Talk:Planck's law/Archive 5

Archive 1

←

Archive 3

Archive 4

Archive 5

Archive 6

Archive 7

→

Archive 10

statements hopefully more consensual

There were two main bones of contention in the version "Common forms". One was about editorial ambit, the other about the currency of expressions.

Some editors seemed to think that even if a calculation takes an expert such as Headbomb several postings and several days to get right, it can still be regarded as a "routine calculation" for the present purpose. With wide editorial ambit, there is potential here for own research with uncorrected mistakes to be posted by less expert editors than Headbomb and the present company (who did not correct Headbomb's work). The form most precious to Headbomb's heart is not common currency, but is it worth the time to resist his desires?

The new section heading removes the currency problem and capitulates to Headbomb's desire to include his B_k form, and sets a precedent for accidents waiting to happen.Chjoaygame (talk) 17:47, 1 November 2011 (UTC)

This just does not sound right to me. I think some WP:AN/I action or mediation is called for, it should not be that you are just putting things in you feel are wrong just to have a quiet life. That is not consensus. Dmcq (talk) 18:02, 1 November 2011 (UTC)

The "calculation" takes 5 minutes to perform at most, and can be done with anyone that knows how to substitute and derive [aka 1rst years students of science]. Want to express B_λ in terms of B_k? Substitute λ by 2π/k, and multiply the result by |dλ/dk|. Want to express it in terms of ν? Substitute λ by c/ν, and multiply the result by |dλ/dν|. Changing formulas from one variable to another is not original research by any stretch of the imagination, this has been explained to you many times and you still refuse to get the hint.

As a side note, I'm not really following Dmcq's post. Headbomb {talk / contribs / physics / books} 18:32, 1 November 2011 (UTC)

Results should not be original research, WP:CALC does not allow for this sort of stuff. Intermediate stages can be an editors work with a little of their own working but anything of substance in particular any conclusions should be based on sources. It sounds from this that you are putting in your own work. Surely there are enough sources around on Planck's law? Why are you insisting on doing things like this? Dmcq (talk) 19:12, 1 November 2011 (UTC)

WP:CALC certainly does allow for this. See WP:Scientific citation guidelines#Examples, derivations and restatements. This is no more original research than converting miles into kilometers. As for the "why", there are two paradigms; people who work with h, and those who work with ħ. The article should cover both, because both paradigms are used [e.g [1], which reports u_ω instead of B_ω, but the difference between the two is a factor of 4π/c]. Headbomb {talk / contribs / physics / books} 19:21, 1 November 2011 (UTC)

Conclusions and final statements are not examples, derivations or restatements. Please work from sources if you are going to present conclusions. Source not your own ideas of what is important are what are needed. 'Should be presented' is not what Wikipedia is about, it is about what reliable sources actually say. If you have ideas about what should be presented and other people have not done it then go write your own book and get it published and then Wikipedia can reference it. Dmcq (talk) 23:29, 1 November 2011 (UTC)

I disagree about the calculation being original research, it is a calculation that anyone who has taken a course in radiation theory learns to do mechanically, in their sleep. Unfortunately, the simple mathematical reasoning for this routine calculation can fade over time, along with the references which describe it. Is 2+2=4 original research? No. What is the mathematical reasoning which makes it so? I used to know (Dedekind cuts and all that), but I can't duplicate it now off the top of my head, and that doesn't worry me. Can anyone name a reference off the top of their head for 2+2=4? I can't, and it does not worry me. If you write 2+2=4 in a Wikipedia article, how do you feel about someone who accuses you of original research, demanding to know the mathematical reasoning behind it and multiple references for that particular statement? Restating Planck's law in terms of different variables is hardly as obvious as 2+2=4, so take that feeling and divide by ten, and that's how I feel about this.

I agree that "common forms" should list common forms only. I have not researched the literature to find out the relative frequency of the various statments of Planck's law that are listed in the table but if they are not common, then they should not be listed. PAR (talk) 00:21, 2 November 2011 (UTC)

If a calculation is done using an established formula for the purposes of an example or illustration or as an intermediate step in deriving an established result then that is fine. However there shouldn't be a need for other calculations in Wikipedia. Dmcq (talk) 00:33, 2 November 2011 (UTC)

'Original research' on Wikipedia has a special meaning. It means the policy WP:OR. I agree it isn't original research in any reasonable real world sense. Producing a form which nobody else has produced is original research in Wikipedia terms even if it involves a routine transformation. Only simple conclusions are allowed with new results like changing kilometers to miles, WP:CALC covers this. Also it doesn't matter much in intermediate steps and examples. If someone has to work out a 'common form' rather than look it up then it is original research in Wikipedia terms. Dmcq (talk) 01:33, 2 November 2011 (UTC)

It's not OR in the Wikipedian sense either. Again, see WP:Scientific citation guidelines#Examples, derivations and restatements. Changing e.g. ν to ω (or differently put, changing ν for chuncks of 6.238 ν") is no different than changing miles to kilometers (or differently put, changing miles to chunks of "0.622 miles"). This is both well documented are non-controverisal. Headbomb {talk / contribs / physics / books} 02:45, 2 November 2011 (UTC)

According to WP:SCG, "It is also important, however, for our articles to clearly indicate the person who first discovered an astronomical object, first proved a theorem, first performed an experiment, or was otherwise responsible for the idea being discussed." Who first introduced the conventions B_ν̃(T), $B_{\omega }(T)$, $B_{y}(T)$, and $B_{k}(T)$ in the table in the Planck's law article? Near as I can tell it was those Wikipedia editors who invented them on their own say-so. These notations don't exist in the literature for the same reason that functions like $\pi \sin(x)$ and $e\cos(x)$ don't have their own notations, namely that they're merely constant multiples of functions that already have names.

Those working in the field know this, making it an embarrassment when Wikipedia doesn't know it and implies otherwise by suggesting that these are all distinct functions each with their own peaks. It should be obvious even to an eighth-grader that a constant multiple of a function has the same peak as the function itself. It seems to me that Wikipedia should not be allowing editors who don't know elementary facts like that to be editing an article at the level of sophistication of Planck's law.

Adding insult to injury, the editor in this case has in the past two weeks been unilaterally deleting the contributions of the regular editors of this article, seemingly without comprehending the content of what he's been deleting, as evidenced by his outrageous technical claims on the rare occasions he's deigned to say anything on the talk pages.

I will be so relieved when the regular editors are permitted to resume their usual editing of this article without this sort of interference from someone who is so obviously way out of his depth in this area. And I should add that I deeply resent the interpretation by some of the more qualified experts here of what I wrote in the previous sentence as a "personal attack" on that editor. I have read Wikipedia's definition of "personal attack" and as I understand it this comes nowhere near it. If you disagree then we need to focus on that question, the resolution of which could well undermine how Wikipedia is supposed to work. --Vaughan Pratt (talk) 08:35, 2 November 2011 (UTC)

Notation should be consistent. It matters little if we call them B_foobar or I_foobar, or whatever, as long as we are consistant. People wanted to go with B above, so B it is. Personally I prefer I, but that is a completely trivial thing. As for the peak question, you seem to have a fundamental reading comprehension problem. The article says that the distribution peaks depend on which variable is chosen, e.g. B_ν does not peak at the same place [w.r.t. energy] as B_λ does, likewise for B_k which does not share the same peak as Bω], not that each distribution has a different peak w.r.t. energy. And, for the umpteenth time, YOU DO NOT OWN THE ARTICLE because you were here "first". Headbomb {talk / contribs / physics / books} 10:17, 2 November 2011 (UTC)

Since we seem to flat out disagree on what WP:Scientific citation guidelines#Examples, derivations and restatements covers I will have to take this to WP:ORN for a decision. Dmcq (talk) 12:17, 2 November 2011 (UTC)

I believe I intrduced the $B_{\nu }$ notation, because it prevalent in the literature I have read, it is a superior notation, it is used in Chandrasekhar, which is practically the bible of radiative transfer, and Rybicki and Stiles, which is a heavy duty book on the same subject. PAR (talk) 14:02, 2 November 2011 (UTC)

Three shapes

PAR is quite right, $B_{\nu }$ is the standard notation for the frequency shape, just as $B_{\lambda }$ is standard for the wavelength shape.

Dicklyon agreed with me that there are exactly three shapes, the third being standardly notated as either of $\nu B_{\nu }$ or $\lambda B_{\lambda }$ (which don't even differ by a constant).

Other notations like B_ν̃ and B_k are a Wikipedia invention that both denote the same shape as $B_{\nu }$, which is presumably why no one so far has found any source for them.

HB deleted the $\nu B_{\nu }$ notation for the (very important) third shape, along with all mention of it, on the ground that it was just multiplying $B_{\nu }$ by $\nu$ and was therefore "silly."

First, why does he believe readers can easily derive the formula for this important shape themselves?

Second HB's reasoning for deleting the standard notation for this shape would show that the notation $B_{\lambda }$ for the second shape was also "silly" because it is just $B_{\nu }$ multiplied by $\nu /\lambda$.

(We'll see this relationship again below.)

Homogeneity vs. clarity

Incidentally, for those worried about formulas containing all three of c, ν, and λ, it should be pointed out there are three ways of writing any formula using only two of them, namely by using the relation c = νλ to rewrite the unwanted one in terms of the other two and then doing the obvious cancellations. For example there are two ways of rewriting ν/λ, namely as ν²/c and c/λ². Those who object to λ appearing in the formula for $B_{\nu }$ can just replace it with c/ν.

One reason for using this method to eliminate c is that c obscures the meaning of the formula. For example eliminating it from the obscure formula $2h\nu ^{3}/c^{2}$ yields $2h\nu /\lambda ^{2}$ which is obviously photon energy divided by area as suggested by User:Just granpa a year ago. (Notice that $\lambda ^{2}/2$ is simply the integral of $\lambda$.) The same principle justifies leaving $e^{h\nu /kT}-1$ the same in all formulas, where $h\nu /kT$ is clearly a ratio of two energies $h\nu$ and $kT$, instead of changing it to $e^{hc/\lambda kT}-1$ where c enters to obscure matters.

I think of eliminating c as "clarifying" and eliminating either of the other two as "homogenizing" (since then only one spectral variable remains).

Although some people seem to prefer homogeneity, I don't see much benefit, whereas I certainly see the benefit of clarity. Interestingly the clarity happens automatically in every formula when c is eliminated, and equally automatically disappears when it is reintroduced via homgenization.

Slogan: clarity good, homogeneity bad. It's a fully deterministic binary choice, not an ad hoc kludge.

Level playing field

This is even simpler algebra than the kind that has to get the derivatives right (counting the correct number of π's for example) in the formula for $B_{k}$. We're allowing hard algebra to derive formulas that don't exist in the literature (other than by substituting variables, such as k for n_p which HB has argued strenuously for), while objecting to simpler algebra that doesn't get involved with the differentials (the third rail of this article if you ask me, very dangerous to touch). This doesn't sound like a level playing field.

We even have in the article that $B_{\lambda }=B_{\nu }|d\nu /d\lambda |$ (in fact $B_{\lambda }=-B_{\nu }(d\nu /d\lambda )$, no need to discard the minus sign which has meaning here because of the differentials), and since $d\nu /d\lambda =-\nu /\lambda$ we immediately obtain $B_{\lambda }$ as $(\nu /\lambda )B_{\nu }$ (the two minuses cancel), the point I was making earlier in response to HB's "silly." This can then be expanded as desired to whichever of $(\nu /\lambda )2h\nu /\lambda ^{2}$ or $2hc^{2}/\lambda ^{5}$ you prefer (clarity vs. homogeneity), along with Planck's famous denominator of course.

Chain rule

Incidentally has anyone been checking HB's extensive edits lately for accuracy? I suggested the chain rule earlier, but what he's got in the article about the chain rule looks nothing like my suggestion. Source, please.

HB, good cop PAR suggested back on your first day on this job that you slow down and discuss things. Bad cop Vaughan Pratt thinks this is an excellent idea, especially when wrestling the third rail to the ground like that. --Vaughan Pratt (talk) 22:21, 10 November 2011 (UTC)

Vaughan, as I said earlier, spectralcalc is the source for Bν̃, which they present as L_σ. Q Science (talk) 08:17, 11 November 2011 (UTC)

I was waiting for someone to second that source before commenting. Those aren't the same function, since σ is not a synonym for ν̃ but is defined by σ = 100ν̃. The real "spectroscopist's function" is the one Spectracalc calls L_σ, which depends on the input being CGS units and the output being mixed MKS-CGS units, not the function Chjoaygame calls B_ν̃ which works for any units, even British FPS, and (in MKS) outputs values a hundred million times smaller than L_σ at the corresponding point in the spectrum, or (in CGS) ten thousand times smaller.

Incidentally the Spectracalc source raises the question of whether a company's anonymously authored literature supporting their proprietary products is a legitimate source. Does Wikipedia have anything to say about either anonymous sources or corporate literature as sources? --Vaughan Pratt (talk) 23:36, 11 November 2011 (UTC)

The form with inverse wavelength ${\tilde {\nu }}=1/\lambda$ as argument has been sourced from two reliable secondary literature sources, Caniou and Paltridge and Platt, besides the internet source stated just above. The literal notations differ, but the function is the object of interest.Chjoaygame (talk) 10:28, 11 November 2011 (UTC)

Those sources are fine for the functions, which in any event are not a Wikipedia invention since they've been around for billions of years. I was objecting to the names for them you and Headbomb have been coming up with, which have been around for less than three weeks.

Clearly names are important to Headbomb, because when we first switched from I and I' to $B_{\nu }$ and $B_{\lambda }$ respectively (exact same functions), his very first edit here of 2011 reverted them back to I and I'. We had to argue that the new names were in wide use in the literature before he would accept them.

But now he and Chjoaygame seem to have changed their minds about the need to source the names from the literature and see no harm proliferating their own homebrew names. Currently Headbomb's six-entry table is missing $B_{\sigma }$ where σ = 100ν̃ is wavenumber in units of cm⁻¹, which is what real-world spectroscopists use. (See e.g. the HITRAN database and its documentation. As can be seen from the latter, only quiche-eaters write either σ or ν̃ for wavenumber in units of cm⁻¹, real spectroscopists write ν.). The angular version would be σ' = σ/2π where σ' should be replaced by a suitable symbol for the angular counterpart of σ, which those who brought us $B_{y}$ should have no trouble coming up with. --Vaughan Pratt (talk) 01:04, 12 November 2011 (UTC)

Dare I make a joke here? No, I think discretion would be wiser! I am not fussed about the name. I just wrote "spectroscopists' wavenumber" for clarity, referring to the relevant section in the Wikipedia article on Wavenumber, and copying the notation used there. If you have a better way to make it clear, please do it. I did not detail the units because it seemed unnecessary, but I have no objection to you doing it if you like. No comment on diet!

Ah, now I see I think by "names" you refer to the mathematical notations, not the words "spectroscopists' wavenumber"? Again I am not fussed. As seems appropriate to the assembled company will do me fine.

On a slightly more general note, I would like to see the table format abolished. I suppose there are those who would like it kept. I think the table makes it hard to deal with particular cases, and nearly forces undesirable entries, and does not really help. In particular, to be explicit, I would like to see the angular wavenumber form go. I do not like the reference to Kramm and Mölder which seems to be generally unnecessary apart from being made necessary by the table format that more or less requires the angular wavenumber form with $\hbar$. I don't recall why Sharkov is referred to; I guess there is a reason; perhaps you can recall how it got there and tell us?Chjoaygame (talk) 02:49, 12 November 2011 (UTC)

Query raised at Original research noticeboard

I have raised a query at:

Wikipedia:No original research/Noticeboard#Can new formula be put in to a scientific article?

about the new common form and the interpretation of Wikipedia:Scientific citation guidelines#Examples, derivations and restatements Dmcq (talk) 12:38, 2 November 2011 (UTC)

I don't quite understand the issue, and the text above is very long. Can someone give a one-paragraph summary? Different people can give different summaries with their take on it, of course. — Carl (CBM · talk) 12:56, 2 November 2011 (UTC)

Here is an example for comparison. If someone had a formula which uses radians, and we want to convert it to use degrees, we may need to add some multiples of π and 1/π in various places (for example, if the formula is based on a derivative, we would have to correct for the different derivative rules that apply to degree-based functions. Is that the sort of change that is being made here, or is it something deeper? — Carl (CBM · talk) 13:00, 2 November 2011 (UTC)

Not too much deeper, there is a problem that angular wavenumber for instance never seems to be used in this context, but yes you could consider it as if we had the formula for sine of twice an angle and nobody had written down a formula for the cosine of twice an angle and we worked it out from the sine formula and stuck that into an article as a usual formula and cited it to the sine formula. Headbomb took a few attempts to do the similar thing here and I guess someone could make similar mistakes converting the sine formula. Dmcq (talk) 13:35, 2 November 2011 (UTC)

I think my initial concern is why it is necessary to include the formula here if the standard textbooks manage to get by without it. Is the formula that was added here really not present in any of the standard texts? At least with the law of cosines we know we could find a book that includes it, even if we derived it ourselves. — Carl (CBM · talk) 13:51, 2 November 2011 (UTC)

Exactly. Dmcq (talk) 13:54, 2 November 2011 (UTC)

These are not "new unsourced" formulas, these are well-known form of Planck's law (and yes, they are sourced, see arXiv:0901.1863, equation 40 which you somehow refuses to read). You believes that making variable substitutions (such as replacing ν by ω = 2πν) consists of original research, when it clearly does not. Converting formulas from ν to ω = 2πν is no different than converting them from miles to meters. Substitute and multiply by |dν/dω|, a method which is both uncontroversial, and yes, sourced ([2], bottom of page).

The level of WP:IDIDNTHEARTHAT here is astounding. Headbomb {talk / contribs / physics / books} 19:00, 2 November 2011 (UTC)

So your justification for this 'common form' is a paper submitted to a maths society but not yet published? Not even a physics society? Moreover that paper uses k for the Boltzmann constant rather than what you insisted in using and which you said had to be the way you wanted it and put that in big bold letters in 'You cannot use the same variable two different quantities, so you need to distinguish both. k_B for Boltzmanm constant and k for wavenumber'. Moreover it is pretty obvious you had to scrape around for this paper after you figured it out yourself with help. How can you justify calling this a common form? And why do you go around insulting other editors who point out things like this to you? Dmcq (talk) 19:22, 2 November 2011 (UTC)

The paper is published (J. Calc. Math. Soc. 5 (1–2): 27–61)! It's written black on white! In what alternative reality do you live? And arguments such as "you worked it out yourself, then sourced it, so it doesn't count" and "it's a mathematics journal, not a physics journal), are pretty much the very definition of petty trolling. Headbomb {talk / contribs / physics / books} 20:01, 2 November 2011 (UTC)

Well you obviously have access to better sources than me. I could only find 4 issues before the last produced, I've done another search and I see now that they produced an issue 5 in 2009 and that seems to be the last they've done but I can't find the contents list. In the alternative universe I live in common forms of physics equations appear somewhere in some noted Physics journal or book, they do not only appear in a maths journal that seems to be having problems finding or publishing papers even though it publishes things wildly outside the remit of it aims in [3]. Dmcq (talk) 20:30, 2 November 2011 (UTC)

The paper in question was published by an editor of the journal it appeared in. I read all six sections of the paper. Five of them were true but not new, one was new but not true, namely the one (section 5) that proved the statement "the Stefan-Boltzmann constant is not a universal constant of nature" by showing that if you replace Planck's law by the Rayleigh-Jeans law you get a different constant. If Headbomb believes this is not crackpot physics then he's a crackpot himself who has no business screwing up this article, which was fine until he came along three weeks ago and completely reorganized it while reverting all attempts of others to edit it except for edits fixing his obvious errors. In view of his violations of WP:3RR (at least one if not two), and his many antisocial antics on Wikipedia, all of which ANI should be able to relate to, along with a long list of idiotic statements about physics that ANI might prefer not to engage with as being too technical, I believe taking this to ANI is fully in order at this point. If I'm the only one with this opinion I will not push it, but if there is support from others then I will. --Vaughan Pratt (talk) 03:10, 6 November 2011 (UTC)

It is not obvious to me that Wikipedia at present has good facilities to deal with the present problem. But I will here state my concerns, which are about expressing things in a Wikipedia article. There are some things that ought to be reasonably easy, but are in fact, difficult to express in the present Wikipedia article, because of the editorial process, not because they are intrinsically difficult to express.

One of my concerns is that it seems very difficult to express in the present Wikipedia article the fact that very few if any genuinely established reliable sources use the eponym 'Planck's relation'. That it is used cannot be denied, but so far as I can see, it is used by sources that are not well recognized reliable sources; an example is the present one: the source is a book that I do not wish to denigrate, but one which is nevertheless not an established scholarly text, and lies more in the realm of popularization. If one wishes to let numbers on a Google search be the main basis of one's beliefs, then one is free to Google for oneself. But I think it no duty of Wikipedia to be a mirror of Google. If one wishes to say somehow that the eponym is used by such popular sources as the one that stands at present, then I accept that it should be possible to do so, but not in a way that hides the distinction between established scientific usage and more popular usage. In the present circumstances, it seems that to express this may need an elaborate explanation, almost a section in itself, when the matter is hardly important enough to justify that. The obvious course of letting the reader see the mention of Planck's relation in the Wikipedia article on Planck's constant is apparently not easily available, and in any case might be objected to on the grounds that if the editors of the article on Planck's law don't endorse the eponym, it seems odd to let it stand in another very closely related Wikipedia article.

The other of my concerns is that it seems difficult to express in the present Wikipedia article that the angular wavenumber form of Planck's law is not nearly as widely used as many other forms.

The problem here is that the difficulty is due to editorial behaviour and attitudes. One editor has considerable useful talent and perhaps in effect appears to have some degree of support, direct or indirect, explicit or implicit, expressed intentionally or unintentionally, amongst other editors for their various personal reasons. But I don't think that quite adds up to a good reason to make it or allow it to be so difficult to express things that would be otherwise far easier to express, and that ought to be expressed in the present Wikipedia article.

I think something should be done about this, but I am not experienced in the relevant Wikipedia processes, and I do not know whether they are suitable for the present problem or whether they even exist. I don't feel I know the best way to go from here.Chjoaygame (talk) 04:56, 6 November 2011 (UTC)

Derivation of Plank's Law

I am new to discussing topics so hopefully the following will not be regarded as spam. I find it confusing that eq. (1), the energy of an electric field dependent on the number of photons includes the zero point energy whereas for the derivation this zero point energy isn't used (especially calculating Z(beta) gives some problems if you include this \hbar \omega * 1/2). — Preceding unsigned comment added by 128.131.52.143 (talk) 14:19, 2 November 2011 (UTC)

It factors out of the partition function, perhaps it can be included to make this clear better. You then do have to deal with the complication that the ground state energy, which the sum over all modes of 1/2 hbar omega, is infinite. Count Iblis (talk) 17:32, 2 November 2011 (UTC)

Different forms

I agree describing these as different forms is far more accurate than calling them common forms. The bit about 'Forms on the left are most often encountered in experimental fields, while those on the right are most often encountered in theoretical fields" is rather problematic. There is also the problem that actual common forms have been removed when putting in the reduced forms. Putting them back would mean rather more ... searching ... for the forms on the right hand side so I think it might be best to split the two sides into separate sections and that would allow for the usual other forms to be included without implying more hassle need be gone through. The separate section could then just be titled something like forms using the reduced Placnk constant without getting into when they are ever actually used. Having the table is a problem at the moment. Dmcq (talk) 13:15, 3 November 2011 (UTC)

NOFI, but your suggestion that the forms using hbar, hardly get used is a bit laughable. Especially the form using hbar and ω is fairly common in theoretical physics books. For good measure these books might also replace 1/(kT) with β.

As for the table, I think it may be best to just present it as a single column. Other forms that are some times used can then easily be added. (We can also get rid of the angular wavelength one, unless somebody provides a reference actually using this.)

As for the sentence: "Forms on the left are most often encountered in experimental fields, while those on the right are most often encountered in theoretical fields". That is indeed problematic. It could be approximately true, but it is also almost impossible to source reliably.TR 14:19, 3 November 2011 (UTC)

I did not say the hbar forms were not used or even were hardly used, I did not mean to put back the wording of 'common forms' as a heading just split the two. I agree the hbar and omega form does occur quite often. However the k form only has one found instance and not even using k. A single column would also get rid of the problem of forms being calculated to fill the empty spaces, or else just have text with formulas interspersed as there was before. On actual usage I tried a quick couple of google searches of Planck's law and looked at the first few pages. In the web search no use of hbar came up in the first three pages, I didn't look further, except for very first top entry which was Wikipedia. In the books search I finally came across a use at the bottom of the second page. I thought separating the two would also make it more accessible to people who use just one or the other. Dmcq (talk) 14:30, 3 November 2011 (UTC)

Regarding Especially the form using hbar and ω is fairly common in theoretical physics books, this is very helpful. What are some example books using that form? I've been unable to find a single one.

Ashcroft & Mermin's Solid State Physics for example. Or Reif's Fundamental of Statistical and Thermal Physics for another. These are dime a dozen. Headbomb {talk / contribs / physics / books} 18:51, 3 November 2011 (UTC)

I would suggest removing the two standard forms from the table and replacing them by other alternative forms. I would suggest Planck's law in terms of energy. In astrophysical context like radiation transport in stars this is used, although the convertion from the frequency form is, of course, trivial. Another one could be Planck's law in terms of momentum. Count Iblis (talk) 15:37, 3 November 2011 (UTC)

A short explanation in the article of why each different form is used would be good. Dmcq (talk) 15:53, 3 November 2011 (UTC)

I agree with Count Iblis. There are nontrivial reasons for why the two forms $B_{\nu }$ and $B_{\lambda }$ should be listed separately, in particular because they along with $\lambda B_{\lambda }$ are the three natural shapes for Planck's law as depicted in Figure 2.3 on p.29 of Goody and Yung, with exactly those three names. All other forms are merely constant multiples of one of these three, hence the same shape with the same Wien peak, there are exactly three peaks. No one gives them separate names like $B_{\omega }$ or $B_{k}$ (so Wikipedia should not be implying that they do).

Dmcq's suggestion is also good. The choice of multiple is governed solely by the desired spectral units for each of ν or λ input and dν or dλ output (usually the same units, though the MATLAB software I'm using treats them as independent). Units for Planck's law that we're likely to find examples of would include microns and nanometers for wavelength, and for frequency cm⁻¹ (33 GHz or 0.29 K temperature), radian/sec (0.16 Hz or 1.6 picoKelvins), electron-volt (242 THz or 2000 K), joule (1.51×10³³ Hz or 14 zettaKelvin, very hot!), etc. (Temperatures are for the Wien peak $\lambda _{\max }$.)

A table of spectral units actually encountered in the field, giving for each whether to use $B_{\nu }$ or $B_{\lambda }$ and the constant factor to multiply it by, would be more useful than a long list of formulas, which only burden the article unnecessarily with trivial variants of the two main formulas that no one in their right mind is going to bother with when the alternative is just to have a constant times one of $B_{\nu }$ or $B_{\lambda }$.

In the unlikely event that every used frequency unit had a matching wavelength unit the table could have two columns pairing them up, but if not then just a list of units and their associated constants ought to be enough. Most units arising in practice seem to be for frequency (with quite a variety), with wavelengths it's mainly just microns and nanometers for temperatures in any practical range (14 zettaKelvin doesn't sound like a practical temperature). --Vaughan Pratt (talk) 18:48, 3 November 2011 (UTC)

• If the only sticker here is the word "common" forms, then I got no problem changing that to "different" forms.
• Agree energy should be listed as well
• There's no reason to have tables with holes in it. Reduced wavelength is one of the least used quantities, but there's no reason to not present it alongside the other forms using hbar. We should aim to be comprehensive in our coverage.
• νB_ν are just silly forms to list. If you want νB_ν, take B_ν and multiply by ν.
• Re: Count Iblis. When would someone express radiance in terms of momentum? Headbomb {talk / contribs / physics / books} 19:03, 3 November 2011 (UTC)

I did not have to look at the signature of this comment to tell that it is was from Headbomb. All of Headbomb's earlier supporters would appear to have fallen by the wayside. Headbomb needs to be addressing them, not us. As a lone duck he's a sitting wolf, or however that goes. --Vaughan Pratt (talk) 07:51, 6 November 2011 (UTC)

We started with a fair range of common forms, including one for spectral radiant emittance. And with a distinction between radiation from a surface and radiation in a cavity. And all fairly well referenced. And in a format that facilitated commentary on the individual forms. Now we have a table format that makes it hard to know whether and how the various forms are referenced; and that tablulates forms of dubious provenance included merely for the sake of "comprehensiveness"; and that does not facilitate commentary on individual forms; and that reasonable editors are asking, perhaps rightly or wrongly, to make exceptions from, or even to entirely reconstruct; and that does not offer a way of expressing a distinction between radiation from a surface and radiation within a cavity; and that does not allow for a distinction between spectral radiance and spectral emittance.

On the one hand, conversion from one form to another is so elementary that a schoolchild could do it and it is by no means a non-routine calculation, so that only one form is really needed, together with the rule for conversion; on the other hand, the present table shows six selected from the many forms which could have been obtained from one form by said schoolchild. The new table format is, to some objectionable extent, inflexible and a product of a one-size-fits-all mentality. And perhaps the new table format may require specialist expert supervision (we do not have an outcome from the query at the noticeboard, and so do not know exactly where we are now.) The change to the new table format, though it has taken up much editorial time and talk, is perhaps hardly an improvement.Chjoaygame (talk) 20:40, 3 November 2011 (UTC)

My sentiments exactly about getting rid of the table form and having explanations. However converting between the forms was not totally elementary and Headbomb amply demonstrated. Even so Wikipedia is for summarizing what is actually out there, not for sticking in our own calculations just because we feel they should be in whatever about what the textbooks say. What is in should be reasonably close to what is in commonly used textbooks, we don't have to scrape the barrel for a subject like Planck's law and it would be far too large an article if every last thing said about it was included. If people do want an unusual form it isn't our job to help them out just because they are liable to make a mistake in the substitutions, we should just summarize what is the general state out there. Dmcq (talk) 00:42, 4 November 2011 (UTC)

Caniou has a table with 10 different forms. I added a link in the citation, to the relevant book pages. He omits using wavenumber k as an independent variable, possibly to avoid dealing with the notational inconvenience. If the only source we have for that one is the obscure Calcutta journal, I agree let's not use it. If there's something more mainstream, or evidence that someone uses it in that form, we should consider adding it; we can also use something other than k for wavenumber, so we can still with k for Boltzman's constant, as many sources do. I also agree with Vaughn Pratt that there are really only 3 relevant shapes, and the rest is just scaling. Maybe we could show a curve, temperature-parameterized family of curves, for each, with a couple of different axis labels, or text that explains what some alternative axis labels would be. The lead image presently doesn't have a useful y-axis label; doesn't even say if it's proportional to intensity per wavelength, or intensity per Hz, or what. Dicklyon (talk) 03:33, 4 November 2011 (UTC)

I would stick with using k_B regardless whether we show the form in terms of reduced wavenumber/momentum. The potential for confusion is there nonetheless, since the use of k as the label for wavenumber is very common. Textbooks avoid this by using a consistent set of conventions across the book. Across wikipedia there never will be consistent set of conventions, and a reader has to get used to new conventions every article. It is therefore helpful if the notation is as unambiguous as possible. (This similar to the reason that in most cases we do not present formulas in natural units, even if that is the most common form in the literature.TR 08:07, 4 November 2011 (UTC)

It is therefore helpful if the notation is as unambiguous as possible. I don't know about anyone else, but personally I find it hard to take this seriously unless you're willing to change k to k_B in the Boltzmann constant article in order to make the notation as unambiguous as possible. Are you? --Vaughan Pratt (talk) 08:23, 12 November 2011 (UTC)

It might be worthwhile putting in more about the dimensions, lots of books say something about them. Does Wikipedia cover changing the dimensions well? I notice the article just glancingly mentions steradians and it probably would be a good idea if an article covered the business of factors of that in Planck units with dimensions properly. I've always liked giving the tools to do a job if possible. Dmcq (talk) 10:27, 4 November 2011 (UTC)

Headbomb, I don't know if momentum is ever used (you could imagine that in some contexts it would appear when radiation pressure becomes relevant, but I haven't seen it). If it isn't then, of course, there is no point in including this. I'm still in favor of k instead of k_{B}, but I don't feel that strongly about it. If I were writing an article like this myself, I would prefer to make the formulas that appear the most frequent carry the least baggage and look the best. That would motivate me to use k, or even better, put 1/(k T) = beta. That would be the first priority for me. Then, if I also wanted to use a wavevector form and I had already used k for Boltzmann's constant, then I would just use q for the wavevector. Count Iblis (talk) 15:30, 4 November 2011 (UTC)

There's several ways to do things, but the usual way in thermal/solid state physics is using k_B for the Boltzmann constant. Using β is just moving the issue at a different place, since we'd also need to say that β = 1/k_BT. Headbomb {talk / contribs / physics / books} 10:11, 5 November 2011 (UTC)

Clearly Headbomb (HB) should take a break from editing Planck's law and turn his attention to Boltzmann's constant, where all those occurrences of k urgently need to be changed to $k_{B}$.

Until HB arrived at this article three weeks ago, the article's notation k for Boltzmann's constant had conformed to that of Wikipedia's article on the constant. For reasons that no one including HB has explained, he changed it to $k_{B}$.

There was an immediate outcry prompting several people to change it back. This got HB's back up, and he resolved to ensure that the Planck's law article would henceforth use the notation he, HB, wanted, and not the notation the article's editors, or the editors of Boltzmann constant, preferred. If one were to use HB's preferred profane idiom to summarize HB's reaction to this attempted reversion of his notational change, it would be "f*** you all, I own this article." (Not all commentators appear to have noticed this right away, even physics seems to have its Neville Chamberlains. At the risk of violating some Wikipedia guideline I would analyze this as Chamberlain:Churchill::England:Australia.)

For HB to get his way clearly required strategy on his part. On 13 October Headbomb replaced all occurrences of ${\tilde {\nu }}$, the standard symbol for wavenumber defined as ${\tilde {\nu }}=1/\lambda$, by k, without making any other changes whatsoever, nothing involving π for example. With this substitution of k for ${\tilde {\nu }}$ he could then argue for a notational conflict that would prevent anyone changing $k_{B}$ back to k.

Failing to notice he'd made this change, but noticing the requests to change $k_{B}$ back to k, 90 minutes after HB's substitution of k for ${\tilde {\nu }}$ I made the requested change.

Immediately HB reverted my change on the ground that it would conflict with the other use of k. I had to admit, this is ingenuity we can all admire.

But there was still a problem. HB appeared to be unaware of the distinction between wavenumber and angular wavenumber, and had naively assumed that all references to "wavenumber" were to the non-angular kind.

When this distinction was pointed out to him, in order to prevent the impending change of $k_{B}$ back to k, his next project was to find a way in which using k for "wavenumber" was consistent with current usage, so that no one could change $k_{B}$ back to k.

This turned out to be more challenging. Eventually he figured out that if one took k to be $2\pi /\lambda$ instead of $1/\lambda$ as he'd originally proposed, people would stop objecting.

One might assume that he'd finally understood the distinction between "wavenumber" and "angular wavenumber," the latter term appearing over thirty times on the talk page. Perhaps he did, though this is not apparent from the talk page since not a single use of this term there was by HB, who has scrupulously adhered to the unqualified term "wavenumber" in all discussions of this controversy.

All that mattered for HB was that replacing the equation $k=1/\lambda$ by $k=2\pi /\lambda$ turned out to stopped the complaints.

But now HB faced yet another problem. People were now complaining that no one had ever expressed Planck's law in terms of angular wavenumber k. They accordingly deleted his invented notation $B_{k}$ and the formula for it that he'd finally gotten right after several tries.

With so much effort invested in this formula, naturally HB reverted these deletions. And with so much opposition to them, naturally they were reverted back.

Now HB faced yet another problem, the WP:3RR rule. (I've totally lost count of the number of problems HB has faced to ensure the survival of the notation $k_{B}$ in this article.) It's fine to go on edit warring like this up to a point, but eventually Wikipedia gets tired of this sort of thing and says no to it.

So, what to do? Well, HB is nothing if not resourceful. Taking a leaf out of Greek history, he constructed an elaborate Trojan horse consisting of a table of six variants of Planck's law, one of which was the $B_{k}$ formula that he was so determined to get back into the article so that no one could change $k_{B}$ back to k.

HB's Trojan horse worked! Everyone was so grateful that these six formulas now clarified the previously obscure Planck's law that they failed to notice that HB had succeeded in getting $B_{k}$ back into this article for the third time.

While I admire HB's resourcefulness, I am unable to say that I admire what he has done to this article. Looking over his edits since he started here three weeks ago, and the profanities with which he responded to the objections to them, any improvement to the article would be comparable to how Genghis Khan improved China by wholesale massacres.

I leave to ANI the question of whether HB's wholesale massacres of the extant editors of this article constitute a net benefit to Wikipedia. Given that those editors clearly know what they're talking about, and HB clearly does not, I would be inclined to vote in the negative. If Polyamorph believes the opposite as he has implied earlier he needs to join this discussion or his vote may carry less weight at ANI (not that I have a clue about ANI politics, mind you, which are far above my pay grade as a mere editor of technical Wikipedia articles).

On Wikipedia, everyone can edit, but that doesn't mean everyone should. --Vaughan Pratt (talk) 05:46, 6 November 2011 (UTC)

deceiving the reader

Unproductive discussion

The stimululs to Headbomb's construction of the table of forms was that other editors thought that the angular wavenumber form was uncommon and unsourced and should therefore be removed from a section about common forms. The table provided a kind of shielding for the angular wavenumber form. Some time ago I proposed a remedy of changing the section heading from "Common forms" to a more general name that would not demand commonness, so as to allow Headbomb's beloved angular wavenumber form. My suggestion was not taken up. Later I actually edited the section heading to simply "Statements" with the same intention. Headbomb reverted my change, which if it had stood would have allowed his beloved angular wavenumber form. I am not exactly sure why. But now he has permitted a change to "Different forms" which has the same effect. It is clear that he wants his beloved angular wavenumber form to stand, and that he now sees that this will help that. The text that Headbomb wrote, in the section that introduces the table of different forms, ends with the following sentence: "Forms on the left are most often encountered in experimental fields, while those on the right are most often encountered in theoretical fields." The purpose of this sentence, combined with the table format, is to endorse the entry of Headbomb's beloved angular wavenumber form. When, for the sake of accuracy, I edited the words "most often encountered" to "may be encountered", Headbomb reverted my edit. He insisted on his wording "most often encountered". The sentence he wrote and on which he insisted, however, has the effect of deceiving the reader. For the angular wavenumber form is not often encountered, while the sentence says it is often encountered. The only source we have for the angular wavenumber form is one journal article, not a secondary source such as is preferred for reliable sourcing, in which the form angular wavenumber form is mentioned as an academic possibility, but not actually used. True, "encountered" covers mention as well as use. But the "often" is false. Making a false statement to the reader, for an ulterior motive such as endorsing a beloved form, is deceitful because of its ulterior motive. Deceitfulness is perhaps permitted by Wikipedia policy, I do not know, I am not deeply into Wikipedia policy. Whatever Wikipedia policy, and no matter how clever or diligent Headbomb might be in Wiki-justifying his edit, it is not good to deceive the reader.Chjoaygame (talk) 12:47, 4 November 2011 (UTC) I agree with you overall. However I also think people who try and systemize things like Headbomb can be useful. The main overall problem I see here is the lack of civility and unwillingness to consider other people may also be right. Systemizing has a real problem like here where it has led to throwing away actually used formulae, putting things in tables where they can't be described easily, and filling in the gaps with stuff just to make up a pretty table. You can see the problem elsewhere with people insisting in filling all the entries in a bio template so they stick in religion ethnicity and nationality for instance when nobody has shown the slightest interest in them in any secondary source. We really need to try being more civil and dealing with the actual content of the article more and not respond to insults. It may not do the job but it has a better chance than the alternative. Just think 'Wikipedia is the encyclopaedia anyone can edit' and stop worrying. Dmcq (talk) 14:25, 4 November 2011 (UTC) I'm not concerned about deception here, but stating that something is "common" is seldom verifiable; we would be better to not. As for filling in a table of possbilities, we need need to negotiate how far to go. Going way beyond what's in sources is probably not a good idea, but collecting things from sources into a logical organization seems OK. Dicklyon (talk) 19:24, 4 November 2011 (UTC) Thank you Dicklyon for letting us know that you are not concerned about deception here. The description of the table is deceptive because it makes the false claim that the angular wavenumber form is often encountered. My edit to remedy that was reverted by the editor who wrote the false claim. Is it part of Wikipedia policy of "not responding to insults" that this allows the false claim to stand?Chjoaygame (talk) 19:41, 4 November 2011 (UTC) I think you are being unnecessarily negative and dramatic by calling it "deceptive" and a "false claim." Just fix it while trying to get along, and maybe you'll have more luck. Dicklyon (talk) 21:21, 4 November 2011 (UTC) Thank you for your advice, Dicklyon. I did try to fix it but was reverted. I do not wish to enter an edit war, but neither do I wish to let it pass unnoticed.Chjoaygame (talk) 00:51, 5 November 2011 (UTC) From the various discussions, I think it's clear that it won't pass unnoticed. Thanks. Dicklyon (talk) 01:17, 5 November 2011 (UTC) You people really are something else. Would it be possible to stop trying to actually read what is written for two seconds, rather than misinterpret things in the least charitable way possible? If you actually bothered to READ WHAT IS WRITTEN you'd see the sentence is about where one would encounter these forms. It says absolutely nothing about the absolute frequency of occurrence, it's a basic breakdown of where each form is more popular. Experimental physicists don't usually work with hbar. Theorethical physicists often do. Thus when you encounter an hbar, it's probably because you're dealing with theoretical considerations. Concerning the header, the sticking point seemed to be the word "common". And you know what, the word "common" isn't there anymore. Everyone should be happy, but no, you insist on bringing resolved issues back on the table, and somehow shoehorn this into some convoluted scenario that I'm "manipulating the reader", or that I'm being disruptive and "wikilawyering" by agreeing with you on something. If I disagree, I'm disruptive. If I agree, I'm disruptive. Woop-dee-fucking-doo. Headbomb {talk / contribs / physics / books} 10:28, 5 November 2011 (UTC) Putting things in capitals is called shouting. Dmcq (talk) 10:51, 5 November 2011 (UTC) In hindsight, seeing the course it took, I tried to remove this section, but it seems that is not considered a reasonable thing to do. So now instead of removing it as I would like to do, I will say I am sorry I started it, and that I now see I made a mistake in doing so.Chjoaygame (talk) 22:25, 5 November 2011 (UTC) I'd have ulcers if I worried like that! Dmcq (talk) 22:41, 5 November 2011 (UTC) Hiding this section benefits Headbomb to the extent that it reduces the visibility of his profanity. Do we have any clear goal here? --Vaughan Pratt (talk) 07:37, 6 November 2011 (UTC) Sometimes it is said that all publicity is good publicity. Therefore I don't think its prominence or hiding would necessarily have a predictable effect. My reason for wanting to suppress this section is that it turned out to be unproductive, and that its obvious appearance would tend to waste people's time. That would make my goal to avoid unproductive waste of people's time.Chjoaygame (talk) 08:31, 6 November 2011 (UTC)

Kramm and Mölders

This Kramm and Mölders article has been mentioned and commented on, and perhaps another comment may be justified. I am working from the arXiv version. One editor here has used the phrase 'crackpot physics', in a way that does not quite make clear what is intended by it.

The article in question is not manifestly and unequivocally crackpot. Relevant to our present Wikipedia article, I think it makes a mistake, but not a crackpot mistake. It seems to assume that Planck's oscillators have quantized energy states. This interpretation is common enough in the literature, I think, but I think it is not justified by the historical sources, and I would not regard Kramm and Mölders as a reliable source for it. In general, there is in nature no special justification for this mistake.

I may also comment that Kramm and Mölders do not use the eponym 'Planck's relation'. In fact they write: "Thus, it is not surprising that the basic relationship Planck referred to was the Boltzmann equation he proceeded to write in the form (Planck [1901]) S_N = k log W + const. ; (74)." My observation is that they are not alone in linking Planck's name to this famous formula. For example, Kuhn 1978 on page 98 writes: "Planck, who must have discovered the combinatorial definition in Sections 6 and 8 of Boltzmann's Gas Theory, appears to have been the first man other than its author to acknowledge even its existence." Planck can be credited with the invention of "Boltzmann's constant" that appears in the equation cited by Kramm and Mölders.

Kramm and Mölders, in their Section 4, distinguish the usage of our ${\tilde {\nu }}$ from our angular wavenumber k on the basis not of the difference between "experimental" and "theoretical" fields, but instead on the basis of the difference between "spectroscopy" and "physics".

Coming now to the sharp business end of the matter. Carefully reading Section 5 of Kramm and Mölders, I would summarize it for the present purpose as very eccentric and very confusing, and probably partisan (in a pejorative sense), in its carefully though oddly defined usage of language, rather than simply as "crackpot". But the section does provide further very good reason, relevant here, that this Kramm and Mölders paper is not an unreservedly reliable source, for the present purpose more likely not a reliable source at all.Chjoaygame (talk) 06:58, 6 November 2011 (UTC)

Fair enough. I don't know where the boundary for "crackpot" lies, so let me just repeat that everything in the K&M article that is true is not new (if we accept that "spectroscopy" is not "physics," which my biology and linguistics colleagues might dispute), and everything that is new is not true. In particular the only novelty I found was the argument supporting Gerlich and Tscheuschner's thesis that the Stefan-Boltzmann constant was not a universal constant. Whereas G&T argued this on geometric grounds, K&M argued it by computing the constant from the Rayleigh-Jeans law. That may well be another constant, and even a "universal" constant if the the Rayleigh-Jeans law were to be taken seriously, but it is not the Stefan-Boltzman constant, which is 5.6704 when the temperature is given in units of hectoKelvins (so 288 K = 2.88 hectoKelvins). Scale appropriately for Kelvins. --Vaughan Pratt (talk) 07:33, 6 November 2011 (UTC)

You two never cease to amaze me about just how much wrong one can be. A few things.

1) The paper is used to source the B_k et al. formulas. It has absolutely nothing to sourcing that E = hv is known as the Planck relation. Why you bring in S_N = k log W + const. is something beyond me. Or why you keep brining in Kuhn.

2) It is not uncommon, and certainly not looked down upon for an editor to publish an article in a journal, if that journal is a society's journal. This is a paper offering historical perspectives on Planck's law, and really contains little material that can be considered original thought. There is no claim of discovery, nor any claim of novel result either.

3) The physics of the K&M paper are completely mainstream. Regarding the constant, K&M write "As recently pointed out by Gerlich and Tscheuschner [2009], the Stefan constant is not a universal constant of physics." If you read G&T2009 (arXiv:0707.1161/doi:10.1142/S021797920904984X, hardly a mainstream article, but that is irrelevant to this particular claim), you'd see that G&T2009 refer to geometrical considerations, to quote "The constant σ appearing in the T⁴ law is not a universal constant of physics. It strongly depends on the particular geometry of the problem considered.[8]" with [8] reading "For instance, to compute the radiative transfer in a multi-layer setup, the correct point of departure is the infinitesimal expression for the radiation intensity, not an integrated Stefan-Boltzmann expression already computed for an entirely different situation." Again, completely mainstream. We say the exact same thing in our article on the Stefan-Boltzmann law.

"An alternative form of the Stefan–Boltzmann constant, more fundamental to physics:

$\sigma ={\frac {\pi ^{2}k^{4}}{60\hbar ^{3}c^{2}}}$

Finally, this proof started out only considering a small flat surface. However, any differentiable surface can be approximated by a bunch of small flat surfaces. So long as the geometry of the surface does not cause the blackbody to reabsorb its own radiation, the total energy radiated is just the sum of the energies radiated by each surface; and the total surface area is just the sum of the areas of each surface—so this law holds for all convex blackbodies, too, so long as the surface has the same temperature throughout."

You can knock off K&M for a lack of clarity and poor wording in what they meant, but what they meant is indeed mainstream physics [even if they chose an incredibly poor reference for it].

Headbomb {talk / contribs / physics / books} 08:23, 6 November 2011 (UTC)

K&M obtain 1.75×10⁻⁸ for the Stefan-Boltzmann constant (a universal constant of physics whose value is 5.67×10⁻⁸) by integrating a law substantially different from Planck's law. Your statement that the Wikipedia article on the constant says "the exact same thing" says the exact same thing about your physics expertise that I've been saying for three weeks. I find it ironic to be told by an administrator that I'll be blocked for saying something as obvious as this, when your insults, profanity, inattention, and repeated reverts get a free pass in that regard. --Vaughan Pratt (talk) 23:02, 6 November 2011 (UTC)

As to (1), the reason for bringing in the formula E=hν is that this paper might be seen as a relevant sample in which that formula is not called 'Planck's relation'. This paper is very prolific in the use of the phrase "the Planck function", referring to his law, but does not use the phrase 'Planck's relation'. The reason for mentioning the other formula is that this paper comes very close to calling it 'Planck's relation' and that it is not the formula E=hν . The provenance of the eponym 'Planck's relation' is indirectly relevant to the immediately present discussion.

As to (3), the relevant question here is not whether an idiosyncratic terminology is compatible with the thinking of mainstream physics, but whether an article that generates it is a reliable source for a particular purpose. The grounds that the article uses for its proposed statement that the Stefan-Boltzmann constant is not "universal" are definite grounds for making it not a reliable source for a Wikipedia article that used such a statement.

More to the point, what matters here is not the things that are discussed in your comment, but rather is something which is not explicitly addressed in your comment, namely, whether for the present purpose this Kramm and Mölders paper is or is not a reliable source.Chjoaygame (talk) 09:12, 6 November 2011 (UTC)

Correct me if I've misunderstood, but we seem to have at least Chjoaygame, Q Science, Count Iblis, Dmcq, and myself on one side of this edit war (any others?), and Headbomb on the other (any others?). One should not show up at ANI without one's ducks in a row. Bearing in mind that ANI is generally not interested in subject matter disagreements per se, do we have a consensus on the relevant non-subject-matter facts to allow us to make a case to ANI for our desired action? Even if Headbomb is the only one with his facts straight, that's not going to make any difference at ANI, where they are unlikely to care whether the mantissa of the Stefan-Boltzmann constant is 5.67 or 1.75, but only whether the minority opinion in that and other matters is disrupting the majority. --Vaughan Pratt (talk) 23:02, 6 November 2011 (UTC)

I think that PAR (talk) sides with Headbomb. Q Science (talk) 23:20, 6 November 2011 (UTC)

And I'll be happy to side against both sides, for their general pigheadedness. Dicklyon (talk) 23:23, 6 November 2011 (UTC)

Touché! We should all be blocked. ;) --Vaughan Pratt (talk) 01:02, 7 November 2011 (UTC)

It seems to me that we do not have just one unequivocally definite, concise and precise statement of what we think should be done and why. I think such a statement would be needed in order to have all our ducks in a row.Chjoaygame (talk) 00:03, 7 November 2011 (UTC)

Agreed. Contributions solicited. --Vaughan Pratt (talk) 01:02, 7 November 2011 (UTC)

Possible simplification of the table

Those areas using hbar for which Planck's law is relevant presumably involve interaction of thermal emission with matter. As I've pointed out above, e.g. at 18:38, 18 October and elsewhere, this article treats simple thermal emission, where angular units complicate the formulas unnecessarily with no benefit.

Here's my understanding of what determines whether to use periodic or angular units (2π radians being one period) in applications of Planck's law. If anyone has a different understanding I'd be very happy to compare notes.

Personal preferences don't appear to play a big role. Rather, when thermal emission interacts with matter, a wavevector perspective becomes important and trig functions enter, often in the form of exponentials of complex variables. In this case angular units are more convenient to work with because the derivative of $\sin _{\omega }(\omega )$ (what we normally call sin, with period 2π) is simply $\cos _{\omega }(\omega )$, whereas the derivative of $\sin _{\nu }(\nu )$ (the version with period 1) is the more awkward $2\pi \cos _{\nu }(\nu )$ (1 is less than 2π so $\sin _{\nu }$ has to rise faster to cycle through a period).

Absent interaction with matter, trig functions don't arise. In that case it's much simpler to use periodic units which avoids all those powers of π in the denominator and the exponent of e.

That said, I see no harm in pointing out the existence of contexts where angular units are preferable, and to simply state that the angular counterpart of $B_{\nu }$ is obtained by making the obvious substitution, namely $\omega /2\pi$ for $\nu$, and dividing the result by $2\pi$ (since the spectral quantity B is a derivative). That's a lot easier to grasp, and is also more explanatory than forcing the readers to infer this simple connection for themselves by staring at a wall of formulas.

If we find reputable sources for Planck's law in terms of either angular wavelength or angular wavenumber those can be handled the same way, always remembering to divide by $2\pi$ after the obvious substitution.

Is this better or worse than the current table? (No need for Headbomb to answer this, we know his answer already.) --Vaughan Pratt (talk) 08:10, 6 November 2011 (UTC)

And as has been repeatedly pointed to you, the angular units only "complicate" the formulas in your mind. They should be present, because they the hbar version represent a widely-used theoretical framework, and you cannot exclude them on the basis that "thermal interaction with matter" is unimportant to Planck's law. And what in the world are you talking about with your $\sin _{\omega }(\omega )$ and $\cos _{\omega }(\omega )$ or that trig mumbo jumbo? And, again, as repeatedly pointed to you, the sources have been provided. So please stop your disruption.Headbomb {talk / contribs / physics / books} 08:31, 6 November 2011 (UTC)

The formulas in the right column are more complex than those on the left on two counts: (a) each formula on the right has two more symbols than its counterpart on the left, and (b) even though that's only two symbols you took hours to get them right when it was obvious from the outset what they had to be: merely substitute $k/2\pi$ for ν̃ in the formula for B_ν̃ and divide by $2\pi$ (because B is a derivative). Why are you so dead set against simply giving that as the uniform procedure for deriving anything on the right from the left instead of carrying it out three times and making a huge table out of it? And why are we selectively omitting some of the entries in Caniou's table, each of which has its angular counterpart? (Oh, we answered that yesterday on the talk page.)

Don't worry about the "trig mumbo jumbo," it was intended for those who might be interested in factors bearing on which version of Planck's constant to use in Planck's law. If you're interested in such factors I'd be happy to clarify $\sin _{\nu }$. (I'm assuming you're already familiar with $\sin _{\omega }$.) But I'd be very interested to know what you find disruptive about explanations of this kind. --Vaughan Pratt (talk) 00:40, 7 November 2011 (UTC)

The formulas are as complex on both sides, unless that constants someone confuse you. It takes 30 seconds at most to switch between the two forms. If it takes you hours to perform, it really just show how completely unqualified you are to comment on the article. Headbomb {talk / contribs / physics / books} 09:45, 8 November 2011 (UTC)

So far, no one has produced a single reference with a graph using that form. Though there are perhaps 2 references that show the form (but with different symbols), I have yet to see a reference that actually uses that form. As a result, the angular forms are not notable enough for inclusion here. Q Science (talk) 20:42, 6 November 2011 (UTC)

Moving the goalposts as usual. If you want one for graphs, see e.g. doi:10.1142/S021797920904984X. The paper itself is shit, but it shows that they are used. Headbomb {talk / contribs / physics / books} 09:41, 8 November 2011 (UTC)

When I clicked on the link I got the following message:

"The DOI you requested --

http://dx.doi.org/10.1142/S021797920904984X

-- cannot be found in the Handle System.

Possible reasons for the error are:

the DOI has not been created

the DOI is cited incorrectly in your source

the DOI does not resolve due to a system problem"

It is probably not good to use a paper of this kind as a Wikipedia source.Chjoaygame (talk) 11:07, 8 November 2011 (UTC)

Fixed the DOI, and the paper not used as a source, so stop misrepresenting the issue. Headbomb {talk / contribs / physics / books} 11:19, 8 November 2011 (UTC)

The common hbar versions shold be presented but I don't believe they should be stuck in a table with the others. That table form causes problems as seen by the reams of discussion here. They don't occur often in this context but they do occur especially in introductory texts where the main impetus is elsewhere and the authors don't want to confuse things with different units. The target for Wikipedia includes introductory descriptions as summarized from reliable sources so the hbar forms that are actually used often should be in. Forms that aren't even vaguely common should not be included , we're not here to publicise what editors think should be in just what is out in the field with WP:DUE weight. Dmcq (talk) 12:58, 6 November 2011 (UTC)

And besides, giving the (very simple) procedure for deriving the angular counterpart of any unit seems to me preferable to carrying it out three times, making a massive table out it, and leaving the reader to guess how to do it for say Caniou's photon energy function (the fifth function in his Table 4.1 on pp.116-117). If Headbomb took hours to do that calculation in the case of wavenumber, it seems unfair to expect readers to do it without assistance. (Gosh, are we just repeating ourselves here? Maybe all this needs to be collected in one place and indexed for easy reference.) --Vaughan Pratt (talk) 00:53, 7 November 2011 (UTC)

Standard photon

The Planck law is used to define a two part photon structure Talk:Photon standard in [4].(HCPotter (talk) 10:20, 6 November 2011 (UTC))

issues

There may be two kinds of issue in play here: (a) what should be in the article; (b) how should editors edit and talk, that is to say, how should editors decide what should be in the article. The direct question of (a) what should be in the article is not one for administration. Administration is about (b) how should editors decide what should be in the article.Chjoaygame (talk) 04:53, 7 November 2011 (UTC)

What should be in an article is what is notable, in reliable sources related to the subject, and with due weight according to the best sources. And a good part of it has to be nice and readable for someone who knows just about enough to start reading it.

If there is a lot of material then an article can be split by having articles on subdivisions but the main article should still be a reasonable cover for the whole thing. After 50k like this article this becomes a thing to think about and at 100k it is something that starts getting recommended. Dmcq (talk) 15:04, 7 November 2011 (UTC)

As I think this over, I get the idea that administration is not likely to be able to deal with our current problems. It is more likely that a solution may be found if we work things out together here.

The internet way of communication has its advantages and disadvantages. I think it will be useful to reflect on those, a reflection that I think each of us can make for himself.

Physics is a human undertaking, and only very few humans are perfect, the present company excepted of course, because here we are all perfect. But for the rest out there in outer cyberspace and in other even more remote reaches of the universe, there are many opinions, some of them more and some less established than others. Fortunately, each of us knows perfectly which exactly of those various opinions is perfectly and uniquely right and correct. That is why we are better qualified than others to edit the Wikipedia. I do not fit in here, being the author of silly, stupid, and obscurantist edits, requiring sighs, and that I be called a troll, and sometimes more direct expletion.

There is a feeling of aggrievance in the air here.

Vaughan Pratt at 15:05, 9 October 2011 expressed a concern that a long-standing paragraph was preachy and redundant. A brief look shows that the preaching began before May 2007 (I haven’t chased it right back.) and kept growing. He has since contributed material that bears on the same questions.

Perhaps the current round of edits may be dated from Vaughan Pratt’s expansion of the previously minimal lead, at 20:40, 9 October 2011. He used a rather unusual form of the law in his new lead. He added a section "Introduction" at 05:34, 13 October 2011, which was quickly deleted by PAR.

Dicklyon is concerned that some of us are generally pigheaded, and he will be happy to side against both sides.

IRWolfie, at 20:23, 16 October 2011, felt aggrieved, in that it was "silly" to include in the article some material about the concept of a photon, primarily referenced to Willis Lamb. IRWolfie felt that he recalled that that Loudon had mentioned that Willis Lamb was expressing an extreme point of view, that should not be represented in this article, an article in which the origin of the concept of photon is relevant. If IRWolfie had checked his source he would have found that Loudon wrote: "It is difficult […] to disagree with some of [Lamb’s] concerns." Presumably IRWolfie also felt that Mandel and Wolf and others were silly to express the same thoughts as did Willis Lamb.

Headbomb feels aggrieved perhaps more than others of us. He entered the present game at 21:50, 12 October 2011, restoring the older $I$ notation in place of the new $B$ notation. On 13 October 2011 he began a more vigorous editing exercise. He feels that some of us try our best to make the conflict personal and take every opportunity to depict him in the least charitable light possible. At 15:25, 16 October 2011 he felt patronized by Vaughan Pratt, and thought that Vaughan Pratt ought to be blocked on grounds of civility. At 05:56, 17 October 2011 Michael C. Price called Headbomb a "nutter". At 06:20, 17 October 2011 Headbomb wrote that Vaughan Pratt was busy attacking him at every opportunity. At this point, at 06:28, 17 October 2011, Polyamorph closed an acrimonious discussion.

Perhaps this is enough of a wall of text from me for the present.Chjoaygame (talk) 22:21, 7 November 2011 (UTC)

How does this help us to improve the article? PAR (talk) 05:08, 8 November 2011 (UTC)

It does not. It's a wall of text for the purpose of a wall of text. E.g. what does it matter that I "[re]introduced the I notation". It's not used in the article now. No one is arguing to include it in the article now. It has no purpose, other perhaps than to again depict me in unfavourable light. Headbomb {talk / contribs / physics / books} 09:49, 8 November 2011 (UTC)

The purpose of my sentence "He entered the present game at 21:50, 12 October 2011, restoring the older $I$ notation in place of the new $B$ notation" was to say when and how you entered the present game.Chjoaygame (talk) 11:30, 8 November 2011 (UTC)

Chjoaygame, regarding your I get the idea that administration is not likely to be able to deal with our current problems, if your take on those problems is what I've understood it to be then I would have to agree with you. Where I would disagree with you is whether there exists any take on our current problems that ANI would find easy to deal with.

I think there does.

Put yourself in ANI's shoes. If someone comes crying to ANI that they're being maltreated on Wikipedia, they'll say they've heard that before and ignore them. But if that person comes with a bunch of people carrying signs and torches it's a different story.

Look at it this way. If I'm the only one suggesting HB be put before a firing squad then I'll just bow out and edit other articles. But if there are others concerned about the negative aspects of his impact on the article then something needs to be done about it. What you have to decide is whether you're ok with HB's modus operandi. If you like it, you don't need to do anything. If not, you need to ask yourself whether your "walls of text" will magically solve the problem, or whether you need to start considering how other people who are being impacted by HB are proposing to deal with the problem.

HB's friends are those who focus only on their own opinions. The only risk HB faces is that two or more editors might decide to work together against him. So far this hasn't happened. HB wins. You ok with that? HB certainly is. --Vaughan Pratt (talk) 06:32, 10 November 2011 (UTC)

Chjoaygame, your proposal to solve this ourselves would have been a good one two weeks ago, but given that this is what I was proposing then and it accomplished nothing, we're now in the sort of situation that ANI exists to deal with. Moreover ANI won't want to wade through a blow-by-blow account of what everyone said while the ideas were gradually taking shape, they'll be looking for a coherent description of the problem as we see it today.

One approach (by no means the only one) is for me to describe how HB's presence here is impacting my own ability to edit this article, and to see who feels I'm being unreasonable. If there's a consensus that my complaints are reasonable then this may be enough for ANI. So let me list my complaints here and see who objects to any or all of them. I try not to be as pig-headed as Dicklyon implies, so if you do have any objections I'll be happy to respond to them. I just want to get this war ended in the cleanest way possible, and I have no objections to suggested changes to the following so long as the eventual outcome is to get this article back to the normalcy of other Wikipedia articles.

1. Ever since HB arrived here three weeks ago, he's reverted all my edits on the ground that he's right and I'm wrong. He has no answer to the obvious question of how to decide which of us is right, other than to continue to insist that he is. In effect I've been banned from editing this article because he prefers to resolve differences by reverting edits whereas I prefer to talk differences through on the talk page. He has repeatedly characterized my approach as "disruptive."
2. He makes many controversial edits. He fights all attempts by various editors to revert them by reverting those attempts. Just before he reaches the WP:3RR rule he figures out some creative way around it to get his way. This is an ongoing pattern.
3. He frequently commits elementary physics errors. In order to prevent editors from reverting his $k_{B}$ notation to k he created a notational conflict by changing ${\tilde {\nu }}$ to k because he thought they were synonyms and could therefore be used interchangeably. He has no idea what Lambert's cosine law is about, claiming in the article that radiation is emitted equally in all directions from a surface element, and is unable to understand objections to his claim. He shows no respect for convention, dismissing commonly used notations such as $\nu B_{\nu }$ as "just silly forms" (thereby insulting all authors who use them) while inventing his own notations having not a single precedent in the literature. He cites crackpot literature to support his position, for example claiming that Kramm et al's proof that the mantissa of the Stefan-Boltzmann constant could just as well be 1.75 instead of 5.76 is "mainstream physics" and claiming that the Stefan-Boltzmann constant article says "exactly the same thing." There are many other instances but this is a good start.
4. He has some cognitive disorder (whether deafness or something else is hard to say) that makes an intelligent discussion with him completely impossible. He just talks right past you. Complaints about his physics errors have no impact, he insists at the top of his voice that he's right and others are wrong. If he passed freshman physics I'd like a word with his instructor.

A list of those willing to back up my complaints is all that should be needed to bring a compelling case to ANI. I don't care who's against me on this because it's HB's problem to round them up, not mine. --Vaughan Pratt (talk) 08:16, 8 November 2011 (UTC)

Vaughan, please tone down number 4. HB claims to be working on a master's in physics and we should AGF. As to why he so strongly disagrees with the rest of us, I don't know. It certainly is not because he has references to support his POV. Wikipedia is based on collaboration and HB simply ignores everyone else. BTW, my biggest problem is with the references, what he did is not something I support. His use of non-standard notation is also a big problem. Q Science (talk) 09:01, 8 November 2011 (UTC)

1. I merged several of everyone's edit with mines and constantly sought to incorporate good suggestions in the article. You can spin this however you want, but you seem to think that being reverted is evidence of you being right, or that reverts automatically imply making the article worse. This is obviously not the case.
2. I do not make controversial edits. Scaling a function by a factor of 2π is not controversial, neither in physics, nor in mathematics. That you claim is is shows just how much out of touch you are with physics in general.
3. Complete horseshit. I introduced the wavevector [and other hbar variants] because they are forms encountered in theoretical physics. You have an irrational hatred of k_B for some unknown reason, and this clouds your judgment with all others issues, and make outrageous claims just so you can somehow justify restoring k for the Boltzmann constant at the expense of clarity and comprehensiveness. I still have no idea what in the world you're referring to with your Lambert's cosine law, I can find no mention of it in the article (nor any revision within the past two weeks), and you would rather go "WAAAAAAAAAAAAAAAAAAAAAH HEADBOMB COSINE LAW!11!1 BAD HEADBOMB!1!" than propose an edit to rectify whatever problem you think there is, or even go so far as to make a damned edit. You are not interested in improving the article, you're only interested in complaining about me. νB_ν IS a silly form to report, it's just B_ν multiplied by ν, and I invented no notation.
4. K&M is not crackpot physics by any stretch of the imagination. They poorly worded for what they meant, but what they meant is mainstream. But even if that section about the SB constant was crackpot physics, IT WOULD STILL BE IRRELEVANT to the article, because the citation isn't used to make any claim about the SB constant. It's only there because you can't derive.

Headbomb {talk / contribs / physics / books} 10:04, 8 November 2011 (UTC)

Re: Q Science. I do not ignore everybody else. If I did, I wouldn't have merged several edits with mine, or taking feedback into account, etc... All my edits are support with references, you just ignore them for some reason, and the notation issue is completely irrelevant (it just follows the convention the article adopted. Calling something B_ν(T) or B(ν, T) is trivial (my personal preference is I(ν, T), but as I said, this is trivial). As long as the notation is consistent, it's all that matters. Headbomb {talk / contribs / physics / books} 10:09, 8 November 2011 (UTC)

I'm unable to agree with even one sentence in HB's list of 4 points. If someone else can (besides HB of course) I'd be happy to address that sentence, if not then 'nuff said. (Walls of text begone!) --Vaughan Pratt (talk) 02:43, 9 November 2011 (UTC)

The major thing wrong with K&M as far as this article is concerned in my view is that it has far too low weight. See WP:DUE. Topics should be dealt with giving due weight as given by the best sources. I did not see the business about the Lambert cosine law used here but I view it as more fundamental to think of it as looking the same from any angle rather than how much is emitted at each angle from a unit area. I agree wth the business that people should edit in the changes they wish more rather than just complaining. However WP:BOLD is part of WP:BRD and other people are allowed to revert and then the business should be discussed and a consensus arrived at. Can I suggest that the items of dissention in the article be listed with preferred solutions of each person? Then a decision can be made on each point in accord with WP:CONSENSUS.

A possibility is to ask for WP:MEDIATION, I think that might be a good step so people can start over in a more structured environment. Dmcq (talk) 11:43, 8 November 2011 (UTC)

Whatever works. I have no experience with either that option or ANI, how are they different? Normally I try to resolve differences at the article, but this edit war is turning out very different from the ones I've encountered in the past. Anyway if you're fine with one of these options then I'm more than happy to support you. What's the next step? --Vaughan Pratt (talk) 16:04, 10 November 2011 (UTC)

Mediation is what you do if you want to work it out. I'd go for that. AN/I is what you'd use if an editor's behavior is so awful that you need to seek admin intervention for a block or ban. I can't see that as productive or appropriate here, in what is essentially an ordinary content dispute with bad behavior on both sides. Dicklyon (talk) 17:10, 10 November 2011 (UTC)

direction of radiation

It is convenient here to state that a perfectly black surface is Lambertian. No actually real physical material has a perfectly black surface. A perfectly black surface can be imitated by a small hole in the wall of a cavity that sustains black-body radiation in its interior.

There are three common ways of expressing diffuse spectral emission of radiation from a surface into a vacuum. Using SI language, they are in terms of (1) the spectral radiance, (2) the spectral radiant emittance, (3) the spectral radiant intensity.

(1) for spectral radiance, one must specify an imaginary straight line from the given point of emission to the given point of detection. The orientations of the actual emitting and actual detecting surface are specified with respect to that straight line.

Spectral radiance is then spectral power per unit emitting area per unit detecting solid angle; it is in general a function of the angle between the emitting surface and the straight line. For thermal radiation from a Lambertian surface, the spectral radiance has the same value for every value of the angle between the emitting surface and the straight line; in that sense its value is a constant-valued function of said angle; one might say that it is the same in every direction. The table as at present in the article tells the spectral radiance.

(2) for spectral radiant emittance, one does not specify an imaginary straight line from the given point of emission to a given point of detection, because the detection is to be collected over an imaginary hemisphere centred on the point of emission and based in the emitting surface. There is no question of an "angle of emission or observation or detection" for the eventual value of the spectral radiant emittance.

Spectral radiant emittance is then spectral power per unit emitting area. It has no concern with angle of emission or angle of detection.

(3) for spectral radiant intensity, one must specify an imaginary straight line from the given point of emission to the given point of detection. The orientations of the actual emitting and detecting surfaces are specified with respect to that straight line.

Spectral radiant intensity is then spectral power per unit detecting solid angle; it is in general a function of the angle between the emitting surface and the straight line. For thermal radiation from a Lambertian surface, the spectral radiant intensity has values that depend on the said angle. The dependence is described by Lambert's cosine law.

A warning about the difference between spectral radiance and spectral radiant intensity was in the body of the article at some stages. It is currently in a footnote.Chjoaygame (talk) 05:54, 9 November 2011 (UTC)

Yep I do find it all confusing. I think it would be better to avoid all mention of detection surface angle, does that actually come into it for these definitions? Also I thought the radiant intensity was independent of the area and that this was independent of viewing angle for a Lambertian surface. Dmcq (talk) 09:45, 9 November 2011 (UTC)

Agree with assessment that there is a risk of confusion. The terminology varies. Paltridge and Platt on page 35 remark that "Certain terms mean different things to different authors."

Much of the literature (for example Planck, Chandrasekhar, Mihalas and Mihalas, Goody and Yung, Liou; not Paltridge and Platt) speaks of specific intensity; this understands that it is specific (radiative) intensity; it means what the current SI system calls spectral radiance; this is the quantity stated in the table with the variations of spectral argument. The SI spectral radiance = literature-common specific intensity from a perfectly black surface (which is by definition required to be Lambertian) is independent of direction, because it is per unit emitting surface area as apparent from the detector.

The SI system talks also of another quantity, the (spectral) radiant intensity which is what a detector sees without regard to the nature of the source. But if the angle of the actual emitting surface is known and is to be taken into account then one has to use Lambert's cosine law to calculate what the detector will "see", and this of course depends on angle; the quantity is then SI spectral (radiant) intensity. Born and Wolf complain on page 195: "It is unfortunate that the same word is used to denote two different quantities."Chjoaygame (talk) 10:57, 9 November 2011 (UTC)Chjoaygame (talk) 10:59, 9 November 2011 (UTC)

I always think of radiance as being independent of angle. Planck's law is stated in terms of spectral radiance. So I would state it something like "The radiance of a black body is independent of angle, which means that the energy detected at an angle to the normal is proportional to the projection of the emitting area, which is proportional to the cosine of that angle as per Lambert's cosine law. This radiation is furthermore unpolarized." This avoids nomenclature other than spectral radiance. I will put this in soon unless there is a better idea. PAR (talk) 14:16, 9 November 2011 (UTC)

So we have the radiance is what an observer sees per solid angle when they look at a surface, and for a Lambertian surface is independent of the viewing direction. The radiant intensity is what comes from a unit area in a given direction and can be measured by multiplying the radiance by the cosine of the angle between the normal to the surface and the angle. And the emittance is the integral of the radiant intensity and so is pi times the radiance for a Lambertian surface. That about right? Dmcq (talk) 16:45, 9 November 2011 (UTC)

From a distance, one can directly measure the radiant intensity because that is what one sees from a given direction, per unit solid angle of detection. Having measured from a distance the radiant intensity, in order to go on and find the radiance, one needs to find out also the viewing angle of the emitting surface. The radiance is the radiant intensity divided by the cosine of the angle between the emitting surface normal and the line of view. For a perfectly black surface, the radiant intensity depends on the viewing angle according to Lambert's cosine law, but the radiance is independent of the viewing angle.

Kirchhoff defined a perfectly black body as having a perfectly black surface, which reflects nothing and absorbs everything though it is infinitesimally thin. Planck recognized that no real physical material can have an infinitesimally thin perfectly black surface, so he dealt with material bodies that have a perfectly black interior, and dealt with the imperfectly black surface; the walls of physically real black cavities are made of such materials.Chjoaygame (talk) 19:52, 9 November 2011 (UTC)

I'm getting more confused. You say 'One can directly measure the radiant intensity because that is what one sees from a given direction'. I would have thought it is the radiance that one sees, if the sun was a Lambertian surface one couldn't tell the difference between the limb and the centre and I thought what one measured there was radiance. Are you perhaps talking about looking at a unit area of the surface rather than per solid angle at the viewer? Dmcq (talk) 22:11, 9 November 2011 (UTC)

It depends on whether you are looking at it directly with your eye, seeing photons per solid angle, or just collecting photons. If you have a Lambertian surface area element, no matter which way you look at it with your eye, it looks just as "bright", but the apparent size of the element varies as the cosine of the angle that a line from the element to your eye makes with the normal thru the element. If you are just collecting photons, you are integrating over that apparent area, and that goes down as the cosine of the angle. Radiance is the "brightness" - independent of angle. Check out Lambert's cosine law for some good diagrams. PAR (talk) 00:49, 10 November 2011 (UTC)

The difference is a matter of context. If you examine the total radiation from a huge surface in the vicinity of a tiny surface element, then the additional radiation from that larger contextual surface in any given direction, up to a distance small compared to the dimensions of the huge surface, makes correct the statement that in every direction as measured from that tiny surface element the radiation in that direction equals that in the normal direction. (If you followed that long sentence you may proceed.  ;) )

However some of that radiation is not coming from the tiny surface element but from its neighbors. If you calculate the Stefan-Boltzmann constant from Planck's law under the assumption that the radiation is the same in all directions, then you will be counting the radiation from those neighboring surfaces as well. The upshot is that the net radiation from each infinitesimal surface element gets counted twice, and you end up with 11.34 instead of 5.67 as the mantissa of the Stefan-Boltzmann constant.

There are a number of things like that that are confusing about thermal radiation. Another is the matter of the geometry of the radiating surface that Gerlich and Tscheuschner referred to as the basis for their claim that the Stefan-Boltzmann constant is not universal. The logic of why it is universal is comparable to the logic of why a triangle with two equal sides has two equal angles, Euclid's fifth proposition from Book I of his Elements, which has a very famous Latin name that I won't repeat here for fear of being blocked by an administrator. --Vaughan Pratt (talk) 02:10, 10 November 2011 (UTC)

I should clarify that the discussion subsequent to my "makes correct" was intended to make clear that it merely "makes seemingly correct." The point of view that makes the radiation appear to be uniform in all directions is unwittingly counting every radiated photon twice. Just another of those little paradoxes of thermal emission. --Vaughan Pratt (talk) 02:38, 10 November 2011 (UTC)

Another thought occurs to me (and I apologize for the resulting "wall of text" but again I don't have time to shorten it). One perspective that removes some (but certainly not all) of this confusion is to consider the energy radiated from one small surface to another, small relative to their separation, and to define the areas of the transmitting and receiving surfaces in terms of the area each end of the "transmission line" or ray sees at the other end. This area is smaller than the physical area of the surface at each end by an amount equal to the cosine of the angle between the surface normal and the ray, also known as the zenith angle. Since the area is in the denominator it exactly cancels Lambert's law. (This was how I defined things in my trial introduction, whose timing was pessimal in that I made the edit the same day Headbomb started editing this article. I did not contest its deletion, thank goodness.)

For example when you look at a spherical Lambertian radiator, Lambert's cosine law does not make the limb of the radiator any darker than the center because to the observer, every point of the radiator seems to be emitting the same intensity because Lambert's cosine law is offset by the decrease in apparent area as seen by the observer. While limb darkening is a well-understood phenomenon, it is not on account of Lambert's law.

Whether Lambert's law applies therefore depends on whether surface area is measured absolutely, or as perceived by the receiver of the radiation. That one can miscompute the Stefan-Boltzmann constant by a factor of two shows the necessity of using the definition appropriate to the task at hand. One cannot simply say naively that the radation is uniform in all directions.

PAR is quite correct that radiance radiant exitance, aka radiant emittance, is independent of angle. However exitance is π times spectral radiance, and this article gives Planck's law only for the latter. As evidence for why this is confusing, Q Science suggested a while back (it would be a tedious search of the archives to find it) that Planck's law should be for radiance spectral exitance (though I believe he was not specific about terminology) instead of spectral radiance. Personally I'd prefer that myself, because it simply removes sr⁻¹ as part of the units, and it sounds like PAR would too. However the standard meaning of B_λ is spectral radiance, not spectral exitance, and if this article is going to stick with spectral radiance it is mandatory to use Lambert's law for that meaning. Not to do so is to sow confusion about the difference between radiance spectral radiant exitance and spectral radiance.

The subject is confusing enough already without Wikipedia amplifying the confusion. --Vaughan Pratt (talk) 03:17, 10 November 2011 (UTC)