Template talk:SI light units

SI
the term "SI" is used but not defined in this article, nor does a content search of Wikipedia define this term. Will someone who knows what this means please add it to this article, or place a link here to another area of Wikipedia that defines this? 68.62.20.247 01:19, 18 October 2007 (UTC)


 * See SI. Dicklyon 03:00, 18 October 2007 (UTC)


 * SI is short for Le Système international d'unités, or in English the International System of Units. It is what is commonly called the metric system. Note that this is not an article, but rather a template to be used inside articles. Articles that include this template should normally have a link to the article on SI. A content search for "SI" does produce the correct article, but note that searches are case sensitive.--Srleffler 04:33, 18 October 2007 (UTC)

Circular definition?
The definition for luminous flux and luminous intensity are circular. Isn't there a better way to do it that is not? It would be like cracking open a dictionary and seeing "Redundant: see superfluous" and then "Superfluous: see redundant." You get nowhere. —Preceding unsigned comment added by AZylman (talk • contribs) 08:46, 3 January 2008 (UTC)


 * No, they're not. But you need to go where they are defined.  The base unit is candela, as the table shows, so that one is not defined in terms of any of the others. Dicklyon (talk) 16:00, 3 January 2008 (UTC)


 * As Dick notes, they aren't circular because they are not definitions. See the individual articles for definitions of these quantities. The relation cd=lm/sr is redundant, as you noted. This redundant equation is there for convenience.--Srleffler (talk) 18:04, 3 January 2008 (UTC)


 * I think they are. If Candela is the base unit then it isn't defined in terms of other units. I've deleted the definition from Candela. If you want to know the relationship between cd and lm then look at lm.filceolaire (talk) 20:52, 7 June 2009 (UTC)


 * I reverted you. Yes, the information is redundant. So what? The template provides a collection of useful information. Sometimes it is useful to provide the same information in multiple forms. Redundancy is OK. Storage is cheap.--Srleffler (talk) 04:32, 8 June 2009 (UTC)


 * They are circular. Radiant Quantities are all clearly expressed through the watt, and here all should be expressed by the lumen (the equivalent of the watt). Both the lumen and the watt are SI Derived Units. Note also the rest of the entries of the table are not SI Derived Units, but Quantities expressed on SI Units (Base and Derived): the Emittance W/m2 or lm/m2. The Watt is a Derived SI Unit. The Emittance has no SI Unit. Hyprwfrcp (talk) 00:26, 2015-04-30 (UTC)


 * "Circular" is irrelevant here. All the relations given are correct. Expressing everything in terms of the lumen would be strange, since the candela is the base unit.--Srleffler (talk) 05:35, 2 May 2015 (UTC)


 * OF course it is relevant. Hence one should find defining all radiometrics units with 'watts' which is NOT a base unit. The main focus of the table is present quantities, and not reflect the candela is the Base Unit. Which is, but one is not worried about to know that. We are talking about Radiometric Power and Luminous Power. Radiometric power IS watts. Luminous Power IS lumens. I suggest to follow standard technical usage on this sense. Radiometry under watts and Luminous under lumens. That is the everyday usage.
 * We also could teach newbies on this table which is the candela, and express everything in candelas because this is the first time they use the unit, and they wish to get used with the fact that is the Base unit. And then also expressing on the same table that we have another unit, which happens to be another new unit for them too on the same day, and also showing that all quantities on the table are strangely and unexpectedly very related with lumens -Derived- and not with the unit they just learned seconds ago. Hence the table is teaching how to use units and have a reminder on the table, creating a little circular, but who cares, if they are not going to use the table nevermore on your life. I insisted. Lumen is the unit here. Candelas is just a newbie reminder, which could be placed as note. Hyprwfrcp (talk) 05:35, 2 May 2015 (UTC)


 * The table lists the most common unit for each physical quantity, and where it isn't obvious it says how that unit is related to other common units. That's all there is to it. A newbie encountering a spotlight with output specified in candelas needs to know what physical quantity is being described. Encountering several related units on the same day ought not to be a problem. Anyone for which that is a problem is probably not going to get far with photometry anyway.--Srleffler (talk) 02:04, 7 May 2015 (UTC)

What about radiometric quantities?
Unlike the name of the template, this is actually a collection of "SI photometry units", as per the table's caption. Compare to the table at Light intensity (as current at this time). Should radiometric variables be included, or the title of the template amended? It is inconsistent/incomplete as it stands now. —DIV (138.194.12.32 (talk) 00:15, 18 December 2009 (UTC))
 * The radiometric quantities are in Template:SI radiometry units. The title of this template is inaccurate, but this isn't really important since the title is only relevant to editors, not readers.--Srleffler (talk) 01:01, 18 December 2009 (UTC)


 * I changed that. These are NOT SI Units. Nor Base nor Derived. They are Photometric Quantities (or magnitudes, variables, etc) expressed in SI Units (Base or Derived). The only two SI units are the Lumen (lm) and the Candela (cd). I leave the title SI Photometric Quantities, and the label on the table Quantity. Hyprwfrcp (talk) 23:30, 30 April 2015 (UTC)

Proposal: Add units in photons
I've referred to this table frequently over the past several years; for some reason, I have trouble making the meanings of these quantities stick.

If no one objects, I'd like to add a column listing the unit in terms of photons. I think many people have intuition for a photon as a little packet of light and then can build on what that means for other things. That is, luminous energy is a number of photons, luminous flux is a number of photons per second, luminous intensity is photons per second per steradian, illuminance and luminous emmitance are photons per square meter per second.

Luminance is photons per steradian per square meter, which I still have trouble wrapping my mind around... I'll ask on Talk:Luminance.

For this to be qualitative, it would obviously be in number of 555nm photons.

Thoughts? —Ben FrantzDale (talk) 15:49, 2 December 2010 (UTC)


 * You cannot correctly express photometric quantities in terms of number of photons. There is no fixed relation between the two. For the very restricted case of monochromatic light one can relate them, but what you propose would be highly misleading if not outright wrong.


 * Everyone has trouble getting their heads around these quantities, especially luminance. You're not alone. Photometry is a bit subtle. Probably the best way to get an intuitive understanding of them is to compare them to the corresponding radiometric units. --Srleffler (talk) 17:42, 2 December 2010 (UTC)


 * I guess, though accurate it could be confusing. Just an opinion -- Edited by Hyprwfrcp (talk) 00:26, 2015-04-30 (UTC)

Template SI photon units
FWIW, I created a template named Template:SI photon units in analogy to Template:SI radiometry units and SI photometry units aka Template:SI light units. It's still "raw", but perhaps useful to some. --Matthiaspaul (talk) 16:56, 17 August 2011 (UTC)


 * I'm all for having that template (see my question above) so long as these are actually SI units. Can you provide a reference that these are "real"? —Ben FrantzDale (talk) 13:23, 18 August 2011 (UTC)

Too wordy
I disagree with this edit. It is sufficient to say that the suffix is v for "visual". "Visual" does not start with a Greek letter nu, so there is no danger of confusion. Adding "small Latin letter v" is unnecessary, overly wordy, and is a good example of overlinking. This is different from Template:SI radiometry units, where it is necessary to specify that the subscript is a nu, and to provide a link to the article for readers who may not be familiar with Greek letters. Readers of the English Wikipedia can be safely assumed to be familiar with the Western alphabet, and with the fact that the English word "visual" does not start with a Greek letter nu.--Srleffler (talk) 23:23, 19 August 2011 (UTC)
 * Answer and potential solution given here:
 * Template_talk:SI_radiometry_units
 * --Matthiaspaul (talk) 00:51, 21 August 2011 (UTC)

Dimension column incorrect?
This table reports "J" (Joules, i.e. energy) as the dimension for luminous flux. Surely this should be power (energy per unit time) instead? Lumen (unit) says that one lumen = 1/683 watt at 555-nm green. See also radiant flux in Template:SI_radiometry_units, which is in watts. This error is repeated all the way down the table, i.e. all the "J"s should be "W"s. Am I missing something here? Reedbeta (talk) 17:27, 4 October 2011 (UTC)


 * I'm not sure that "J" stands for "Joules" there. Dimensions are usually denoted by describing the physical quantity not the units: M for mass, L for length, T for time, etc. Not sure what "J" stands for in this context.--Srleffler (talk) 03:17, 5 October 2011 (UTC)


 * Hi! "J" does not stand for the unit joule here at all. This column does not discuss units (joule, candela, lumen, watt, etc.), but dimensions. In the context of dimensions, "J" is the symbol for the dimension of the quantity luminous intensity Iv, measured in units of candela (cd). I was living under the impression that it is defined as such in the SI system (see, for example, German WP article Internationales_Einheitensystem or Lichtstärke (Photometrie)). Quite interestingly, I cannot find this covered in the English WP right now. --Matthiaspaul (talk) 11:38, 5 October 2011 (UTC)


 * Very interesting. If "J" really is the accepted symbol for the dimensions of luminous flux/intensity, then the table is correct.  However "J" seems a very bizarre symbol to pick for this purpose, so I wonder whether this is correct for English or is perhaps a mis-borrowing from another language? --Reedbeta (talk) 17:38, 5 October 2011 (UTC)
 * Hm, given that the letter I was already used up for the dimension of the electrical current I, J does not look like a bad choice at all to me, as the letters I and J are often used as alternatives. --Matthiaspaul (talk) 19:08, 5 October 2011 (UTC)


 * The symbol doesn't seem obvious or well-known enough to use in this table without explanation. I have replaced "J" with "Iv" in the table, as a quick fix to the problem. While probably nonstandard, the latter at least has the merit of already being defined within the table, and eliminates the redundancy of using two different symbols for the same thing in one table.--Srleffler (talk) 02:33, 6 October 2011 (UTC)
 * I reverted your change and added a footnote, as I found an official source clearly stating that the SI standard recommends the usage of symbol J for the dimension of luminous intensity: [] It's in German language, but the Physikalisch-Technische Bundesanstalt (PTB) is Germany's official organ to publish this kind of information, so this is basically from the horse's mouth. We should be able to find the same information in the actual SI standard. Since it is difficult to implant references into footnotes (WP limitation), I suggest that we incorporate the actual references into the International System of Units article. --Matthiaspaul (talk) 14:11, 6 October 2011 (UTC)
 * Here is the English and French version of the same document: [] --Matthiaspaul (talk) 14:47, 6 October 2011 (UTC)
 * Thanks for adding the footnote, and for looking into the source of this notation.--Srleffler (talk) 23:50, 6 October 2011 (UTC)

The table is wrong!!! 1/683 Watt per steradian equals watts, not joules. The steraidian is dimensionless. — Preceding unsigned comment added by 68.84.232.225 (talk) 02:26, 8 April 2013 (UTC)
 * Hm, where in the table do you see watt per steradian equaling joule? Are you perhaps mixing up Luminous intensity (a photometry quantity in lumen per steradian) and Radiant intensity (a radiometry quantity in watt per steradian) as well as the concepts of units (watt, joule) and dimensions? If this doesn't solve your problem, please be more specific. --Matthiaspaul (talk) 03:06, 8 April 2013 (UTC)
 * I have to say, Matthias, that I am very tempted to remove the "dimension" column. It's not clear to me who it serves. It makes the template larger and more complicated, and as the discussion above shows it creates a lot of confusion, mostly due to the unfortunate choice of "J" as the symbol for the dimension of luminous intensity.--Srleffler (talk) 00:34, 9 April 2013 (UTC)
 * Hm, I like your improvement of the footnote, but I don't think removing the column would be a good idea. Let's face it, radiometry and photometry are difficult to understand subjects for many people, no wonder that some will mix up a few terms until they really dig into it. Studying the dimensions can help them a lot recognizing the differences and similarities - and much easier than by comparing the units. Therefore, I think, this column is really useful - and important for completeness (this is exactly the kind of information why I would consult this Wikipedia table).
 * Yes, it is unfortunate that the letter J is used for multiple purpose, but that's how it is standardized and used internationally. Symbols for quantities, units and dimensions do not need to agree at all (in fact, they don't), as they are used in different workspaces. Perhaps the problem was that some readers did not recognize symbols as being associated with either quantities, units or dimensions; I hope, the modified table header will make these dependencies more obvious. Greetings. --Matthiaspaul (talk) 03:41, 9 April 2013 (UTC)
 * I like the change you made in the header a lot. It's a simple thing, but to my eye it seems to clarify the relationship between the columns. Hopefully these changes will reduce reader confusion about "J".--Srleffler (talk) 04:12, 9 April 2013 (UTC)
 * Thanks for the feedback! I'm glad that you like it. :-) --Matthiaspaul (talk) 04:34, 9 April 2013 (UTC)

Typeface for dimensions
Currently, bold is used for differentiating quantity symbols. This, as far as I know, is completely nonstandard. Should we not use a more usual convention, such as square brackets? Example:. —Quondum 02:54, 19 May 2015 (UTC)
 * Bold is used here for dimension symbols, not quantity symbols. The style used here seems to follow that used in publications by NIST and BIPM, both of which specify that the dimensional symbols are to be written in sans serif roman capital letters. Neither mentions bold, but in the NIST publication (page 22) the dimension symbols appear heavier than the surrounding text, perhaps just due to the particular sans serif font that was chosen. On Wikipedia, we use bold because the default font is sans-serif, so the dimension symbols aren't clearly distinguished otherwise.
 * The template you used supports this convention too: . It has the advantage of being more compact, and easier to read.--Srleffler (talk) 03:47, 22 May 2015 (UTC)
 * A closer examination shows that the brackets are used to mean "units of", and would lead to confusion if used to mean "dimensions". My impression of it being a "more usual convention" was a misconception.  I would agree that bold is less unsuitable than brackets.  I have stripped the brackets options from the template (after checking all uses).
 * The typeface in the NIST document is not properly describable as a "heavier typeface". That is merely the side-effect of the font being required to be sans-serif, and is noticeable because of the particular choice of serif and sans-serif fonts in that document.  The ISQ also speaks of roman sans-serif type for this use (JCGM p. 5: "The conventional symbolic representation of the dimension of a base quantity is a single upper case letter in roman (upright) sans-serif type.").  I interpret this to mean "not bold".  I hear you about the surrounding typeface being sans-serif, but this is still not really a satisfactory workaround.  I'll mull on this.  —Quondum 18:58, 24 May 2015 (UTC)
 * I agree that the sources at hand don't support the use of bold. I'm not sure if this is a convention that is actually used elsewhere, or if it was just invented for Wikipedia. Certainly on this template the dimensions do need to be visibly set off somehow. Readers get confused between dimensions and units.
 * Does your template force the font to sans-serif? If not, it should, in case the reader has set a serif default font for Wikipedia.--Srleffler (talk) 17:08, 25 May 2015 (UTC)
 * No, dimanalysis does not force sans-serif. It should force both sans-serif and roman (upright) font.  What markup can be used to do so?
 * Incidentally, the middots used in this template SI light units) are possibly nonstandard. Would it not be better to remove these?  —Quondum 18:30, 25 May 2015 (UTC)


 * Don't think it's possible to force a font to be serif or sans serif - how does WP template syntax know what is serif and sans serif? It should be possible to force it to Arial though, by explicitly including the font in the template (maybe use HTML blocks or similar, or the font).
 * Yes remove middle dots between dimensions, no point and doesn't seem to be standard.. M&and;Ŝc2ħεИτlk 21:11, 25 May 2015 (UTC)


 * Specifically,  to give  ...entire content of the template dimanalysis inside... M&and;Ŝc2ħεИτlk 22:20, 25 May 2015 (UTC)
 * I managed to force sans-serif, using HTML inside font – thanks, both, for the pointers. Perhaps we should be discussing that template on its own talk page.
 * I'm removing the dots in this template, for now. —Quondum 22:37, 25 May 2015 (UTC)
 * And I have put them back in. The standard convention in mathematics is to indicate multiplication by a dot-operator. Omitting the dots is a shorthand notation that can be used when it is obviously clear to anyone that the arguments still need to be multiplied. I don't think we can assume this here, as, judging from past discussions, a lot of people (pupils?) seem to be unfamiliar with SI unit and dimension symbols in general and some units specifically. That's why we explicitly use dots in the units column as separators. While there's less risk for misinterpretation in the dimensions column (as those symbols use a single letter only), using the dot in one column and omitting it in another would create an inconsistency which may make it more difficult than necessary for readers to digest the information. Since omitting the dots in the dimensions column does not give us any advantage, we should keep it. --Matthiaspaul (talk) 15:41, 4 June 2015 (UTC)
 * CSS (HTML cascading styles) allows the font designations "serif" and "sans-serif" in a font list, so that if none of the specific fonts called for is available it falls back to some sans-serif font chosen by the client web browser. I'd have to look up the exact syntax; it's been awhile.--Srleffler (talk) 00:59, 26 May 2015 (UTC)

Before it comes up, I want to mention that I would probably oppose using the dimanalysis template within this template. I don't see what value it adds; it produces the same output but makes the edit-window text longer and more obscure. I presume the template is useful for something, but I can't quite see what.--Srleffler (talk) 01:04, 26 May 2015 (UTC)
 * – Agreed, it would not really make sense here; it would just slow down page rendering. I think the template simply helps with uniformity: it kind of sets a formatting standard, including the ordering of the symbols.  Useful for those of us with fuzzy memories.  —Quondum 01:13, 26 May 2015 (UTC)
 * That can be useful, and the addition of font control adds some value too.--Srleffler (talk) 01:34, 26 May 2015 (UTC)

Dimension symbol order MLT or LMT?
I have no personal preference but I changed the order of dimension symbols back to MLT rather than LMT as almost all examples I could find use this order (this is also why I used this order originally). Even the English dimensional analysis and the German de:Dimensionsanalyse articles have it this way (until partially changed recently by Quondum). Does someone know the reason for why most examples are given as MLT? --Matthiaspaul (talk) 16:37, 4 June 2015 (UTC)
 * Just to answer this: the SI brochure, which is really the source and reference, uses the LMT order (when referring to dimensions, as opposed to units – SI seems to be blind to this subtlety). I do not particularly like that order, but if there is to be a convention on the order of dimension symbols, it seems to me that the SI is the only one that will ever have traction.  Note that units are more complex, since there are also derived units.  Though if we want to synergize the order of base units and dimension symbols, I'd be happy to support this as a proposal for a MoS principle.  —Quondum 16:10, 15 December 2017 (UTC)
 * Update: in the 9th edition (2019) the order of dimensions has been changed to T,L,M,... - similar to the base units, which are now s,m,kg,... (see SI Brochure 9th edition chapter 2.3.3) -- Wassermaus (talk) 10:50, 5 January 2020 (UTC)

Positioning of footnote
I'm a little confused by. I did not remove the footnote; I merely moved it to the top of the column and generalized it. Repeating it on the first few examples instead seems inappropriate; it really does apply to the entire column. —Quondum 16:00, 15 December 2017 (UTC)
 * I didn't notice the change in footnote text; I was responding just to the change in location. The reason the footnote appears on the first three occurrences of J rather than just the first one is that hard experience has taught that readers will miss it and complain that the dimension of luminous flux or luminous intensity is given as "Joules". We can try it and see, though. The layout of the table has improved since the early days; maybe that is enough.--Srleffler (talk) 18:22, 15 December 2017 (UTC)
 * Yes, I guess there is always the tension between the understandability for first-time readers and uncluttered exposition for reference. I haven't tracked (and cannot remember) whether some footnotes were present before the queries.  As you say, we can see whether the problem still manifests.  If it really continues to be a problem, I'd suggest adding mouse-over text to each symbol instead of the more obtrusive repeated footnote, for example: T ⋅J.  —Quondum 19:26, 15 December 2017 (UTC)