you're left wondering what exactly it's there for. Should be an essay. (2) There was no consensus on making it a semi-guideline and then a guideline. (3) There appears to be ownership going on there, by a small clique, or worse, one person. (4) It's in contradiction with MOS, MOSLINK and MOSNUM in some respects. Tony (talk) 02:38, 20 September 2007 (UTC)
Funny how you see WTB as ownership going on there, by a small clique, or worse, one person - the fact is that I am afraid the idea of limiting links is also the work of a clique. I would like to know exactly how many people support either view. NerdyNSK 03:53, 20 September 2007 (UTC)
  • (outdent) "WTB"? You mean BTW? Please provide a link to the consensus. Fact is, there was none. Now, if there were a way of linking years that didn't splatter more bright underlined blue over the text (thus diluting valuable links and reducing readability and neatness), I wouldn't mind if every year were linked. Just click on any (black) year and up comes the year page: sure. It's feasible, technically, but getting the developers to act on these matters is like trying to push an ocean liner. Until that happens, linking years is a real problem. Tony (talk) 05:22, 20 September 2007 (UTC)
    • Sorry, it was WP:BTW. NerdyNSK 05:43, 20 September 2007 (UTC)
      • For me, blue links do not detract from readability. Therefore I follow a policy of "if in doubt, link first use". Not for years particularly - I mean quite generally. But this is subjective. How do we find out what *readers* think? Thunderbird2 06:59, 20 September 2007 (UTC)
I first sorted out the issue in my mind shortly after I started on WP, when a friend who was a casual reader of WP queried why every year was blue, and expressed irritation at the feature. Then I delved into the matter and realised that those who invest time and energy into writing year articles were very reluctant to give up the privilege of having many many links to their articles. Thus, factions had formed, and continue, although the matter seems to have been settled in favour of not linking trivial chronological items for the sake of it. I'd love it if the WikiMedia Foundation were lobbied to overhaul the linking and autoformatting systems: they need to be decoupled and made more flexible. There are at least half a dozen issues. Tony (talk) 09:01, 20 September 2007 (UTC)
I can sort of understand why we would want dates to be linked. It does enable people to quickly figure out if two events coincide. However, this is a large world and events that only coincide temporally are usually not related. I don't think clicking linked dates is common practice. So here's a suggestion: add a configuration option in the preferences called "link dates" that is by default turned off. That way we can have properly autoformatted dates everywhere, the links only show up for people who like them, and everyone is happy. If the developers are too busy or lazy to program this, we can do it ourselves if necessary. I hear it's written in PHP. A scripting language, how hard can it be? And even if we can't get this functionality in MediaWiki, we can still as a last resort use Wikipedia's own site-wide JavaScript to remove link from dates. Shinobu 16:31, 26 September 2007 (UTC)
And when we finish this discussion, let's submit a feature request. The developers are no mindreaders, and no one can expect them to implement something if no one tells them what we want. Shinobu 18:53, 26 September 2007 (UTC)
In principle, I don't mind Gerbrant's suggestion. The problem, however, would be that the default "no date autoformatting" will display all of the shabby inconsistency in raw date formats that have been entered over the years within the same articles. It would be a significant task to convert them all to the appropriate format for each article. On second thought, I guess it wouldn't make any difference to the 99% of readers who see them linked but not autoformatted now. Tony (talk) 12:08, 18 December 2007 (UTC)