Wikipedia:Articles for deletion/Derivation of the partition function

The following discussion is an archived debate of the proposed deletion of the article below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.

The result of the debate was delete. Mailer Diablo 08:04, 8 June 2006 (UTC)

[edit] Derivation of the partition function

article is barely coherent and full of misinformation. some examples:

1. it gives at least two fundamentally erroneous statements re ergodicity. it confused ergodicity with the a priori equal probability assumption in the first section. and the statement "... (otherwise, we would have ergodicity with respect to particle number). ..." is simply nonsensical.
2. the derivation is awkward and unilluminating.
3. the last expression in the first section is precisely the canonical partition partition. yet, immediately below, the article claims " The previous derivation is too restricted..." then apparently proceeds to "derive" canonical ensemble.
4. the last section proposes to derive the grand canonical ensemble, so why not give the grand partition function explicitly?

In summary, the article is possibly correct information that is at least awkwardly arranged or misarranged (one can probably find pieces of it here and there in some texts), with incorrect statements and no clear understanding demonstrated throughout. Mct mht 21:21, 21 May 2006 (UTC)

my response to a prev removal of prod tag where an apparently generic justification is given:

"someone has removed the deletion tag and gave the following edit summary: (rm prod tag. mistakes in content aren't a reason to delete. if the material is irrelevant, double check that it's all covered elsewhere then make it a redirect.) if this kind of incoherence doesn't warrant deletion, then what does. one shouldn't weigh in unless 1. one has the background and is wiliing to rewrite and salvage this article, or 2. has knowledgable rebuttal to the points raised above. furthermore, as stated in the tag added, same material is covered in other relevant pages in much more legible fashion. Mct mht 21:09, 21 May 2006 (UTC)"

Mct mht 21:26, 21 May 2006 (UTC)

• keep, and if possible speedy keep. If you don't think the quality is good, edit the article. Change the mistakes to make it correct. If there's another article that already covers this material, consider merging to that article, or redirecting. You have given no reason to delete this per the deletion policy. There is no way this article is so bad that deleting it would be better than editing it. Mangojuice^talk 23:34, 21 May 2006 (UTC)

the above seems to be another generic response. again, the article gives bad information, in a technical sense. as long as that's properly noted, doesn't really matter whether it is deleted or not. it's also redudant to have a separate page, esp. a questionable one, on that subject. it is somewhat funny that people enter the discussion for what seems to be purely bureaucratic reasons and give no indication whatsoever that they understand the reasons given (e.g. "There is no way this article is so bad that deleting it would be better than editing it...", then no specific reason.) if that happens to be the policy, fine, as long as the reader is notified the information is unreliable. as noted above, any rebuttal from someone familiar with the subject in question is surely welcome. Mct mht 00:17, 22 May 2006 (UTC)

it would be kinda funny if the article is retained and bad information is left out there because people ignorant in the subject matter decide to have their bureaucratic say. seems pretty obvious the only sensible objections are from someone who's competent to judge, and who either disagrees on the specific points raised or on the worthiness of the article. Mct mht 00:34, 22 May 2006 (UTC)

• Comment. It's possible that the article is so bad that it should be deleted. If it is wrong, then in AfD language, it would be unverifiable, original research and unencyclopaedic. I also see that it was been around for three years without major improvements. On the other hand, correct information merely being awkwardly arranged is not sufficient reason for deletion. More later. -- Jitse Niesen (talk) 07:30, 22 May 2006 (UTC)
• Keep. If there is incorrect information, simply delete that material and move it to the talk page. If the material is covered elsewhere, remove those bits with {main} tags as appropriate. If we end up with an article that has nothing left to say, we'll delete it then. It's harmful to delete an article that someone's put a lot of time and effort into on such shakey grounds. Very, very few articles are so "bad" that they cannot simply be stubbed. Stevage 07:32, 22 May 2006 (UTC)
• Delete - if the experts say it's nonsense and the material is already covered better elsewhere, then this article is beyond hope, and could never be redeemed. Stevage 11:35, 2 June 2006 (UTC)

shaky in what sense? you even read the article? by "possibly correct" i meant surely pieces of it here and there can be found in various references. as whole, article is incorrect and incoherent. also, i do realize mere awkwardness is far from being sufficient to warrant deletion. Mct mht 07:46, 22 May 2006 (UTC)

something should be clarified: is the ability to read the article and the reasons given for deletion at least somewhat intelligently required to enter the discussion (perhaps as an unwritten understanding somewhere)? this seems to be absent in the two votes given so far. if that's not the policy, then my mistake. i'd assumed whatever objections, if any, the proposal is to encounter would be knowledgable and informative. this hasn't been the case so far. Mct mht 08:12, 22 May 2006 (UTC)

The "ability to read the article ... somewhat intelligently?" I find that insulting. I read the article, and I read your objections, and I understood them and stated my response. If you're so insistent, point by point: 1: there are a couple of errant uses of "ergodic" in the article that, if removed, solves this problem. 2: awkward and unilluminating? this is general criticism, and a good reason to improve the article, but no reason to delete at all. 3: I can't tell if you're objecting to a sentence in the article or the article's structure. If the former, it's easily fixed. If the latter, it's still fixable, but requires more rewriting. 4: The article doesn't have to be written the way you would write it unless you actually work on writing it. My fundamental question for you is, why not just work on the article? Maybe there's a misunderstanding here: you don't have to take it to AfD to erase parts of the article; AfD is for full deletion, which removes the article and its entire edit history from Wikipedia. You don't have to get central approval for editing, even major editing of an article. When other editors object to your edits, you should stop and discuss. I see no reason why this article is unsalvageable; it contains useful introduction text, for instance, and I personally didn't find the derivation that confusing, it seems pretty well-explained from a mathematical perspective. Hence, keep it. My suggestion is, take this to Wikipedia:WikiProject Physics or Wikipedia:WikiProject Mathematics to find collaborators to help improve the article. And while we're at it, how about a little good faith for the editors who come here and comment? Mangojuice^talk 13:39, 22 May 2006 (UTC)

"...3: I can't tell if you're objecting to a sentence in the article or the article's structure. If the former, it's easily fixed. If the latter, it's still fixable, but requires more rewriting..." begs one to question whether there's actual understanding of article behind that comment. Mct mht 16:43, 22 May 2006 (UTC)

The real purpose of this article seems to be to derive the Boltzmann factor, not just the partition function (see the context in Partition function (statistical mechanics); this article was likely forked from there. However, from the title, that shouldn't be the goal. Thus, the last two sections should be cropped out (actually, the last section should maybe be merged into Grand canonical ensemble since the material doesn't seem to be there), and the links to this article put in the proper context. But honestly, you don't seem to take a stand on whether it's just an awkward transition or whether we shouldn't be talking about the ensembles at all, which explains my comment. Mangojuice^talk 17:52, 22 May 2006 (UTC)

No, it is unreasonable to expect people voting on AfD to thoroughly read an article before voting. AfD exists to answer the question: "Is this article so inappropriate or so far beyond hope that it should be nuked?" - any discussion about actually fixing the article should take place on the article's talk page. Could a good article on this topic be written? Does the current article contain some basic information to get a rewrite started? Yes? End of discussion: Keep the article, but remove as much as you need to to the talk page while a rewrite is in progress. Deleting the article itself erases history and won't help a rewrite proceed any faster. Stevage 11:38, 22 May 2006 (UTC)

I agree that voters are only required to read as much as necessary to answer the question "Is this article so inappropriate or so far beyond hope that it should be nuked?" However, you said yourself that the article should be deleted if, after removing incorrect and duplicated bits, "we end up with an article that has nothing left to say". I'd argue this could well be the case here. I also think that redirection is not a good option, since there is also no sensible target to redirect to. -- Jitse Niesen (talk) 12:48, 22 May 2006 (UTC)

In response to Mct mht's request for clarification: regrettably, such a requirement is not observed here.

In response to Stevage: This is not a typical article, because it's a derivation. This means that one cannot simply delete parts of it.

I am of the firm opinion that we should look at the technical points before deciding what to do with the article. I numbered the nominator's point above, so that I can easily refer to them.

1. is a valid point as far as I can tell, but easily corrected by removing all references to ergodicity. I don't see what it has to do with the derivation anyway. 2. and 4. are not important enough in my opinion to warrant deletion. I'm not sure what is meant with 3. It seems that the article first derives the partition function for a closed system, then notes that interaction with the environment needs to be taken into account ("The previous derivation is too restricted..."), and then proceeds that this does not change the result. However, I cannot square this with the sentence "Because of its assumption of the independence of the molecules, it only really applies to ideal gases." in the article. The discussion that follows does not seems connected with this assumption.

I had a look at the talk page and I saw the statement:

"… it would be obvious that there would be a maxima in the distribution (somewhere near total energy of system divided by number of molecules). Yet the derived equation gives an exponetial decay in the relative occupancy of energy states i.e. more molecules would be in the lowest energy state than in any other state."

I can see this point. There must be an assumption being used here which is not made explicit (energy of excited states is large compared with total energy divided by number of particles?). However, on the talk page I also see User:Vzlatic defending the derivation, at least up to a point. As far as I can see, the mathematics is okay, except that I'm not sure whether Stirling's approximation can be used (surely, N is large, but what about N_i?). Perhaps this can all be fixed by taking the limit N → ∞.

In summary, the four points raised by the nominator are not enough to warrant deletion, but they do indicate a problem, and if there are too many problems then the article should be deleted. -- Jitse Niesen (talk) 08:23, 22 May 2006 (UTC)

• Comment. I would like to see a more respect for experts in extremely difficult academic subjects, even if they are not fully up-to-speed on the usual comings-and-goings of AfD. If the derivation is no good, and nobody has time to improve it, then I do think it's better than delete than to leave it—or perhaps, to make things more palatable to everyone, we could simply redirect to the partition function article. -- SCZenz 08:47, 22 May 2006 (UTC)

Reply ok, finally some technically sensible responses. it was getting rather disappointing.

reply to Jitse Niesen: yes, Stirling's approx is a common trick in this context. N is typically on the order of 10^(23).

reply to Jitse Niesen: regarding reason 3 for deletion, clearly there's a problem there. after a long-winded calculation to obtain the partition function for the canonical ensemble, it immediately claims what's been done is too restrictive then "considers the canonical ensemble." this is a fundamental mistake in realizing the two approaches (one by analysing the multiplicities directly and the other by considering a system coupled with a reservoir) are equivalent and both give the canonical ensemble.

reply to Jitse Niesen, regarding the exponential decay, the result is right. higher energy states are less likely to be occupied.

reply to Jitse Niesen: as far as deletion policy goes, well, deletion proposals shouldn't become a forum for the un-informed, for articles of this nature.

general comments: derivations for the microcanonical ensemble and canonical ensemble are already given, IMO in a much better fashion, are given in their respective pages. grand canonical ensemble is a stub, but at least everything is correct and one can just add to it. now the article in question apparently proposes to give separate derivations for these ensembles, and is clearly confused, IMO. along with the mistakes i cited, i thought the overall awkwardness and the redundancy warranted deletion. if someone thinks my judgement of the article is overly harsh and think it's salvagable, that's fine with me. as long as the information on there is correct. Mct mht 10:10, 22 May 2006 (UTC)

• Delete. The majority of the page duplicates the derivation of the partition function given in canonical ensemble (I wished I had realized that immediately), except that the derivation on that page is clearer. The other points made on the page can be found in other articles. The usual solution would be to replace the content of derivation of the partition function by a redirect to canonical ensemble; this would be correct if the page were titled "derivation of the partition function for the canonical ensemble", but it is not, and in the current situation all possible targets for a redirect are misleading. Hence, it should be deleted. -- Jitse Niesen (talk) 13:46, 22 May 2006 (UTC)

another comment on redundancy, the article Maxwell-Boltzmann statistics contains 2 derivations (second one is definitely correct, first one seems to be as well) of the canonical ensemble. Mct mht 17:35, 22 May 2006 (UTC)

• Delete per Jitse Niesen. I urge the closing admin to take into account the technical expertise of the voters in closing this AfD, and in particular pay attention to the effort made by Jitse to explain why the usual approach to a flawed article (e.g. a redirect) is inappropriate in this case. -- SCZenz 00:29, 24 May 2006 (UTC)

• Delete - the article by itself is too singular and confusing, and would be difficult to clean up. Some of the material involving W is already present in information entropy, albeit in a different form. I don't think it would benefit the organization of topics related to the derivation by having the article. Most (if not all) of the content is replicated elsewhere already. --HappyCamper 03:51, 24 May 2006 (UTC)

• delete - I'm changing my vote; I don't think the nominator has given much reason why deletion is necessary, but I agree with Jitse Niesen that the contents of this article are included in the encyclopedia elsewhere and are better written. Also, that this article can be in such a bad state after so long on Wikipedia indicates that this isn't a topic people look for very much. Mangojuice^talk 20:39, 24 May 2006 (UTC)

• Delete — normally I'd say that redirects are cheap, but like Jitse Niesen, I can't come up with an appropriate redirect that wouldn't be misleading. — Laura Scudder ☎ 13:59, 25 May 2006 (UTC)

This AfD nomination was incomplete; listing now. - Liberatore(T) 18:48, 1 June 2006 (UTC)

• Delete without prejudice. There's very little there which is both correct and not included in one of the other related articles. (Yes, there's more than one, so a redirect would be inappropriate.) Also, the title is misleading with or without a redirect. Sorry, there's nothing that should be under that title. — Arthur Rubin | (talk) 19:51, 1 June 2006 (UTC)
• Keep solid content. Xyrael ^T 19:59, 1 June 2006 (UTC)
• Delete. I am particularly influenced by the views of HappyCamper and Jitse Niesen both of whom are experts in statistical mechanics. I ask the closing admin to take into accounts that at least those two are experts, although of course some other editors commenting here may also be. --Bduke 23:26, 1 June 2006 (UTC)
• Delete - Most of the content is covered elsewhere in a better manner. I believe that the article is mistake-riddled: such as that {N, 0,0,0,0} is less likely than {N-1, 1,0,0,0} - this would imply that each config is equally likely, irrespective of energy - when the result derived indicates that it is energy dependent. eg, in an infinite energy gap between ground and 1st excitation, the probability of the second state is zero, even though there are N times more configs. IMHO, the extension to the generalizations are written and seem dubious, so the benefit of keeping them together is minimal at this stage. Also the redirect issue is well explained above.Blnguyen | Have your say!!! 02:37, 2 June 2006 (UTC)
• Keep - If you think the article contains mistakes why don't you try yo correct them?--Pokipsy76 08:36, 2 June 2006 (UTC)
  • Comment - I could but as already stated, there is no point.Blnguyen | Have your say!!! 08:39, 2 June 2006 (UTC)
• Delete. After removing everything that is handled better elsewhere or is original unverifiable research, only a stub would be left. There is no point in having a stub for an article that by its name promises to deliver a derivation, but that states something like: several things are called partition function, and they have derivations, and to find out about one of them click here. --Lambiam^Talk 13:45, 2 June 2006 (UTC)
• Delete per the talk page, happycamper and various other comments above. Kotepho 16:44, 2 June 2006 (UTC)
• Keep no need to delete useful content.  Grue  10:51, 3 June 2006 (UTC)

The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.