Misplaced Pages

Revision as of 16:43, 25 August 2009

Manual of Style

This page falls within the scope of the Misplaced Pages:Manual of Style, a collaborative effort focused on enhancing clarity, consistency, and cohesiveness across the Manual of Style (MoS) guidelines by addressing inconsistencies, refining language, and integrating guidance effectively.Manual of StyleWikipedia:WikiProject Manual of StyleTemplate:WikiProject Manual of StyleManual of Style This page falls under the contentious topics procedure and is given additional attention, as it closely associated to the English Misplaced Pages Manual of Style, and the article titles policy. Both areas are subjects of debate. Contributors are urged to review the awareness criteria carefully and exercise caution when editing. For information on Misplaced Pages's approach to the establishment of new policies and guidelines, refer to WP:PROPOSAL. Additionally, guidance on how to contribute to the development and revision of Misplaced Pages policies of Misplaced Pages's policy and guideline documents is available, offering valuable insights and recommendations.

Archives

General 1, 2, 3, 4, 5, vote, 6, 7, 8, 9 (AD/CE), 10, 11, 12, 13, 14, 14a, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 49a, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161 Binary prefixes B0, B1, B2, B3, B4, B5, B6, B7, B8, B9, B10, B11, B12, B13, B14, B15, B16, B17 Years and dates D1, D2, D3, D4, D5, D6, D7, D8, D9, D10, D11, D12, D13, D14 See also Misplaced Pages:WikiProject Dates Misplaced Pages talk:Date debate Misplaced Pages talk:Naming conventions (calendar dates) Misplaced Pages talk:Timeline standards Misplaced Pages:Measurements Debate Misplaced Pages talk:Manual of Style (dates and numbers)/Date autoformatting Complete rewrite of Units of Measurements (June 2008) Dec. 2008 RfC on Date linking/autoformatting (three proposals) Dec. 2008 RfC on Date linking/autoformatting (detailed)

This page has archives. Sections older than 7 days may be automatically archived by Lowercase sigmabot III.

Anomalies in the template for converting measurements.

I have discovered an anomaly with using the 'convert' template. Look at this:

4,700 square miles (12,000 km)

4,699 square miles (12,170 km)

When converted into square kilometres, 4699 sq miles comes out 170 km more than 4700 sq miles. Lose one square mile, gain 65.6 square miles!

I think the problem may occur because of rounding when a number ends in two zeros.

Therefore it is risky to rely blindly on the conversion template. It may give anomalous results.

Please let me know if this issue should be raised elsewhere. Michael Glass (talk) 06:45, 8 August 2009 (UTC)

{{convert|4700|mi2|sigfig=4}} gives 4,700 square miles (12,170 km). You just have to use the significant figures setting. For many articles that rounding is appropriate; if you write 4,700 nobody is going to expect it to be exactly that. If you write 4,699 they will expect it to be exactly that. SimonTrew (talk) 12:07, 8 August 2009 (UTC)

Well, in my humble view, this is another reason not to use a conversion template, but to allow fellow editors, including newbies, maximum control over all aspects of the construction. I'm all for keeping it simple, and if that means using a calculator, so be it. Tony (talk) 12:57, 8 August 2009 (UTC)

I disagree. If a newbie just throws the numbers in, a more experienced editor can use {{convert}} later. The thing about using the template, if e.g. something changes in MOS then EVERY article that uses it will get that change. SimonTrew (talk) 13:45, 8 August 2009 (UTC)

Well I do agree to the extent that it can be fiddly to get right and there are some things it simply can't do. But on the whole, if you can do it with the template, you should. SimonTrew (talk) 13:53, 8 August 2009 (UTC)

I think the trouble is that as the rounding cuts in automatically unless it is overriden. The default position can be a trap for the unwary. An over-precise conversion can be overridden; the opposite may not be noticed by the uninformed.Michael Glass (talk) 14:08, 8 August 2009 (UTC)

The template reads the precision of the input number and matches the precision of the output accordingly. This is in accord with the MoS and normal mathematical practice. There is no anomaly. An over-precise conversion can be overridden, sure, but are the uninformed any more likely to fix an over-precise than an under-precise one? In the majority of cases the template will not give the wrong precision. 4,700 sq mi ≈ 12,000 km is correct as is 4,699 sq mi ≈ 12,170 km. It is the conversion templates without this type of default rounding which are more prone to produce output with incorrect precision. To those unwary I can only say "Get wary." JIMp _talk·cont 15:11, 8 August 2009 (UTC)

It's a bit of a circular argument that it conforms with MoS since this is a discussion on a subpage of the MoS.

I do agree however that yeah if you put in 4,700 it is unlikely it is going to be exactly 4,700 (of whatever) and so misleading to a ridiculously over-precise value. If you pur in 4,699 presumably you mean that (and not 4,698 or 4.701) and should be converted more accurately. It comes down to common sense. My little book here gives logs and other things such as trig functions to 8 sig figs. Most floating point maths used in computers is at double-precision and has about 12 sig figs (decimal) in the mantissa. If it is good enough for rocket science and subatomic modelling (and it is cos I have programmed it) then I think it ridiculous to be more precise than that. There are all kinds of more precise representations but in real science and engineering that is plenty: in fact single precision (5 or 6 sig figs decimal) is usually good enough, but with modern processors using a double is as fast, if not faster, than a single. And the same applies to humans as computers. Of course propagation errors may occasionally require higher representations in intermediate results, but on Misplaced Pages that hardly applies, unless the underlying templates are so off that they preduce an obviously bizarre result such as 1 cm = 1.00001 cm or whatever.

I see no problem here. These templates are a bit fiddly and I am sure if we were starting from scratch we would change a lot of things, but we aren't and we can't. It is one of those things an editor just has to do. The {{cite}} templates are also trick but everyone expects those to be used. SimonTrew (talk) 04:55, 9 August 2009 (UTC)

If we put the template in question up against a statistically representative sample of typical editors with calculators, I'd put my money behind the template to give the most appropriate levels of precision for the typical measurements you find on Misplaced Pages. JIMp _talk·cont 08:06, 9 August 2009 (UTC)

The behaviour contradicts the principle of least surprise, but as a mathematician I think this is one of the situations where that is actually justified. Besides, I don't see how we can avoid it. Consider:

Notice that in the last column we always write 4,700, whether we mean 2, 3 or 4 significant digits. Similarly, in the penultimate column it's not clear whether 470 has 2 or 3 significant digits. The template needs to guess. We can't make it guess correctly in all situations, but if it makes sure to under-, rather than overestimate the number of significant digits in doubtful cases, then it's more likely to be wrong right than if it does the opposite. And even if the template gets the intended number of significant digits wrong, then except in situations where a human reader can infer it from the context it is usually correct and encyclopedic to round the numbers.

But there is an unrelated anomaly in the top right cell of the table. I am taking this to Template talk:Convert, because it's clearly a bug that needs fixing. Hans Adler 09:40, 9 August 2009 (UTC)

¶ I don't see the anomaly at first glance, Hans. If you're being so precise as to specify 4,700.0 square miles, then you are giving 5 significant digits, which are converted to 5 significant metric digits. The range is between 4,699.95 sq. miles and 4,700.05 sq. miles; otherwise the figure would be either no more than 4,699.9 square miles or no less than 4,700.1 square miles —— Shakescene (talk) 21:05, 10 August 2009 (UTC)

Do we really need this whole conversion business at all? As stated above, the conversion is not 100% reliable and it's not "official info" anyway. Maybe we should just use English units in American/British articles and metric units everywhere else. Offliner (talk) 10:29, 9 August 2009 (UTC)

But this is the English-LANGUAGE Misplaced Pages, not tied to any one country.

In the UK and Ireland metric is commonly mixed with Imperial units in everyday life; in the UK some things such as road distances and heights, and pints of beer served on draight, MUST be measured in Imperial; in Canada metric is most commonly used, with occasional use of US Customary simply because of the close ties with the US; in Australia and New Zealand exclusively metric and so on.

Last night I was talking to someone from mexico and said that Cambridge was about 80 km from London, whereas I would never say that to an English person (I would say about 50 miles). Yes, the conversions need to exist. Again, if using {{convert}} and the MoS then says "don't use conversions except in this or that circumstance" we can probably change the templates and 90% of articles will immediately conform (the remainder being for things like historical use of units, or quoted sources, etc). So, even if the conversion is not particularly useful in itself, simply as a marker that "this is a measurement" is. I know {{val}} also stands for that purpose but the same applies, mutatis mutandis. It is also extremely helpful for people translating across different Misplaced Pages (what am I supposed to do, for example, if I translate a French article about an American car?) SimonTrew (talk) 10:51, 9 August 2009 (UTC)

To Hans: I would not write 4,700 if only the first two digits are significant: instead, I'd write 4.7 thousand; likewise, I'd not write "50" if only the first digit is significant (third cell of bottom row): I'd write "fifty" (or even "about fifty", if appropriate). And I'd like the template to assume that the zero in "50" is significant, for this reason. I once read about an editor converting numbers such as 187, 190, 191, and 194 and being surprised that the second one was converted with less precision. --A. di M. 13:12, 9 August 2009 (UTC)

Just noticed a bug: {{convert|470.00|mi2|km2}} gives 470.00 square miles (1,217.3 km), with one 0 after the point instead of two. --A. di M. 13:25, 9 August 2009 (UTC)

I might say 4.7 thousand square metres too, though many don't, and the template has the capacity for a number of such constructions. However, if you see "4,700", the safest assumption is that it's precise to the nearest 100. As for the bug mentioned above, it should be fixed as soon as an admin puts in the new code. JIMp _talk·cont 17:48, 9 August 2009 (UTC)

Hum... maybe for 4,700 I agree, but in cases such as 5,000 I think that assuming that none of the zeroes is significant is excessive. When giving numbers with two sigfigs, there is a 1-in-10 chance that the second one is 0. So I'd assume that numbers such as 500,000 have two significant digits. What do you others think? --A. di M. 22:15, 9 August 2009 (UTC)

Which should generally boil down to giving an output of at least two significant figures which is just what the template does. So, yeah, I think that's just about right. JIMp _talk·cont 00:32, 10 August 2009 (UTC)

No, it should assume at least two significant digits in the input, not in the output. 14 kilometres (8.7 mi) is excess precision. --A.  di M. 23:48, 10 August 2009 (UTC)

I think a template that automatically treats a zero as not a significant figure is a bit of a worry. Take the areas of the states of the United States. it would be bizarre to think that the areas of the states of Washington and North Dakota were less accurately surveyed than the other 48 states simply because the areas in square miles happened to end in two zeros . Michael Glass (talk) 13:02, 10 August 2009 (UTC)

So, your first step is to compare these with the areas of other states and base your rounding on that. Without this comparison how else could you justify assuming the zeros were significant? Zeros may or may not be significant whether you use the template, a calculator or do the conversion in your head, you have to deal with this. With nothing else to go on standard practice is to regard the zeros as not significant (and generally they aren't). JIMp _talk·cont 13:42, 10 August 2009 (UTC)

My 2003 Statistical Abstract of the United States (Table No. 359) converts North Dakota's 70,700 sq. mi. (yes MOSNUM fanatics: sq. mi. is their official U.S. Government statistical abbreviation, periods and all, not some archaic idiosyncrasy of mine) to 183,112 sq. km. (their usage again) and Washington's 71,300 as 184,665. (I might be able to dig out the actual acreage, square feet or square meters from somewhere in my desktop's archives.) Just as in the debate over what to introduce between quotation marks, sometimes there's no substitute for checking the source, nor for making your degree of precision clear to the ordinary, non-technical reader. —— Shakescene (talk) 21:05, 10 August 2009 (UTC)

Treating a number ending in a zero as being automatically less accurate than a number ending in any other numeral strikes me as some kind of magical thinking. Obviously there are times when the zero is not significant, but by taking it for granted that it is always non-significant is like some kind of superstition, where inaccuracies come in even tens and hundreds just as bad luck is supposed to come in threes. These things need to be judged on a case by case basis. Michael Glass (talk) 14:00, 12 August 2009 (UTC)

It may seem like magic but it's a pretty standard approach which in the vast majority of cases gives appropriate results. Of course there will be cases where the zero is significant but the template makes allowance for that. If you prefer the calculator, there's nothing wrong with that but you'll still have to know what you're doing. JIMp _talk·cont 15:30, 12 August 2009 (UTC)

I agree with Jimp. There is a clear difference in implied precision between using 4000 and 4000.0 when specifying a measurement. The default should be to use standard scientific convention for specification of precision. Any "anomalous behavior" can be corrected on a case-by-case basis. Often this is caused by the choice of units, e.g. 4 km vs. 4000 m. However, there are some cases where uniformity in units is desired and there can be unintended results. (For example, the areadisp in {{infobox settlement}}) Plastikspork (talk) 16:36, 12 August 2009 (UTC)

It's not magical thinking, it's user friendliness. The only reasonable alternative would be for the template not to guess a precision and simply print a red error message when no precision is provided explicitly. That wouldn't exactly encourage editors to use it.

A better example of magical thinking in this context is nonsense such as the claim in the article Vienna that the city had 1,680,266 inhabitants in the 1st quarter of 2009. I hope that's not true, because a completely constant number of inhabitants over 3 months can only be achieved with very rigorous methods that I would probably suffer from when I move there. The same editors who insist on writing out even the last digit in such a case (it happens in all the best encyclopedias, not just Misplaced Pages) would shy away in horror from the claim that when computed using a certain well-defined algorithm based on well-defined official data, the average population of a certain village in 2008 was "1680.266 inhabitants". Hans Adler 20:40, 12 August 2009 (UTC)

Hans, I cannot express how much I enjoyed your comment. Well put. Studerby (talk) 01:30, 21 August 2009 (UTC)

WP:OTHERDATE

This section has been moved to Wikipedia_talk:Manual_of_Style_(dates_and_numbers)/Datestempprotectedsection, as this page is more relevant to the request.

"Adopting suggestions from standards bodies"

I have put the two paragraphs under this heading in their own subsection. The reasons I did this are as follows:

• The original placement broke the style of what preceded and followed it.
• The original placement split two related subsections.

The passage was out of place where I found it. However, someone might be able to find a better place for it than where I have put it. Michael Glass (talk) 08:03, 15 August 2009 (UTC)

I made some changes to the tone and phrasing of that paragraph—it was almost aggressive, and belaboured the idea of 'the real world', without adequately referring to the role of consensus. I also managed to simplify things a little bit. Do the changes look reasonable? TheFeds 20:42, 15 August 2009 (UTC)

The "real world" was use to emphasize that the overnight consensus of 20 people should not force an obscure standard into the Manual Of Style. The consensus has to reflect the writing style of the real world not some ideal dream world. It took 3 years to rid Misplaced Pages of the Kibibytes and Mebibytes crusade. -- SWTPC6800 (talk) 20:58, 15 August 2009 (UTC)

The rephrased version was reverted by Greg_L (talk · contribs), with the edit summary "No consensus for this change, which was extensively discussed". Since the rephrasing didn't significantly alter the meaning of the paragraphs—though it does adjust the tone—there's no case of violating prior consensus here. (The lengthy prior discussions resulted in consensus leading to the banishment of things like " %" and "MiB" from the MOS—these outcomes were not challenged by the revision.) We've had "B", and now "R"; let's have the "D". TheFeds 07:27, 18 August 2009 (UTC)

Definitions needed?

One of the first pieces of advice in the section on units of measurement says:

• Aim to write so you cannot be misunderstood.

Unfortunately this section undermines this aim by using two terms that may defy understanding. These are

• region-specific topics, and
• internationally accepted units.

To clear up confusion we need to define what we mean by region-specific topics and internationally accepted units.

In plain English, a region-specific topic, for example, may refer to any region and any topic, e.g., the Pavillon_de_Breteuil, the home of the International Bureau of Weights and Measures. But of course it is referring in a roundabout way to US and UK based articles that happen to use Imperial or US Customary units. In this case it may be better to find some other term that won't be so ambiguous or confusing or simply write US-specific articles and some UK-based articles.

Internationally accepted units may need explanaion, perhaps like this:

Internationally accepted units are:

• SI base and derived units
• Non-SI units accepted for use with the SI
• Units based on fundamental constants
• other non-SI units that are used internationally

What do others think?Michael Glass (talk) 08:48, 15 August 2009 (UTC)

That all sounds pretty reasonable to me. I certainly agree about region-specific topic. My only minor comment there is that, I imagine, some US-specific articles may also use metric/SI, e.g. those that are writing about science or engineering projects that are based in the US, but use metric/SI units (e.g. NASA projects, and, in theory, all US government projects). So perhaps "most US-specific articles and some UK-specific articles". And why "specifc" for one but "based" for the other?

Cut "base and derived" - just put "SI units"

"other non-SI units that are used internationally; I think you could cut "non-SI" here as it is implied (it is more appropriate for the ones "accepted for use with (the?) SI"). Perhaps give examples, e.g. the degree (angle), nauticl mile? And for constants e.g. Planck's constant (physics), Pi (mathematics), etc. For other non-SI units perhaps carat (gemology). No need for an exhaustive list, just a couple of examples for each.

Since in a sense US Customary/Imperial units are also accepted internationally, perhaps anyway this term is inappropriate. Suggest "widely used worldwide" but that is not brilliant either. SimonTrew (talk) 21:10, 15 August 2009 (UTC)

It's intended to refer to the unit in most common use worldwide for the type of measurement being mentioned. For example, the diagonals of cathode-ray tube displays are by far most commonly measured in inches around the world; exceptions are South Africa and Australia where centimetres are used. So, the diagonal of a CRT display should be given in inches and followed by a parenthetical conversion to centimetres; but an article specifically about South African CRTs would use centimetres, with a conversion to inches between parentheses. I'd use "for a given measurement, use the unit which is most commonly used worldwide for that type of measurement", or some less wordy equivalent thereof if anybody can find one. The "for that type of measurement" part is essential, both in non-regional and in regional articles: the unit commonly used for the energy of airsoft pellets is not the same used for ultra-high energy cosmic rays, despite being measurements with the same dimension and roughly the same magnitude; likewise, as Trew said, the units commonly used in US engineering are not the same units commonly used in US household items. (As for "Non-SI units accepted for use with the SI", that's the name the SI uses verbatim, so I'd keep the "non-SI" even if it's redundant. And a list of internationally accepted units including "SI units" and "other non-SI internationally accepted units" looks tautological to me.) As for "region-specific topic", that's as in WP:ENGVAR. --   A. di M. 22:05, 15 August 2009 (UTC)

I agree.

We clearly need some units that are not accepted by SI at all for use throughout Misplaced Pages per the principle you brought up above. Obviously, we aren't going to do away with the month or the year, even though neither is recognised by SI for the very good reason that neither is of consistent length. There is also clear benefit in using units such as inches, feet and nautical miles in contexts where they are common internationally, even if they are not defined by SI (I think the nautical mile is, but not the other two). On region-specific topics where there are region-specific deviations from these units (and this is not just in the US and UK), we should adopt the region-specific deviations. Pfainuk talk 22:51, 15 August 2009 (UTC)

Non-SI units accepted for use with SI lists the minute, hour, day, degree of arc, minute of arc, second of arc, hectare, litre and tonne as non-SI units accepted for use with SI. It doesn't seem to mention multiples of the litre (millilitre, kilolitre, megalitre, etc.). It doesn't mention the week, month, year, decade, century, annum, kiloannum, megaannum, etc. It doesn't mention the kilometre per hour, litre per hundred kilometres, etc. These should be allowed without SI coversions.

The electronvolt (kiloelectronvolt, megaelectronvolt, etc.) is not SI nor is it based on fundamental constants but a hybrid of both but these should be allowed and we generally won't need to convert them to SI.

What about the kilowatt-hour, debye, astronomical unit, lightyear and parsec? I'd be inclined to convert them to SI depending on context.

"other non-SI units that are used internationally" is a little vague. Certainly we'd want the nautical mile and knot in certain contexts but a conversion to kilometres and miles (per hour) would be desirable. As noted above, we could agrue that imperial/US units are used internationally but we'd want these converted. Many units (e.g. the carat, calorie, ton of TNT, oil barrel, millimetre of mercury, atmosphere and Troy ounce) are used internationally but should be converted to SI.

We'd be better of ditching the ångström, bar, millibar, etc. entirely but there probably is little hope of that; however, we don't really need to convert these to SI (since it's just a matter of moving the decimal point). JIMp _talk·cont 23:02, 15 August 2009 (UTC)

Surely millimetre, kilolitre, kilometre per hour etc are SI derived units (or, at least for the latter, a non-SI derived unit accepted within SI, by transitivity)? I suggested butting "base and derived units" but surely e.g. millimetre is a dervied unit and thus SI.

As for angstrom, you will never get rid of this; in the field I work for (molecular modeling) it is pretty much a standard unit, and to write as tenths (or tenths?) of nanometres would be just odd. By the way my dictionary lists it as "angstrom" with no ring above, though of course its symbol is Å.

calorie poses a unique problem in that of course it should be kilocalorie, and using calorie in the convert template (for typical values for foods) puts the result in joules, not kilojoules i.e. it takes a calorie to be a calorie, not a kilocalorie. Yet the main text may use calorie and it looks clumsy or unduly pedantic to write kilocalorie. I almost had this problem with Bacon today; I used the convert template (with kcal) but elsewhere in the text it used "calorie" and it would have seemed pedantic to change it. Fortunately I escaped that one first as it was quoted (aw "zero calorie") and second since zero calories = zero kilocalories I could avoid the issue.

bar and millibar I would have more support for ditching, although certainly for weather forecasts in the UK the pascal is unheard of, and lines of the same pressure are isobars not isopascals.

Gravity is probably another one to add to the list e.g. defining things as zero G, 1 G, etc. Obviously gravity does vary slightly with longitude, latitude, various geophysical effects and altitude, but for all practical purposes 9.81 m/s is good enough, and to again a conversion is useful but simply to abandon giving it in G at all would be odd. And since, obviously, standard atmospheric pressure has rather a lot to do with gravity then those, by extension, could be argued to come under that wing.

i've rather rambled off the point. But I suppose what I am arguing is that the list of internationally accepted units in various fields is almost limitless, and really the context of the article should drive what is appropriate, not some more-or-less arbitrary rule. SimonTrew (talk) 14:36, 16 August 2009 (UTC)

SimonTrew asked, by way of a question mark, "surely millimetre, kilolitre, kilometre per hour etc are SI derived units (or, at least for the latter, a non-SI derived unit accepted within SI, by transitivity)?" Meter is a base unit; adding an SI prefix does not change the status of a unit among base, derived, or accepted for use with SI. Litre is a special name for cubic decimeter, which is a derived unit. Kilometer per hour is a unit accepted for use with SI. --Jc3s5h (talk) 15:38, 16 August 2009 (UTC)

Regarding which are base units, which derived, and which non-SI, I suggest a study of the official BIPM document on that. See section 2; prefixes (as multipliers) are in section 3; the litre (or liter) is in section 4 being "a non-SI unit accepted for use with the International System of Units".

A derived unit is the product of base units: whilst the metre is a base unit, the square metre is a derived unit. Metre per second is a derived unit; but since the hour is "a non-SI unit accepted for use with the International System of Units", the kilometre per hour is also.

It's not a good idea to encourage the use of "litre". Whilst the literal definitions of virtually all SI units have changed over the years, they have all retained their practical values - except for the litre. The 1901 definition was 'the volume occupied by a mass of 1 kilogram of pure water, at its maximum density and at standard atmospheric pressure'; in 1960 they noted 'the cubic decimetre and the litre are unequal and differ by about 28 parts in 10', whilst in 1964 they declared 'that the word "litre" may be employed as a special name for the cubic decimetre' and recommended that 'the name litre should not be employed to give the results of high-accuracy volume measurements'. See p.141 of the PDF doc linked earlier. According to my physics teacher, they made a mistake when cutting the original prototype kilogramme. --Redrose64 (talk) 17:04, 16 August 2009 (UTC)

"It's not a good idea to encourage the use of 'litre'." That's nonsense. This depends entirely on the purpose. Litres (cubic decimetres) and millilitres (cubic centimetres) are absolutely standard units throughout most of the metrised world. I have never heard anyone use "kilolitres", though. It's clear what is meant, but everybody calls it cubic metres. But a 2-litre bottle is a 2-litre bottle, not a 2-cubic-decimetre bottle. Trying to forbid such standard units as the litre looks to me like an attempt to introduce a problem that otherwise exists only in the minds of proponents of pre-metric systems: that the metric system is "unpractical" because it doesn't have all the necessary "natural" units such as the pint. We already have the litre and as the UK is slowly crawling towards full metrication I predict that use of the "metric pint" of 1/2 litre = 0.88 Imperial pt = 1.06 US pt will become standard in the same way that the metric pound of 1/2 kg = 1.1 lbs avdp has been standard in large parts of Europe for a hundred years. It's probably as easy as the pubs beginning to call a pint of beer 1.136 metric pints once they are allowed to do this.

In science and technology, cubic centimetres and cubic decimetres are also used under these names. But only very rarely in normal life. Hans Adler 18:28, 16 August 2009 (UTC)

jch: yes, my bad about the derived units; of course you are right, a prefix does not make it a derived unit. I was just nodding there. Hans: All you say is good and true as always; but I reckon we will get 500 ml pints and call them pints. There is no way in this country they would even make it 550, let alone 570 or 600, but yes, we will have a metric measure and it will be called a pint, I reckon. But my bet is on 500 ml. And we will bitch about it to our grandchildren. When they changed from 1⁄6 gill measures to 25ml the pubs advertised "NEW LARGER MEASURES; the Imperial measure 94+2⁄3% the size of the metric one. woop dee doo. (The difference between 1⁄4 gill and 35 ml is a little larger, the Imperial measure being 35.5 ml and thus slightly larger, which the pubs did not shout about so much). SimonTrew (talk) 01:22, 19 August 2009 (UTC)

By the way, in Nineteen Eighty-Four the chapter where Winston goes into a prole pub, he finds a man complaining about his beer being in litres not like the pints he had in the old days. This of course was written about 1946-1948 (it took him a while) and, if I recall correctly, the complaint is that half a litre is not enough and a litre too much. This always seemed odd to me since half a litre is not far from a pint but a a litre is much much more. I presume Orwell was taking artistic licence here (or trusting on the ignorance in his audience) since, having lived in Paris, he must have been aware on how big a litre was. I always wondered how this was translated for audiences who only ever had metric. SimonTrew (talk) 01:30, 19 August 2009 (UTC)

I'll buy or borrow the Italian translation of that book sooner or later; BTW, I would have translated it "as is". Even in Italy, most people who have ever drunk in a pub (at least, in a British- or Irish-themed one) know what a pint is, and those who haven't can figure out from the context that it is the traditional unit for beers in pubs and that it is larger than half a litre but smaller than a litre, which is all one needs know to understand Winston's point. --___A. di M. 01:45, 19 August 2009 (UTC)

I can't imagine that we actually ever do use the litre for high-accuracy volume measurements. And for most applications, an inaccuracy of 28 parts per million is so much smaller than the margin for error inherent in the measurement that it's totally irrelevant. To put it into perspective, that's a difference of nearly 16 litres when measuring the amount of water in Sydney Harbour. I see very little reason to avoid the litre normal (non-scientific) circumstances. Pfainuk talk 18:21, 16 August 2009 (UTC)

Besides, it's a problem stemming from an error that was corrected more than 40 years ago. For more than 40 years litre has been an exact synonym for cubic decimetre. We would only ever have to worry about this problem should we encounter pre-1964 sources with high-precision litre-based measurements. They would have to be converted into modern litres. But exactly the same problem exists with inches and whatnot, since very roughly around the same time the inch was defined as precisely 2.54 cm, etc., after it was previously slightly different in various parts of the world. Hans Adler 18:35, 16 August 2009 (UTC)

28 parts in 10 ain't that bad for two centuries ago and greater accuracy than you mostly find here. We surely shouldn't discourage the use of the litre and millilitre in cases where they are used in the world out there. Ask for a pint of beer in the pub & you shouldn't expect a 28-parts-per-million accuracy. JIMp _talk·cont 18:32, 16 August 2009 (UTC)

OK, OK, maybe I overstressed one tiny aspect which doesn't really matter. It began with my possible misreading of what had gone earlier. There appeared to be disagreement concerning which are base SI units, and which derived SI units, and somebody mentioned litre (or kilolitre, or something like that).

My intended point was that the BIPM document has already done all the categorisation, and that based on what it has on page 124 "Table 6. Non-SI units accepted for use with the International System of Units", litres are not SI units (base or derived). --Redrose64 (talk) 20:00, 16 August 2009 (UTC)

I would argue that there are some units that are not accepted by the BIPM but that we still have to accept, depending on context. The most obvious are the month and year, which are not defined by the BIPM for the very good reasons that they are not of consistent length. In other contexts, there are units that are generally used internationally - the barrel of oil, the inch for measuring the sizes of television screens (as cited above) - that are specifically not accepted by the BIPM.

We should generally use the most commonly used unit in a given context (and I would suggest that may well include using different units in different regional contexts). Pfainuk talk 20:25, 16 August 2009 (UTC)

I don't know why Michael Glass is continuously starting this kind of discussion. The intent of the rules about use of units is absolutely clear: Use those units that will be expected in this context by the greatest fraction of the expected readership. If some readers will need a conversion to understand a measurement properly, provide it.

The metric system in the sense of SI units + units accepted for use with SI is a useful first approximation for the practical result of this rule. It's not quite correct for a number of reasons

• SI does not give sufficient guidance to choose, e.g., between millilitre and cubiccentimetre.
• In specific contexts, certain units that are explicitly not approved are dominant and must normally be used, e.g.:
  • Years for longer periods of time.
  • Light years for distances between stars.
  • Inches for TV and computer screens.
  • Gallons for raw oil.
  • Typographic points in printing.
  • Metric carat for jewels.
  • Kilometres per hour for car speeds.
  • Litres per 100 kilometres for car fuel consumption.
  • Nautical miles for distances at sea, especially for the definition of territorial waters etc.
• In some regions – especially, but not only, the not yet fully metric countries – the usage patterns for units diverge from the international ones. This must also be taken into account whenever we have reason to believe that most readers will be from a specific region:
  • Inches for TV and computer screens in Australia. (It's hard to see how that might become relevant, though.)
  • Miles for road distances in the US and the UK.
  • Miles per hour for car speeds in the US and the UK.
  • Miles per gallon for car fuel consumption in the US and the UK.
  • Kilocalories instead of kilojoules for food energy in some countries.
  • Dekagrammes instead of grammes for cold cuts in Austria. (Again, this is more theoretical. I am sure one can find better examples.)

I see no need to change the current text. Hans Adler 21:06, 16 August 2009 (UTC)

In some countries, such as Italy (and, I think, most of the EU), food energy is "officially" measured in kilojoules, but about 99% of people would normally use kilocalories for that (calling them "calories" in casual speech, except when they want to make the quantity sound bigger, as in a TV ad claiming that their product can help you burn "up to 1000 kilocalories"). (And cold cuts are measured in hectograms in Italy.) --   A. di M. 21:20, 16 August 2009 (UTC)

My aim is the clarify the wording, not change the policy. If the intent of the wording is clear, then it might be possible to have the wording equally clear. For example, 'some regional topics' might be preferable to 'region-specific topics', and I can't see the problem of defining 'internationally accepted units'. For example:

Internationally accepted units are:

• SI units
• Non-SI units accepted for use with SI (e.g., the nautical mile)
• Units based on fundamental constants (e.g. Planck's constant)
• other units that are widely used internationally.

The basic problem is that we have the metric system that is used in most countries of the world, the US Customary system that is widely used in the US and the Imperial System, parts of which are still used in the UK and to a lesser extent in other English-speaking countries, and also in aviation and some specialised measurements such as computer screens. To cater for the needs of an English-speaking readership we need to provide both metric and traditional measures in a wide range of contexts. I think if we concentrate on the needs of readers we might make more progress. Michael Glass (talk) 23:39, 16 August 2009 (UTC)

According to the article the nautical mile is not accepted for use with SI. The phrase "other units that are widely used internationally" is pretty vague, in contrast to the narrowness of the three preceding points. Yes, that some of us use the metric system and others use either the US or imperial system is a problem. How does the change you're suggesting solve this? JIMp _talk·cont 00:29, 17 August 2009 (UTC)

The knot ant the nautical mile are listed here . Table 8 is appended to Chapter 4 of the BIPM brochure which is entitled, "Non-SI units accepted for use with the SI, and units based on fundamental constants". However, I do agree that "other units that are widely used internationally" is too vague. It could be rephrased as "other units that are widely used internationally for specific purposes" but if we tried to be more specific than that, someone is sure to come up with some measure that isn't covered. That said, I would welcome a better phrase, if anyone could come up with it. Finally, Misplaced Pages can't solve the issue of English-speaking people using different weights and measures; what we might be able to do is work out how to cater for these differences and how to express the guidelines in a way that is clear and helpful. Perhaps something like this would be the way to go:

Internationally accepted units are:

• SI units
• Non-SI units accepted for use with SI (e.g., the nautical mile)
• Units based on fundamental constants (e.g. Planck's constant)
• other units that are widely used internationally for specific purposes.

Michael Glass (talk) 01:40, 17 August 2009 (UTC)

If we're talking about the introduction of the section (the three-bullet list immediately under the header "Units of measurement"), you might replace "internationally accepted units for the topic at hand" with "the units in most widespread use worldwide for the type of measurement in question". As for the first bullet of "Which units to use", I don't think it's broken and doesn't need fixing. (As for your list, it'd be close to tautological if all SI units were "internationally accepted", and even that isn't the case: the megasecond isn't internationally accepted—and my browser's spell-checker even underlines it in red.) If something should be fixed, I'd replace "region-specific" with "regional and historical" (you might want to use cubits first in Noah's Ark), and the point about conversion should be added to the general principles (first three bullets), too. --   A. di M. 10:47, 17 August 2009 (UTC)

Cubits at Noah's Ark is a good example for a problem I was always sure must exist. For history/archaeology articles historical units are sometimes the best choice. In history because we may know the precise number in an obsolete unit but have no certainty about the conversion factor. In archaeology because measures may be exact integer multiples of a well known obsolete unit. In such situations the obsolete units are internationally the most accepted ones. Hans Adler 12:42, 17 August 2009 (UTC)

Yes, and indeed for Goliath's height etc (four cubits and a span wasn't it?) I forget how many spans there are in a cubit but it is well-defined, but nobody really knows how big a span was. A similar problem occurs with Roman stadia and for that matter inches/ounces (uncia); obviously we have a rough idea but not an exact one, which doesn't stop maths working but if quoting Latin mathematics that gives an example in these units, it is pretty pointless to convert them (e.g. one might say – I make this up – "if a right triangle has sides by the right angle of three uncia and four uncia, the the third side will be five uncia" and that is good regardless of how big the inch is.) SimonTrew (talk) 14:51, 17 August 2009 (UTC)

I am pleased to see that the wording of the policy has been revised in the light of this discussion. I think that something could be done about "country-specific" and "region-specific" and I'll come back to this discussion with a further proposal later. Michael Glass (talk) 11:41, 18 August 2009 (UTC)

Automatic currency converter button idea

Wondering if someone's smart enough to write a template to do automatic currency conversions? I'd love to be able to put in something from which Misplaced Pages readers automatically choose what currencies they want the data in. Suppose NZ GDP is $42,052 (NZ$). And I put this in an article. Is there some way a user could click on a button next to the amount to switch the currency? Like, click, and it's $28K USD. Or, click, it's $42K NZD. Or, click some other currency? It would be really cool to have. Simpler variant: assume no inflation and its easier but less accurate. A simple template that translates NZ$ to US$ or vice versa based on today's exchange rate, and ignores inflation considerations or the passage of time. Complex variant: Suppose a fact about money was added on date X. But today it's date Y. So, information needed would be: money amount in NZ$ on date X; conversion rate NZ$ to US$ on date X; inflation (or deflation) of US currency between date X and date Y; lastly, date Y. Boom -- up-to-date accurate amount information. No way Encyclopedia Brittanica could ever do that. That would be really cool! As far as I can tell, Village Pump doesn't have any converter tools for inflation or currency conversions. Tomwsulcer (talk) 01:03, 17 August 2009 (UTC)tomwsulcer

There are good reasons why not. It depends so much on whether you are quoting an historical price at its historical value or current value: and, if at its current value, how you adjust for inflation (which inflationary index you use, e.g. retail price index, consumer price index, inflation based on the rise in cost of houses or mars bars or bread or any particular item); second, that since most currencies are on floating exchange rates the article will constantly change every time it is accessed, or, in the alternate, will need to state when and where the rate was taken from; third, that it would require use of a currency conversion site, and (assuming permission was granted to do that on a grand scale) which to choose?; fourth, that many currencies are not widely traded and so, for example, to convert danish krone into kenyan shillings is almost entirely meaningless; fifth, that even freely traded currencies such as US dollars have a variety of exchange rates: the spot rate given for today is not what you will get at a bureau de change, so which do you choose?

It is best to have the editor make those decisions, adding references to where the conversion came from if necessary, rather than make WP do it. In general prices are quoted ether in US dollars or in the prevailing currency of the article's subject (e.g. the local currency or historical currency). If a reader really wants a currency conversion not provided, they can look up a conversion site themselves, surely; and if they can't find it, then an automated tool is not likely to either. SimonTrew (talk) 14:45, 17 August 2009 (UTC)

I agree there can be problems with a currency converter template, particularly as time span lengthens from date of entry to date of readership. A time span of twenty to fifty years can seriously begin distorting values as you say, and any calculator would be problematic. But I still think a currency converter toggle switch, next to money figures, is a good idea, particularly for relatively-recent information (ie time between info added and info read), to convert currencies. Then inflation is miniscule. I see it primarily as an aid to readers. Some readers think in terms of US$ or a commonly used currency like the Euro; others will think in terms of their native currency. Why not permit readers the option to choose which figure they'd prefer? (And I don't think anybody would seriously want to convert a rarely-traded currency with another rarely-traded currency -- I doubt readers would expect Misplaced Pages to even try that). Stick to a pure currency converter (forget inflation). For example, in the article on New Zealand, there are many references to dollar amounts -- sometimes US$, sometimes NZ$ (technically, the policy is to use native currency like you say, but I bet many New Zealanders think in terms of US dollars, and foreigners won't know which is being referred to -- since NZ calls their money "dollars" too). I think there is consensus about particular exchange rates -- there's some variation, but not much. For example, $1 New Zealand dollar is worth about $.67 US dollars, and there are different rates today which vary only slightly from that amount, and I don't think such variation is a good excuse to ditch a good idea. At first, when I read the New Zealand article, I thought the figure $28,000 average GDP of New Zealanders was in New Zealand dollars -- it happened to be in US dollars so it threw me off -- the actual GDP figure is closer to $42,000 NZD, or about $28K US (numbers slightly off here -- I'm working from memory). But I'm saying that a simple toggle button next to money amounts, letting a reader a choice to switch from a native currency to a commonly traded currency (USD, Euros, pounds, yen perhaps) would be a (1) helpful (2) more accurate than letting readers mentally guess the exchange rate (3) easy to program (4) a nice extra which differentiates Misplaced Pages from book-bound static encyclopedias which has (5) numerous applications.Tomwsulcer (talk) 04:16, 18 August 2009 (UTC)tomwsulcer

¶ I'm too sleepy to go through all these points seriatim, but (with the blithe obliviousness of the technically-ignorant), I don't see insuperable obstacles on either side:

1. After preliminary discussion here, this proposal should probably move over to one section or other of Misplaced Pages:Village pump, or perhaps a template project or talk page. Let us know where to follow the discussion once it's moved.
2. What's needed by most readers is something that would convert a cited amount of money of whatever era into the reader's preferred currency now. When the number of pounds sterling, say, that a Victorian Englishman would exchange in 1850 for a U.S. Double Eagle of 1850 is a significant fact in itself, the conversion will usually already be in the text. The average reader would probably want to know the rough value of that 1850 double-eagle in the pounds (or euros or dollars) that he uses today, rather than having to do a second conversion from 1850 pounds to 2009 pounds.
3. What's also called for is just a rough indication of a sum's value, not a precise conversion.
4. When there's little bilateral exchange of two little-traded currencies, just let the template convert each into some relatively-universal currency, be it euros, US dollars or Special Drawing Rights, and then calculate a synthetic result that won't differ too much from what happens on those occasions when Kenyan shillings are traded directly for Danish krone. (That's the function of arbitrage, to exploit and thus flatten any discrepancies.) This kind of conversion happens all the time in the real world (e.g. translating Afrikaans texts into Catalan via some third language).
5. Similarly for converting that 1850 double-eagle. First let the template convert its value then into today's US dollars and then into pounds sterling (etc.) of 2009.
6. Because both the Danish krone and the Kenyan shilling have a real value in today's US dollars, even a hypothetical conversion — which may not represent the average of bilateral transactions in an active free market — isn't meaningless for the limited purpose of giving the Danish or Kenyan reader some idea of what a sum of shillings or krone would be worth to her. (What would be meaningless is pricing an 1850 double-eagle in imputed 1850 euros.)
7. There are many things that Misplaced Pages changes daily, weekly or monthly. A table of currencies wouldn't be a big challenge so long as someone or some project is prepared to commit to meeting it regularly.
8. But, on the other hand, one shouldn't be too sanguine or blasé about the slowness of inflation or the stability of exchange rates. Not so very long ago, the pound sterling jumped above US $2, and almost achieved parity with the euro, while the Canadian dollar reached near-parity with the U.S. dollar, before they sank back much closer to their historic relative values. And a few successive years of 5-8% inflation, as opposed to 1-3%, can make a great difference in a currency's value, both at home and abroad. So the conversion tables need to be relatively fresh. —— Shakescene (talk) 08:09, 18 August 2009 (UTC)

Shakescene, thanks for your comments. User Ohconfucius found a template which does currency conversions. But you have to be in edit mode to see it working: As of this week (data updated weekly), the exchange rate between the New Zealand and United States dollars is User:7/Template:fx.

I only gave the NZD amount; I didn't write the US$ value -- the template fished it from somewhere. The template is User:7/Template:fx It converts a number of different currencies into US dollars. And it seems to work; it picks off an exchange rate which seems right. And there's no button for users to click. Plus, there may still be bugs in it (if a space follows the second closing parenthesis, weird stuff seems to happen). Rather, it just converts currencies (doesn't account for inflation etc). And it's on a user talk page as opposed to an official Misplaced Pages page; so I had problems convincing other editors to use it. User Gadfium thinks the community needs to come to consensus before it can be widely used; Gadfium was skeptical that the conversion rates weren't being updated enough (last update = May; I'm writing this in August). Last, instead of asking a Misplaced Pages community member to constantly update tables, why can't we fish off currency conversion rates from an established non-Misplaced Pages site, and quote the site as the source? Tomwsulcer (talk) 19:25, 18 August 2009 (UTC)Tomwsulcer

Shakescene, thank you also for your comments. I am glad that maybe enumerating some of the problems let you, perhaps, come up with answers to them... that is called being productive! Thank you! As a small qualification, I was not suggesting the danish Krone was not widely traded, only that it is not traded much with the Kenyan shilling (and I better make sure I got the right spelling of Krone there since pre-euro all the Scandinavian countries had krona/krone etc but were spelt differently – of course it means crown – but let's correct it here before we move it to wherever you think it should go.

Yes, getting a currency feed each day (or hour, whatever) technically is no problem of course, it is what obligations does that place upon Wikimedia to get that feed? Or is the feed done client-side? Does the reader get a choice of where he gets the feed from (xe.com, msn money, etc)? Which rate to choose do you take the spot rate? I guess so, probably, since articles are likely to be talking about reasonably large amounts of money (in the thousands or tens of thousands of dollars, at the very least) not holiday cash.

My concern with it changing all the time is for those taking printouts etc. The printout will almost inevitably differ from that of some other's printout, i.e. the article is not stable. Now of course, articles are edited all the time and so any printout is no more than a snapshot of what it was at a particular time, but it seems this adds a new level of instability that, as far as I can tell, does not exist elsewhere on WP. So not only are we making a new little app, but I think fundamentally changing the expectatins of wikipedia: although we expect pages to change, we don't expect the same version of the same page to produce different results each time we look at it.

Tomwsulcer well, it seems I am the gainsayer and perhaps it has legs. I still think it is too complicated to do, but it might be worth a try if some developer wants to take it. I am quite willing to try (I am a software developer but have not developed for WP before and would not know where to start.)

Best wishes to you both SimonTrew (talk) 01:49, 19 August 2009 (UTC)

Thanks for your intelligent thinking. SimonTrew, I was wondering about the issue of changes and printouts and instability too. And I'm thinking that we're still focused on encyclopedias like Brittanica (printed, fixed, static) being what we're trying to emulate, but Misplaced Pages is computer-based as we know with much richer possibilities. So, I think the way to think about it is not to presume that the "snapshot" static fixed-page Brittanica-type photocopy is accurate -- this presumes a static world. But of course we live in an ever-changing world in which accuracy should mean "approximation to the real state of events" and not something fixed. Currencies change in value. Inflation happens. And what Misplaced Pages should strive for, in my view, is trying to be as accurate as possible, as up-to-date, reflecting the true state of affairs when the reader reads. In that sense, a currency converter button is much more accurate than a fixed-page or "stable" article if it can get the converted currency values right. If reality is unstable, Misplaced Pages should reflect that reality. Dynamic is good; static is problematic. We want instability. Tomwsulcer (talk) 14:52, 19 August 2009 (UTC)Tomwsulcer

Wondering what's the next step here. One problem -- this discussion happens in several places. Should we make a project page which becomes the main focus for this stuff? Or should we move it to the Village Pump? Tomwsulcer (talk) 14:52, 19 August 2009 (UTC)Tomwsulcer

Also, I agree with comments that these efforts might be parts of larger initiatives such as translation tools and such. In general, I favor interactivity. So Misplaced Pages could become more like a science museum exhibit -- click on this, X happens; push that, Y lights up. This would be so much more interesting and informative than flat static articles.Tomwsulcer (talk) 14:52, 19 August 2009 (UTC)Tomwsulcer

For present amount of money, there would be little problem having a converting template, but I'd suggest that: 1) it rounds conversions down to one significant figure (or two, if that lone one would be a "1" or a "2"—this can be implemented via {{#expr: {{{1}}} round -floor(ln(0.4*{{{1}}})/ln10)}}) and precedes them with "approx."; for example, 10,000,000 Swedish kronor (approx. US$1.4M, €1.0M, or £800k as of Aug. 2009), or 42,052 New Zealand dollars (approx. US$30k, €20k, or £17k as of Aug. 2009); 2) that, if possible, whoever updates the conversion rates uses a moving average over a sufficiently long period of time rather than an instantaneous value (if we can find a source giving them). This way, the converted figures don't change whenever the wind blows. --___A. di M. 16:18, 19 August 2009 (UTC)

Yeah, I like the idea of a moving average. Presumably these would be stored essentially in a table on a subpage of the template or something; this would be good because it would give the historical record for those going back through revision histories etc (I can imagine the fun when someone says "It printed completely the wrong rate" and it's not reproducible cos the rate has changed since). A month would seem to be about right, I think, for "a sufficiently long period of time" and by no coincidence also fits with the way you put the date (I know you wouldn't abbreviate the month in a real article as it is forbidden by MoS, but even so it is probably best simply to put month-year rather than day-month-year and update the averages about once a month. Of course if there is a currency crash or something some special measures might be needed to reflect that, but since one can get a dead cat bounce and so on, it is probably no big deal, especially with the deliberately vague "As of month-year").

I must admit I'm warming to the idea, though as A. di M. implies I like that it is just in plain text and not a button. Personally I like that WP is not so cluttered with eye candy etc. xe.com seems to have suffered that fate; it was much nicer when you could just go to xe.com and get the rates for common currencies right there on the front page.

I still am a bit worried about what licensing etc Wikimedia might need to use a conversion site on a wholesale basis; other sites such as for Bible links etc are used in templates so presumably there is some process to obtain permission, but I don't know of it, it's not my area of expertise at all.

Now, A. di M. gives examples in USD, GBP and EUR, but where will it end, Australian and New Zealand dollars, Canadian dollars, Indian rupees etc? I suppose the template will have options for which to show and which not to, but that opens another can of worms, i.e. what on MOSNUM (or MOSCUR or whatever) do we recommend? USD and EUR would seem to be essential; perhaps GBP considering the extensive UK readership, but NZ and AU dollars are perhaps the grey area that would be argued over. It would be useful to get feedback from our antipodean editors about whether they are used to seeing conversions just to say USD or whether it is essential to give it in Australian dollars, and if so whether in NZ dollars (are NZers used to seeing prices in Aus. dollars and doing a mental conversion anyway?) Of course for some articles it will be obvious that Aus. or NZ. dollars are needed, but I can see this being a big grey area.

I agree all these discussions should be wrapped up and put in one place. I don't know where best to put it but for myself I have no trouble if you move my comments of this page to wherever it goes (let me know please!). SimonTrew (talk) 17:12, 19 August 2009 (UTC)

Yes, I was thinking of something such as conversion rates updated on the 1st of each month to the average rate over the previous month. As for plain text vs. button, maybe with Javascript it would be possible to show that as 10,000,000 Swedish kronor (approx. US$1.4M as of July 2009V) (or even 10,000,000 Swedish kronor (approx. US$1.4MV as of July 2009) if they are always all updated at the same time), with as many currencies as practically possible in the drop-down menu. As a fall-back for users without Javascript, or if it'd be technically infeasible to do that, and when the page is printed, etc., I'd use US dollars, euros, and pounds sterling, as being three of the four most used currencies in the world (the remaining one being the yen, which isn't very familiar outside Japan, where there are few native English speakers), as well as the ones already suggested in MOS:CURRENCY. --___A. di M. 18:05, 19 August 2009 (UTC)

As for "Aug" (in what I called the fall-back), the MoS says, "Abbreviations such as Feb are used only where space is extremely limited, such as in tables and infoboxes." Now, here space isn't actually limited in the sense that there is an upper bound, but it is in the sense that you want the parenthesis to be as small as possible. --___A. di M. 18:14, 19 August 2009 (UTC)

I did a little fiddling around and came up with something that (I think) accomplishes what you are talking about, at least in the sense of the non-javascript version. Pretty barebones but I believe functional (and pretty easy to modify to suit), but feel free to test it out a little and see if this is a good foundation for what you are looking for and can be built upon. It is currently at {{User:Shereth/Templates/Currency|number|currency}}, where "number" is the amount, obviously, and "currency" is the ISO code for the currency in question. There is also an optional parameter to link to the currency's page. Thus 20,000 Brazilian reals can be converted with {{User:Shereth/Templates/Currency|20000|BRL|link=on}} gives you 20,000 Brazilian reals (approx. US$10,000, €7,000 or £6,000 as of August 2009). Should work for most currencies. Shereth 21:18, 19 August 2009 (UTC)

I would have implemented it a different way, with a subtemplate for each currency (as {{convert}} does, if I can understand at least the outermost level of that beast correctly). That way, you'd avoid giving e.g. conversions from pounds to themselves, and you might link obscure currencies but not famous ones by default. I'll give a try. --___A. di M. 11:23, 20 August 2009 (UTC)

(On further thought, that'd be a maintenance nightmare. --___A. di M. 12:09, 20 August 2009 (UTC))

A couple of conditional statements should weed out GBP -> GBP conversions and the like. I figured consolidating the currency conversion values all in a single subtemplate would make for simpler updating of the values from month to month. Shereth 14:19, 20 August 2009 (UTC)

BTW, you could use a /round subtemplate, instead of writing the code for the rounding thrice and the code for each conversion twice. --___A. di M. 18:23, 20 August 2009 (UTC)

I'm wondering what the status of this project is. Is the toggle button idea (active click by the user to switch currencies) realistic? Or a template-based converter realistic (several currency values appear chosen by the editor and not the user)? Wondering what the status is. Tomwsulcer (talk) 14:47, 21 August 2009 (UTC)Tomwsulcer

Template-based conversion is pretty easy to implement. I've already half-done it, it just needs a little tweaking, formatting, and so-on in order to be put in to use. A toggle button is somewhat beyond me (I'm not proficient in Javascript) but I don't think the idea is too far out of reach. Shereth 14:49, 21 August 2009 (UTC)

Template-based conversions I think work but I'm wondering whether the Misplaced Pages community must first recognize them as valid. I've had a case in the New Zealand article in which an editor rejected one of the currency-converter templates on the grounds that it had not yet received approval by the Misplaced Pages community. It was on a "user page", according to editor = Gadfium, and therefore invalid. Is a next step to get currency-converter templates deemed "official" by the Misplaced Pages community? Tomwsulcer (talk) 17:45, 21 August 2009 (UTC)Tomwsulcer

If we want to move forward with the idea of using the template, I'd suggest a little more discussion to hammer out any issues with it and then simply move it to template space once it is at a state that we like. Shereth 17:53, 21 August 2009 (UTC)

I don't think templates have to be "approved" for use or something; after creating {{radic}} I sought no-one's approval before using it in articles. But I would definitely not use a template in user space in an article; templates used in articles should be maintainable by the community (except protected ones, but that usually only happens to already-established ones), but most editors will refuse editing pages in someone else's user space. --___A. di M. 18:40, 21 August 2009 (UTC)

Querying "nineteenth-century painting"

The guideline for naming centuries (here and at WP:MOS) has been hotly contested, and I do not think we are ready to go back to that topic yet. But I am interested in just one provision:

Centuries are named in figures: (the 5th century CE; 19th-century painting); when the adjective is hyphenated, consider nineteenth-century painting, but not when contrasted with painting in the 20th century.

I would like to change the provision to this, to remove what I regard as an unsourced and probably unprecedented invitation to inconsistency:

Centuries are named in figures: the 5th century CE; 19th-century painting.

Can anyone cite a reputable guide that allows for nineteenth-century painting even in the same text (let alone contrasted with) as painting in the 20th century? If no one does, I will proceed with the change. (Even if someone does, I would invoke more major guides that do not support such an inconsistency.)

–_⊥Noetica!– 06:19, 18 August 2009 (UTC)

I would have thought, even when it is to be read by the Masters of MoS, that "but do not consider nineteenth-century painting when contrasted with painting in the 20th century", which is the plain meaning of the key clause, would have been condescending and verbose. Guess not. I will amend accordingly. Septentrionalis PMAnderson 06:29, 18 August 2009 (UTC)

• In the process, I settled on "but do not consider it when..."; the referent of it ought to be plain. Septentrionalis PMAnderson 06:36, 18 August 2009 (UTC)

PMAnderson, the matter has been raised for discussion, and a specific question was put forward for editors to respond to. Please do not pre-empt such discussion; and please focus on the question that I have raised. While the matter is under focused discussion, it is not appropriate to shift or dissipate the focus. For that reason I am reverting your edit, and I await your response to the clear point that I raise in this section.

–_⊥Noetica!– 07:34, 18 August 2009 (UTC)

I must say that your original question is not at all clear to me. Can you check it for typos (e.g., did you make the shift from "19th" to "nineteenth" because it's your main point or is it accidental???) and clarity? Hans Adler 07:57, 18 August 2009 (UTC)

Well, I must allow that the question may not have been clear. It is free of typos, though, and I cannot see how it can be construed in any way other than what I intended. Hans, I can only think that the confusion arises because the guideline as presently worded is really strange!

Let me put it this way:

The guideline in its present form (correctly cited above) suggests that nineteenth-century painting is acceptable instead of 19th-century painting. It only excludes nineteenth-century painting when this would be "contrasted with painting in the 20th century" (to quote verbatim). What I ask is this: why should we ever want nineteenth-century painting in an article? The central point of our guideline proposes these forms: the 5th century CE and 19th-century painting. Should an article have nineteenth-century painting at all, then? That would be inconsistent with other usages in the same text that do follow our guideline, perhaps like painting in the 20th century at several paragraphs distance from the spelled-out form we are talking about.

I asked, and still ask, is there any reputable style that permits nineteenth-century painting as well as, somewhere far from that phrase, something like painting in the 20th century and 21st-century computer art? In the same text? (Never mind "when contrasted with"!) I suspect there is no such style guide, but I am waiting for an answer.

–_⊥Noetica!– 08:26, 18 August 2009 (UTC)

See A. di M.'s comments above; it is unlikely for a style guide to mention such a point at all. But this begs the question; show us one that permits both forms - as we should - and forbids their presence on the same page. Septentrionalis PMAnderson 15:56, 19 August 2009 (UTC)

You are right. You were clear and I simply wasn't sufficiently concentrated. Sorry! And yes, I support making it completely uniform for simplicity. Hans Adler 08:40, 18 August 2009 (UTC)

Noetica is clear, and clearly mistaken. The guideline, as it stands, means:

Centuries are named in figures: (the 5th century CE; 19th-century painting); when the adjective is hyphenated, consider nineteenth-century painting, but do not consider it when contrasted with painting in the 20th century.

I am perfectly willing to add the bolded words, since "but not when..." seems to be confusing; indeed I did, and Noetica reverted me. But, short or long, this advises against using nineteenth-century painting and painting in the 20th century in the same context.

Noetica's question therefore is like "when did you stop beating your wife?": tendentious, inflammmatory, and irrelevant to the issue at hand. Septentrionalis PMAnderson 12:35, 18 August 2009 (UTC)

(outdent) i support changing the provision to just "Centuries are named in figures: the 5th century CE; 19th-century painting." it's simple, clear and consistent. Sssoul (talk) 08:36, 18 August 2009 (UTC)

• Strongly oppose, also wrong. This is the besetting problem of MoS; the declaration that "I don't happen to write this way, therefore let's make a rule that nobody else can". Septentrionalis PMAnderson 12:35, 18 August 2009 (UTC)

PMAnderson:

I have now twice clearly asked my question. Because of the confusing way the guideline is worded (with an addition that you are responsible for some time ago, as I recall), the guideline itself is hard to make sense of. For everyone's benefit, and especially for you, I will make the point another way, and pose two distinct questions, with some preliminaries:

The guideline certainly allows that the 5th century CE and 19th-century painting may occur in the same article. Call these standard forms, for our convenience here, OK? Now, the guideline also seems to allow that the spelt-out form nineteenth-century painting may occur in the same article as those standard forms, provided only that nineteenth-century painting is not near those standard forms. That is the best sense I can make of the current wording: "but not when contrasted with painting in the 20th century". If another sense is intended, it is utterly obscure; and your recent suggestion merely confirms that this is the sense of the current guideline. Now, here are two good questions:

1. Why should nineteenth-century painting be allowed anywhere in the same article as the standards forms? That is plain inconsistency, and therefore against one of our major principles. What style guide recommends such an inconsistent practice? I have not seen one, and I would like someone to produce evidence of such a guide.
2. Beyond the matter of consistency within an article, why should the form nineteenth-century painting be allowed at all, anywhere in any article? Most of us want consistent, simple guidelines to settle needless disputes, and to guide editors. Look at the many fine articles in Category:Centuries, where editors strive for elegant and efficient uniformity in these matters. Why undo that effort?

If you can't provide precedent from any reputable style guide for your complex old wording (or your complex proposed new wording, which has the same meaning), and if others prefer a plain simple guideline, we should change the wording to this:

Centuries are named in figures: the 5th century CE; 19th-century painting.

–_⊥Noetica!– 14:01, 18 August 2009 (UTC)

I have just responded to both those points, but let me try recasting - especially since there is a third red herring, . I will await a little more evidence before resorting to dispute resolution on this confused and obnoxious thread.

• The present wording says nothing, for or against, on using painting of the 20th century at one end of an article and nineteenth-century painting at the other. Why does Noetica assume it does? Whether those are a clash depends on the taste of the individual editor. Why do we need to rule on it, except to satisfy a will to power? That's the red herring.
• Does Noetica deny that nineteenth-century painting is English usage, which a literate editor e may well write, and a literate reader see, without complaint? If xe does not, then we have no reason to prevent it, save the base satisfaction of compelling all Misplaced Pages to follow the tastes of a handful of meddlers here.
• Since the present text advises against what everybody agrees is undesirable - a meaningless failure of parallellism - what's the problem with it? I have no objection to striking the whole clause, if it confuses people. Septentrionalis PMAnderson 14:18, 18 August 2009 (UTC)

What or who exactly is "obnoxious" here? No one has rushed to edit, apart from you! Three times I ask my questions, and three times you refuse to answer – or cannot answer. What reputable guide supports anything like the inconsistency you suggested when you tampered non-consensually with this guideline in the first place, some months ago?

To answer your question, even if you ignore mine:

PMA's question: The present wording says nothing, for or against, on using painting of the 20th century at one end of an article and nineteenth-century painting at the other. Why does Noetica assume it does?
Noetica's answer: I don't assume that it does that! I think it should make it clear that the two quite different forms are not to be used in one article; but it suggests that it might be all right, by saying in effect only that the two forms should not be in close proximity. That's what "contrasted with" must involve, if it means anything at all.

–_⊥Noetica!– 14:41, 18 August 2009 (UTC)

You have received three paragraphs of answer (which are themselves not new); you acknowledge this by responding to them; you then complain that you have gotten none. What you have failed to get is agreement - but this is because you are insisting on two (inconsistent) useless and meddlesome conventions, which many competent writers will simply ignore, as most of MOS's half-educated wikicreep should be ignored.

• That Misplaced Pages should never use nineteenth-century painting. Why not? It's perfectly good and natural English.
• That Misplaced Pages shouldn't use nineteenth-century painting, if painting in the 20th century happens to occur in the same article. Arguable but silly.

Make up your mind which you support, or -better- abandon both, and let editors write any respectable variant of English. Septentrionalis PMAnderson 14:57, 18 August 2009 (UTC)

My two cents: 1) It's not like style guides must explicitly permit some constructions for them to be valid English; you just need that they don't forbid them. I'm quite sure there's no style guide stating that "the pronouns somebody and someone may be used in the same article", or even that "the number of items in a bulleted list may be a multiple of six", but this is no good reason to forbid using "somebody" and "someone" in the same article (or even banning "somebody" altogether), nor to forbid six-, twelve- and eighteen-item bulleted lists. 2) At least to me, all other things being equal, very small numbers look better when spelled out: e.g. "third century" rather than "3rd century". (But I would still be consistent with numbers of the same type, avoiding e.g. "third century" and "17th century", in the same section.) Anyway, beauty is in the eye of the beholder, and if all beholders around here believe otherwise, I'll follow suit and always use "3rd century", should I ever have to mention that period of time in an article. --___A. di M. 19:50, 18 August 2009 (UTC)

I concur with Hans, Sssoul, and Noetica that the new suggestion is ideal. We should change the text. --Andy Walsh (talk) 15:42, 18 August 2009 (UTC)

I also concur. One of the main point of a manual of style is to select among all the plausible variants a standardized one, to allow for greater consistency across articles. It seems that PMA is arguing that any form of standard English is acceptable here and that we ought not further constrain the variants to be used. I disagree; while I have no particular preference vis-a-vis "19th" versus "nineteenth", I think that the consistency of picking one and sticking to it in well-edited articles enhances the encyclopedia. Editors of course remain free to not follow the MOS in their submissions; and other editors will come along and make the important articles MOS-compliant. Studerby (talk) 19:34, 18 August 2009 (UTC)

Adding my "concur" to Noetica, Hans, Sssoul, Andy Walsh, and Studerby. Goodmorningworld (talk) 20:19, 18 August 2009 (UTC)

Adding my "concur" to Noetica, Hans, Sssoul, Andy Walsh, Studerby, and Goodmorningworld.  HWV258  22:30, 18 August 2009 (UTC)

However, per A. di M.'s edit above (giving "third century" as an example) I can see leaving wiggle room for the occasional exception. Does it need to be made explicit? I think not: overdrafting makes a guideline harder to absorb. Remember the caveat at the top of the MOS page, that should be sufficient. Goodmorningworld (talk) 22:45, 18 August 2009 (UTC)

PMAnderson writes above: "You have received three paragraphs of answer ...". Whether those many paragraphs of prevarication qualify as "answers" is a matter for semantic analysis. One thing is clear: they are not answers to my questions.

I thank other editors for their clear responses. My comment on them: simple consistency is usually the best policy, and all the most influential style guides aim for that. But yes: there is always the provision for exceptions in practice, stated at the top of our MOS pages. Editors at an article can negotiate such things on their merits, aided by clear consensual guidelines from their Misplaced Pages Manual of Style. Outside of Misplaced Pages I myself prefer to use the fully worded forms like in the nineteenth century and twelfth-century French kings. But at Misplaced Pages, I adapt. So do we all. Almost all, I mean.

If there is no substantial support for the present unprecedented and obscure guideline, we should amend the text to the simpler form in a couple of days.

–_⊥Noetica!– 23:05, 18 August 2009 (UTC)

Any answer to my question: "what's wrong with nineteenth-century?". Septentrionalis PMAnderson 23:47, 18 August 2009 (UTC)

Sure PMAnderson. If you will look immediately above what you just wrote, you will see that I have no objection to nineteenth-century paintings. In fact, I prefer it! I use that style, outside of Misplaced Pages. I have just said so. But the firmly established style at Misplaced Pages is otherwise. I say once more: look at the articles listed in Category:Centuries. Some of us adapt; nearly all of us do. A day might dawn when you understand the virtues of joining in this cooperative endeavour.

–_⊥Noetica!– 01:29, 19 August 2009 (UTC)

Thank you. But I am still discussing the adjective, with which we began, as in nineteenth-century painting; those are substantives. Septentrionalis PMAnderson 02:02, 19 August 2009 (UTC)

What on earth are you talking about, PMA? I am discussing all forms, explicitly including the adjectival. Look! Here, again:

Noetica wrote: Outside of Misplaced Pages I myself prefer to use the fully worded forms like in the nineteenth century and twelfth-century French kings.

And that twelfth-century is adjectival, is it not? On the other hand, your barely articulated rejoinder might refer to "the articles listed in Category:Centuries", which I now draw to your attention for a third time. Well, I did not say "look at the titles of the articles"; I said "look at the articles". The clear preference is for the adjectival form nnth-century xxxxx. It is not adhered to with complete uniformity, and that is a bad look. To improve those articles (and innumerable others), we have a Manual of Style with clear, rational, consensus guidelines – reflecting the evolved style on Misplaced Pages, where there is one.

Is there anything you feel like retracting at this stage, PMA? It wouldn't be the end of the world, you know. Please pay attention. You're wasting far too much of everyone's time. You expect answers, and you get them. You are asked questions, again and again: but you refuse to answer them.

–_⊥Noetica!– 02:58, 19 August 2009 (UTC)

Now your point is clear; I looked at the list of articles in Category:Centuries and saw nouns. A useful Manual of Style, which this will never be, would let you use nineteenth-century, and HWV258 use 19th-century, even if his only reason is that he doesn't like the longer form; but even if the preference were overwhelming, and it is not, it would suffice to say "most Wikipedians use". Septentrionalis PMAnderson 15:03, 19 August 2009 (UTC)

And, having looked at several of the century articles, I see neither nineteenth-century nor 19th-century (nor any other number), so I fail to be persuaded by Noetica's evidence. Section links? Septentrionalis PMAnderson 15:56, 19 August 2009 (UTC)

"wrong" is a pejorative way of looking at the issue. A Manual of Style provides guidelines for how editors prefer to see text in a publication. You should be asking the question: "why is consensus forming that prefers '19th' over 'nineteenth'?". Because one method is preferred doesn't make the other method "wrong". Cheers.  HWV258  00:09, 19 August 2009 (UTC)

No, there isn't. This page is too obscure and too ill-frequented to have any such consensus. But in fact, the real question is "what do we do when someone doesn't like what some small pool of editors like?" (in this case. 19th-century). Some people say "away with it: the six of us don't like it; you must use what we like," but the useful parts of MOS say what GregL says below: "don't use it then." That way we will find out those few cases in which Misplaced Pages as a whole has consensus; when article-space as whole (unmeddled by bots) overwhelmingly does something one way or the other. (We can then say "almost all Wikipedians do X," which is valuable information for editors, while it remains true.) Septentrionalis PMAnderson 00:31, 19 August 2009 (UTC)

Dictionaries might work the way that PMA wants (descriptionist instead of prescriptivist) but no Manual of Style ever has. Goodmorningworld (talk) 00:55, 19 August 2009 (UTC)

Then Manuals of Style are contrary to policy. GMW might also try consulting Otto Jespersen's A Modern English Grammar on Historical Principles. But first he might try the paragraph I just wrote, which does not say "why be prescriptivist?", it offers two alternatives when - as will happen with advice (like this proposal) not supported by English usage, but by the prejudices of six editors- other editors disagree. Septentrionalis PMAnderson 01:07, 19 August 2009 (UTC)

Please assume good faith. You use the word "prejudice" when you should use "point of view". (For the record, I guess I can assume that your "point of view" is not prejudiced?)  HWV258  01:16, 19 August 2009 (UTC)

No. A prejudice is a judgment reached before reasoning and evidence - I will modify the word to the extent that a preference for 19th-century is supported by either. So far it has not been. Septentrionalis PMAnderson 01:31, 19 August 2009 (UTC)

I'm all for giving this issue time (in order to see which format prevails). Personally, I prefer "19th" to "nineteenth". The reason being that if you prefer "nineteenth", then I guess you also have to prefer "twenty-first-century", and I simply prefer the brevity of "21st-century". If I was writing in the style of a Victorian novel, I would use "nineteenth"; however I feel a briefer style is appropriate when imparting information in an encyclopaedia (hence "manual of style"). Nothing prejudiced about it.  HWV258  04:59, 19 August 2009 (UTC)

The expected response. You persist with your pejorative view of the "world". The real point being (apart from what I raised above is) that the MOS is a styleguide. It allows editors to check what is the established way of constructing something on WP. Of course, if an exception is needed (that is supported by local consensus), then so be it. Regarding, "This page is too obscure and too ill-frequented to have any such consensus"—that is a personal view (and of course anyone is welcome at MOS to add to the debate). Regarding "you must use what we like"—that isn't correct (rather that is what you believe will happen). In actuality, if there is a localised dispute, there are two practical options: start a discussion on the MOS to allow an exception, and/or start a local debate in order to enforce the exception to the guideline. Assume good faith is the key to this issue.  HWV258  01:12, 19 August 2009 (UTC)

Nonsense. I would be happy to see this page guide; what HWV258 wishes is to have it command. In English, at least, these are not synonymous terms. The claim that atarting a discussion on MOS is a practical solution is denied by experience; discussion elsewhere has been generally foiled by disruptive bots. Septentrionalis PMAnderson 01:31, 19 August 2009 (UTC)

"Enjoyable" to get to the stage in the debate where PMa's pronouncements begin with an imperative (Latin can't be far away now). A simple response: I don't wish the MOS to be a "command" (and have never stated that opinion). I think we're done (now that PMa's prejudices have once again risen to the surface). What a pity (especially in the light of calm discussion by everyone else).  HWV258  01:54, 19 August 2009 (UTC)

Noted. To HWV258, "Nonsense" is an imperative. Septentrionalis PMAnderson 01:57, 19 August 2009 (UTC)

I don't understand your point. The use of "Nonsense" as an imperative was precisely the observation I was making.  HWV258  02:42, 19 August 2009 (UTC)

Yes, you don't. You might, if you read imperative mood, find out that it is a form of a verb, and that this "Nonsense" is not one; nor is it a command but a description. . Septentrionalis PMAnderson 14:47, 19 August 2009 (UTC)

More disingenuity by Pmanderson. From imperative mood: "The imperative mood is a grammatical mood that expresses direct commands...". Your use of "Nonsense" was clearly intoned in that mood. I took your opening remark to be a command to all readers of this discussion to treat my comment as nonsense. Please also note from that page: "The use of imperative mood can easily be considered offensive or inappropriate in social situations due to universally recognized politeness rules". Regardless of the grammar, I find your mood to be offensive and uncivil. Please remain civil in these discussions—you will be taken more seriously if you do.  HWV258  22:46, 19 August 2009 (UTC)

Am I misreading this, or did HWV258 just assert that a grammatical mood is an emotional state? It isn't; it's a verb form. "Nonsense" wasn't a verb, let alone an imperative. Septentrionalis PMAnderson 23:26, 19 August 2009 (UTC)

The expected response. You persist with your pejorative view of the "world". The real point being (apart from what I raised above is) that the MOS is a styleguide. It allows editors to check what is the established way of constructing something on WP. Of course, if an exception is needed (that is supported by local consensus), then so be it. Regarding, "This page is too obscure and too ill-frequented to have any such consensus"—that is a personal view (and of course anyone is welcome at MOS to add to the debate). Regarding "you must use what we like"—that isn't correct (rather that is what you believe will happen). In actuality, if there is a localised dispute, there are two practical options: start a discussion on the MOS to allow an exception, and/or start a local debate in order to enforce the exception to the guideline. Assume good faith is the key to this issue.  HWV258  01:12, 19 August 2009 (UTC)

No, you must use what we like is what the present text says: Centuries are.... Experience shows that it will be enforced literally. If you don't mean that, rephrase it. Septentrionalis PMAnderson 15:49, 19 August 2009 (UTC)

A styleguide gives direction to those who are unsure of how to phrase. It also allows other editors to make articles consistent. In this case, and based on the following searches:

• Google search of "nineteenth-century" results in 13,500,000 matches,
• Google search of "19th-century" returns 73,400,000 matches,
• WP search of "nineteenth-century" returns 3,911 matches,
• WP search of "19th-century" returns 10,204,

you might care to have a rethink of your position (and understand why there is such support for the view opposite to yours). Regarding "rephrase it", it is not practical to phrase every part of a MOS in such terms. The overriding message at the top of the MOS gives the direction you require: "This part of the Manual of Style aims to achieve consistency in the use and formatting of dates and numbers in Misplaced Pages articles. Consistent standards make articles easier to read, write, and edit. Where this manual provides options, consistency should be maintained within an article, unless there is a good reason to do otherwise" (underlining is mine).  HWV258  23:20, 19 August 2009 (UTC)

Ah, proof by google; how nice. But what this actually shows is that nineteenth-century is a well-established form, even on the web, where brief forms are particularly favored. Septentrionalis PMAnderson 23:26, 19 August 2009 (UTC)

No, not "proof by Google" (and disengenuity has forced you to ignore that Google wasn't the only source mentioned). PMAnderson has done his usual trick of forming a position at the beginning of a discussion and then stubbornly refusing to shift from that position (despite evidence of any form). I responded with some statistics due to statements from you of the form: "not supported by English usage, but by the prejudices of six editors". I read your words and realised (once again) that you just make things up to suit your agenda. Do we really need to devolve into sub-debates when "not supported by English usage" is shown to be incorrect? Geez, say it isn't so.  HWV258  00:16, 20 August 2009 (UTC)

PMAnderson, I suspect you disagree with this concept, but I think most of us believe the Misplaced Pages editors form a community of practice (or discourse community, in Foucault's language) that will decide what is "right" and "wrong" by consensus. What emerges is the MoS. If we simply leave it wagging in the wind for every individual to decide, we have no community. --Andy Walsh (talk) 01:18, 19 August 2009 (UTC)

A point well worth considering; give me a minute. I doubt that these six "usual suspects" form a community of practice; I believe I agree with the rest of the first sentence. Septentrionalis PMAnderson 01:34, 19 August 2009 (UTC)

• I do not believe that this decision, or most decisions at MOS, are reached by consensus; indeed, this section, and the claims made in the past few hours, are characteristic WP:Consensus#Process violations. "A handful of editors agreeing on something does not constitute a consensus, except in the thinnest sense. Consensus is a broader process where specific points of article content are considered in terms of the article as a whole, and in terms of the article's place in the encyclopedia, in the hope that editors will negotiate a reasonable balance between competing views."
• If Andy Walsh's last sentence were true, then we would not have a community, no matter what MOS does; that is the process by which we write article text, to which this page - and all of WP space - is auxiliary. I do not think it is true, and it is one of the reasons why Foucault is a dangerous guide to praxis; but politics are off topic. Septentrionalis PMAnderson 01:54, 19 August 2009 (UTC)

Proposal to conclude this discussion

The question raised in this section has been settled, I think. No one wants the version we had except PMAnderson; everyone else who comments here explicitly prefers the simpler, clearer guidance. I would have waited longer (as I say above), but the simpler form is now in place. PMA, will you now accept this with good grace? I call on you to do so, and to remove the tag you imposed on the page, since it is obvious that nothing new will be transacted here.

Please see my talkpage also, where I have responded to your post.

–_⊥Noetica!– 00:54, 20 August 2009 (UTC)

I don't particularly want the version we started with; I offered to change it twice, and changed it once. I would settle, indeed, for something phrased as advice; since the choice between nineteenth-century and 19th-century is a matter of taste, I don't care what approach we offer those who genuinely have no opinion, as long as we don't demand that those who do have an opinion change it.

Since advice is more convincing with a reason, I propose something like

the 5th century CE and 19th-century painting are shorter than spelling out the numbers.

or whatever other reason people would like; but

Many Wikipedians prefer the 5th century CE and 19th-century painting to the fifth century CE and nineteenth-century painting.

is fine for those who don't like to explain themselves. Septentrionalis PMAnderson 01:30, 20 August 2009 (UTC)

PMA, what in the expression good grace continues to elude you? If you don't understand what the community of MOS editors is trying to achieve here, it might be best for you to stay away longer than ArbCom has required of you. Just accept the reasoned consensus. None of your attempts to finesse a muddled solution out of an elegant and reasonable one will succeed, and you are wasting your time and everyone else's. And now see more at my talkpage.

–_⊥Noetica!– 03:31, 20 August 2009 (UTC)

Comment I don't want to derail the current proposals, but why can't we simplify this by using the same guidelines for writing normal numbers: numbers under 10 are spelled out; otherwise, use numerals. So, fifth century, but 19th century. Dabomb87 (talk) 03:40, 20 August 2009 (UTC)

Not a bad point, Dabomb. But as I hinted at the top of this section, we are not ready to deal with some things. I would add this matter of numbers under 10 versus over 10, which has caused much wrangling in the past. The practice on Misplaced Pages (and most other places, let me assure you) does not lean toward applying that principle to the naming of centuries (see the articles I refer to above: those in the centuries category); there are sound reasons for that, and even if there were not objective reasons, it would be far too disruptive to move and alter internally all those articles to comply. As it is, a much smaller number of articles are overdue for moving in any case.

–_⊥Noetica!– 03:51, 20 August 2009 (UTC)

Ah, that's fine then. I missed the valid point about the naming of centuries. I have no horse in this race, but to prolong this discussion much longer would be undesirable. Dabomb87 (talk) 03:54, 20 August 2009 (UTC)

If you don't understand what the community of MOS editors is trying to achieve Very well, explain it to me. What sublime goal is at the end of this tunnel, and how will insisting that all articles use 19th-century contribute to it? Septentrionalis PMAnderson 17:20, 20 August 2009 (UTC)

Consistency. (Ignoring the snark of "sublime"). Why is consistency valuable? Several reasons, IMHO. First, people who notice that there is a consistent style to the Misplaced Pages, from their surface observations of it, will immediately realize that Misplaced Pages is a structured project, not an Internet free-for-all. Second, editors who tend to pay attention to details in one area will, more likely than not, tend to pay attention to details in other areas. Most importantly to me, a consistent look to the work is valuable to me personally. While I personally have never more than barely touched an FA or GA, I take personal pride in the quality of those articles and the knowledge that I am contributing to a project that can produce work of that quality. I think that the community of MOS editors also values consistency; I think that's implicit in a manual of style, which of course describes the style choices that a project or work has chosen to adopt. Most editors also seem to value consistency; many who don't spend much effort on the MOS itself still refer to it regularly, and produce work that attempts to conform to it to some extent. Studerby (talk) 20:12, 20 August 2009 (UTC)

We will never have a consistent style, even if MOS - as it tends to - grows larger than article space. We are an Internet free-for-all; we are inconsistent; and MOS encourages us to be inconsistent in far more visible matters than the spelling of 19th century. Most of our editors don't read MOS - and the longer it gets, fewer will.

editors who tend to pay attention to details in one area will, more likely than not, tend to pay attention to details in other areas Not in my experience.

The quality of an article which conforms to MOS and has never been reviewed for content or writing is open to doubt. All too often, the noise of the Crusaders for one or another section of MOS drowns out or substitutes for actual evaluation of genuine merit. Septentrionalis PMAnderson 20:27, 20 August 2009 (UTC)

"even if MOS - as it tends to - grows larger than article space". Um... I've been laughing about that statement for several minutes now. Do you seriously mean that, or was that pure rhetoric? Studerby (talk) 01:06, 21 August 2009 (UTC)

• And, before we close, we should answer #Centuries, below. Septentrionalis PMAnderson 17:25, 20 August 2009 (UTC)

PMAnderson, you are not addressing the topic. You hardly ever do, it seems. I started this section for the specific purpose of dealing with one provision in a guideline. There was not even an implicit invitation to discourse on the nature of consensus, or any such broader issue. We have now dealt with the specific topic, even though you steadily refused to answer questions (while others answered yours). What you are now doing is an abuse of the talkpage. Take your generalities elsewhere. At the very least start a new thread – if the topic is localised to WP:MOSNUM and its associated pages.

Since the discussion is now over, and you are on your soapbox concerning other matters, I am now removing the tag {{disputedtag|section=yes}} that you applied. Yes, we could turn our attention to the new section below (#Centuries).

–_⊥Noetica!– 22:10, 20 August 2009 (UTC)

I answered your question: No, this is not acceptable to me. I offered two variants which are; if you have a preference between them, let me know. If you have reasons that either of them is undesirable, do let everybody know.

I then asked a question as to what you meant by ''what the community of MOS editors is trying to achieve, which you have not answered. No, I do not understand why MOS is heading towards a ruling on every word in the dictionary; some conjectures do suggest themselves, but I await an explanation.

What questions have I left unanswered? Septentrionalis PMAnderson 23:12, 20 August 2009 (UTC)

Unanswered, above:

• "Can anyone cite a reputable guide that allows for nineteenth-century painting even in the same text (let alone contrasted with) as painting in the 20th century?"
• "I asked, and still ask, is there any reputable style that permits nineteenth-century painting as well as, somewhere far from that phrase, something like painting in the 20th century and 21st-century computer art?"
• "Why should nineteenth-century painting be allowed anywhere in the same article as the standards forms?"

You see PMAnderson, when you say "contrasted with", or "in the same context", we can't tell what you mean. We need to know about the whole article (which MOS is normally concerned with), not the immediate vicinity in which a "contrast" may be visible, or some ill-defined "context". Unless you say "in the same article", there is no indication that you intend it.

This is not to suggest that we take up the discussion again. All the action is over. Can we move on, now?

–_⊥Noetica!– 04:46, 21 August 2009 (UTC)

• Your first two questions have been answered by A. di M. above; they ask for the unlikely, that a style guide will expressly give permission on such a point.
• Your last question begs the issue, twice:
  • by a unsupported claim that 19th-century is the standard form; if there is one, it is nineteenth-century, which several style guides recommend, but there are more likely to be two.
  • That English forms, commonly used by good writers, need the permission of a dozen editors. Is this so important to you that you have to rely on flimsy arguments like this, and the unsupported claim that 19th is "easier to maintain"? Why?
• The discussion is not over; it's just moved down. Septentrionalis PMAnderson 20:50, 21 August 2009 (UTC)

Markup for examples and mentions: italics, quotes, and the xt template

Markup for examples and mentions at WP:MOSNUM remains inconsistent, even more than at WP:MOS. I have done a little housekeeping editing just now, but I did very little with such markup.

Some points are clear: for good reasons we generally use italics for a mention (see WP:MOS for discussion of the use–mention distinction) as opposed to an example, so we should be consistent with that:

The word approximately is preferred to approx.

Not:

The word "approximately" is preferred to "approx".

But also, I should say, not this:

The word approximately is preferred to approx.

I don't say that exactly these cases turn up; I merely illustrate. Sometime we will need to discuss more subtle cases, and then go through these pages making all such markup rational and consistent. I propose that the topic be dealt with at WT:MOS, rather than here, for three reasons:

1. The development of template:xt was managed there more than here.
2. The implementation of that template is more advanced at WP:MOS than here at WP:MOSNUM.
3. WP:MOS is the central page for the whole Manual of Style, and the decisions made there can reasonably be applied to all the other affected pages.

Do editors agree to centralising the discussion there?

–_⊥Noetica!– 07:26, 18 August 2009 (UTC)

I'd agree to discuss it there. Anyway, since no thread about this was created there yet (move/copy this when you do):

According to that, we would write that the symbol of the metre is m, but that would be a false statement; italic m is the symbol of mass. The symbol of the metre is a roman small em. (This was the reason why I created {{xt}} in the first place.) So I'd prefer the use of quotation marks for mentions. (I don't think this could cause problems, because I can't see a situation where you'd mention a string containing quotation marks itself other than as an example, in which case {{xt}} is appropriate, or as computer code, in which case <code> is appropriate.) --   A. di M. 08:56, 18 August 2009 (UTC)

Thanks, A di M. I am well aware of the protracted difficulties we had concerning these symbols and italics. That gives yet another reason for centralising the discussion and dealing with all such issues together in orderly fashion. It will not be easy; but with goodwill, flexibility, and rational analysis we can sort it out. I propose that we postpone it for about a week, now that we have signalled the discussion here.

Are there any more broad procedural points from editors?

–_⊥Noetica!– 09:13, 18 August 2009 (UTC)

"Large numbers" section

What do you think about this edit? The text before this edit was discussed in /Archive 123#Comparison of texts in "Large numbers"; that version was essentially the one by TheFeds, who had bothered the gargantuan nasty task of wading through all the archived discussions about this. The edit removed, among other things, the permission to use commas in numbers 1000 ≤ x < 10,000 which are not years or page numbers, which is the current behaviour of {{convert}} and some other templates I can't remember right now. --___A. di M. 10:33, 18 August 2009 (UTC)

Unwise. It should at least be tweaked to permit such forms when an author wishes to discuss 1,944 guns and would like to be clear that he does not mean military production in the year of Normandy. Septentrionalis PMAnderson 12:43, 18 August 2009 (UTC)

I (obviously) prefer the version prior to the change. I've left some comments in the "Flurry of edits" section below. TheFeds 17:43, 18 August 2009 (UTC)

Defining 'internationally accepted units' and getting rid of the phrase 'region-specific'

I propose that the following wording be considered for the policy:

In place of this:

Current wording

Which units to use

• In articles which are not region-specific, prefer internationally accepted units. Usually, they are the units of the International System of Units (SI) and non-SI units accepted for use with SI; but there are various exceptions for particular classes of measurements. However, on region-specific topics, use the units used in the place the article is about, for example US customary units for US-related articles.
  • When different parts of the English-speaking world use different units for the same type of measurements, add a conversion so that all English-speaking readers will be able to understand one of the units. For example, the Mississippi River has a length of 2,320 miles (3,734 km); the Murray River has a length of 2,375 kilometres (1,476 mi). (See {{Section link}}: required section parameter(s) missing below.)

I propose considering this wording:

Michael Glass's proposal

Which units to use

• Apart from US and some UK-related articles where US or Imperial measures are used, prefer internationally accepted units. These are units of the International System of Units (SI), non-SI units accepted for use with SI, units based on fundamental constants (e.g. Planck's constant), and other units that are widely used internationally for specific purposes.
  • In general, add a conversion so that all English-speaking readers will be able to understand one of the units. For example, the Mississippi River has a length of 2,320 miles (3,734 km); the Murray River has a length of 2,375 kilometres (1,476 mi). (See {{Section link}}: required section parameter(s) missing below.)

My intention is not to change the policy but to express it more clearly and concisely. Any comments? Michael Glass (talk) 12:34, 18 August 2009 (UTC)

Pmanderson's proposal

I would split the difference:

Which units to use

• Apart from US and some UK-related articles where US or Imperial measures are used, prefer internationally accepted units. These are units of the International System of Units (SI), non-SI units accepted for use with SI, units based on fundamental constants (e.g. Planck's constant), and other units that are widely used internationally for specific purposes.
  • When English-speakers use more than one unit for a given type of measurement, it is generally advisable to add a conversion so that all English-speaking readers will be able to understand one of the units. For example, the Mississippi River has a length of 2,320 miles (3,734 km); the Murray River has a length of 2,375 kilometres (1,476 mi). (See {{Section link}}: required section parameter(s) missing below.)

This provides a rationale for the ruling, and allows for exceptions; there aren't many, but I foresee MGlass's text being used to demand conversions between calendar and tropical years, and other totally silly demands. Septentrionalis PMAnderson 12:51, 18 August 2009 (UTC)

I don't want to go round the houses, but I don't like "Apart from US and some UK-related"... many US articles use metric units, and using exceptio probat regulam this could suggest that SI is actively discouraged in US and some (unspecified) UK articles. This just does not cut the mustard; you can't get rid of the "regional" bit altogether, ugly though it be. For a start, write "Apart from most US-related" (since some use SI etc); and indeed since it simply says "prefer" why not cut that qualifying clause and put elsewhere? Below is not perfect but an attempt to show what I mean:

Which units to use

• Prefer internationally accepted units. These are units of the International System of Units (SI), non-SI units accepted for use with SI, units based on fundamental constants (e.g. Planck's constant), and other units that are widely used internationally for specific purposes.
  • In general, add a conversion so that all English-speaking readers will be able to understand one of the units. For example, the Mississippi River has a length of 2,320 miles (3,734 km); the Murray River has a length of 2,375 kilometres (1,476 mi). (See {{Section link}}: required section parameter(s) missing below.)
• Articles about US or UK subjects should use US or Imperial measures when appropriate, with conversions to SI unless the context makes that ridiculous.
  • Take care to consider Canada and Ireland, which although largely metricated still use US or Imperial measures in some parts of daily life.

SimonTrew (talk) 12:56, 18 August 2009 (UTC)

Certainly an improvement. I would make the first sentence Misplaced Pages generally prefers internationally accepted units - although the phrase internationally accepted is both tendentious and incorrect; units accepted by the US and Canada, or Britain and Canada, are internationally accepted. If it is left as it is, some good soul will go through and switch George Washington to kilometers, quoting the first sentence in isolation. Septentrionalis PMAnderson 13:17, 18 August 2009 (UTC)

Yes, I did consider hoisting the "Articles about US... to the top of the example (even before "prefer internationally...", possibly). I agree about "internationally" and said so in the earlier discussion, but perhaps not as clearly. I think we can cut "internationally accepted units" completely since we immediately give its definition. For the other uses of "internationally" we can say simply "worldwide" (or is that equally contentious? Surely nobody will expect penguins in Antarctica to be getting out their slide rules?)

How about this? (Again, not intended as a final suggestion more something to bite on.)

Which units to use

• Articles about people or places strongly associated with one place or time should use the units appropriate to that place or time. For the US this will generally mean US units, for the UK, sometimes Imperial units.
  • Take care to consider Canada and Ireland, which although largely metricated still use US or Imperial measures in some parts of daily life, and used non-SI units for much of their past.
• With that considered, Misplaced Pages prefers units of the International System of Units (SI), non-SI units accepted for use with SI, units based on fundamental constants (e.g. Planck's constant), and other units that are widely used worldwide for specific purposes.
  • In general, add a conversion so that all English-speaking readers will be able to understand one of the units. For example, the Mississippi River has a length of 2,320 miles (3,734 km); the Murray River has a length of 2,375 kilometres (1,476 mi). (See {{Section link}}: required section parameter(s) missing below.)

SimonTrew (talk) 14:06, 18 August 2009 (UTC)

How about

• For articles not associated with a place or time, especially scientific articles, Misplaced Pages normally uses units of the International System of Units (SI), non-SI units accepted for use with SI, units based on fundamental constants (e.g. Planck's constant), and other units that are widely used worldwide for some specific purpose.

which is self-contained. Septentrionalis PMAnderson 14:28, 18 August 2009 (UTC)

(edit conflict) I hadn't thought about the possibility of misunderstanding the phrase "internationally accepted" that way . I still find it's unlikely that a reasonable person could accidentally misunderstand it, but we'd better prevent unreasonable wiki-lawyers from purposefully misunderstand it. I still don't like the "shopping list"-like explanation, which suggests that all SI units, all units based on fundamental constants, etc. are accepted (while the megasecond isn't accepted, and the Planck mass is only accepted in advanced theoretical physics contexts). Also, I'd rather go with "parts of the English-speaking world" than with "English speakers": no conversion could make all English speakers understand a measure such as "4.7 microfarads", for there are many who don't know what electric capacitance is; but anyone who knows what it is would measure it in submultiples of the farad, regardless of where they're from; so a conversion for that measure is unnecessary and useless. Let me give a try:

----

Which units to use

• Except in the cases mentioned below, prefer the units in most widespread use worldwide. Usually, they are the units of the International System of Units (SI) and non-SI units accepted for use with SI; but there are various exceptions for some measurements, such as inches for display sizes and years for long periods of time.
  • When discussing topics strongly associated with one place or time, use the units appropriate to that place or time. In articles about the present, for the US this will usually mean United States customary units, and for the UK this generally means Imperial units for some classes of measurements and metric units for others (see, for example, the Times Online style guide under "Metric").
• When some parts of the English-speaking world would use a different unit than the one used in the article, generally add parenthetical conversions so that readers from those regions can understand the measurement: for example, the Mississippi River has a length of 2,320 miles (3,734 km); the Murray River has a length of 2,375 kilometres (1,476 mi). (See {{Section link}}: required section parameter(s) missing below.)

----

I'm not sure about "general articles"; I think clearer alternative could be found. As for cases such as kilojoules v kilocalories in Italy (and I suppose Canadians and Irishmen can find other examples of that), it says "locally used", not "locally recommended". Per WP:BEANS, let's wait until some editor cites this guideline and some obscure law requiring in to replace with in an -related article before we explicitly make that point. --___A. di M. 14:47, 18 August 2009 (UTC)

I like the general principle Simon put forth: Articles about people or places strongly associated with one place or time should use the units appropriate to that place or time. Arago used toises, and we should describe in toises - translating into meters and yards. Septentrionalis PMAnderson 15:31, 18 August 2009 (UTC)

Emended to incorporate that point. BTW, I think that "people or places" is too restrictive, as beer glass is neither a person nor a place (not in the obvious sense of "place", at least); also "articles" should be toned down, because an article discussing several topics can use different units for each one of them, as in the Irish road speed limits example (mph for historical limits, km/h for modern ones). --___A. di M.

I agree with A. di M. that, yes, the phrasing around "internationally accepted" is to avoid unreasonable wikilawyering (and I would love to get rid of the "internationally" and prefer, say "accepted worldwide"); but wikilawyering is exactly what that wording was intended to avoid, i.e. deliberate misinterpretation. I also agree that "people or time" (I didn't say people but places, btw, but it holds) is unduly restrictive, but of course we don't want to enumerate every possible kind of association. I also see the point about "articles", what I was trying to do here was restrict the recommendations to article space, not to e.g. MOSNUM talk space where obviously it is necessary to mix units, nor (which I think you had more in mind) to articles where it is right and proper to mix units.

How about "Articles or parts of them that are about things strongly associated with one place or time..." i.e. substitute "things" for the first "place or time" but keep the second "place or time" since that, I think, *is* what essentially defines which units are used, e.g. the size of beer glasses depends on the size the beer is served in which depends on the place. There may be a case for putting "people" in there too, since i can imagine for example the article on Anders Celsius is so strongly associated with the unit of measure that it is the person who causes the association, not particularly where or when he lived. But I don't want to enumerate it much further than that, if we can help it.

"Things" perhaps sounds vague, but I think is perfectly adequate. I suppose there will be wikilawyering about whether concepts are things and so on (yes they are. A thing is a noun. And most WP articles are about nouns. Even an article, say, Wikilawyering (this is in WP space but the redirect is from article space) is a noun, it may look like a verb but is an agentive noun (some would call it a gerund). I suppose there are some articles that are not in any sense about things, though.

The kilocalorie thing should probably be dropped from this thread and maybe start another for this special (unique?) case. I hate to want to write a special case into MOSNUM simply for it, and surely a reasonable editor would take WP:COMMON (as I did at Bacon in the example I gave above, and although that article is reasonably frequently edited, it has not been reverted). My only concern here really is that a picky GA or FA reviewer will point out an inconsistency here, where the inconsistency really is there in daily life and must be fairly reflected in WP.

Best wishes SimonTrew (talk) 12:30, 19 August 2009 (UTC)

Maybe "topics" is the word you are looking for instead of "things". --___A. di M. 16:54, 19 August 2009 (UTC)

SimonTrew's proposal

Yes, "topics" is much better. I just start a new subsection since we've not had a complete proposal written out for a while, and this is very similar to one above (has it been edited? I don't recall it being quite so similar) so here goes (the bold and strike are to indicate changes that may otherwise be too suble to be noticed; and are not intended to indicate proposed markup):

----

Which units to use

• Except in the cases mentioned below, prefer the units in most widespread use worldwide. Usually, they are the units of the International System of Units (SI) and non-SI units accepted for use with SI; but there are various exceptions for some measurements, such as inches for display sizes and years for long periods of time.
  • When discussing topics strongly associated with places, time or people, use the units appropriate to them. In articles about the present, for the US this will usually meanbe United States customary units, and for the UK this generally means Imperial units for some classes of measurementstopics and metric units for others (see, for example, the Times Online style guide under "Metric").
• When some parts of the English-speaking world would use a different unit thanfrom the one used in the article, generally place conversions afterwards in parentheses so that all readers from those regions can understand the measurement: for example, the Mississippi River has a length ofis 2,320 miles (3,734 km) long; the Murray River has a length ofis 2,375 kilometres (1,476 mi) long. (See {{Section link}}: required section parameter(s) missing below.)

----

I am not that het up about "different than", but since others may insist on "different from", we might as well do that and avoid needless argument. The other rewordings are basically just simplifications or to remove being over-specific (e.g. "readers from those regions" -> "all readers", surely the intent is that anyone reading the article can undertand the measures in it.) SimonTrew (talk) 12:16, 20 August 2009 (UTC)

Minor points: I'd still rather go with "readers from all over the world" or something like that, than with "all readers"; see the example about microfarads somewhere else in the thread. (BTW, is "time" rather than "times" in the inner bullet a typo?) --___A. di M. 12:42, 20 August 2009 (UTC)

Excellent proposal, explains the spirit very well and makes clear how all the details flow from it. Hans Adler 12:40, 20 August 2009 (UTC)

Headbomb's proposal

Leave it as is. It ain't broken, and at this point all we're doing is listing WP:BEANS scenarios. The original wording of "Use international units usually this means SI, SI-related & units accepted with SI" is as both as clear and as vague as it needs to be. The subtleties are covered in the bullets. Headbomb {_{κοντριβς} – WP Physics} 11:31, 21 August 2009 (UTC)

General comments on region vs. internationally accepted

All these wordings are confusing in that they say "use", but in most cases, what is really meant is "list first", because conversions are usually provided. It would be nice to think that editors would read the manual from end to end and remember everything, but that just isn't going to happen, so wording that does not require the editor to read a different part of the manual to understand that "use" usually means "list first" would be better. --Jc3s5h (talk) 14:34, 18 August 2009 (UTC)

That's why I added a sub-bullet about unit conversions to the first bullet in the "Which units to use" subsection , where there was none. --___A. di M. 15:06, 18 August 2009 (UTC)

Adjusting the wording in the light of the previous discussion

I agree that the wording is confusing in saying 'use' but meaning 'list first'. It is also inconsistent with later dot points. This is how the passage reads at the moment

Which units to use

• Except in the cases mentioned below, prefer the units in most widespread use worldwide. Usually, they are the units of the International System of Units (SI) and non-SI units accepted for use with SI; but there are various exceptions for some measurements, such as inches for display sizes and years for long periods of time.
  • When discussing topics strongly associated with places, times or people, use the units appropriate to them. In articles about the present, for the US this will usually be United States customary units, and for the UK Imperial units for some topics and metric units for others (see, for example, the Times Online style guide under "Metric").

• When some parts of the English-speaking world use a different unit from the one used in the article, place conversions afterwards in parentheses so that readers from all over the world can understand the measurement: for example, the Mississippi River is 2,320 miles (3,734 km) long; the Murray River is 2,375 kilometres (1,476 mi) long. (See {{Section link}}: required section parameter(s) missing below.)

Which units have priority

• Except in the cases mentioned below, put the units in most widespread use worldwide first. Usually, they are the units of the International System of Units (SI) and non-SI units accepted for use with SI; but there are various exceptions for some measurements, such as years for long periods of time.
  • With topics strongly associated with places, times or people, put the units most appropriate to them first. In US articles about the present, this will usually be United States customary units; for the UK Imperial units for some topics and metric units for others, and a mixture of units for others (see, for example, the Times Online style guide under "Metric"). The most appropriate units for articles may be deduced from the units most used in the sources for the article.

• When some parts of the English-speaking world use a different unit from the one used in the article, place conversions afterwards in parentheses so that readers from all over the world can understand the measurement: for example, the Mississippi River is 2,320 miles (3,734 km) long; the Murray River is 2,375 kilometres (1,476 mi) long. (See {{Section link}}: required section parameter(s) missing below.)

Comment

This version replaces 'prefer' with 'put.....first'. I have also removed the reference to screen sizes as I thought this was a rather trivial example. Years are important, the use of feet for the altitude of aircraft is important, but the size of a television or mobile phone screen is not so important. With UK articles a useful guide to what unit has priority may be to follow the majority of sources of information for the article. Michael Glass (talk) 13:53, 22 August 2009 (UTC)

"Put first" only makes sense when conversions are used, which won't be the case whenever something is measured with the same unit throughout the English-speaking world (hours, volts, picofarads, ohms, typographic points, dots per inch, yadda yadda yadda). "Use" always makes sense; the bullet about conversion makes clear than in "the Mississippi River is 2,320 miles (3,734 km) long", by "the one used in the article" we mean the mile. The addition of the point about source sounds fine. --___A. di M. 15:01, 22 August 2009 (UTC)

Based on your suggested changes:

I don't think it is necessary to change the title - we do already have a unit conversions section directly below this and we do already mention unit conversions in the text. For the same reason, I prefer prefer to put... first, but am not overly fussed about it.

I think leaving it at years is not really enough. It is, IMO, useful to include a measurement that is an exception to the SI-only rule in most of the world but not all of it - to show that this applies to more than just the most obvious cases. Inches for screen display size is a good example. If you can think of a better one, then I don't object to it being used instead.

Your sources bit appears to require source-based units in all circumstances on UK-based article. I don't think this is a good idea. Sources are often written for a non-native audience, and may reflect a units that would not ordinarily be used in the context of the topic concerned. Taken to a relative extreme, requiring source-based units could mean we have to compare a 2,000-foot (610 m) hill with a 650-metre (2,130 ft) hill because those are the units the two (completely different) sources use - something which I do not believe would look particularly professional.

Better to be a bit vaguer, IMO, to allow editors to determine the best units in the context of the article. If consensus is that converting the sourced units is the best idea, then we should convert the sourced units. We do still have WP:V and the section on sourcing below, so it's not like sourcing is not part of the equation. Pfainuk talk 15:09, 22 August 2009 (UTC)

My responses:

• I agree that 'Put first' only makes sense when conversions are used. In other words, the policy would only apply where it was relevant. If the language is more precise, this is an advantage, not a defect.

• if we mention unit conversions in the text then it is only sensible to mention unit conversions in the title, since this is what the text is about.

• I think the use of the foot to measure altitude in aviation is a better example than the screen size of television sets.

• My mention of sources The most appropriate units for articles may be deduced from the units most used in the sources for the article offers guidance to editors; it is not a straitjacket. Note that it says may be deduced not is to be deduced.

• My mention of sources is consistent with present policy: 'If editors cannot agree on the sequence of units, put the source value first and the converted value second. If the choice of units is arbitrary, use SI units as the main unit, with converted units in parentheses.'

Michael Glass (talk) 22:57, 22 August 2009 (UTC)

I see that you say "may be" as opposed to "is to be", but this remains the implication to my mind.

You argue that this is consistent with current policy. I don't totally agree, but I'll take this for the sake of argument. If it is consistent with current policy, why does it need reiterating specifically for British units? Better, surely, for the current guideline to remain and apply to all cases equally. This would avoid any implication that the rules for UK-related articles are any different from the rules for other articles. If the guideline is unchanged, why does this need to be added? Better, in my view, to avoid repeating ourselves too much.

My understanding was that the foot was universal (or nearly) when doing plane heights. Looking at Altitude, apparently not - so I will accept the foot in this case. Pfainuk talk 08:49, 23 August 2009 (UTC)

Another refinement of the suggested wording

Taking into account the criticisms above, here is another refinement of the suggested wording and the title:

• A more comprehensive title is provided.
• The wording has been tweaked to make it clearer in one or two places.
• The words 'in the present' have been removed from the direction about US articles because I couldn't see that they said anything useful.
• The dot points have been rearranged into two lists. This enabled me to remove one redundant sentence and to tidy the rather ragged layout in the present policy.

Which units to use and how to present them

• Except in the cases mentioned below, put the units first that are in the most widespread use worldwide. Usually, they are the units of the International System of Units (SI) and non-SI units accepted for use with SI; but there are various exceptions for some measurements, such as years for long periods of time or the use of feet in describing the altitude of aircraft.

• With topics strongly associated with places, times or people, put the units most appropriate to them first. In US articles, this will usually be United States customary units; for the UK Imperial units for some topics and metric units for others, and a mixture of units for others (see, for example, the Times Online style guide under "Metric").
• If editors cannot agree on the sequence of units, put the source value first and the converted value second. If the choice of units is arbitrary, use SI units as the main unit, with converted units in parentheses.
• Generally, use units consistently (e.g., write a 10-kilogram (22 lb) bag of potatoes and a 5 kg (11 lb) bag of carrots, not a 10-kilogram (22 lb) bag of potatoes and a 11-pound (5 kg) bag of carrots).
• Nominal and defined values should be given in the original units first, even if this makes the article inconsistent: for example, When the Republic of Ireland adopted the metric system, the road speed limit in built-up areas was changed from 30 miles per hour (48 km/h) to 50 kilometres per hour (31 mph). (The focus is on the change of units, not on the 3.6% increase.)
• Avoid ambiguous unit names (e.g., write imperial gallon or US gallon rather than gallon). Only in the rarest of instances should ambiguous units be used, such as in direct quotations, to preserve the accuracy of the quotation.
• When some parts of the English-speaking world use a different unit from the one used in the article, place conversions afterwards in parentheses so that readers from all over the world can understand the measurement: for example, the Mississippi River is 2,320 miles (3,734 km) long; the Murray River is 2,375 kilometres (1,476 mi) long. (See {{Section link}}: required section parameter(s) missing below.)

• In scientific articles, use the units employed in the current scientific literature on that topic. This will usually be SI, but not always; for example, natural units are often used in relativistic and quantum physics, and Hubble's constant should be quoted in its most common unit of (km/s)/Mpc rather than its SI unit of s.
• Some disciplines use units not approved by the BIPM, or write them differently from BIPM-prescribed format. When a clear majority of the sources relevant to those disciplines use such units, articles should follow this (e.g., using cc in automotive articles and not cm). Such use of non-standard units are always linked on first use.
• Use familiar units rather than obscure units—do not write over the heads of the readership (e.g., a general interest topic such as black holes would best be served by having mass expressed in solar masses, but it might be appropriate to use Planck units in an article on the mathematics of black hole evaporation).

Michael Glass (talk) 13:05, 23 August 2009 (UTC)

I'll accept this. I think it would be better if the sourcing point was second last in the sub-list, so as to list the more general rules first, and then to proceed to what to do if unsure. But this is a fairly minor point - not worth too much hassle. Pfainuk talk 19:57, 23 August 2009 (UTC)

Thanks for the input. I'll go with the word order above and see what the result is. Michael Glass (talk) 22:51, 23 August 2009 (UTC)

Flurry of edits

There seems to have been a flurry of edits over the last day or so from A di M (5) Noetica (9) and GregL (6). While I know these are all good faith edits from good faith editors, it suggests to me that this has not properly achieved the consensus we should expect before changing WP:MOSNUM, where stability is incredibly important. May I suggest we hold off and perhaps use the talk page more rather than the guide itself, since a guide that is constantly changing is no use to anyone. SimonTrew (talk) 12:46, 18 August 2009 (UTC)

SimonTrew, at PMAnderson' talkpage you say that I have "jumped the gun" with my editing. On the contrary: my recent edits are all like these, the most minor uncontroversial tidying. The only ones that go beyond such housekeeping are to revert PMAnderson's premature editing in response to my raising a point for discussion (see above), and a conservative clarifying response to a point raised by A di M.

Please refrain from incautious accusations. I am not disrupting anything at all, and I have explicitly called for discussion rather than hasty editing.

–_⊥Noetica!– 14:16, 18 August 2009 (UTC)

I did not accuse you of anything. I stated my own opinion, and said as such. It is not a question of how minor the edits are: any edit to the MOS impacts, in theory, on every article in the encyclopaedia, if only in so much as now the article has to be checked again against the MOS to see if it still conforms. When I have had even a minor "housekeeping" edit (change of wording etc) I have taken it to discussion; and most of the edits are not marked as minor (so, are they minor as you claim, in which case mark them minor, or not, in which case discuss them)? I am not prepared to enter into discussion of personalities here; I am also trying to make the MOS better, because I edit articles and try to make them conforming, or at least more conforming, and continual changes to the MOS, however small, are counterproductive to that aim. Since there have been (using my above figures which are a little dated) 20 edits to MOSNUM in the last day or so, in fact more now, that averages a little under 1 change an hour. How is a poor article editor like me supposed to keep up with that? Better to get consensus for one big change, and this goes against my usual reasoning at WP:OWNFEET, because here we are not talking about an end article but something that affects millions of articles.

To repeat: I stated that I thought you had jumped the gun; I think you did. I stated it on a user's talk page which, while not private, does not oblige me to have NPOV. That is not an accusation of anything, it is my opinion. SimonTrew (talk) 15:45, 18 August 2009 (UTC)

You do not say which edit of mine "jumps the gun". Surely not the ones in which I change the dashes and quotation marks so that they conform to MOS styling! Don't make scattershot assertions. I have responded to your comment at my talkpage. If you raise an issue, expect it to be discussed till all is clarified.

–_⊥Noetica!– 22:45, 18 August 2009 (UTC)

• I had been away from MOSNUM for a while. While gone, there were several edits made that seemed to take MOSNUM away from the long-standing ways things have always been done on Misplaced Pages with regard to numbers. For the most part, these changes seem to have been the product of a tag-teaming by two like-minded editors over a period of one week. A consensus of two editors does not a consensus make. What I’ve now restored (and better organized) reflects widely observed, common practices on Misplaced Pages that have long enjoyed a true consensus. These time-tested practices, which were the product of much discussion over the years, are intended to yield the most important thing on Misplaced Pages: result in the least confusion for the greatest portion of our readership. Sometimes, editors come here to MOSNUM to change things in order to lend legitimacy to their particular way of doing things in articles they’re working. However, this is often done with an insufficient understanding of the ramifications. Greg L (talk) 16:45, 18 August 2009 (UTC)

• I don't object to the reorganization—I have no issue with moving things to different subsections for clarity. It's the reversions that I don't particularly understand

  (In the interests of disclosure, with regard to the following, it looks like Greg_L worked on the current version, and I worked on the versions that were recently reverted. It's not personal, but I do think that my edits were better.)

  It's pretty clear that the consensus (which extends far beyond MOSNUM) with respect to things like regional usage variations is to allow them in context (and the text I proposed and eventually inserted into MOSNUM on digit grouping upholds that). Furthermore, the recently-reverted section clearly articulates that there are two methods of digit grouping in standard English usage (and that neither is to be mandated exclusively), noting some contexts in which it is common to find one or the other. The previous version of that text is also clearer, because it organizes these things into bullets and uses more precision in explaining the technical details.

  With regard to another recent change—that of the reworded adoption of international standards section—I still object to the aggressiveness and essay-like qualities of the current (reverted) version. My version retains the core message that certain things (i.e. " %", "MiB") are not valid on Misplaced Pages despite the existence of various international standards, but avoids the repetition of "real-world" and removes the commentary about the objectives of Misplaced Pages. (Those things are rightly found in the policy documents and user essays, but don't need to be reinterpreted here, especially not in the context of two long-running editing disputes.)

  Besides, the consensus on Misplaced Pages is not that real-world usage always prevails—though it often does, justifiably—just look at the citation system for evidence of consensus in favour of an invented system not found externally. The point is that Misplaced Pages can choose to follow whatever the community wants, and isn't necessarily bound to the real-world norm as a matter of policy. If doing something differently makes the encylopedia better, then it's a valid course of action. But if following someone else's lead (be it BIPM or traditional American usage) leads to a better encyclopedia, then that's appropriate as well. If we want to decide what Misplaced Pages's broad objectives ought to be, we should discuss that at WP:VPP. TheFeds 17:38, 18 August 2009 (UTC)

• TheFeds, I think my edit regarding the percent sign and MiB, etc., was in haste. I misunderstood the impact of your change when I was looking at the edit-diff. Looking at the actual text, I think your version was an improvement and have no objection at all if you change that section back. I would offer to do it myself, but I will give you the liberty of changing it so it is sure to meet with your satisfaction. Greg L (talk) 17:42, 18 August 2009 (UTC)

• P.S. That was easy enough. I restored it to the version you made. Sorry for the inconvenience. Greg L (talk) 17:54, 18 August 2009 (UTC)

• Thanks, Greg. I'm glad that the misunderstanding is behind us. Looks like you beat me to the change, and that section is definitely alright now. TheFeds 17:57, 18 August 2009 (UTC)

I actually end up back where I started. I asked for a stand off of edits and to take it to talk page rather than continually change the policy article itself. I am sorry Greg L that you are offended that a mere article editor like me might have an interest in the guidelines under which he is supposed to edit articles, and impinging on your space by it not going the right way to discuss changes while you were away (and frankly it was going quite nicely, with very few edits and plenty of discussion before changes were made). As for history etc etc, well, who cares? I look at the article and see the problems NOW, not as they were seven years ago. Since 85% of my edits are in article space not WP space or template space or whatever, I just want to come here, note a problem, get consensus, etc. While it is useful (sincerely) that other kinds of WP editors take time to make sure MOS etc are correct, I simply am not going to get bound up in this, but it annoys me that it smacks of WP:OWNERSHIP almost. The veiled thing about "New editors" I assume refers to me. If it does, just say so, I can take it. I didn't realise longevity was a criterion ("take MOSNUM away from the long-standing ways things have always been done" – excuse me while I bring the boy down the chimney and teach him to type a response to that).

MOSNUM actually had a period of stability where I could actually rely on it for a bit. I think now I give up and will just stick to, say, the convert template talk where, if there are problems or additions needed for articles, User:Jimp and many other helpful folk there actually sort it out and, if us poor article editors are mistaken, kindly and politely tell us so. What is the problem here? Have I hit the nail on the head? It seems to me, right now, that there is a kinda warring faction with Greg L, Noetica and PMA some long-standing editors that none of us mere mortals are privy to, and only they have the right of an opinion on MOSNUM? Can you point me to maybe a meta-policy that says so?

Perhaps I am not in the best of moods, so for that I apologise in advance. But MOSNUM is supposed to be here to HELP PEOPLE WRITE ARTICLES, not as some navel-gazing activity. It does not help me write articles if it is constantly changing under my feet, and what in other contexts would be characterised as an edit war has taken place in the last couple of days. C'mon, folks, you are supposed to be the best of the best to edit something as crucial as this. Live up to that responsibility.

Best wishes. SimonTrew (talk) 00:47, 19 August 2009 (UTC)

Please abide by consensus

Jc3s5h, your edits were contrary to the results of the RfC you yourself conducted here on Archive 123. The {val} template was the result of very lenghty, months-long discussions by very many editors on both WT:MOS and WT:MOSNUM.

{Val} (originally known as “Delimitnum”) had its functionality described here in WT:MOSNUM Archive 94…

…it was extensively discussed and voted upon here in WT:MOSNUM Archive 94…

…and was well received here on WT:MOS Archive 97…

…where its functionality was tweaked to achieve a compromise solution that made everyone happy on an issue regarding the look of scientific notation.

Then a number of developers and template authors worked on it.

Please don’t presume that you can come along and change it without a proper consensus. Greg L (talk) 17:17, 18 August 2009 (UTC)

Greg L misinterprets the results of the RFC. While the Val template is generally acceptable in that it can be altered to accommodate almost any format for which there is consensus, Greg L is the only one who considers the commas-left, thin-spaces-right format to be the best choice. A few others considered it acceptable but not superior to other choices. The outcome of the discussion was to avoid the commas-left, thin-spaces-right format. Val should be modified accordingly.

I also call upon Greg L to abide by the argument he has often made with respect to binary prefixes (that is, follow external consensus rather than advocate formats that have not achieved external acceptance, and choose a format that has at least limited acceptance outside of Misplaced Pages to a format that has no acceptance at all outside Misplaced Pages. --Jc3s5h (talk) 18:34, 18 August 2009 (UTC)

• You must not have read the links I provided, above; they show widespread, overwhelming support for {val}. The consensus (not just me) on Misplaced Pages is that the techniques {val} uses for scientific notation and long strings is a good one that causes zero confusion. If you don’t like it, don’t use it. But please stop trying to delete mention of it or discourage its use. It serves a valuable purpose in technical articles. Greg L (talk) 18:44, 18 August 2009 (UTC)

Delimiting numbers

• (no longer beating around the bush here): One source of periodic friction on MOSNUM is the delimiting of numbers. There are four or even five different ways of doing so. In Sweden, they teach school children three different methods (and two of them are “Swedish 1” and “Swedish 2”). To make a long story short: Misplaced Pages allows both British and American-dialect English (spelling) in its articles, so long as they are consistent within an article. However…

  This practice of “your way / my way… it’s all just six of one / half a dozen of the other” doesn’t apply to numbers. Why? Misplaced Pages has gone through all this before many times, and new editors who come here don’t have the benefit of all that history and discussion. But it comes down to this: English-speaking Europeans are familiar and comfortable with a many different ways of delimiting numbers and the American style causes them no confusion whatsoever. Comma-delimiting might not be the most common practice for English-speaking Europeans, but they recognize what it means and fully understand the numbers. However, Americans are familiar with one and only one method; as a group, they have had no exposure whatsoever to other ways of delimiting numbers. So, especially for general-interest articles, in order to cause the least confusion, the American method of using commas to the left of the decimal point is to be used on Misplaced Pages. Scientific articles; particularly ones directed to a professional readership, are the only exception.

  The argument that “Well, Misplaced Pages will just start using the Euro/BIPM method and dumb-ass Americans will simply learn” just doesn’t fly and it never will. Misplaced Pages doesn’t have that kind of influence; all that sort of attitude does is produce confusion. Our aborted attempt to push the world into the adoption of the IEC prefixes (kibibytes and KiB) amply demonstrated that. After three long years, the practice was no more well adopted throughout the world than before. All Misplaced Pages accomplished by letting itself by hijacked by a handful of editors who wanted to push the world into a new and brighter future with warp drive and membership in the United Federation of Planets™®© was to make our computer-related articles needlessly confusing. We follow the way the world works and can not presume to lead by example.

  We can’t have MOSNUM subtly edited in a fashion that tacitly allows numbers in articles, other than science-related ones, to be delimited with thinspaces in place of commas to the left of the decimal point; it is unnecessarily confusing to too many readers. This is the way it has long been done and there has been no decision to change the practice.

  As for delimiting with gaps to the right of the decimal point using {val} on high-precision numbers, particularly in engineering and scientific-related contexts where the distinction between numbers is important and the values actually have to be parsed and understood, that confuses absolutely no one—even “sheltered Americans”. A value like 1.6162523625×10 meters is no more confusing than the decidedly non-SI-compliant, five-digit delimiting used for mathematical constants (particularly in tables of constants), such as 3.141592653589793238462643383279...; everyone instantly “gets” it. It is a much-appreciated and much-needed touch that makes long strings of digits much easier to parse. Moreover, its technique of using thinspaces to the right is the only possible method that could be employed to solve the problem of long strings without upsetting the apple cart; it was either do nothing (and have absurdly long strings that can’t easily be parsed) or utilize the only logical available technique. Greg L (talk) 18:28, 18 August 2009 (UTC)

• My position is that using thin spaces to both the left and right of the decimal when an article contains one or more numbers with 5 digits to the right of the decimal is preferable to Misplaced Pages making up its own format. It also my position that the practice of using thin spaces to the right and commas to the left looks especially asinine in the case of a number like 4,046.8564224. I will not change my position unless a reputable external source, such as a major style guide, which supports the 4,046.8564224 format, is cited. --Jc3s5h (talk) 21:22, 18 August 2009 (UTC)

• That’s perfectly fine, Jc3s5h. Don’t write it that way then. A number like 4046.8564224 is rare anyway. With that much precision, such numbers will often be in scientific articles where scientific notation might be more appropriate. Amongst all the above-cited discussions in the archives, there was another editor who felt as you do. We go with the consensus here on Misplaced Pages and the approval of {val} was (very) lopsided. The guideline advising editors that numbers like 1.6162523625 × 10 meters are hard to parse and they should consider using 1.6162523625×10 m is a sound one because it makes Misplaced Pages easier to read and understand. Greg L (talk) 21:44, 18 August 2009 (UTC)
  • Thank you; as long as this is phrased so as to be clear that the answer to "I don't like 4,046.8564224" is "Don't use it then", not "MOS breach! MOS breach! shun this start-class article", a recommendation backed by a lopsided majority should be fine, and harmless. The present use of may be seems to achieve that, but I would value Jc3s5h's opinion. Septentrionalis PMAnderson 00:09, 19 August 2009 (UTC)

(unindent) When Pmanderson asks whether the present use of may be, I surmise he is asking about this section:

• Numbers with more than four digits to the right of the decimal point, particularly those in engineering and science where distinctions between different values are important, may be separated (delimited) into groups using the {{val}} template, which uses character-positioning techniques rather than distinct characters to form groups. Per ISO convention (observed by the NIST and the BIPM), it is customary to not leave a single digit at the end, thus the last group comprises two, three, or four digits. Accordingly, the recommended progression on Misplaced Pages is as follows: 1.123, 1.1234, 1.12345, 1.123456, 1.1234567, 1.12345678, 1.123456789, etc. Note that {{val}} handles these grouping details automatically; e.g. {{val|1.1234567}} generates 1.1234567 (with a four-digit group at the end). The {{val}} template can parse no more than a total of 15 significant digits in the significand. For significands longer than this, editors should delimit high-precision values using the {{gaps}} template; e.g. {{gaps|1.234|567|890|123|45}} → 1.2345678901234567.

Let us set aside my objection to the 4046.8564224 format for the moment; whether my interpretation or Greg L's interpretation of consensus prevails will become apparent in due course. The present version of the guideline states, or at least strongly implies, that the Val template conforms to "ISO convention (observed by the NIST and the BIPM)", but the template at present does not conform to the ISO convention. Furthermore, every example in this section has exactly one digit to the left of the decimal, so the non-conformance is concealed.

This is a falsehood. I don't think it has been pointed out until now, so it is an innocent falsehood. But over time, as the falsehood becomes better understood among editors of this guideline, it could ripen into a lie. --Jc3s5h (talk) 00:28, 19 August 2009 (UTC)

If there is a question of fact (assertions of fact are generally undesirable on guideline pages, because it provokes exactly this sort of discussion), would both of you provide citations? Septentrionalis PMAnderson 00:34, 19 August 2009 (UTC)

• Alternatively, can Jc3c5h propose a text which supports the use of {{val}} - commonly but not without exception - but does not make those claims of fact?
• Is GregL willing to consider such a text? Septentrionalis PMAnderson 00:37, 19 August 2009 (UTC)

• In response to Pmanderson's request for a citation, the BIPM brochure setting for the International System of Units states "for numbers with many digits the digits may be divided into groups of three by a thin space, in order to facilitate reading. Neither dots nor commas are inserted in the spaces between groups of three."

As for proposing a text supporting the use of {{val}}, I don't support that template at all until it is modified to not allow commas and thin spaces in the same number. (I have no objection to a version that provides a parameter to choose between the BIPM format and the customary American format.) --Jc3s5h (talk) 00:50, 19 August 2009 (UTC)

I see that citation does address how to handle 4 digits (make a single group of 4); but one issue between you is how to handle 8 digits. {{Val}} divides them IIUC 3, 3, and 2; what would you do, and can you cite it as ISO? Septentrionalis PMAnderson 01:14, 19 August 2009 (UTC)

Septentrionalis, I suspect you are missing the point. Looking at 8 digit numbers, 1234.678 is fine. The format 12345678 is fine if the article does not contain any numbers with 5 or more digits to the right of the decimal. The formats 12345678 and 0.12345768 should not appear in the same article. The formats 12345678 and 0.12345768 in the same article are fine. --Jc3s5h (talk) 01:29, 19 August 2009 (UTC)

Let me get this straight. Either commas or gaps, not both. That still leaves several possibilities to object to. Do you object to

• The use of gaps and commas in the same article?
• The use of gaps and spaces in the same number?
• The claim that (which of the above?) is ISO?
• All three?

Are you willing to let Greg use his preferred format, if you can use yours? Septentrionalis PMAnderson 01:41, 19 August 2009 (UTC)

My two cents: there's no problem if the lead of LHC reads "over 10,000 scientists and engineers" and a paragraph deep down inside it says, "At this energy the protons have a Lorentz factor of about 7500 and move at about 99.9999991% of the speed of light." They are different contexts: the former is something even my grandmother could read (if he spoke English), the latter is an advanced technical detail. But I'd prefer avoiding both styles in the same contexts: after all, contexts in which you want to use the traditional American style not to confuse American readers are very, very, very unlikely to contain any number with a large number of digits after the decimal point. (Personally, when I see a number with more than three significant figures I ask myself where the hell the last ones were taken from, and if I can find no answer, that confuses me much more than an unfamiliar format would.) These generally occur in scientific and engineering contexts, where using thinspaces is OK. Of course, these are generalizations and exceptions may occur. --___A. di M. 02:16, 19 August 2009 (UTC)

In response to the items questioned by Septentrionalis,

• The use of gaps and commas in the same article? Object Neutral.
• The use of gaps and spaces in the same number? Strongly Object
• The claim that (which of the above?) is ISO? The claim that {{Val}} conforms to ISO/BIPM/NIST format is not compatible with the use of the comma as a separator at all. This wording will have to change if {{Val}} is capable of using commas as a separator.

I do not support the use of made-up formats when an acceptable format is already in use in non-Misplaced Pages publications. Gaps and spaces in the same number is certainly a made-up format. Mixing numbers with commas and numbers with spaces in the same article is, in my view, also a made-up format, but I suppose others could argue to the contrary. --Jc3s5h (talk) 02:09, 19 August 2009 (UTC)

Greg? The third bullet point above is a call for citation; please give one. I cannot help with this otherwise without proposing a compromise text, and I do not see one. If one of you bends to the breaking point, the other may see the result as marginally acceptable. Septentrionalis PMAnderson 02:14, 19 August 2009 (UTC)

• Quoting Jc3s5h: The present version of the guideline states, or at least strongly implies, that the Val template conforms to "ISO convention (observed by the NIST and the BIPM)". Fact: It does (and says) no such thing. Read the text. It says as follows:

Per ISO convention (observed by the NIST and the BIPM), it is customary to not leave a single digit at the end, thus the last group comprises two, three, or four digits.

It’s 1.2345678, not 1.2345678. That’s what it says. That’s all it says.

Following the logic of Jc3s5h, if there are any numbers in an article that have the right hand side delimited (for clarity of parsing), then that gives him carte blanche freedom to use spaces to the left of the decimal point. Perhaps. But only if the article is on a scientific subject. There is no reason for this majorly-conflicted dilemma. It seems Jc3s5h rather wants to expand the Euro-style way of delimiting and is exploiting the existence of {val} to push this view. That would be a mistake if it’s the case. But, if it’s just a matter of WP:IDON'TLIKEIT, then don’t use it, because many others do like it.

Count the number of readers who are confused by the use of {val} on Kilogram. Zero. Zip. Count the number of readers who would be unnecessarily confused if we expanded the use of comma delimiting outside of the realm that it is currently limited to (science). How many American readers are going to be baffled if we started delimiting with thinspaces to the left of the decimal marker?? A metric butt load; that’s how many. It would be foolhardy to contemplate any such change. Like A. di M. wrote, above, …there's no problem if the lead of LHC reads "over 10,000 scientists and engineers" and a paragraph deep down inside it says, "At this energy the protons have a Lorentz factor of about 7500 and move at about 99.9999991% of the speed of light." Even his grandmother wouldn’t be confused.

And, Jc3s5h, don’t bother quoting how the BIPM says numbers ought to be delimited to the left of the decimal marker in documents for a world-wide audience. That isn’t the first advise from the BIPM that Misplaced Pages (and the rest of the world ignores) and it won’t be the last. Notably—perhaps very unfortunately—America ignores that advise. Perhaps you find that to be an unfortunate shame. I might agree with you. Perhaps you think we will Make the world a better place by leading by example.®™© No. That is not what Misplaced Pages can or should do. This flouting of standards applies to the IEC and many other standards bodies: sometimes proposals fly like a wet noodle. However, the good ones that solve a problem rapidly gain traction in the world. And when the rest of the world changes, Misplaced Pages will follow suit. Note also, en.Misplaced Pages is not intended to be read by all cultures throughout the world; just English-speaking ones (either first or second language). As I’ve patiently explained above, the rest of the English-speaking world “gets” the American-style way of delimting. However, Americans don’t “get” other ways of delimiting. Not… in… the… slightest. That’s why en.Misplaced Pages has adopted the comma to the left. Please accept that and stop seizing on {val} as a reason why thinspaces ought to be expanded. Greg L (talk) 03:07, 19 August 2009 (UTC)

• The idea behind Misplaced Pages allowing the BIPM method of delimiting in its science-related articles was borne out of the fact that professional, peer-reviewed articles in science are typically written that way, as required by the manuals of style of the respective journals. They have an international audience and the journals try to make the papers as easy to digest for an extraordinarily wide, professional audience. The “science” exception has often been used as a window of opportunity to be exploited by some Wikipedians in an effort to expand the practice into just about anything (“well… water is a ‘sciency’ subject.”).

  IMHO, the objective on Misplaced Pages should always, always be about writing in a way that is clearest and causes the least confusion. There are many distasteful practices that come with that philosophy, including using feet and inches and pounds in American-related articles like Boston Red Sox. Too often, editors want to do things a certain way because… well… editors simply like doing it that way and always have done it that way. It ends up being more about writing to please the Wikipedian rather than do what is best for the readership. Greg L (talk) 04:35, 19 August 2009 (UTC)

A break after the boxed sentence puts ISO and {{val}} into different and shorter paragraphs; that should remove any implication of connection where none is intended. Septentrionalis PMAnderson 00:59, 21 August 2009 (UTC)

I was trying to think of something that would remove (or at least greatly reduce) the implication without being too bulky. This is an elegant change. --Jc3s5h (talk) 01:05, 21 August 2009 (UTC)

Proposal

To resolve the issue I offer a two part proposal:

Part 1: {{Val}} is altered so that by default, it produces commas to the left and no separation to the right of the decimal point (customary American style). A parameter is added, BIPM=y or yes which produces thin space separators on both sides of the decimal point.

Part 2: The "Delimiting (groups of digits)" be altered to read as follows:

• Numbers with five or more digits to the left of the decimal point; i.e. 10,000 or more, should be delimited (separated into groups so they can be easily parsed visually) using commas every three digits; e.g. 12,200 and 255,200 and 8,274,527 etc. Exception: articles containing numbers with five or more digits to the right of the decimal point, e.g. 3.141593 (as such articles should delimit numbers as for scientific articles—described below).
• Numbers with four digits to the left of the decimal point may be delimited with a comma; that is, there were 1250 head of cattle and there were 1,250 head of cattle are both acceptable. The same exception as for five digits to the left of the decimal point applies.
• Numbers are not delimited when they are part of mailing and shipping addresses, page numbers, and years with four or fewer digits. Years with five or more digits (e.g. 10,400 BC) shall use commas.
• In scientific articles, particularly those directed to an expert readership, numbers may be delimited with thin spaces using the {{gaps}} template. Coding {{gaps|8|274|527}} produces 8274527 (note: the thin space character and its HTML entity,  , do not render correctly on some browsers). Articles containing numbers with five or more digits to the left of the decimal point should delimit numbers with thin spaces.
• The style of delimiting numbers should be consistent throughout an article.
• Mathematical constants in math-oriented articles may be grouped into groups of five; e.g. 3.141592653589793238462643383279....
• Numbers with more than four digits to the right of the decimal point, particularly those in engineering and science where distinctions between different values are important, may be separated (delimited) into groups using the {{val}} template, with the BIPM parameter set to "y" or "yes"; {{val}} uses character-positioning techniques rather than distinct characters to form groups. Per ISO convention (observed by the NIST and the BIPM), it is customary to not leave a single digit at the end, thus the last group comprises two, three, or four digits. Accordingly, the recommended progression on Misplaced Pages is as follows: 1.123, 1.1234, 1.12345, 1.123456, 1.1234567, 1.12345678, 1.123456789, etc. Note that {{val}} handles these grouping details automatically; e.g. {{val|1.1234567}} generates 1.1234567 (with a four-digit group at the end). The {{val}} template can parse no more than a total of 15 significant digits in the significand. For significands longer than this, editors should delimit high-precision values using the {{gaps}} template; e.g. {{gaps|1.234|567|890|123|45}} → 1.2345678901234567. Also note that when the BIPM parameter is omitted, {{val}} does not conform to the ISO/NIST/BIPM format; instead it uses the customary American format of grouping digits to the left of the decimal with commas, and not grouping digits to the right of the decimal.

--Jc3s5h (talk) 03:01, 19 August 2009 (UTC)

• As I said above, there is well established reason why en.Misplaced Pages uses commas to delimit numbers to the left of the decimal point: it results in the least confusion in our readership. Please stop using {val} as an excuse to advocate that we diverge from this well-entrenched principle. If you don’t like {val}, don’t use it. It solves a legitimate problem when you have really long, hard-to-parse strings of numbers to the right of the decimal point. It is not a vehicle to be used to try to use Misplaced Pages to enlighten Americans to the One True Way that numbers really ought to be delimited. Greg L (talk) 03:12, 19 August 2009 (UTC)

• I reject the claim, unsupported by any scientific survey or by adoption by any external publication, that commas to the left of the decimal point and thin spaces to the right are less confusing than a format that uses the grouping method, thin spaces, on both sides of the decimal. --Jc3s5h (talk) 03:17, 19 August 2009 (UTC)

• Well, you clearly aren’t familiar with America, are you? How many methods of delimiting numbers do you think American school children are taught? Just one. In Sweden, they teach elementary kids three different techniques, including the American method. You are certainly free to choose to believe what you want to believe. Greg L (talk) 03:23, 19 August 2009 (UTC)

  P.S. I’ve been to England, Scotland, Wales, Ireland, Antigua, Moroco, Spain, Monaco, France, Italy, Austria, and Hungary—and Canada, if you can count that as not being part of the U.S. ;-). Notwithstanding that I am American, I do all my engineering in hard metric. Have you been to the U.S.A? If so, how long? Because, as an engineer, I can assure you I have keen insight into what causes confusion in Americans. The current practices of Misplaced Pages were no accident. Greg L (talk) 03:42, 19 August 2009 (UTC)

• I have found that the current edition of the U.S. Government Printing Office Style Manual, in rules 12.9e and 12.14, calls for the use of commas to the left of the decimal and thin spaces to the right. (Although the manual does not say "thin spaces", the spaces in their example look thin to me.) Therefore I withdraw my contention that this is a made-up format. This manual does not address the question of what to do when one number has 5 or more digits to the left of the decimal, and also 5 or more digits to the right of the decimal. In most cases this could be handled by putting the number in scientific notation. --Jc3s5h (talk) 04:31, 19 August 2009 (UTC)

• Jeez! I didn’t know that. That was big of you. You did the homework and declared an “inconvenient truth” that I didn’t even know about. Damned big of you; I’ll remember that. Like the guideline says, editors may use {val}. If you think it is important for readers to be able to easily parse the digits to the right of the decimal point in a particular value, and if scientific notation doesn’t seem suitable for some reason, then use {val}. Otherwise, don’t. Cheers. I’m done for the evening. Greg L (talk) 04:43, 19 August 2009 (UTC)

• I think that the proposed changes to {{val}} would be straightforward and reasonable.

  As for the proposal, it's on the right track—but let me raise some issues. I think the current text is a bit cumbersome, in that it has a lot of exceptions and places where the text diverges into subcases. I think we can state the exceptions a little more clearly, and make them general in application.

  There's also the question of whether to treat the U.S. and BIPM delimiting schemes as similarly-acceptable, or not. I don't buy the idea that Americans don't understand the BIPM format—any literate English speaker ought to be able to figure out the meaning of either the U.S. customary or BIPM formats (or for that matter the meaning of an undelimited number), with minimal effort. (And for subsequent occurances, zero effort.) It's not a significant issue of understanding at all—it is straightforwardly stylistic. And given that an English-speaking Canadian or European would probably default to the BIPM format, and would certainly understand it, it's not appropriate to direct those users to use an unfamiliar method despite the conventions of English-language style that are typically employed in their regions. (It's the same can of worms as telling them to spell it color instead of colour, or vice versa.)

  It also follows that there needs to be an exception for special regional contexts that would be unintelligible to the majority of English speakers, but which are still in widespread use. (This particular exception could use further discussion, and might be relevant in so few cases as to be omitted, but I'll suggest something below anyway.) For example: English is a second language of hundreds of millions of Indians, Pakistanis and Bangladeshis, and they often use the Indian numbering system when communicating in English. For articles specifically about an Indian topic, if we're to be faithful to the idea of avoiding regional bias, we have to acknowledge that their method is widespread enough to consider in the MOS.

  There's also the matter of splitting the technical details out into their own subsection. I think things were getting a bit convoluted with the mixing of implementation details with stylistic concerns, so I think that we should separate them from the main text.

  So with that in mind, I've spliced portions of the various versions of this section (by Greg L, Jc3s5h and myself) together, and come up with this:

===Digit grouping===

• In numbers with many digits, digit grouping symbols (inserted at intervals from the decimal point) are used to subdivide the number into easily readable groups. The acceptable digit grouping schemes are:
  • Commas every three digits to the left of the decimal point, and thin, non-breaking spaces every three digits to the right of the decimal point (e.g. 8,274,527 or 0.12345). This is traditional usage in many English-language contexts, and recommended by the U.S. Government Printing Office.
  • Thin, non-breaking spaces every three digits (e.g. 8274527 or 0.12345). This format is suggested in BIPM and NIST style guides for scientific and engineering works, and is in common use in interlanguage contexts.
  • Mathematical constants in math-oriented articles may be grouped into groups of five digits on both sides of the decimal point; e.g. 3.141592653589793238462643383279....
  • Other traditional digit grouping schemes, when relevant to the subject matter of the article; e.g. 82,74,527 in the Indian numbering system. (An explanation of the digit grouping scheme should be provided, typically by way of a link or brief summary. Also or instead, parenthetical conversions to another acceptable scheme may be used.)
• The style of grouping digits within numbers should be consistent throughout an article.
• When grouping digits by threes, and a number has exactly four digits on any side of the decimal separator, those digits may optionally be expressed as a group of four instead of three; e.g. both 9876 and 9,876 are acceptable.
• Digits are not grouped when they are part of addresses, page numbers, and years with four or fewer digits. However, years with five or more digits (e.g. 10,400 BC) are delimited as any other large number.

====Technical implementation====

The {{gaps}} template uses CSS to output thin spaces (using the syntax {{gaps|8|274|527}}). Using HTML entities for this purpose (e.g.   or  ) may cause rendering problems in some browsers, and should be avoided when practical.

The {{val}} template handles digit grouping automatically, in the American style (by default) or in the BIPM style (by using the BIPM=y parameter). In both cases, CSS is used to output the thin spaces. For example, {{val|1.1234567}} generates 1.1234567 (with a four-digit group at the end). The {{val}} template can parse no more than a total of 15 significant digits in the significand. For significands longer than this, editors should manually delimit values using the {{gaps}} template; e.g. {{gaps|1.234|567|890|123|45}} → 1.2345678901234567.

TheFeds 07:08, 19 August 2009 (UTC)

Good, except for a few points:

• To address Greg's concerns about the "BIPM" system, you might consider to add something like In some major English-speaking countries, this format is unfamiliar most persons without advanced scientific or engineering education, so avoid using it in general-interest topics.
• I'd replace "throughout an article" with "within a context": as I said above, the lead of LHC is a general-interest topic (as it made major appearances in the media last year), and so you'd want to use the "American" style; the sections about technical details are advanced scientific topics, and so you'd want to use the "BIPM" style. (Probably a better example than that could be found.) BTW, in the former type of contexts you'd seldom use numbers with more than three or four significant figures anyway (if you do, you're likely to be breaching the second bullet in "Large numbers"), so issues of long strings of digits won't occur; and laymen are unlikely to want to read stuff in the latter type of contexts.
• Don't see the point of making {{val}} use the "American" style by default. It is a template with facilities for scientific quantities and it is far more likely to be used for the mass or magnetic moment of the electron than for the population or GDP of the US: the latter type of numbers are normally hard-coded in articles without using templates. (BTW, look at the last example. I'm going to fix it in the actual MoS text where it's also found, but I don't generally edit other people's comments.) --___A. di M. 12:24, 19 August 2009 (UTC)

I hadn't read that the "American" style does include spaces after the point. Given that, the "long string of digits" issue is moot, so the grounds for replacing "throughout an article" with "within a context" are weaker than I believed: 99.9999991% conforms to the "American" style, too. --___A. di M. 15:07, 19 August 2009 (UTC)

(unindent) As one of the guys who take care of the {{val}} template, the sensible solution as far as the template is concerned is to leave the default behavior of {{val}} as-is, because otherwise you will screw up a million articles and remove the ease of using {{val}} in >90% of cases. Then for all those who hate gaps, or hate commas, two parameters can be added, |gaps=no (producing 1,232,345.0023234) and |commas=no (producing 1232345.0023234). You can then write exactly what you want, and you'll have the choice of picking whichever format is best suited for the article.Headbomb {_{κοντριβς} – WP Physics} 15:18, 19 August 2009 (UTC)

"Scientific articles; particularly ones directed to a professional readership, are the only exception."

• ... plus mathematical articles, articles about technology (computing, etc.), articles about measurement, etc.

I have to say that I've never really been a fan of the commas+gaps hybrid formatting that val uses. I'd rather have either gaps+gaps or commas+nogaps consistently throughout the article perhaps with a few exceptions based on context as mentioned above.

Rather than having the default behaviour of {{val}} produce 1,232,345.0023234, I'd have it produce 1232345.0023234 but that's my preference & would probably not be what many would consider appropriate in many cases.

However, I'd be happy with the suggestion by Headbomb. Between three and four thousand pages use the template, many maths/science/technology related, many not. It wouldn't be impossible to send a bot round adding |gaps=no and |commas=no where appropriate.

On the other hand, instead of a hybrid format that happens to turn out perfectly acceptable in 90 or so percent of cases (i.e. not too many numbers on WP are both greater than a thousand and have more than four digits after the decimal point), we might consider a hybrid default. For example, the template could produce gaps+gaps if there are more than four digits after a decimal point and commas+nogaps otherwise. That shouldn't screw up a million articles and it might even make the template a fraction easier to use. JIMp _talk·cont 20:49, 19 August 2009 (UTC)

• That’s what {gaps} is for. Nothing is broken. This “dump on {val}” thing was just a vehicle to promote the point of view of an editor who wants to use gaps left and right—perhaps even outside the strict field of science. Now I see “engineering” being proposed to add into the mix, which is most unwise and unnecessary. And Jimp just described it as “maths/science/technology”—almost like “smart-like articles.” We often get that on Misplaced Pages—encroachment because it makes editors happy because they can edit in a fashion they are accustomed to in their country; the practice seems truly “right” in some way to them. That’s fine with dialect/spelling. It is a can of worms with numbers for the reasons I’ve stated above and unnecessarily creates confusion in our readership. The exception of using thinspaces to the left of the decimal marker is currently limited to “science” and needs to stay that way for good reason. Greg L (talk) 20:58, 19 August 2009 (UTC)

That's what I use {{gaps}} for but I'd rather use {{val}}. I'm not trying to water things down to "smart-like articles". Perhaps gaps is not appropriate for technology articles, I'm sure its not appropriate for all of them, it may be appropriate for some. I'm just suggesting that the scope is broader than just science articles. Here's a maths article using spaces both sides. It looks fine to me. JIMp _talk·cont 21:18, 19 August 2009 (UTC)

• IMO, that math article ought to be conforming to the general rule of using commas to the left. Fortunately, it is an obscure enough of an article (only 700 hits per day). Was there ever a manual of style on Misplaced Pages saying that that math articles ought to to be delimiting numbers to the left of the decimal marker with gaps or did that article just happen that way? Just because there are math articles that have been done this way does not mean it is a wise idea. I can point to articles that have countless quantities of ending sentences with a preposition; that doesn’t mean we need to change WP:MOS. Greg L (talk) 21:30, 19 August 2009 (UTC)
  • Go talk to WP:MOSMATH. But in context of the article, which depends on the contrast of long strings left and right of the decimal point, it seems a reasonable deviation for clarity. Septentrionalis PMAnderson 23:42, 19 August 2009 (UTC)

Actually that article is itself inconsistent, using sometimes commas and sometimes spaces. I'm going to fix it to all-commas before and all-spaces after, as I think it was intended for a lay readership. (I don't think they'll ever appear in the same number or even section, given the structure of the article.) --___A. di M. 18:11, 20 August 2009 (UTC)

Please change it to all gaps, which seems to predominate; that works better to my eye. Septentrionalis PMAnderson 18:30, 20 August 2009 (UTC)

To me it seems that the powers of ten immediately below the section headers mostly have spaces, but the actual examples mostly have commas. --___A. di M. 18:52, 20 August 2009 (UTC)

It looks to me like that double rule is exactly true, except for the scientific notation and one glitch at 1 000, which I fixed. Let's leave it alone -unless there are other glitches. The gaps in the headers are what is useful. Septentrionalis PMAnderson 19:24, 20 August 2009 (UTC)

It used normal space characters, which are too wide (at least with my font and in my eye), and it had a couple numbers with spaces within the actual lists, which I fixed until the 10^39 section. But there's another issue: the last sections have gargantuan exact integers, currently delimited with normal spaces. Using commas or {{gaps}} would blow up the line width, forcing horizontal scrollbars on any window of any reasonable width. --___A. di M. 10:48, 21 August 2009 (UTC)

I'll go along with TheFeds proposal, once one technical issue is ironed out. This bullet:

• When grouping digits by threes, and a number has exactly four digits on any side of the decimal separator, those digits may optionally be expressed as a group of four instead of three; e.g. both 9876 and 9,876 are acceptable

is not compatible with the example 1.1234567 because the example has exactly seven digits to the right of the decimal, not four.

Also, the style implemented by {{Val}} should really be called the GPO (Government Printing Office) style, because the prevalent American style is to not group digits to the right of the decimal point. (The Chicago Manual of Style is one publicaton that advocates no separation to the right of the decimal outside the realm of science and technology). --Jc3s5h (talk) 22:36, 20 August 2009 (UTC)

• And the Chicago Manual of Style probably gives the same guidance as the Associated Press’ Manual of Style; both wouldn’t delimit to the right for non-technical articles directed to a general-interest readership, now, would they? The {{val}} template probably isn’t a good fit for an article on Wham‑O’s Super Ball: Super Balls have amazing rebound kids! Yet it has a specific gravity of only 1.25864(16) g/ml (which is how much a fifth of a teaspoon of a Super Ball weighs). High-precision values will most frequently be found in technical articles, won’t they? The only noticeable legitimate exception that comes to my mind at the moment is high-precision numbers intended to impress, such as French scientists in 1799 measured the density of water to within 99.997495% of the currently accepted value. Wow!

  The current guideline reads, in part, as follows:

The whole section seems clear enough to me—even for editors who are disinclined to bring much common sense to the party. It reflects what we ought to be doing in order to communicate without confusion, and, after A. di M. got through with it, seems to be capturing the all the known exceptions to the general rule. Greg L (talk) 03:57, 21 August 2009 (UTC)

• Regarding the groups of four issue, recall that stemmed from an unclear recommendation from the BIPM manual. They said groups of four were the exception, but didn't explicitly provide for a group of four followed or preceded by a group of three on the same side of the decimal separator. (But gave examples in-text that seemed to support that style.) I don't think it's a huge deal, so I'd be fine with a revised version of that sentence. Do you want to suggest the wording? It would be good to consider whether a case like 9876543 is permissible or too ugly to let stand—contrast that with 9876543 (which is grouped 1/3/3 instead of 4/3). TheFeds 03:34, 21 August 2009 (UTC)

• Are we having a “consensus of two”-thing going here again? This issue of number delimiting, as I wrote (abundantly) above, was thoroughly discussed by a very wide group of participants. Moreover, A. di M. and I got deep into the ISO and BIPM recommendations and sorted out—and documented—why things are being done that way. The current MOSNUM guidelines reflect proper practice and there is no good reason to diverge from the practice, notwithstanding your musings as to what sorts of alternative delimiting strikes you as “too ugly to let stand.” Greg L (talk) 03:59, 21 August 2009 (UTC)

  The problem with what BIPM say w.r.t. numbers with seven digit after the point is that it differs from what it does. They say that you should use 0.1234567, but their examples use 0.1234567. As for ISO, their standards aren't available for free; a 2008 draft of ISO 80000-1 I found somewhere on the Web explicitly forbids 0.1234567 in favour of either 0.1234567 or no delimiting at all, but judging at this I think it hasn't been accepted. So we'd better ditch the question of what they say and think about what is better. The IUPAP Red Book clearly says "Instead of a single final digit, the last four digits may be grouped." (But above it has unclear wording which doesn't seem to allow delimitation with four digits as in 1987 or 0.1234.) In actual usage, 0.1234567 seems to be more common than 0.1234567 (even because it's not clear how you'd use the latter with two digits uncertainty as in 0.1234567(89)). So I'd keep the convention as it is, but I wouldn't call it "ISO convention" unless someone has access to ISO 31-0 and can confirm that they say that, and is willing to buy ISO 80000-1 when it's officially published to check if it continues saying that. Let's just drop "According to ISO convention (observed by the NIST and the BIPM)"; most people won't even actually care. Also, is "to within 99.997495%" supposed to mean "to within 0.002505%"? BTW, you could impress the reader by writing "99.9975%" or "0.0025%" as well, and neither would have more than five digits after the point. --___A. di M. 10:20, 21 August 2009 (UTC)

  BTW, I'd support this:

A. di M.'s proposal

===Digit grouping===

• In numbers with many digits, digit grouping symbols (inserted at intervals from the decimal point) are used to subdivide the number into easily readable groups. The acceptable digit grouping schemes are:
  • Commas every three digits to the left ofbefore the decimal point, and thin, non-breaking spaces every three digitsno delimiting to the right ofafter the decimal point (e.g. 8,274,527 or 0.123450.12345). This is traditional usage in many English-language contexts, and recommended by the U.S. Government Printing OfficeThe Chicago Manual of Style and the AP Stylebook for non-technical articles.
  • As above, but with thin, non-breaking spaces every three digits after the point (0.12345). This is recommended by by the United States Government Printing Office.
    • Usually there is no difference between the two styles above, because non-technical articles should avoid using excessively precise values: see the second point in {{Section link}}: required section parameter(s) missing, below.
  • Thin, non-breaking spaces every three digits (e.g. 8274527 or 0.12345). This format is suggested in BIPM and NIST style guides for scientific and engineering works, and is in common use inand is in common use in scientific and engineering works and interlanguage contexts.
    • In some major English-speaking countries, this format is unfamiliar most persons without advanced scientific or engineering education, so avoid using it when discussing general-interest topics.
  • Mathematical constants in mathematics-oriented articles may be grouped into groups of five digits on both sides of the decimal point; e.g. 3.141592653589793238462643383279....
  • Other traditional digit grouping schemes, when relevant to the subject matter of the article; e.g. 82,74,527 in the Indian numbering system. (An explanation of the digit grouping scheme should be provided, typically by way of a link or brief summary. Also or instead, parenthetical conversions to another acceptable scheme may be used.)
• The style of grouping digits within numbers should be consistent throughout an article.
• When grouping digits by threes, and a number has exactly four digits on anyeither side of the decimal separator, those digits may optionally be expressed as a group of four instead of three; e.g. both 9876 and 9,876 are acceptable. Additionally, after the decimal point a final group of four digits can be grouped together, e.g. 0.1234567 or 0.1234567. The latter style is more common, and is recommended on Misplaced Pages.
• Digits are not grouped when they are part of addresses, page numbers, and years with four or fewer digits. However, years with five or more digits (e.g. 10,400 BC10,000 BC) are delimited as any other large number.

====Technical implementation====

The {{gaps}} template uses CSS to output thin spaces (using the syntax {{gaps|8|274|527}}). Using the thin space character or its HTML entities for this purpose (e.g.   or  ) may cause rendering problems in some browsers, and should be avoided when practical.

The {{val}} template handles digit grouping automatically, in the American style (by default) or in the BIPM style (by using the BIPM=y parameter)in the U.S. GPO style automatically. In both cases, CSS is used to output the thin spaces. For example, {{val|1.1234567}} generates 1.1234567 (with a four-digit group at the end). The {{val}} template can parse no more than a total of 15 significant digits in the significand. For significands longer than this, editors should manually delimit values using the {{gaps}} template; e.g. {{gaps|1.234|567|890|123|456}} → 1.234567890123456.

Differences from TheFeds' version are marked. --___A. di M. 11:15, 21 August 2009 (UTC)

"This format is suggested in BIPM and NIST style guides for scientific and engineering works" misrepresents the position expressed in the BIPM brochure (and the NIST brochure is, with minor variations, just a translation). The brochure states on page 133

The practice of grouping digits in this way is a matter of choice; it is not always followed in certain specialized applications such as engineering drawings, financial statements, and scripts to be read by a computer.

So as far as NIST and BIPM are concerned, the thin space format should be used for everything except "certain specialized applications". --Jc3s5h (talk) 12:16, 21 August 2009 (UTC)

Amended to reflect this. --___A. di M. 12:24, 21 August 2009 (UTC)

• A. di M., thank you for stepping in here. Please insert, somewhere in your suggestion, something along the lines of this: “In technical articles such as engineering and science, where distinctions in the magnitude of numeric expressions are important, editors should consider using the {{Val}} template to delimit numbers that have many digits to the right of the decimal point in order to make them easier to parse.” Greg L (talk) 19:21, 21 August 2009 (UTC)

• I strenuously object to the clear objective of the proposed verbiage that reads

Thin, non-breaking spaces every three digits (e.g. 8274527 or 0.12345). This format is suggested in BIPM and NIST style guides and is in common use in scientific and engineering works and interlanguage contexts.

The current wording, (“scientific articles, particularly those directed to an expert audience”) is the proper limit. The clear objective with that proposed wording is to expand the “Europeanizing” of Misplaced Pages’s numbering into anything remotely ‘technical’ in nature. Unlike dialect (spelling), delimiting with gaps to the left of the decimal point is far too confusing to American readers who are not taught any other methods of number delimiting. Conversely , the typical Swedish school child is taught three different ways to format numbers and are not in the least bit confused when they come across numbers with the American-style of delimiting.

I’ve made this point numerous times, above, and the best counter-argument I’ve yet to see is a demand to see scientific proof or studies showing that alternative numbering formats would be confusing to Americans. Commons sense does not need to be legislated. This is hopeless. This is simply nothing more than an attempt by some Europeans, who are absolutely convinced that their BIPM-endorsed, European way is superior (perhaps it is) and should be adopted here on Misplaced Pages and dumb ol’ Americans will learn it here, if nowhere else. These editors need to drop this agenda to change what has long worked here. Misplaced Pages is not to be exploited as a vehicle to promote change by educating Americans to conventions with which they are entirely unfamiliar. The current guidelines are fine.

And, a final note: Just because something is endorsed by the BIPM doesn’t mean it enjoys world-wide adoption. The BIPM wants 75% written 75 %, but Americans (and as far as I know, no one else) don’t do it that way so Misplaced Pages goes with the flow and ignores the BIPM so we don’t confuse people for God’s sake. Greg L (talk) 23:35, 21 August 2009 (UTC)

I have to say, having read this thread I am not at all persuaded by the BIPM-rules-all feeling that's about. What is being proposed here seems controversial and is unlikely to gain broad consensus. I see no reason to change the current guidelines, which have long been on MOSNUM and have served us well. Tony (talk) 00:12, 22 August 2009 (UTC)

As someone who has lived in France for many years, I am familiar with that convention. It appeared bizarre to me when I first arrived there. Now it is pointed out that this is also adopted in some scientific contexts. Before this is imposed on the unwitting WP public, I would point out that it is a convention I have never seen before in the Anglo-Saxon world. Being able to understand it is one thing, but to insist that this professional technical standard be adopted in any form in a general knowledge encyclopaedia is quite another. - BTW, the French also use the comma in place of the period as the decimal delimiter. Even if we were not to adopt this last 'quirk', it is unlikely to find widespread acceptance. Implementation is likely to meet with bewilderment, and be universally reverted to the comma delimiter which all English-speakers are familiar with. Ohconfucius (talk) 06:37, 22 August 2009 (UTC)

• Regarding Greg's oft-repeated assertion that Americans would be confused by the BIPM style, let me also reiterate my feelings. I don't believe that someone with a working comprehension of written English would be so hopelessly confused by that format as to fail to comprehend the meaning after a short pause to think about it. To me, the worst reasonable case is that someone will assume that a previous editor neglected the commas.

  To consider an analgous case, an American reading about a boot and a bonnet might picture articles of clothing, rather than car parts—yet we would not expect articles to generally defer to the American's preferred vocabulary, even though Britons are somewhat accustomed to American vocabulary (e.g. because of the pervasiveness of American culture and media in Britain, vs. the opposite), and would grudgingly understand "trunk" or "hood" in the context of an automobile. In fact, with numbers, at least there's a contextual clue to the meaning—the numerals are the same, after all. With "boot" vs. "trunk", we have to resort to things like linking the first occurrance, to avoid misunderstandings; the case of dialect is more, not less confusing than numbering style. Yet despite that slight inconvenience, the well-established principle is to accept such regional eccentricities as part of the language, and accomodate them in the interest of avoiding the imposition of one region's style over another. Just because the Swedes often understand either format—to recall Greg's example—does not mean that their preference ought to be be moot.

  And let me just point out that I was not the one who demanded a study or scientific proof of the alleged incapacity of Americans to use other formats—I simply expressed my doubt. My common sense appraisal doesn't agree with Greg's, and I suspect that when it's nothing but assertions of common sense on either side, there's nothing in particular to refute. (And of course, repeating something doesn't make it true; neither argument is any stronger upon retelling.)

  I've also got to take issue with the idea that Europeanization is the motive. ("This is simply nothing more than an attempt by some Europeans, who are absolutely convinced that their BIPM-endorsed, European way is superior (perhaps it is) and should be adopted here on Misplaced Pages and dumb ol’ Americans will learn it here, if nowhere else.") I doubt that it was an intentional strawman, but for the sake of keeping this discussion on track, let's drop the idea that there's any effort to force-feed Americans with foreign concepts, because there's no evidence that any editors presently commenting on this matter are attempting to engage in any such scheme.

  To me, the convention used by the BIPM is a necessary inclusion in the MOS because it is the dominant English-lanugage usage in some regions—not just in science or engineering, but in any English-language usage. That's the way things tend to work in Continental Europe (and maybe that's why this "Europeanization" allegation arose), where several states and languages have their own incompatible digit grouping schemes, and where English is the de facto language of international commerce and collaboration. There is no ulterior motive: it's simply an acknowledgment of European English style. Consider the European Commission English Style Guide, which calls for BIPM-style digit grouping. That guide is "intended primarily for English-language authors and translators" and "serve a wider readership as well" (speaking in terms of the EC and of EU institutions)

  The fact that things like scientific journals and engineering handbooks have also adopted the BIPM convention is a secondary component of the issue—when we say "in common use in scientific and engineering works", that's just an acknowledgment that scientific usage of this convention is worldwide, versus the regional adoption of the convention for general use. If we wanted to capture that reasoning in detail, we could say that BIPM style is limited to topics in the fields of science and engineering, plus topics dealing with Continental Europe and other regions that enjoy significant adoption of that style.

  But what's the point of being needlessly precise? It's much simpler to acknowledge that both the U.S. and the BIPM convention are in general use in large regions of the world, and each preferred by major fields of study (as evidenced by style manuals). On balance, the MOS is clearer by just allowing either one (though calling for consistency within an article), without complicating the issue with topic-related stipulations (and the consequent discussion of what belongs to which topic, or what dominates in which region). This is easier to understand, and easier to enforce—that's a useful thing, given the level of complexity of the MOS and Misplaced Pages policy in general. TheFeds 19:46, 22 August 2009 (UTC)

• Quoting you, TheFeds: I don't believe that someone with a working comprehension of written English would be so hopelessly confused by that format as to fail to comprehend the meaning after a short pause to think about it. You might well be right. After a moment of looking at the unfamiliar-looking number, many—perhaps most—Americans would eventually figure it out. So what? Do you think that is the litmus test the Associated Press uses in judging what to put into their manual of style(?): whether readers, who are initially confused by the written word, will for the most part eventually figure it out??? The point of all technical writing is to write with the least confusion. Your arguments of how the BIPM recommended this ‘n’ that so their recommended method of delimiting numbers has been anointed with holy oil and ought to be adopted here is entirely beside the point.

  Frankly, your arguments that we should follow whatever the BIPM says the world ought to do comes right out of the four-year-old playbook used by proponents of the IEC prefixes (see the B0–B16 archives, above). That whole stink (seventeen entire archives dedicated exclusively to that one fiasco) was because a small handful of editors banded together and decided to start using terms like “256 kibibyte (KiB)” instead of the “256 kilobyte (KB)” used by computer manufacturers and computer magazines throughout the rest of the planet. Then one of those editors ran around and changed hundreds of articles over a period of a few weeks and the group blocked anyone from reverting all those edits, citing “lack of consensus”. That this could have occurred that way, that such stupidity lasted for three years, and that it took three months of bickering to put an end to it speaks volumes to how broken MOSNUM can be at times. And all because it was a proposal by some vaunted standards organization. The end result(?): needles confusion in readers who had the misfortune for three long years to come here and read many of our computer articles. I’m sure there were a number of people read up on computers here on Misplaced Pages and then went into computer stores where they announced they were looking for a computer with “512 mebibytes of RAM.” They were no-doubt met with blank looks (at best) or snickers.

  Quoting you again: To me, the convention used by the BIPM is a necessary inclusion in the MOS because it is the dominant English-lanugage usage in some regions. Yeah, I got that much. I figured that bit out in the first four seconds after seeing what you were trying to do. And, as I’ve explained, it unfortunately isn’t a two-way street. There are many ways the Europeans format numbers. In Sweden, the “Swedish1” technique (there’s yet another) is to write the population of America as “285.865.855” so why don’t we just go ahead and let articles here use that system too(?), right along with the BIPM method (which Swedish school children are also taught)? That’s a rhetorical question, please don’t answer it. The answer is because Swedish school children are also taught to recognize the American-style of delimiting numbers. In fact, Europeans by and large are not in the least bit confused when they encounter numbers with American-style delimiting. The trouble is that American’s know of only one way; they’ve been taught no other. Your argument that Americans’ confusion will be short-lived because *Misplaced Pages* will just teach them using an “oh… didn’t-cha know?”-fashion (and only the galactically clueless and retarded will be left in the dust) is wholly uncompelling.

  With regard to how numbers are written here on Misplaced Pages, it doesn’t matter in the least what the proposal is or who proposed it. When the American style of delimiting numbers is no longer the dominant numbering format universally recognized throughout the English-speaking world, and when Americans have as much familiarity with alternative numbering styles (like “Swedish1” and “BIPM”) as Europeans do with the American style, then Misplaced Pages can change over. Misplaced Pages follows the way the world works and never, ever tries to promote change in the way the world works by presuming we can somehow lead by example. Misplaced Pages wisely decided to use American-style delimiting in our numbers so there is minimal confusion. Greg L (talk) 21:50, 22 August 2009 (UTC)

• With regard to your first point, we're writing with a different set of rules than the AP (to use your example). They are free to insist that their writers use "truck" instead of "lorry", or group digits with commas rather than thin spaces, because they have no mandate to consider the issues summarized in WP:ENGVAR, and choose instead to cater to one region's preferences. Misplaced Pages is written for an international, English-reading audience. That means that we need to accept that occasionally, a user may encounter a term or convention from elsewhere, and be briefly confused. But this is not a critical flaw, because most readers don't just shut down at that point; they keep reading the sentence or paragraph. And the context generally makes clear what is meant, without loss of information. I submit that if we presented a number all alone in the BIPM format, there might be unacceptable confusion among Americans, but that if it was placed within prose or a table, the context would make the meaning accessible to all.

  Let me stress that there is no implication of "teach them using an “oh… didn’t-cha know?”-fashion". Readers are free to assume that the formatting was a stylistic error, rather than another valid convention, and be none the wiser. The content of the article still gets across. (In other words, I am absolutely not advocating some sort of patronizing policy that would educate the savages of the world about Europe's genteel ways.)

  My concern here that by mandating a convention that represents the preference of a subset of English speakers, we would imply that other accepted English-language usage is inappropriate. It would be like saying "because the British spelling of words like 'harbour' is recognized throughout the world"—despite the American spelling "harbor"—"this spelling convention is mandatory (unless quoting a source or explaining foreign usage)". Misplaced Pages has not gone this route, instead permitting either form.

  Regarding your Swedish example, I assume that Swedish-1 and Swedish-2 are used in Swedish-language works? This being the English Misplaced Pages, one would expect them to use the style that is typically used in their region for writing in English. (Which is most probably the BIPM style.) If the Swedes don't use their native styles when writing in English, then there's no need to deal with them here.

  When you say with certainty that "Misplaced Pages follows the way the world works and never, ever tries to promote change" by example, I think you're misinterpreting what's going on here. Maybe during the binary prefix debate, some editors advanced the idea of promoting IEC usage to the world. Nobody's said anything of the sort here. Furthermore, we usually follow bits and pieces of what influential parts of the world do, but only because consensus among editors often ends up settling upon that course of action. Sometimes that means following or ignoring a particular standard or widespread convention, but the key is that Misplaced Pages does what the editors agree upon. It's coincidence rather than design that consensus usually settles upon permitting what's popular in the world.

  Lastly: you're definitely mischaracterizing my motives. I'm not waging some promotional campaign that serves to advance the interests of the BIPM, or Europe, or any other entity. Specifically, despite your assertion above, I have not stated or implied that "we should follow whatever the BIPM says the world ought to do". (I have previously stated that I tend to use the BIPM style in my outside work—which is immaterial to your allegation.) TheFeds 00:18, 23 August 2009 (UTC)

• Quoting you: I am absolutely not advocating some sort of patronizing policy that would educate the savages of the world about Europe's genteel ways.) {{smiling, very amused emoticon}}. Perhaps you aren’t. But it seems you are mightily crossing your fingers that mixing things up here and allowing editors to use different types of numbering formatting if they feel like it won’t cause needless confusion with our readership. Better cross two fingers.

  I’m an American engineer. I do all my design in hard metric. In fact, only a half hour ago, I was busy calibrating a celsius-only thermostat that I’m going to retrofit into my otherwise-gaugeless, Italian-made Rancilio espresso maker. The boiling point of distilled water was 97.8 °C at that moment (barometric pressure) at my altitude (about three meters above the sewer cover outside my house). I’m also installing a pressure gauge calibrated only in bar. I am as “BIPM” as they come for an American. However, as an engineer who has done far more than his share of technical writing (owners manuals, white papers, etc.), I am fully aware of what Americans know and don’t know technically. I am also keenly aware of how utterly stupid it is to needlessly confuse readers.

  Here’s how to speed America’s adoption of all-things-BIPM (e.g., their SI (metric) system, their numbering system, etcetera): Just run for elected office while promising that if America adopts BIPM into their way of life, they’ll loose weight while they sleep and their taxes will go down due to less government waste. You’ll get into office. Moreover, you’d stay there; all you’d have to do is repeat the promise each election cycle; they’ll buy into your promise each time.

  America is big and homogenous; a drive that would take you clean across the whole of Luxemburg won’t even get you across Harris County, Texas. This is a source of strength and weakness. For one thing, we can drive across Harris County and not require a pocket-full of plug adapters to plug our iPhones into the outlets in the next county (or state for that matter). We could design a big-ass airplane in the late 60s and not get all confused because Californians can’t speak French whereas Oregonians right next door can’t speak German. But this homogeneity also means that Americans would be surprised to get into an elevator on the ground floor and see that they are at floor “0”.

  Indeed, Americans are certainly not as “genteel” and worldly as Europeans (that’s a friend of the guy who owns a tool & die shop I frequent); there’s this big-ass pond that shelters us from being exposed to other languages and other ways of doing things. Therefore, it’s a damned-seldom American indeed who is fluent in two or more languages and easily recognizes two—even three—ways that numbers can be delimited. Europeans “get” “285,564,125” in a nanosecond. The typical American would be needlessly confused by “285564125”. So, you Europeans have a leg up on Americans when it comes to understanding other cultures and practices.

  Please take note of the following three expressions. The first is “A civilized man can convincingly imitate a barbarian, but a barbarian can not convincingly imitate a civilized man.” The second is “Never try to teach a pig to sing; you only waste your time and annoy the pig.” The third is “The goal of all technical writing is communicate with minimal confusion.” What you are proposing handily ignores all three. Greg L (talk) 03:52, 23 August 2009 (UTC)

Arbitration-related comments

For everyone's information, I have initiated a discussion here about behavior I have witnessed since editing restrictions were lifted for parties to the delinking case. --Andy Walsh (talk) 19:08, 18 August 2009 (UTC)

Thanks for that, Andy. It seems there is no prospect of reasoning with that editor, who persists in slandering us and hindering our valuable work to maintain standards in Misplaced Pages's articles. Let's see what remedies there may be.

–_⊥Noetica!– 22:19, 20 August 2009 (UTC)

A link to Andy's other notification concerning this issue: Arbitration/Requests/Enforcement#Pmanderson.

–_⊥Noetica!– 12:08, 21 August 2009 (UTC)

Centuries

I'm not sure why but there is a pull towards using numerals in the MoS, maybe because of the preponderance of "computer types". The MoS used to recommend words for centuries, then it was changed, when I noticed I had, by agreement, the option for words put back in:- then it was taken out again...

AS far as it goes Oxford recommends words as does Chicago "In nontechnical contexts, the following are spelled out: whole numbers from one through one hundred, round numbers, and any number beginning a sentence. For other numbers, numerals are used." (also here Strunk and White 1918 has "It was chiefly in the eighteenth century that a very different conception of history grew up."

To me, numerals looks like note-taking in this context, I expect to see a hastily scrawled C around the number.

However if the three main (?) MoS concur - and the other reasons for abbreviation, saving time, keystrokes and paper are absent - surely we should go with spelled out centuries?

Rich Farmbrough, 14:52, 20 August 2009 (UTC).

I can't see why not ... or at least as an option. JIMp _talk·cont 19:54, 20 August 2009 (UTC)

I'd agree with that...—MDCollins (talk) 22:03, 20 August 2009 (UTC)

This issue has a long history. It is not a question simply of what people do "out there", though that is one important consideration. Some points in response:

• Rich, what Oxford source are you referring to? The Oxford guide that is used for serious publishing is New Hart's Rules. Here is how the current edition begins its treatment of centuries:

11.6.2. Centuries
Depending on the editorial style of the work, refer to centuries in words or figures; Oxford style is to use words. (p. 191)

That guide (one of the world's most influential and respected) does not say "use either"; it suggests that there be "an editorial style of the work". Our work is Misplaced Pages.

• Please: let's cite the relevant provisions of guides, and give accurate information on the source. Here is what the Chicago Manual of Style has to say:

9.36 Centuries
Particular centuries are spelled out and lowercased.

• Other sources with other rulings could be cited; but all agree: set a style, and use it consistently.
• Rich, you say " used to recommend words for centuries, then it was changed". How long ago did MOS make such a recommendation? The practice that predominates – especially in systematic treatments in Misplaced Pages's historical projects and the relevant categories for centuries, decades, and years – is to use figures. Personally, I like the Oxford style well enough, and I use words for centuries elsewhere. But on Misplaced Pages we need consistency, and we need stable guidelines. So I happily adapt. So do most other editors, who are grateful not to have to weigh up such a relatively trivial matter each time, at each article they tackle.
• Apart from long practice and the need for stability, it is objectively easier to maintain articles using figures for centuries. They are easier to type in, being free from the spelling errors or typos that beset eighteenth, and the punctuational uncertainties that editors have with in the twentyfirst century.
• Yes, there is a modern popular bias toward the simpler, easier way. It is well-founded and inevitable. Working against it would foster instability, needless confusion and contention at articles, and (wait for it) the inevitable harangues from the enemies of MOS, here at WT:MOSNUM and also at WT:MOS.

I move that we leave things alone. It works well. Don't fix it.

–_⊥Noetica!– 22:55, 20 August 2009 (UTC)

• Misplaced Pages is not a crystal ball; we are not required or encouraged to predict the wave of the future - and especially if this is what is happening naturally, why do we need a sentence on it? If Noetica is correct, silence will also have 19th-century proliferate; why do we need to discourage nineteenth-century further? Septentrionalis PMAnderson 23:16, 20 August 2009 (UTC)
• It seems unconvincing that 19th should be easier to maintain that nineteenth; a typo in 19th could produce 17th, whereas the worst a single typo in nineteenth can do is sineteenth; even a spell-checker will recognize that as error, and anybody can figure out what has gone wrong. Septentrionalis PMAnderson 23:22, 20 August 2009 (UTC)

In the context of WP, I prefer the numerical form as we are trying to convey information concisely; however if I'm reading (for example) a novel, I prefer the word form. As background to current usage, I got the following results yesterday:

• Google search of "nineteenth-century" results in 13,500,000 matches,
• Google search of "19th-century" returns 73,400,000 matches,
• WP search of "nineteenth-century" returns 3,911 matches,
• WP search of "19th-century" returns 10,204.

 HWV258  22:59, 20 August 2009 (UTC)

If you prefer it, use it. Let others do likewise. May the best usage win.

But why not have silence in this guideline, and permit evolution to take its course? Septentrionalis PMAnderson 23:16, 20 August 2009 (UTC)

PMAnderson, yet again you are not addressing the topic of the section. The question is not the one you return to without fail every time a specific move is contemplated. The question is not "Why can't we just allow everything?", or its variant "Why can't every minority practice be accepted equally with the majority practice?", or your perennial "Why have guidelines that actually guide, at all?"

Pay attention for once. The topic is "Should we introduce latitude in how centuries are referred to, rather than maintain the existing guideline?" I have addressed the topic with precision; so have others. Join in, or go somewhere else and do something else. Find a forum for your abiding, single-minded concern. This is not your anti-MOS forum; this is MOS.

–_⊥Noetica!– 23:33, 20 August 2009 (UTC)

Yes, we should. The "existing guideline" has only existed for a few days; it does not benefit the encyclopedia, it lengthens this page (which makes it less forcible and effective), it is not agreed (even that 19th is preferable, much less than it is the only thing preferable), and it is not in accord with other manuals. Septentrionalis PMAnderson 23:42, 20 August 2009 (UTC)

PMAnderson, as I point out we are perfectly in accord with other manuals, which unanimously call for (or assume as obvious) the use of one chosen style consistently. As for the present guideline being new, I believe it is merely a restoration of the simple guideline we had until you, without discussion let alone consensus, slipped in an unprecedented invitation to inconsistency some time ago. I opened a discussion on that addition (see above), and you refused my repeated requests to provide just one precedent from any style guide to support it. To support, that is, a guideline that would acquiesce in an article having both nineteenth-century painting and photography in the 19th century. Rather than focus on the issue, you made certain that several of us wasted hours of our time countering your destructive agenda and your recalcitrance. You've started on that tack again here, by false representations and soapboxing. Stop it.

–_⊥Noetica!– 23:59, 20 August 2009 (UTC)

Thank you PMAnderson for once again missing the point of the discussion. This section is not to do with how the MOS operates at WP; instead, it is to do with the wording of one particular section on the MOS. If I were to be tempted to get sidetracked into your crusade, I would feel inclined to point out that the Chicago Manual of Style (section 9.36) defines this issue as "Particular centuries are spelled out and lowercased"; and if I wanted to labour the point, I would observe that there is no option (numerical or otherwise) permitted at that MOS. If I really wanted to labour the point, I would hammer the conclusion that on some issues it is fine for a MOS to be "dictatorial" as to style (obviously based on the target audience of the publication). As I'm not inclined to get sidetracked, I'll simply respond that all I have done (above) is inject an opinion and some facts into the discussion. At that point I feel I've done my duty here, and whatever consensus arises will be just fine by me. I will point out that I don't intend to belittle or badger anybody's else's opinion on this issue, and would expect a similar courtesy.  HWV258  00:00, 21 August 2009 (UTC)

And if we were to require one and forbid another, I would prefer that we forbid 19th and recommend nineteenth; that's my second choice. Can we compromise on that? If we can't have what would be useful for the encyclopedia, let's have what Noetica and I both prefer to use. Septentrionalis PMAnderson 00:48, 21 August 2009 (UTC)

I'm still terribly worried that you have a weak grasp as to how things work on WP. It's not up to what I, or you, or Noetica want; instead, it is a consensus based on the input of everyone who comments on the issue. Let's wait for a while longer and see what other input is created in this section. You have now stated your preference, so leave it at that while other people state preferences (hopefully with reasoning). (To be frank, I'm not sure how "compromise" works with someone who peppers their input with pejorative phrases such as "If we can't have what would be useful..." and "..and forbid another".)  HWV258  01:22, 21 August 2009 (UTC)

"compromise"... "You keep using that word. I do not think it means what you think it means." Studerby (talk) 01:37, 21 August 2009 (UTC)

My second choice; his second choice. Sounds like a compromise to me. Septentrionalis PMAnderson 01:47, 21 August 2009 (UTC)

Hell no! It may be OK for centuries from the 20th century BC to the 20th century AD, but "twenty-first-century music" looks real gawky! If we need to have consistency throughout WP at all, let it be the "19th century". --___A. di M. 09:46, 21 August 2009 (UTC)

A valid point. Does the proper twenty-first–century (with an emdashendash) look any better? Septentrionalis PMAnderson 14:58, 21 August 2009 (UTC)

If you ask my own preference in writing, independently of Misplaced Pages, I would say: for centuries in "isolation", I have a moderate preference in favour of letters for "small" ones (i.e. up to "tenth century", mostly monosyllabic); a very slight preference in favour of figures for "intermediate" ones (i.e. from "11th century" to "20th century", mostly polysyllabic but single-word), and a strong preference for figures in "large" ones (from "21th century" onwards, mostly containing hyphens or spaces). But I would be consistent within any given context; and would prefer figures in places such as notes, tables, etc.

If you ask my own preference about what the MoS should say, independently of other editors, I would say: remove the sentence about centuries altogether; the general rule would apply.

If you ask me what the MoS should say, in light of the opinions of other editors, I would say: "Centuries should be written in figures". That suggests what the consensus here seems to be, but without looking like a descriptive statement about the English language, which would be false. (How long will it take before an editor will add after "twentieth century" in a quotation?)

___A. di M. 18:54, 21 August 2009 (UTC)

• I concur with Noetica's reasoning above for recommending figures in centuries. The figures are easier to maintain and far less prone to error. "Twenty-first-century music" and "twenty-first–century music" are unthinkable. Additionally, I'm interested in the language of guides that state "in nontechnical contexts". Really, WP is a technical text. We're not writing persuasive essays and novels here, where Pablo sweeps his wife over the threshold of his eighteenth-century villa. This is a reference work, quite technical in nature. --Andy Walsh (talk) 16:59, 21 August 2009 (UTC)

• Easier to maintain how?
• But this is characteristic of one of the problems of this page; different editors focus on different articles: We may well say that someplace is an "eighteenth-century villa" or "building"; indeed, we do. Septentrionalis PMAnderson 20:37, 21 August 2009 (UTC)

There is no need to hyphenate twenty-first to century; except in pathological cases, "Mike Gatting's twenty-first-century scores were not as good as earlier." perhaps. But to argue WP is technical is to have a broad definition of the term. Indeed I would aver that the only narrowly technical use of "19th century" might be calibrating a time machine. The Oxford I refer to is The Oxford Style Manual which incorporates The Oxford Guide to Style. Proof by Google is not helpful either technically ( try for example "nineteenth century" site:.edu ) or stylistically ("fuck" 158 million, "copulate" 380,000 - and that's my excuse for typing "fuck" into search engines). As far as typos go, it is amusing, if not instructive to see "21th century" typed in this discussion, not demolishing the argument, but certainly denting it. The typical use of a century is in a similar manner to a reign, "Jacobean, Elizabethan, nineteenth century" - I have even seen statements where the term has been redefined to fit the needs "when we say eighteenth century in this context we mean from about 1692 to 1780". It is not a measure in the sense that we can say (or we could but we would be ill advised) "the 21.12th century". The numbers are effectively cleansed of their numerical meaning. Rich Farmbrough, 21:05, 21 August 2009 (UTC).

And normal American, except for the minority who have read too much Fowler, is probably twenty-first century music, even if it is illogical. Septentrionalis PMAnderson 21:12, 21 August 2009 (UTC)

YYYY-MM-DD = <3

May I add encouragement for the use of the YYYY-MM-DD date format in articles relating to, at least, computer programming / languages? We like it. --Darxus (talk) 16:04, 20 August 2009 (UTC)

What is the reason for this? Maybe programmers "like" it, but human-readable formats (DMY or MDY) are probably preferred by most of the general readership. There is no reason to make date formatting rules more complicated than they already are. Dabomb87 (talk) 18:58, 20 August 2009 (UTC)

Programmers "like" it because it's a sortable date format. I fail to see what that has to do with writing an article though. --Malleus Fatuorum 19:13, 20 August 2009 (UTC)

But we can use {{dts}} for sortable dates without ISO format; so that's a moot point. Dabomb87 (talk) 19:16, 20 August 2009 (UTC)

I meant programmers when they're writing programs, or designing database tables, nothing to do with wikimarkup. --Malleus Fatuorum 19:19, 20 August 2009 (UTC)

OK. Pardon my ignorance. Dabomb87 (talk) 19:31, 20 August 2009 (UTC)

We also like it because it's unambiguous. DMY and MDY can all too often be unclear. While 01-01-09 doesn't matter, 01-02-09 can be a problem, especially if the century might be significant as well. YYYY-MM-DD is the only format I use everywhere I use a date if I have a choice in the matter. And it's sortable, which is icing on the cake. - Denimadept (talk) 19:45, 20 August 2009 (UTC)

When I referred to DMY and MDY, I meant January 1, 2001 and 1 January 2001, neither of which is ambiguous. Dabomb87 (talk) 19:53, 20 August 2009 (UTC)

We're writing articles not programs. ISO might be liked by programmers but dmy or mdy is understood and felt appropriate by all. Nor are dmy nor mdy ambiguous if the month is spelt out (or abbreviated to three letters). JIMp _talk·cont 19:51, 20 August 2009 (UTC)

If you spell out the month, it's not numeric, which is a whole different format. We seem to just be talking about numeric formats at the moment. - Denimadept (talk) 19:54, 20 August 2009 (UTC)

Yes, but Misplaced Pages doesn't use DD-MM-YYYY and MM-DD-YYYY formats, anyway. Dabomb87 (talk) 19:58, 20 August 2009 (UTC)

At times, I've seen various forms. Remember December 1, 2009 gets reformated, to ... what? I realize that date linking is depreciated, but my point is the same. - Denimadept (talk) 20:02, 20 August 2009 (UTC)

If we're talking about numeric formats (are we?), then ISO is the only one that should ever be seen anywhere on WP. However, if we're talking about using ISO format as opposed to non-numeric dmy or mdy in articles, I'm opposed (though some exceptions might exist). JIMp _talk·cont 20:06, 20 August 2009 (UTC)

I like the YYYY-MM-DD format in a controlled environment. However, the creators of ISO 8601 have adopted and/or hijacked this fomat, and overlaid it with many restrictions (e.g. always use Gregorian calendar, get reader's permission unless 1583 <= year <= 9999). I don't know how popular the format was before the ISO came up with their standard, but I don't consider Misplaced Pages, the encylopedia anyone can edit, to be a sufficiently controlled environment to use this format. --Jc3s5h (talk) 20:26, 20 August 2009 (UTC)

While the computer programming/language articles may at times be heavy going for the non-technical initiate, that's who they should be written for. Material written exclusively for the technorati (of which I'm a middling high priest), should not be in Misplaced Pages. Thus, a general format for dates should be favored, except perhaps in technical examples. Studerby (talk) 20:42, 20 August 2009 (UTC)

So which format do we use? Should we only use dd M yyyy/M dd, yyyy? Might want to fix the date reformatter in that case. - Denimadept (talk) 20:50, 20 August 2009 (UTC)

I expect that once the bot runs through to get rid of the dates that were linked purely for autoformatting, date autoformatting in articles will be turned off. (You will still be able to see some system-generated dates and times, such as on history pages, in your choice of formats.) --Jc3s5h (talk) 20:56, 20 August 2009 (UTC)

We had this discussion very, very recently. Opinion wasn't uniform or unanimous, but I and several other editors said that this format is still unfamiliar to most potential readers of Misplaced Pages; it's not apparent at first glance that it's even a date. Once one understands the logic behind it, then, yes, it's unambiguous, but one first has to understand the logic, which is not by itself obvious or intuitive. The second consideration is that one way Misplaced Pages has avoided ambiguity is to strongly disfavour all-numeric dates (8/20/2009 vs 20/8/2009) and instead insist that months be written out in words, or at least abbreviated alphabetically (Aug., Oct., etc.) A numerical date would certainly be a favour to those whose first language isn't English, and would recognize "9" or "09" faster than "September", but it's outweighed in most people's minds by the unambiguous clarity of an alphanumeric date.

Reading the above discussion more closely, maybe where YYYY-MM-DD might fit in articles about astronomy or computer programming, there be a parenthetic clarifying conversion on its first appearance . —— Shakescene (talk) 21:15, 20 August 2009 (UTC)

I don't see any advantage in allowing ISO in the article space. I'd even rid the ref sections of it. "20 August 2009", "August 20, 2009", "20 Aug 2009" or "Aug 20, 2009" only. JIMp _talk·cont 23:42, 20 August 2009 (UTC)

I can imagine it in examples and some technical tables, but to quote Yoda, "the default, not it should be", even for most specialized material. I'm hesitant to say "never, ever", but it should be pretty rare in the prose text. Just my humble opinion. Studerby (talk) 01:18, 21 August 2009 (UTC)

I'll agree here. There may be rare cases (perhaps in tables), but they can be dealt with by judicious silence or an adverb. Septentrionalis PMAnderson 02:14, 21 August 2009 (UTC)

I was inspired to bring up this subject by the addition of a citation to SVG animation. Anyone checking references on SVG animation entities, if they have not already come to know and love the one true geek date format, will, by the time they finish flicking their eyes up the list. I agree that it is not as immediately obvious as using words, but for those who recognize it, it parses far more quickly and easily in the grey goo than anything involving letters. I actually initially used "20 August 2009" in my citation, as I noticed that is what is automatically generated by Cite web, but then immediately afterward noticed my addition was the only one in the list not using YYYY-MM-DD (and corrected my inconsistency in accordance with this document). I also agree that this date format is far more than a pawn of the ISO.

Communicating with humans vs. computers can be a context switch requiring concerted effort. Date formats involving words are on the wrong side of that context switch for the articles for which I recommend this format. --Darxus (talk) 03:26, 21 August 2009 (UTC)

• I'm not averse to this format, though I personally prefer writing it with slashes (YYYY/MM/DD) when working in the outside world. However, I don't think it makes for great prose. It's a reasonable compromise for the MOS to call for a spelled-out date format within prose, while preferring this form in places like tables, infoboxes and references. If some form of date autoformatting comes to pass, this format is also a good default format for the routine to expect, because it's compact and unambiguous. TheFeds 03:42, 21 August 2009 (UTC)

• By the way, the ISO 8601 version of this format has some restrictions that make it questionable for general use, and the format is ugly to look at when coding times (YYYY-MM-DDTHH:mm:ss). I'm not specifically advocating ISO 8601, but rather the general idea of a year-month-day-ordered short form. TheFeds 03:45, 21 August 2009 (UTC)

TheFeds, Misplaced Pages contains many articles about people and events before the Gregorian calendar was adopted in various places, and for the most part, these articles use the Julian calendar. The ISO specified that if one is claiming conformance with their standard, the Gregorian calendar must always be used. But the ISO has failed to make clear in the minds of the public that there are special conditions attached to the format whenever conformance to the ISO standard is claimed (or implied). One constantly sees people who have no idea about the the special requirements refer to "ISO dates". So if the ISO can't educate people to get it right, why do you believe Misplaced Pages will be able to educate its editors and readers?

Personally, I think Misplaced Pages has a lot more public impact than the ISO, but isn't organized enough to educate readers and editors about the meaning of the format. --Jc3s5h (talk) 12:26, 21 August 2009 (UTC)

• That Gregorian-Julian confusion was foremost on my mind when I wrote the follow-up above ("the ISO 8601 version of this format has some restrictions that make it questionable for general use"). Given that ISO 8601's particular stipulations are not well known to the public—even though many will recognize the basic meaning of the format—I'm not convinced that we should insist on ISO 8601 compliance, except in places where it's specifically called out. (Currently, the MOS says to avoid using YYYY-MM-DD where ISO 8601 could be assumed, but doesn't say that these are actually ISO 8601 dates.)

  In short, I don't expect Misplaced Pages to educate people to get ISO 8601 right, and don't feel that the MOS needs to advocate the use of ISO 8601. On the other hand, I have no problem with a YYYY-MM-DD format in general. TheFeds 17:03, 21 August 2009 (UTC)

In general use, people don't often refer to times for which Julian dates are appropriate. If you want to be that way, you have to use NS/OS on the dates as well. But when you're talking about a meeting last month or a birthday for your kid, Julian dates are beyond irrelevant. Now, if you want, in the British-colonized areas, to talk about dates before 1752, or in some Catholic areas before 1582, or in Russia before 1918, or in other Catholic areas before 1583, or in Germany before 1700, etcetra, etcetra, ad nauseum, that's you business. I personally have no problem using dates like 0300-03-25, as I don't care that much. I'm not attempting to time travel to a precise date, after all. - Denimadept (talk) 17:29, 21 August 2009 (UTC)

There are three acceptable date formats for a given day (excepting of course quotations): "12 December 2004", "December 12, 2004<some piece of punctuation if it is in prose>" and "2004-12-12" but never in prose and only where space is short. This is one or two more than is desirable, but is something we can work with. Certainly if we all changed to "2004-12-12" tomorrow we would get used to it rapidly and wonder how we got on with the clumsy old style, however WP is not a champion of change - one WP:NOT I do agree with. Rich Farmbrough, 21:28, 21 August 2009 (UTC).

Why no punctuation in the mdy? I'm unaware of any practice that allows the day and year numerals to jostle. It's a recipe for mistakes, and editors need to be encouraged to insert the comma boundary where the unfortunate juxtaposition of the numerical elements is used. Tony (talk) 03:49, 22 August 2009 (UTC)

"In general use, people don't often refer to times for which Julian dates are appropriate." sure, in general use people don't often refer to history ... actually it's not that uncommon ... especially on an encyclopædia. I still don't see the use of allowing ISO dates at all. If space is short 12 Dec 2004 or Dec 12, 2004 should be fine. JIMp _talk·cont 13:56, 24 August 2009 (UTC)

Locking down MOSNUM

I propose that MOSNUM be permanently locked down. “Consensus by parties of two” and “consensus by who can make forty edits a day” is not a consensus and just makes MOSNUM unstable. Unfortunately, it seems that if there is no teacher in the room, we kindergardeners can get out of hand. Way too much time is being wasted under the current system, which breeds anarchy and where the only remedies are to start big formal ANIs, WQAs, and ArbComs.

Too often, editors come to MOSNUM to change things in order to lend legitimacy to their particular way of doing things in articles they’re working. However, this is often done with an insufficient understanding of the ramifications. This results in edit wars and instability on MOSNUM.

I propose that there be a gate keeper on MOSNUM. There were some nice (very nice) periods where MOSNUM was locked down due to protracted bickering over IEC prefixes and date linking. And in both cases, the admins (MZMcBride and MASEM) did fabulous jobs watching over WT:MOSNUM discussions. What about those discussions? Well, with MOSNUM locked down, suddenly there was an outbreak of peace and tranquility and awfully civil, good-faith discussion. Check out WT:MOSNUM Archive 120 to see how smoooothly things worked back then. All MZMcBride and MASEM had to do was watch over the discussions to ascertain whether what was being proposed was uncontroversial, minor, or was significant but enjoyed a wide consensus. Then they simply copied the proposed verbiage and pasted it into the spot designated by the proposing editor.

How say others? Greg L (talk) 04:05, 21 August 2009 (UTC)

Ingenious, Greg L. And understandable. Almost tempting! But I think it is a defeatist approach, and something of a perversion of our Wikipedian institutions. When WP:MOS was locked for two months recently, over a needless impasse in wording, there was less acrimony – but also very little useful development. Sometimes small and incremental adjustments need to be made; and to require an intermediary even for those is laborious. If nothing else worked, I might be in favour. But there are more options to try yet: including systematic and concerted peer pressure on "difficult editors", to exercise restraint. If we have to get the MOS pages locked in order to deal with their disruption, they have effectively won.

–_⊥Noetica!– 06:21, 21 August 2009 (UTC)

Much as I enjoy the stability of the locked page (the main MoS was recently locked for weeks due to edit warring over logical quotation), I agree with Noetica that it's counter to the Misplaced Pages spirit. I'd rather people exercise self-control and abide by consensus by choice than be systematically forced to. There are plenty of areas of the MoS that need plain old editing for clarity and cohesiveness. I'm not sure we want to burn the house down to make toast. --Andy Walsh (talk) 17:47, 21 August 2009 (UTC)

1RR rule — Short of full protection, a 1RR rule on a page has benefits. Such a rule is currently working at Fascism and Massacres of Poles in Volhynia. This limits a single editor to at most one revert per day. Though I haven't checked carefully, I see multiple contributions by single editors at WP:MOSNUM on August 20, and if those edits constitute multiple reverts they would be prevented. Such a rule forces people to negotiate for support, or at least to revert more carefully. Any admin can impose such a rule if he or she gets it reviewed at a noticeboard. Of course, it helps if the local editors are in support. If people don't favor a 1RR, then permanent full protection seems to me like the next best option. EdJohnston (talk) 19:08, 21 August 2009 (UTC)

• Fine. My proposal gained no traction. I understand the reasoning. However, as regards arguments here predicated upon notions of “how things are normally done throughout Misplaced Pages”, I would point out that MOSNUM is a unique institution on Misplaced Pages. The temptations it affords and—human nature being what it is—leads to conflict. It should come as no surprise that a LOT of WQAs and ANIs arose from this place.

  MOSNUM guidelines are often the product of long, protracted, evolutionary-like progress. Having editors who just advanced to Junior High and gained access to the computer in the school library come in and change some guideline, like, “everything will be metric-only” (perhaps doing so will smooth our membership into the United Federation of Planets), may seem perfectly rational to that contributor. I can think of better things to do than go to MOSNUM and start counting reverts-per-24-hour period. Or explaining (again and again) what “consensus” means. And, sometimes, some editors in real life don’t have a social bone in their body and that behavior manifests itself here. So we end up reverting a change on WP:MOSNUM and then going through endless and circuitous arguing here on WT:MOSNUM until someone gets shot for a whole 24 to 48 hours. I have no illusions that this will end. (*sigh*) Greg L (talk) 20:02, 21 August 2009 (UTC)

  • And the JHS students won't know about 1RR, so that may actually increase the instability. They'll continue to revert, and the rest of us will be bound. Septentrionalis PMAnderson 20:27, 21 August 2009 (UTC)

But they don't know about 3rr either. Rich Farmbrough, 21:17, 21 August 2009 (UTC).

• But it leaves us equal. We will respect it, and it's easy to enforce against them. Septentrionalis PMAnderson 21:55, 21 August 2009 (UTC)

Greg L, while I don't support this drastic proposal of yours, I am heartened that someone at least is surveying such options. You write: "I would point out that MOSNUM is a unique institution on Misplaced Pages". I can understand your saying that, and I sympathise profoundly with you on all of the above. But I want to correct something. The whole Manual of Style is a unique institution on Misplaced Pages; the difficulties you experience here at WP:MOSNUM beset WP:MOS also – and it is, after all, the central page of the Manual. I have argued this before: we need to work out more accurate terms for the whole and its parts. Sometimes we use "MOS" to mean the whole; sometimes just that part discussed at WT:MOS. Rectification of names, in which I have some research interest, is a core Confucian doctrine. Curiously enough in the context of this present discussion, our article was initiated as a class project, in 2008. It could be better! But I commend it to MOS editors.

There are unique superordinate matters for our Manual of Style to address. I have made this point repeatedly, in one forum or another. But it is one thing to make a point lucidly, and another to turn a sufficient corps of editors away from their immediate detailed tasks to this broader work, so that that everything we attempt can eventually be on a more secure footing.

–_⊥Noetica!– 00:39, 22 August 2009 (UTC)

Also Shang Yang, if I recall correctly. Septentrionalis PMAnderson 04:51, 22 August 2009 (UTC)

In my view, WP:MOS should list the general principles, and then give a summary of its sub-pages, containing the rules which apply most often (as opposed to the ones which you will only ever need to know in very obscure cases). And neither WP:MOS nor any of its sub-pages should be more than 64 KB in length; otherwise it means that either they give the same rules several times due to a bad general organization of the page, or that it mentions stuff which per WP:BEANS and WP:CREEP had better be left out. --___A. di M. 10:47, 22 August 2009 (UTC)

I'd like to point out here a clause from Misplaced Pages:Requests for arbitration/Date delinking:

Greg L topic banned

13) Greg L is topic banned indefinitely from style and editing guidelines relating to the linking or unlinking of dates, and any related discussions.

Because I don't read this discussion page on a regular basis (I only learned about this thru reading the current Misplaced Pages Signpost), enforcing this now would be inappropriate, so I ask my fellow Admins that any time Greg L comments on anything related to MOSNUM he is violating the terms of this ruling & is liable to sanctions. (PS, while he is not the only one who is banned from this topic, he was the one who initiated this thread.) -- -- llywrch (talk) 16:43, 25 August 2009 (UTC)

National variety of English for MOSNUM

This guideline contains a mixture of UK and US spelling, especially with respect to SI units. Apart from any sections that specifically address national varieties of English, shouldn't we make it consistent? (I don't care which variety is used.) --Jc3s5h (talk) 11:45, 21 August 2009 (UTC)

In the actual text of the guideline (black) sure, that should be made consistent. The examples (green) can and should be inconsistent, as they present text as found in various articles, and different articles will use different Englishes. --___A. di M. 12:29, 21 August 2009 (UTC)

• There has long been a tendency for MOSNUM’s guidelines to give a large amount of deference to the style used by the first major contributor. I think that is OK to a point and serves a valuable purpose; it ensures contributing to Misplaced Pages is an enjoyable hobby, which is necessary for an encyclopedia that has an all-volunteer body of editors. I think that if an article has been created or greatly expanded by a particular, primary shepherding author, great weight should be placed on retaining the dialect of spelling used in that article. Certainly, the dialect should be consistent throughout (no “colour” in one paragraph and “realize” in another). However…

  I’ve long held that great weight should also be given to *what* the article is about. If it is a generic article, like water, then we can expect the normal distribution with our readership (which is about 25% American, as I recall). Accordingly, “first major contributor” works fine and should avoid conflict amongst editors. However, if the article is about Spokane River Centennial Trail or Boston Red Sox, there will obviously be a much higher proportion of American readers and the dialect used in the article should be American English. If we had an article on the Bondville Miniature Village (something I visited while in England when I was in my 20s), it would, IMO, be most appropriate to be in British English. This mix of guidelines best serves, in my opinion, what is the primary objective of any encyclopedia: writing in a way that seems most natural, fluid, and least confusing for the target audience Greg L (talk) 19:10, 21 August 2009 (UTC)

• ??? A.'s comment was about the variant of English used within the text of the guideline itself, i.e., the idea was that the guideline itself should follow WP:ENGVAR. Also, "colour" + "realize" is a bad example because it's perfectly good British English as promoted by Oxford University Press. Now "colour" + "analyze", that would be atrocious. What you are describing is already a part of ENGVAR, and I am not aware of any problems with its implementation. Hans Adler 19:45, 21 August 2009 (UTC)

• Maybe there isn’t any problem. I must profess I am not up-to-speed on how this issue fleshed itself out over the last several months. There used to be abundant conflict over the nuances. Maybe it's all settled. Greg L (talk) 20:04, 21 August 2009 (UTC)

  Huh? WP:ENGVAR has been stable ever since I can remember; at least, I can't find any non-trivial difference between its current revision and the latest 2007 version. (And this discussion is supposed to be about the text of the guidelines themselves, not articles.) --___A. di M. 20:45, 21 August 2009 (UTC)

• The ArbCom decision on which this is based (like the text of MOS) gives greater weight to an established period of stability (which it is not worth having Anglo-American wars over). Often this will go back to the first major contributor, but we only look at him when there has been no period of stability. This may appear less arbitrary. Septentrionalis PMAnderson 20:25, 21 August 2009 (UTC)

Most if not all of the style guides happen to use US spelling. Of course, the green examples can and should be in a variety of spellings. Tony (talk) 03:44, 22 August 2009 (UTC)

(slight diversion) For what it's worth, Talk:War of 1812, which naturally has British, Canadian and U.S. readers and editors, with a topic that relates to all three countries, recently reaffirmed a broadly-accepted consensus, following WP:ENGVAR, that British spelling should be followed in the article because its first major contributors used British spelling. The one exception was locations like New York Harbor, whose spelling follows the country they're in, unless it can be clearly shown that the locals used and use a different one. —— Shakescene (talk) 05:46, 23 August 2009 (UTC)

Ranges of values

From the current revision of WP:MOSNUM#Conventions:

• Numerical ranges of values are formatted as lower value-en dash-higher value-non breaking space-unit symbol (e.g., 5.9–6.3 kg, not 5.9 kg – 6.3 kg or 5.9 – 6.3 kg), or can be specified in written form using either unit symbol or unit names, and units can be stated either after both numerical values or after the last (e.g., from 5.9 to 6.3 kilograms, from 5.9 kilograms to 6.3 kilograms, from 5.9 to 6.3 kg and from 5.9 kg to 6.3 kg are all acceptable, but only one of these format should be in use in a given article).

As for the point about numerical ranges, sometimes it is useful to use different units for the endpoints, as IR-C in Infrared#CIE division scheme. So I'd rather write:

• Numeric ranges are formatted with an unspaced en dash if the endpoints have the same unit which is only explicitly written at the end: e.g., 5.9–6.3 kg. They have a spaced en dash if both endpoints are followed by a unit (for example, because the units are different due to very different magnitudes): e.g., 3000 nm – 1 mm.

As for the point about prose (mysteriously called "written form" here), I don't see the point of mandating consistency in the article when it's explicitly not mandated for all other units. Now the MoS suggests that if in an article I write from 2 years to 200 years (to avoid the silly alliteration "... two to two ..." which omitting the first "years" would result in, and to follow the common-sense practice of avoiding using symbols for a unit in everyday usage with a very short name), then I'll have to write from 50 kilowatts per squared metre to 80 kilowatts per squared metre as well, whereas from 50 to 80 kW/m would clearly suffice (provided that the kilowatt per squared metre has been spelled out before). So I'd remove the "but only one of these ..." part, so that the general rule in the first bullet of that section can apply. --___A. di M. 13:26, 22 August 2009 (UTC)

The reasonable thing to do would be to rewrite as from 2 to 200 years..., IMO. Headbomb {_{κοντριβς} – WP Physics} 01:36, 23 August 2009 (UTC)

from 2 to 200 years... doesn't seem to be to silly to me ... nor does the rule seem that sensible since it seems to ban the likes of from 4 to 8 inches (50–100 mm). Generally you don't want to write the unit symbol/abbreviation/name twice though occasionally you might but that shouldn't mean that you must do so throughout. JIMp _talk·cont 13:33, 24 August 2009 (UTC)

Delimiting numbers and screen readers

The guidelines for delimiting numbers are fine for screen readers at the moment. However I think it's worth noting that screen readers don't read numbers with spaces in them properly: for example, 33 300 400 will be read as "thirty three three hundred four hundred" which makes no sense. Therefore I made this edit. I just changed many instances of HTML thin spaces into commas in the Paleolithic article without knowing about the {{gaps}} template; perhaps I should have used it. Graham87 17:54, 22 August 2009 (UTC)

Thin spaces also caused problems with some browsers (e.g. one of the five I use; can't remember which one) or regular computer screens when the number is at the end of a line: the unbroken number just extends waaaaaaaay past the right of the screen, instead of jumping to the beginning of the next line. —— Shakescene (talk) 18:56, 22 August 2009 (UTC)

Suggestion to avoid use of {{val}} in the MoS

The {{val}} template is used to make it easy to format a value according to the MoS. Using the template in the MoS itself causes a "circular referrence" where the MoS depends on the output of {{val}} and {{val}} is supposed to depend on what the MoS says. I suggest the MoS uses hard-coded examples; otherwise anyone could change {{val}} and the MoS would change with it.     — SkyLined (talk) 23:43, 22 August 2009 (UTC)

Below, the top-most expression has the exponent, hard-coded with thinspaces on both sides of the × symbol. The one below it was created with {{val}}:

Tests:

___A. di M. 10:13, 24 August 2009 (UTC)

As you can see, 0.25 x 0.15 em on my platform combinations appears to have a balanced appearance that is narrow so it ought to meet the expectations of those who participated in the WT:MOS compromise between no regular spaces and no spaces at all. How does 0.25 x 0.15 em look on yours? Please post screen shots if necessary. Greg L (talk) 17:58, 24 August 2009 (UTC)

As you can see, 0.25 x 0.15 em on the iPhone also looks really good. Greg L (talk) 18:28, 24 August 2009 (UTC)

(edit conflict) It looks best to me, too. This was taken with Firefox 3.0.13 under Ubuntu 9.04:

I s'pose this should be taken to Talk:Val rather than here, though. --___A. di M. 18:31, 24 August 2009 (UTC)