Wikipedia talk:Manual of Style/Archive (numbers)

Wikipedia talk:Manual of Style

Numbers
Large numbers: is it fair to say large number should be written thus: 1,234,000 when not using exponential notation? -- Tarquin

One disadvantages of using commas is that some people are used to seeing these in place of decimal points. One disadvantage of using spaces instead is that browsers might wrap a number over two lines, which is really nasty. -- Matthew Woodcraft

One solution is to use "&amp;nbsp;", but the drawback to this is that it looks ugly in the wiki markup (which is why I've stopped using it in mathematical formulas). &mdash; Toby 10:57 Nov 11, 2002 (UTC)

Units
See Measurements Debate for lengthy discussion of the superiority of metric (granted), the depravity of American measurements (granted) and the need to communicate with depraved Americans (grudgingly granted).

2.3 km (1.4 mi) seems to be the method of choice for lengths.Sebastian 07:21 Jan 23, 2003 (UTC)

Dimensions
I can't find any mention of a Wikipedia-preferred format for dimensions. Here are several formats for the example of metres per second, ranked in order of my particular preference:


 * m.s-1
 * Pros: can easily tell the difference between each unit as they are separated by full stops (periods).
 * Cons: not that common outside of Europe, afaict.


 * ms-1
 * Pros: saves typing
 * Cons: can lead to confusion, e.g. this could read as "inverse milli-seconds"


 * m/s
 * Pros: seems to be more universally common outside of scientific publishing
 * Cons: is by some believed old fashioned and not standard scientific format

charlieF 18:06 Mar 14, 2003 (UTC)

Difficult....
 * http://www.bipm.fr/enus/3_SI/si-derived.html uses the dot
 * http://physics.nist.gov/cuu/Units/checklist.html says all 3 are ok

--

Not the least bit difficult:


 * m.s-1 [sic] is wrong (the dot should be raised like this: m·s-1), and even if written correctly does not make sense to anyone but scientists, and is not common in Europe either.
 * ms-1 [sic] is completely useless, since it is 1/ms.
 * m/s is commonly understood by everyone, commonly used, exact, scientific as far as I know, and cannot be misinterpreted.

But by all means, km&sup2; should be written thus, and not sqkm.

Finally, there is a typographical/aesthetical problem when used in running text, in that "-1" as a superscript causes excessive line spacing, but superscript "2" is available within the character set and does not have this problem.

-- Egil 07:38 Mar 15, 2003 (UTC)


 * OK, you're right about the raised dot for the first form, but the third form can be equivocated when you have more complex units, e.g. thermal conductivity:
 * W/mK looks like either Watts per metre Kelvin (correct) or Watts per milliKelvin (incorrect)
 * whereas W·m-1·K-1 is unequivocal.
 * or take metres per second per kilogram squared
 * m/skg&sup2; looks 'orrible, whereas m·s-1·kg&sup2;, to my eye, looks much better.
 * Furthermore, anything with a superscript that is not 2, such as internal heat generation (Watts per metre cubed), renders in the third form as
 * W/m3, which causes just as much excessive line spacing as
 * W·m3
 * So there's no way of avoiding unsightly line spaces if you want to use a standard system throughout Wikipedia.
 * Lastly, there should really be some standard symbol table, i.e. if anyone, anywhere, mentions thermal conductivity then they should use the same symbol (e.g. k). Does this already exist?
 * charlieF 11:17 Mar 15, 2003 (UTC)


 * I think we probably should pause, and try to look up what SI has to say on the matter first (on-line SI refs should be treated with caution). W/m&sup3; works just fine. In terms of m/s, then as long as that is fine with SI, then that is what I would prefer since everyone would instantly know what it means (which is a good point, admit it). More exotic units that probably would appear on a total of two Wikipedia pages, I'm not so concerned about. -- Egil 12:00 Mar 15, 2003 (UTC)


 * This looks like the page of choice . See sections 6.1.5 onwards.... charlieF 12:38 Mar 15, 2003 (UTC)

Mathematics and Physics
I found the following in inertial mass:
 * mi&times;ai1 = m1&times;a1i,

My question is: Should we really use the &times; (&amp;times;) symbol for multiplication? To me this looks rather like a vector product.

Sebastian 23:42 Jan 24, 2003 (UTC)
 * Can't you tell by context? If mi is a scalar then it can't be a vector product, can it?  cferrero 09:52 Mar 7, 2003 (UTC)

Use of %
I was going to rattle on at great length about this, but I'll be brief: I'm not really happy with disallowing the use of the % symbol - I think in lots of scientific contexts, spelling out "percent" would look considerably stranger than just using "%". --Camembert


 * (I've fiddled with this now, and am less unhappy, though I still have a nagging doubt (I have a predilection for nagging doubts, though...). --Camembert)


 * It was a bit of a unilateral decision by me so I fully expected to have it reverted! It was just based on the house style I am used to using, but I'd be glad to hear comments from others on whether they regard its use as correct. It is most often in the geography related articles (like the rambot city ones) that I have noticed it. Angela 06:19, Sep 15, 2003 (UTC)


 * Hm, the %s don't really bother me on the rambot pages - I think it's largely a matter of taste, which is probably why I have this nagging doubt. Still, having slept on it, I've decided I don't really mind the rule in its current form. By the way, for the sake of comparison, I think the Chicago Manual of Style says to use % in scientific copy and "percent" in humanistic copy, which is more or less what we've got here now. I don't know if there's anything on it in Fowler. --Camembert

You write: 10% and ten percent. 10 percent is as silly as ten %. Here on the 'pedia, our policy is to always use numbers instead of words, even for single-digit quantities. (I don't happen to like this policy, but no matter - it's the policy we have.) In consequence, we don't have to ponder this question: it's always % Tannin


 * That makes sense to me, but I think Chicago says to always use numerals in percentages, but use "%" or "percent" depending on context (I know Chicago isn't the last word, but it's widely used). I think the whole thing is probably too fiddly for us to be offering advice on, to be honest - it's a matter of taste, and I think we should just let people handle percentages how they see fit. Do we really have a policy that says always use numbers instead of words, btw? If so, where is it? --Camembert


 * I'm willing to let the percentages thing go, especially as the Americans will spell it wrong anyway :), but I think the numbers rule is important. Numeric values less than ten look unprofessional IMHO, so I'd like to leave that one in there. Angela 19:28, Sep 15, 2003 (UTC)


 * I agree, Angela. I'd prefer to see numbers smaller than ten be written out rather than done in numeric values - it does look unprofessional. But (unless someone has changed things while I wasn't looking), that's the Wikipedia rule, so I try to remember to respect it. Tannin 14:23, 16 Sep 2003 (UTC)


 * But where does it say that is the rule? Angela


 * Good question. I've certainly seen it somewhere, because I know perfectly well that I didn't much like it but figured I had to respect it. I'll have a hunt around. Tannin 00:14, 17 Sep 2003 (UTC)

BTW re the % versus percent debate, there is another complication. Many users of British English and Hiberno-English use per cent and see percent as an annoying americanism to be resisted on pain of death. One English teacher I know tells students that Americans use percent because they are in too much of a rush to bother writing two words where they think one will do. The 'rush' argument is also used jokingly used to explain the American 'inability' to spell colour properly, the u just being too much hazzle for all the lazy yanks! :-) Oh the joys of English (or as a surprised American commented - I had the pleasure of hearing it and laughing for days - on having to make an emergency landing at Shannon Airport, the first airport one meets when travelling from the US to Europe, and hearing Irish people speaking in English, "Gee, these guys speak American too!" An Irish traveller responded rather cuttingly, "the difference is, we can spell it too." Meow! lol FearÉIREANN 02:06, 17 Sep 2003 (UTC)


 * Bah. I dream of a world where the Americans learn to speak properly, and the English learn to spell sensibly! (Alas, I confess that I usually spell color as colour, but that's probably just habit. Or general laziness. Or something.) On the numerals vs spell-it-out thing, I can't find that rule anywhere. Maybe it has been changed? Tannin


 * Aha! Another (partial-)convert to the true way of spelling! :-)
 * On-topic, I would suggest that Angela's suggestion should stand; on whether we should use 'percent' or 'per cent' or, indeed, ' per cent ' (as one would use id est or ejusdem generis), I would be more for either the first or last, but more so for the former due to inertial difficulties in converting the dratted Americans ;-)
 * James F. 14:13, 17 Sep 2003 (UTC)

I am not native (although I use English more than other languages in my work), and I do most definitely don't argue that I represent anyone except myself, but let me tell that I as a foreign reader of English texts find that the guidance against the use of percent signs decreases the readability and the scanability of the text. I hold, firmly, that percent signs are superior to any of the two(!) spellings per cent and percent in most contexts where it at all is meaningful to speak about percents, as when the sizes of population minorities are given, or successive changes of something from year to year. On the other hand, there are other and better ways to express that roughly 25% or 50% or 60% is something, or goes somewhere... ;-) --Ruhrjung 18:29, 17 Sep 2003 (UTC)


 * IMHO, using % instead of per cent increases readability if several percentages are mentioned. -- User:Docu

I favour % over percent, spelling out numbers smaller than ten, and never spelling out percentages (1%, five, 14, 12.5%)

This should be at Manual of Style (dates and numbers). Martin 10:09, 24 Sep 2003 (UTC)

Protest to universal addition of metric measurements to US topic articles
I would like to protest the mechanical, universal addition of metric measurements to all US topic articles that is now occurring. I would like to see what the consensus is, and if there is support for my position. Please see my more complete entry at Wikipedia Talk:Manual of Style (dates and numbers). Thank you. --

Gary D 00:09, 1 Sep 2004 (UTC)


 * I'm OK with additions, as long as the English system is also used. But the tables in entries on U.S. states are all metric, which I think is inappropriate to the topic.

Maurreen 11 Sep 2004


 * Forget us. Just think of the reader. Most readers have as hard a time visualizing how high 126 feet is, as we have trouble imagining how much six stone weighs. The metrics should always follow any mention of feet... when you get around to it. Wetman 07:45, 16 Sep 2004 (UTC)
 * Exactly. Unfortunately we all too often forget that there might be readers and that some of them might find the Imperial system incomprehensible. Filiocht 08:37, 16 Sep 2004 (UTC)

Percent
Is there any convention about whether to use "%" or "percent"? Maurreen 11 Sep 2004

Hi, Maurreen. :) I have seen both usages occur with about equal frequency. func(talk) 07:17, 12 Sep 2004 (UTC)


 * ... and it's "per cent" or even "per cent" in British English. Life's so fun.
 * James F. (talk) 08:46, 12 Sep 2004 (UTC)

I seemed to remember this being codified in the MoS, but apparently I was mistaken. Conventional usage, in both AmE and BrE, is to use "%" when enumerating using Arabic numerals ("25%") and to spell out the word or phrase ("percent" or "per cent," according to dialect) along with (usu. shorter) spelled-out numbers ("five percent"). As for when to use which form, you're on your own, but as a general rule it's best to use Arabic numerals for percentages equal to or greater than ten. Austin Hair 00:38, Sep 13, 2004 (UTC)
 * Also spell it out if it begins the sentence. Rmhermen 01:03, Sep 13, 2004 (UTC)
 * Yes, good point, and also if it stands alone. Austin Hair 05:41, Sep 13, 2004 (UTC)


 * Here are some suggestions from a brief search of online style guides:
 * use digits with the symbol (' 5 % ' rather than ' five % ')
 * use decimal forms rather than fractions. e.g. ' 5.25 % ' rather than ' 5&frac14; % '
 * The Economist style guide says "Use the sign % instead of per cent. But write percentage, not %age (though in most contexts proportion or share is preferable)." The symbol form is international and is therefore particularly appropriate for an online encyclopedia. The symbol is easier than the word form for people who do not have english as their first language, and requires less effort for those who are kind enough to translate articles into other languages.
 * Treat the '%' symbol just as if it were any other unit symbol e.g. precede it with a space. For example:


 * Bobblewik (talk) 09:13, 13 Sep 2004 (UTC)

Associated Press style is to use figures even if the percentage is less than 10 and to spell out "percent." I think that's easier to read in body copy (in contrast to tabular matter).

The main reason I brought this up was because of the city listings. In the sections on demographics, I was distracted from the actual words by all the percentage signs. Maybe we could go with using the symbol, but encouragement for text that needs more than perhaps a few symbols to be put into a list or table.

Maurreen 13:03, 13 Sep 2004 (UTC)