Wikipedia talk:Manual of Style (dates and numbers)/Archive 82

From Wikipedia, the free encyclopedia
Jump to: navigation, search
Archive 75 Archive 80 Archive 81 Archive 82 Archive 83 Archive 84 Archive 85

The convert template and MOS

There is a bot going around removing {{convert}} templates from articles due to it causing performance problems. See here. Epbr123 07:44, 31 August 2007 (UTC)

Does it remove the conversion completely? Does this have a bearing on our discussion here? Tony 10:46, 31 August 2007 (UTC)
Sorry, your highness. Please don't hit me. This seemed the best section to mention it. Epbr123 11:54, 31 August 2007 (UTC)
Oh, :-( didn't intend to be bossy—sorry, it's the brevity of the wording that appears so. First question related to my concerns over that template (hard to make it comply with MOS); second question of concern to the process here. Tony 12:38, 31 August 2007 (UTC)
Sorry. The bot coverts "{{convert|10|mi|km|1}}" to "<span style="white-space:nowrap">10 miles (16.1 km)</span>". I'm not certain how it would effect the issues in this discussion, but if users are disuaded from using the convert template, it may discourage some users from adding equivalents due to the extra work in having to convert the units themselves. Also, if users had to convert the units themselves, this may lead to inaccurately calculated equivalents. Epbr123 12:52, 31 August 2007 (UTC)
The convert template either needs technical fixing or much clearer instructions on how to use it. I've seen examples of MOS breaches from its use, and I'm hoping the experts can work out why this is happening. Tony 13:58, 31 August 2007 (UTC)

The bot has been grounded, apparently indefinitely. Epbr123 may be encouraged to see how my implementation of the convert template corrected the inaccurate equivalents in TT Circuit Assen. This article was a poster child for the template on 3 counts: conversions were needed both ways (and you can't blame the Americans for this one) because records were kept in mph and then in kph; the conversions had been done incorrectly; MoS was not followed. Chris the speller 16:23, 31 August 2007 (UTC)

I have some comments, I am a *major* user of the convert template in my monobook metrication script.
  • 1. I don't much care *how* articles are metricated, as long as it is done widely and correctly. I see nothing wrong in principle with the convert template, plain text conversion, or editors switching between them.
  • 2. I am not aware of how the template affects performance and would welcome specialist debate on that.
  • 3. Wikipeda text in edit mode is becoming less accessible to the novice. That is a bad thing. I understand the motivation to go back from templates to plain text. I use the template purely because of its convenience to add metric units where none exist.
  • 4. The bot doing plain text conversion makes excessive use of 'nowrap' and 'nbsp'. The convert template does too. Line breaks *should* be permitted between metric and non-metric units. Line breaks *should* be permitted between each of the following words: 'seventeen', 'square', 'miles'. The 'nbsp' is another instance of the Wikipedia movement away from simple-to-read plain text. Personally, I would remove/downgrade/caveat the nbsp recommendation from the MOS for all but tables. In body text, the comparatively rare instance of the disease is preferable to the cure.
  • 5. Incidentally Chris, the symbol is not 'kph', it is 'km/h'. My monobook corrects that (plus 'kmph', 'km/hr' and other symbolic no-no's). I suppose a good thing about the template is that it guarantees that text gets the correct symbol.
I recommend people look at my monobook metrication script. Feel free to use it or copy the code for yourself. Thanks to Epbr123 for bringing it here. However, debate about the merits of the template versus plain text probably belong on the template talk page. Lightmouse 18:07, 31 August 2007 (UTC)
So where's the "how to use it for dummies" page, so I can test whether it does hyphens, ranges, abbreviations, singular/plural and appropriate decimal places, as required by MOSNUM? My problem with convert templates is that they take a lot of learning to get all of these things right, so most editors are better off doing it manually with a calculator (if you can't get simple calculations right, don't edit, or get someone else to check it for you) Tony 01:32, 1 September 2007 (UTC)
With regard to your first sentence, I am not able to answer those questions. Perhaps people on the talk page can. With regard to your second sentence, I agree with you. This debate is now taking place in three places, can we move non-MOS debate about the template to the template talk page please ? Lightmouse 10:51, 1 September 2007 (UTC)

Improve Extension:UnitsFormatter to solve English WP problems with units! (talk) 11:21, 24 November 2007 (UTC)

Natural numbers

We should not be proscribing natural number; it's idiomatic in both fields concerned. Furthermore, there are cases, such as the construction of the integers out of the naturals, where substituting some form of integer is simply, if subtly, wrong. Septentrionalis PMAnderson 22:35, 25 August 2007 (UTC)

So why not slow down this wagon and wait a day or two, or three? Then you can change it with greater likelihood of "consensus". Tony —The preceding signed but undated comment was added at 01:19, August 26, 2007 (UTC).
Because we should not have open error; and the existence of the error is uncontroversial among those who know the subjects concerned. (It is quite glaring from my point of view; but there are those who regard all mathematics as subtle). Septentrionalis PMAnderson 17:15, 26 August 2007 (UTC)
I must disagree with my esteemed colleague, Septentrionalis; constructing the integers from integers is not subtly wrong, but dramatically wrong. I do agree that the natural numbers are far too important to "write around", as any competent mathematician is expected to know. That's one reason they have their own Unicode code point, U+2115 (a double-struck capital "N"). And I regret to note that the first sentence of our article is misleading: Often the natural numbers are taken to be a model of the Peano axioms, augmented with addition, multiplication, and ordering; we have an algebraic structure, not just a set, contrary to the lead sentence.
The best resolution is to remove the section; either to say nothing, or leave it to the mathematics manual of style. It is dangerous instruction creep. (If a surgeon needs directions for how to tie off a suture, that surgeon has no business cutting anyone open!) If the consensus is to keep it in spite of my recommendation, it should be reworded as follows:

The set of natural numbers has two common meanings: {0,1,2,3,…}, which may also be called non-negative integers, and {1,2,3,…}, which may also be called positive integers. Use the sense appropriate to the field to which the subject of the article belongs if the field has a preferred convention. If the sense is unclear, and if it is important whether or not zero is included, consider using one of the alternative phrases rather than natural numbers if the context permits.

A less awkward wording might be possible, but this cannot be safely abbreviated. For example, we cannot eliminate the phrase "set of", because substitution should not be done for the algebraic structure. We cannot eliminate the word "common", because we have other meanings as well. We do not want to encourage a computer science convention in a number theory article. And so on.
What are the chances this will be read and understood? (Slim.) Do even the brilliant and highly educated readers of this talk page appreciate the fine points? (Probably not.) If we can't say something helpful for a general audience — and I don't think we can — then we should say nothing. --KSmrqT 09:46, 26 August 2007 (UTC)
I support removing the paragraph. The concept of naturals precedes the one of integers. Naming the former in terms of the latter is philosophically wrong. −Woodstone 10:47, 26 August 2007 (UTC)
On the other hand, "a prime is a positive integer which..." is a reasonable phrasing. If the writer is thinking of the primes as a special case of prime ideals in rings, it is natural. (No pun intended.) Septentrionalis PMAnderson 17:15, 26 August 2007 (UTC)

Direct quotes aren't linked?

A silly and unhelpful rule; sometimes a direct quote, like

"Whether a little farmer from South Carolina named Tillman is going to rule the Democrat Party in America - yet it is this, and not output, on which the proximate value of silver depends."

contains a proper name which should be linked to so the reader can tell who we're talking about without looking up our source. Please tell me where we put that piece of unhelpful nitpicking so I can go object to it. Septentrionalis PMAnderson 23:49, 25 August 2007 (UTC)

I strongly agree with the principle; if you link something in a quote you may be changing the intended context or interpretation. WP:MOSLINK#Quotation Altering this guidelie is a slippery slope, and you never know what garbage may be written into the article you link and how that will reflect on the original quote. SandyGeorgia (Talk) 23:56, 25 August 2007 (UTC) Example, you could link to Tillman today, and someone could change it to dab page tomorrow. SandyGeorgia (Talk) 23:56, 25 August 2007 (UTC)
Both of Sandy's arguments work against linking at all; they have no greater force against links in quotations than elsewhere. She may wish to rephrase the second: this is a piped link to Ben Tillman, who is certainly the primary use of that name. Septentrionalis PMAnderson 17:19, 26 August 2007 (UTC)

By the way, this is edit warring taken to a pathetic extreme. Please remove and discuss. We can't have different MOS pages contradicting each other. SandyGeorgia (Talk) 00:02, 26 August 2007 (UTC)

  • Manderson, you're trying the patience of people here, and making the page look unstable. Do what polite, cooperative contributors do: raise all but copy-editing changes here first. Otherwise, I'm not going to bother sorting out your edits—I'll just revert them all as being generally in bad faith. Tony 00:58, 26 August 2007 (UTC)

Regarding conversions in quotations, how about using the good-old square brackets? For example, 'John Doe said, "I ran six miles [10 km] today!". --Itub 06:48, 26 August 2007 (UTC)

Thanks; that's in my first green example above. Tony 07:45, 26 August 2007 (UTC)
  • It changes emphasis at least as much as wikilinking, and does not help in the example cited; the software will make it [[[Ben Tillman]]]. Septentrionalis PMAnderson 17:05, 26 August 2007 (UTC)
    • I was not talking about links, but about conversions, because that was the context where the edit in question occurred. Square brackets are commonly used for clarifications in quotations, and I think that adding a conversion could be a legitimate use. --Itub 08:36, 27 August 2007 (UTC)
      • I agree completely. As above, with the use of only metric units in some articles, this should be a matter of editorial discretion. Septentrionalis PMAnderson 15:07, 27 August 2007 (UTC)
I think that the general rule that links should be avoided in quotations is a good and important one. Sometimes there is need for clarification, and links perhaps should be used at such times, but very sparingly. If the surrounding context can make the meaning clear there should be no linking in the quote. Otherwise, it may be the least evil. CRGreathouse (t | c) 12:21, 27 August 2007 (UTC)
I agree with this too; linking in quotes should be done with extreme caution; but it is one solution to a difficult if minor problem. Septentrionalis PMAnderson 15:07, 27 August 2007 (UTC)
I think the rule should stay - but occasionally be ignored. The few times I have done this it tends to survive later edits. I would only do so if the alternative is a new sentence after the quote to allow insertion of a link.Johnbod 15:32, 27 August 2007 (UTC)
A reasonable test. Perhaps it should be added to WP:MOS (links), as guidance. Septentrionalis PMAnderson 15:57, 27 August 2007 (UTC)
So, CRGreathouse, do you have an example up your sleeve of a rare instance where a link within a quotation is necessary or desirable? Tony 11:27, 28 August 2007 (UTC)
I'm not really a good one to ask -- I always remove links from quotes, and have never added one. I only concede the possibility. CRGreathouse (t | c) 16:36, 28 August 2007 (UTC)
Without an example, it's hard to conceive why the existing policy text needs to be changed. Tony 00:54, 29 August 2007 (UTC)
Of course I don't mind leaving it -- I just didn't want the more extreme version suggested by PMAnderson. CRGreathouse (t | c) 19:17, 29 August 2007 (UTC)

Units on Astronomy Pages

On the page Goddard High Resolution Spectrograph there has been some thrashing on the units. I'm not an astronomer, but I do work with them and from what I've seen, for that instrument, Angstroms is probably the best choice - they're commonly used for UV and visible light astronomy. IR folks tend to use microns. I don't know about the radio folks. Dfmclean 01:27, 28 August 2007 (UTC)

As for Angstrom, it has a 10:1 relationship with nano-metres (nm) so the conversion back is easy. anon 2007-08-30 16:45 UTC —Preceding unsigned comment added by (talk)
What puzzles me is how you measure resolving power in km/s. Doppler effect? But if so, you should say so. WikiProject Physics is more likely to find a radio astronomer than here. Septentrionalis PMAnderson 01:45, 28 August 2007 (UTC)
Uh, there has been one edit and a revert by me, hardly 'thrashing'. Optical astronomy universally uses angstroms, so we should stick with them. Furthermore, the source lists the numbers in angstroms. IR uses microns (mostly) and radio usually quotes in frequency, so GHz and MHz. Yes the resolutions are quoted in km/s due to the doppler effect (hence they are quoted for a given wavelength), because that's one of the main things high resolution spectrographs are used to measure. I'd pull out the manual and grab some more details, but I doubt anyone is going to be that interested in an obsolete spectrograph that is no longer working! Oh and a quick disclaimer: I'm an optical/IR astronomer who has also worked on a radio telescope.
Oh and final thought, perhaps there should be a wikiproject astronomy page on units? Modest Genius talk 14:38, 3 September 2007 (UTC)

Really necessary to have a template showing all units of length?

Please look at the use of Template:Unit of length. Somebody apparently believed it was necessary to have a template showing all units of length in terms of ångströms, astronomical units, light years, inches etc. It seems a bit silly to me. From a metric readers point of view, it is just weird. It may derive from the no-longer-amusing joke about furlongs-per-fortnight but Wikipedia articles are not supposed to be internet jokes. The template assumes readers of the kilometre article need to know how many ångströms it contains. Similarly the reader of the millimetre is supposed to be interested in how many light years it is. I would not mind if this silly template were confined to one or two articles, but it has been put on lots.

Please comment at Wikipedia:Templates_for_deletion#Template:Unit_of_length
Lightmouse 15:59, 28 August 2007 (UTC)

Yes, but the template also tells you how many feet are in a kilometer. It's blatantly obvious that this is useful information. --JayHenry 21:06, 28 August 2007 (UTC)
Blatently obvious information like that is already in the articles. A fixed template that includes obscure or unlikely conversions makes it not useful. Lightmouse 21:21, 28 August 2007 (UTC)
Yep, I'd like examples of where some of these obscure unit conversions are useful to justify their retention. Tony 00:55, 29 August 2007 (UTC)
Oh I agree with that. I'm not arguing to keep the conversion to ångströms, that's a matter for template talk though. Anyways, it's all a discussion at TFD for now, sorry for cluttering things here with it. Cheers! --JayHenry 01:20, 29 August 2007 (UTC)
I've been working on it and have removed the most absurd conversions. There's still a lot of room for improvement though. Jɪmp 08:28, 29 August 2007 (UTC)

Links vs. <font color=darkblue>

First of all, I fail to see how using links, with pipes to make sure it shows what is in fact displayed, is bad. There are several problems with the previous markup - for one thing, it doesn't look like a link even in the default theme (real link color is much brighter blue), and the whole point of having it in any color at all is to simulate a link. Second, it won't change if someone has different CSS, say, in the worst case, what if they have a dark blue background and white text? the "links" will be invisible then. But even if they simply have a different color (gray, green, whatever) for links, it won't match. Also, <font> is a bad tag regardless - and whoever made this seems to think the closing tag is </font color>. These are the reasons I made this change, and I think it should stand. --Random832 23:52, 30 August 2007 (UTC)

I changed the color from "darkblue" to "blue". I suppose a cleaner version would be to set the style so that it would display as if it were a link, but I don't think we want real links. As for </font color>, in some HTML versions, </font color> doesn't end a <font size>, but only a <font color>. I don't know if that's correct HTML, but it seems to work. — Arthur Rubin | (talk) 13:57, 31 August 2007 (UTC)
That's not how HTML works. There is no, and there has never been any, html version where there are separate "font size" and "font color" tags. Can you explain why we don't want real links? They don't do any harm, and they're the only thing guaranteed to display the same as links on everyone's system (what if I have links underlined? what if I have a stylesheet that makes "blue links" red and "red links" orange?) - can you explain your objection to my change in the first place, and why you reverted it? Regardless of whether a font color tag is an adequate solution, I don't see what's wrong with having them as links. Why don't you think we want real links? It's possible that whoever made the markup that way didn't realize pipe links were a solution so thought he/she needed to make fake links w/ font color tags to get around the autoformatting (since, obviously, we _don't_ want autoformatting here) --Random832 17:17, 31 August 2007 (UTC)
I would favour the use of links, so that they appear exactly as they would under the relevant preferences. SamBC(talk) 19:01, 31 August 2007 (UTC)
Are you sure that the pipe links won't be autoformatted in future versions of the Wikimedia software, or automatically removed by reformatting bots? Thinking it over, the real links don't hurt, other than those problems. — Arthur Rubin | (talk) 19:42, 31 August 2007 (UTC)
Well, I'm sure that if that happens it'll be noticed pretty quickly and worked around again. If there's no problem now (as in the case of autoformatting) or only a worry that there might be a problem (as in the case of bots), then I'd say it makes sense to go ahead, rather than worrying about hypotheticals. SamBC(talk) 19:49, 31 August 2007 (UTC)
The correct closing tag to both is </font>. And the best way to implement such a thing is with an inline CSS style, not the deprecated (read: valid but bad) <font> tag. Now, what exactly are we talking about? — Aluvus t/c 05:13, 2 September 2007 (UTC)
Yeah, I'm not following this either (other than your accurate corrections of the HTML issues). The idea of creating fake links strikes me as very "user-hateful", but maybe I'm missing something here... — SMcCandlish [talk] [cont] ‹(-¿-)› 10:58, 8 December 2007 (UTC)

Metric conversions of knots and nautical miles

Although the SI system is used worldwide for most applications, it certainly seems that nautical distances and speeds are still referred to almost exclusively in terms of nautical miles and knots. It seems unnecessary to convert knots to km/h and nautical miles in every instance. Ranges for guns are often expressed in metres or yards, but longer distances are usually nautical miles, and even sources that use kilometres for these distances use knots to represent the speed. Whether this is logical may be debatable, however, it is the standard of many sources. Almost any source for information on warships uses the knot standard. Sacxpert 19:39, 1 September 2007 (UTC)

Nautical and ordinance measurements should be converted to metric equivalents so that people not familiar with knots can easily appreciate the facts presented.
I was very tempted to respond to this by suggesting scientists who use metric should learn knots so that they can understand nautical articles. I fear that the recent exception to conversions for scientific articles has created precedence which will encourage more of these proposals. I hope we don't end up with articles which are only accessible by people already familiar with the subject area. -- PatLeahy (talk) 19:56, 1 September 2007 (UTC)
I'm perfectly comfortable with SI measurements, but I do not convert in several cases:
  1. Only a traditional measurement is stated within a legal document, especially one under UN or other auspices where it is reasonable to assume that metric-oriented people have had substantial involvement in the text
  2. Where a traditional measurement is nominative, as the "10 yard line" or "100 meter dash" is customary in a sport
  3. In cases where several traditional measurement types have a specific interrelation, such as knots, nautical miles, and minutes of latitude
The BIPM, which is the international authority on SI, includes "knots" as one of a number of non-SI units that will be acceptable within the SI system. Howard C. Berkowitz 20:13, 1 September 2007 (UTC)
I would point out that knots and nautical miles are still commonly used in aviation as well. Still, I see no problem with providing metric conversions in parentheses following their use. Askari Mark (Talk) 20:31, 1 September 2007 (UTC)
Yes, they are "common in aviation". They are not, however, universal in aviation, and how common they are varies in different contexts within aviation (indicated air speed, wind speed in weather reports, military vs. civilian, nationality, etc.). Some in aviation do use SI units, and most Wikipedia readers are not involved in aviation: SI conversions should be included. Gene Nygaard 18:40, 13 October 2007 (UTC)
Furthermore, statute miles are also "common" in aviation, especially in most contexts before the middle of the 20th century, in others before the last quarter of the century, and in others continuing until the present. Whenever "miles" are used in an aviation context, it is important to specifically identify them as either statute or nautical. 18:48, 13 October 2007 (UTC) —Preceding unsigned comment added by Gene Nygaard (talkcontribs)
Following this train of thought, can anyone tell me why NASA use nautical miles to measure height? Thunderbird2 18:58, 13 October 2007 (UTC)
Just in a negative sense. It's not because anybody else in aviation or space uses nautical miles to measure vertical distances. Gene Nygaard 19:53, 13 October 2007 (UTC)
And continuing on the what-it's-not train, it's not because they are following the recommendations of their own Inspector General following the Mars Climate Crash-Lander, to stick to SI units. Gene Nygaard 19:59, 13 October 2007 (UTC)


In the continuing "edit feuding" over knots and nautical miles, I've noticed some curious abbreviations being touted. For what it's worth, "knots" is most commonly abbreviated "kt" (not "kn"), and "nautical miles" as "nm" – and these are both widely used on Wikipedia. I haven't changed the recent edits, since I think it proper to first develop consensus that "kt" and "nm" are best usage. Askari Mark (Talk) 01:15, 3 September 2007 (UTC)

I'd definitely agree that kt for knot is probably best usage (although it is similar to kT, which is short for kiloton-TNT-equivalent). I've certainly seen "nm" for nautical mile, but the problem is that nm is also the symbol for nanometre, one thousand-millionth of a metre. Granted, it's unlikely that anyone would confuse the two, but perhaps "nmi" is more precise? According to Wikipedia's own article on nautical miles: "There is no official international standard symbol for the unit nautical mile. The symbols M, NM, nm, and nmi are commonly used in some areas." Of those, it seems that nmi is probably best, to avoid confusion, even if nm is common. Sacxpert 01:25, 3 September 2007 (UTC)
Sacxpert, you make good points. "kt" seems safe enough since it would be rare to use explosive yield in a context of speed. Kinetic energy weapons are a fairly bizarre edge case, and their energy is often specified in joules. I can certainly live with nmi, if for no other reason than avoiding confusion with nanometers. Howard C. Berkowitz 02:41, 3 September 2007 (UTC)
The template {{convert}} uses nmi as the abbreviation already. I'm surprised no one mentioned Nm (Newton meters) above.  MJCdetroit 03:21, 3 September 2007 (UTC)
Actually, I was going to. :-) Also, I’ve never seen “M” for nautical miles. Actually, I’ve never seen “nmi” used either; “n.mi.” or “n. mi.”, but not run together. In any case, I have no problem with opting for “nmi”, although “nm” has come to be the modern preference, but if people really think it’s likely to be mistaken for nanometer (which would seem to be most unlikely IMO), then nmi would be okay by me (though less preferable) … although there are a lot of extant “nm” to change. Askari Mark (Talk) 03:34, 3 September 2007 (UTC)
To be fully honest, I prefer nm to nmi, but I will accept either if that will end the conversion wars in nautical articles and let me focus, again, on content. I would worry very much about someone that measured fishing vessel positions in nanometers. Howard C. Berkowitz 04:12, 3 September 2007 (UTC)
The IEEE guidelines for authors are clear one this: "nmi" for the nautical mile and "kn" for the knot. The guidelines are generally followed in scientific literature. There is a separate discussion on each at SHIPS. Thunderbird2 17:04, 7 September 2007 (UTC)
But other authorities differ in both regards, and our standard on Wikipedia is "kt" for knots, explicitly decided on after considerable discussion in the aviation WikiProject, for example. Note also that when it comes to tons, the symbol "kt" should only be used for units of mass, not for units of energy. If we insist on both "kt of TNT" (or "Mt of TNT" when appropriate) when those energy units are used, always, not kt alone, and a conversion to joules when those non-SI units are used, then there is little chance of confusion. A bigger problem with tons, of course, is that the t is ambiguous, being used for both long tons and short tons as well as metric tons. If I had my druthers, we'd have an all-megagrams (gigagrams, teragrams, etc.) encyclopedia! Gene Nygaard 10:20, 17 October 2007 (UTC)
Forgot to mention that we have very, very few (if any) articles using "kt" as mass units. Hardly anybody uses "kilotons" or "megatonnes" or whatever outside the TNT-equivalent units of energy context; otherwise it is "thousands of metric tons" or "million tons" or "million tonnes" and the like. I don't hear anybody say that Algeria made a deal to "buy 2.5 megatons" of wheat from the U.S., for example, but rather they "buy 2.5 million tons" or whatever precise amount[1] [2] of wheat. (Adding mega- would be clumsy if the U.S. name "metric tons" were used in this context; we'd have people arguing for a hyphen and others insisting on an en dash in mega–metric tons, or whatever. Even adding "million" is clumsy, that's one reason why it is often just "million tons".) Sure would be simpler, and get rid of the ambiguity, if the world's grain traders would start referring to this quantity as 2.5 teragrams. Gene Nygaard 10:52, 17 October 2007 (UTC)
To MJCdetroit: newton (that's lowercase n) meters are only a problem on Wikipedia links, and here only for two reasons: 1) because some Wikipedia editors fail to include the requisite space or centered dot in N m or N·m, and 2) because initial capitalization is turned on in English Wikipedia, so links to Nm and to nm go to the same place. I've never seen anybody use an uppercase N and a lowercase m for nautical miles, so there is unlikely to be any problem in what people read, and a bad link only takes them to the disambiguation page. Gene Nygaard 20:06, 13 October 2007 (UTC)