Wikipedia talk:WikiProject Mathematics/Archive/2007/Sep

Another math article on AfD

Uncertainties of the limits, a recently created math article (without categories, so not found by the bot — I just added one) has been listed for deletion. —David Eppstein 15:48, 3 September 2007 (UTC)

I've redirected it. It was a badly written article on a topic worthy of inclusion by long-established consensus, and there was an already existing article. Michael Hardy 18:38, 3 September 2007 (UTC)

More possible vandalism in mathematics articles.

User:WAREL is adding seriously wrong information to Riemann hypothesis again.

Waxex (talk · contribs) is making almost sensible, although clearly incorrect, edits to Function (mathematics) and Grammatical function.

I've reverted 3 times already today in each. (45 in Function (mathematics), but the last two waswere clear vandalism.) — Arthur Rubin | (talk) 00:46, 5 September 2007 (UTC)

I blocked User:218.133.184.93 for 24 hours. Arthur, I understand your hesitation in blocking the guy (as a person involved in the edit conflict at that article), but next time if you (or anybody else) sees him doing the thing he's been doing for the last year or so, he should just be blocked on sight. I think there will be full support from the community here. Oleg Alexandrov (talk) 01:04, 5 September 2007 (UTC)

How funny, no real number ε is known to make that hold? You'd think ε = ½ would be easy enough to prove, both functions being bounded by 0 and x. Yeah, I think a block has to happen. CRGreathouse (t | c) 14:57, 5 September 2007 (UTC)

Is fuzzy logic math?

I am wondering whether Category:Fuzzy logic should be added to the list of mathematics categories. This would have the effect that all articles in this category would be listed in the list of mathematics articles by the bot.

So the question is, is fuzzy logic math? The answer could be a bit fuzzy, I guess. :) At the core it may be math, but it has a lot of applications outside math, and for example, Fuzzy electronics could not be considered math. Comments? Oleg Alexandrov (talk) 20:19, 4 September 2007 (UTC)

There is a recent proposal that WP:LGC keep two worklists, one philosophy, and one mathematics. The fuzzy logic category is proposed to be among the mathematical logic categories. Gregbard 22:26, 4 September 2007 (UTC)

Fuzzy is not logic. Fuzzy is not mathematics. (Although it might be a technique in control theory.) To the extent that it pretends to be logic or mathematics, it is a scam, a fraud. It would be more accurate to call it an attempt to destroy logic rather than logic. JRSpriggs 04:18, 5 September 2007 (UTC)

Huh ? I am not an expert, but I have always had the impression that fuzzy logic is either a sub-topic or possibly a generalisation of the perfectly respectable mathematical subject of modal logic. Indeed the term "fuzzy modal logic" throws up a lot of academic-looking hits on Google. Am I mistaken ? Could you expand on your comments, JRSpriggs ? Gandalf61 10:48, 5 September 2007 (UTC)

It seems that fuzzy logic has multiple meanings. I think of it mostly as a topic in control theory. In that guise it's an important tool in engineering but has had little to no impact on mathematical logic. The study of modal logic in mathematics is also quite small, although present; modal logic is usually studied by philosophers. — Carl (CBM · talk) 12:50, 5 September 2007 (UTC)

That seems a bit harsh, JRSpriggs. A lot of the probabilistic variables used in models in eg cluster analysis, or data compression and transmission, strictly speaking are actually fuzzy variables -- ie Bayesian variables whose "true" value would still not be definitively known, even given a complete physical description of the state of the universe. And recognising this can sometimes be useful. Where "fuzzy logic" becomes much more questionable is in its many ad-hoc prescriptions and rough-and-ready shortcuts for manipulating such variables, which to Bayesian eyes generally look chancy at best, and often significantly wrong-headed. Nevertheless, fuzzy approaches are widely used in model making, and calling them "not mathematics" seems, shall we say, a bit strong! -- Jheald 13:31, 5 September 2007 (UTC)

Fuzzy logic gained critical attention when it was used in a rule-based artificial intelligence expert system for medical diagnosis, MYCIN, designed at the teaching hospital of Stanford University under the direction of Ted Shortliffe. The entry in the Stanford Encyclopedia of Philosophy may be helpful reading. Serious mathematical content exists. Compare to catastrophe theory, where perfectly respectable mathematics drew kookie attention because of a catchy name. --KSmrq^T 13:24, 7 September 2007 (UTC)

Well, for now I put fuzzy logic in non-math, for practical reasons, since it has plenty of non-math articles and the potential to add more non-math to the list of mathematics articles. Perhaps some of the articles in that category could be put in Category:Mathematical logic, and Category:Fuzzy logic itself could perhaps be better categorized in Category:Logic rather than in Category:Mathematical logic. Oleg Alexandrov (talk) 02:35, 8 September 2007 (UTC)

Mathematics Wikia

One of the annoying things I've found here at Wikipedia is the tendency for mathematical lists and tables to be nominated for deletion as "indiscriminate lists of information" (a recent example). Such information has been transwikied in the past to Wikisource, only to be deleted from there, as well. I suppose the best hope for that kind of stuff now is Wikibooks, but I haven't checked the state of their math section lately... Anyway, the Mathematics Wikia (link is to English version, but other languages exist) hasn't seen a lot of activity since it was created few years ago, and I was hoping we could use it for — at the very least — transwikiing useful mathematical content deleted from the various Wikimedia projects. If anyone has old versions of deleted math-related stuff (that's actually worth keeping) — or wants to expound on mathematical topics more than would be appropriate in a Wikipedia article — please consider putting it up at the Math Wikia. Obviously, not every deleted math article will be worth saving, but if it's legitimate, correct mathematical information, it probably should go somewhere. Before diving in over there, please see wikia:math:Mathematics:Guidelines (and its talk page!) for some ideas about what we should be doing with the wiki(a). - dcljr (talk) 08:13, 5 September 2007 (UTC)

Some of the missing table- or list-articles i intigrated in my book projects in de:wikibooks:

• Carmichael numbers [1]
• Sequence:Prime numbers [2]
• Prime Gaps [3]
• Sequence:Pseudoprime numbers [4], [5], [6], [7], [8]

--Arbol01 21:48, 6 September 2007 (UTC)

Difference Fourier or Difference fourier

I notice this is a requested article in Wikipedia:Requested articles, I can't decide whether to make it a redirect to Fourier transform or Fourier analysis, or is this not a real technique after all? Help and advice requested. Thanks Tim Vickers 23:48, 8 September 2007 (UTC)

The term is used with two very different meanings. One meaning, current in spectroscopy, diffraction analysis, and such, is conventional Fourier analysis & synthesis or Fourier transforms applied to the difference of two signals (see e.g. here). I don't think a simple redirect would address its significance. A quite different meaning is found in quantum algebra, as in the maths preprints found here.  --Lambiam 05:28, 9 September 2007 (UTC)

Brahmagupta

I request Maths people to review the article. "In mathematics, Brahmagupta is considered the father of arithmetic, algebra, and numerical analysis." Sentences like this are not supported. 59.91.253.97 04:08, 9 September 2007 (UTC)

I removed this sentence and the next two ones. As has been the case for a long time, the articles on Indian maths could use some attention from somebody who knows this stuff. -- Jitse Niesen (talk) 07:39, 9 September 2007 (UTC)

Integral and surrounding articles

Integral has been going through a significant collaborative effort, and is emering in a good shape. It should not take too much additional work (I'll add a few comments to the article's talk page) to get it to WP:GA. However, the article is in the middle of a thick web of interlinking articles on measure theory and various types of integral. Unfortunately these "surrounding" articles are often overlapping, use differing conventions, and certainly do not form a coherent set of articles. As the area of measure and integration is nevertheless fairly "compact", it seems quite feasible to edit the key articles refereced to / from Integral into better shape. I've been doing some minor fixes every now and then, but a more thorough effort of putting some structure on the set of these articles is what I think is needed. With the other articles in better shape, it would be also feasible to condence Integral in some places, making references to the more specialised articles instead of reproducing (quite as much) material in slightly different form.

The articles that I have in mind are (at least): Lebesgue integral, Riemann integral, Multiple integral, Line integral, Surface integral, Differential form, Green's theorem, Stokes' theorem, Fundamental theorem of calculus, Measure (mathematics), Lebesgue measure, Haar measure, Radon measure, Borel measure, Riemann-Stieltjes integral, Lebesgue-Stieltjes integral, Daniell integral, Lp space, Measurable function.

Sorting out these articles should in particular help make the structure of Integral clearer as regards the somewhat different if interlinked themes that it covers:

1. Integral in the sense of measure and integration: the quest for more "general" integrable functions, different weights through measures or (in some cases) the Stieltjes method; integration on more general spaces than R ("multiple integrals" is hardly a "generalisation" anymore after the introduction of general measure-theoretic integral); spaces of integrable functions; improper integrals; and
2. Line integrals, surface integrals, integrals of differential forms, Green's / Stokes' theorem

With such number of articles involved, and given that the whole topic seems to get emotional at times, I think the only way to do a more significant overhaul is to discuss it first here and only then start implementing. I'd be happy to develop and propose a more concrete plan for discussion if there is interest in such an effort. Stca74 11:25, 9 September 2007 (UTC)

Where does this belong (if anywhere)?

I added the following paragraph, which originally had a place in our Probability article, to the article on Probability theory:

In probability theory, a probability is represented by a real number in the range from 0 to 1. An impossible event has a probability of 0, and a certain event has a probability of 1. (However, events with probability 0 are not necessarily impossible, and those with probability 1 are not necessarily certain.)

This passage used to be in the Probability article but was removed from there with the edit summary belong in theory.[9]. However, when I added it to Probability theory, it was promptly removed with edit summary deleted confusing paragraph. Asked for an explanation, the editor stated that "The issue of what events are impossible or not is still a matter of philosophical dispute. In any case this belongs to the interpretation or application of probability, not probability theory proper."

• Question 1: Should this information be presented somewhere on Wikipedia, preferably in terms a reader can understand without having to learn about measures and sigma algebras?
• Question 2: If so, what is the most appropriate article?

 --Lambiam 12:29, 9 September 2007 (UTC)

This information ought to be in the probability article. I disagree that there is a "philosophical dispute" - the probability of drawing a red marble from a bag of all blue marbles, in our mathematics model of the universe, is 0. I thought the parenthesized sentence was just trying to say there are nonempty sets of measure zero, which at least makes sense. In any case, whenever naive English reading of a passage makes the passage self contradictory, as with this passage, the text should be clarified. — Carl (CBM · talk) 12:56, 9 September 2007 (UTC)

Quick suggestion for clarification:

In probability theory, a probability is represented by a real number in the range from 0 to 1. An impossible event has a probability of 0, and a certain event has a probability of 1. (However, other events may also have probabilities 0 or 1: events with probability 0 are not necessarily impossible, and those with probability 1 are not necessarily certain.)

—David Eppstein 15:22, 9 September 2007 (UTC)

That seems clearer to me. On second reading, I realize the original isn't contradictory; but the rewording is more direct about what's going on. — Carl (CBM · talk) 18:50, 9 September 2007 (UTC)

I agree that some such remark is appropriate for Probability. I do not agree that the clarification is much clearer, and propose including an example. For instance, suppose we are asked to choose a real number greater than six and less than seven, with every number equally likely; then, with an infinite collection of numbers from which to choose, the probability that the chosen number will be, say, 6.283 is zero, while the probability that it will not be 6.283 is one. Yet 6.283 is just as likely to be chosen as any other number in the interval, such as 6.75; and some number is sure to be chosen. (Please feel free to improve the example!) --KSmrq^T 20:05, 9 September 2007 (UTC)

The difficulty is that nonempty measure zero sets only appear with infinite sample spaces, at which point the idea of "each element equally likely" means something different than with finite sample spaces (and it doesn't even always make sense: there is no nonzero nonatomic measure on any countable set). It takes a lot of space to explain these things truly clearly, more than can be expected in an article's lede.
   Perhaps we should continue this discussion at Talk:probability, since several people seem to think the content belongs there in some form. — Carl (CBM · talk) 20:45, 9 September 2007 (UTC)

I've simply re-added the paragraph at Probability – but it remains slightly strange that the reader entering at Probability theory only finds out that probabilities are represented by real numbers through the process of induction from examples.  --Lambiam 23:03, 9 September 2007 (UTC)

Principles and Standards for School Mathematics

I just wanted to ask some of you to check out the Principles and Standards for School Mathematics article and give your thoughts. Back in February a redlinked user removed significant amounts of content from the article. [10] Just wondering what your thoughts are on the article.--Jersey Devil 06:51, 10 September 2007 (UTC)

Looking over it again it seems that the two users who had made most of that content were disruptive sockpuppets. However, it is still surprising that there is no mention on that article of the debate over "standards based" mathematics.--Jersey Devil 06:54, 10 September 2007 (UTC)

Article protection at the beginning of the school year

The academic year is beginning in the United States, which will probably lead to an increase in IP vandalism from students at computer labs during the school day. There are only three math articles currently semiprotected from editing by IP editors and new editors: Actuary, Mathematics, and Space. We have had problems in the past with vandalism of other articles such as Geometry and Randomness. If you notice an article is getting enough vandalism to warrant semiprotection, contact me or another admin to ask about it. — Carl (CBM · talk) 13:13, 5 September 2007 (UTC)

And just to repeat myself, one can track the changes to the math articles from the list of mathematics articles (now with a link to all the changes combined in one page). Oleg Alexandrov (talk) 14:59, 5 September 2007 (UTC)

I just semi-protected Dependent and independent variables for a week. I don't understand why just that article gets so many useless edits. -- Jitse Niesen (talk) 03:09, 12 September 2007 (UTC)

Wikipedia:Notability (numbers)

This guideline is being discussed at Wikipedia talk:Notability with suggestions that it be deleted or merged into a single guideline for everything. --Bduke 07:00, 11 September 2007 (UTC)

From wikibooks

Hello. I have started a wikibook entitled The Book of Mathematical Proofs on the english wikibooks. It can be used to complement wikipedia articles by containing the proofs of the stated theorems. Please help in expanding the wikibook. In case of any questions kindly send me a message either on my wikipedia account or on wikibooks. Cheers--Shahab 17:54, 9 September 2007 (UTC)

I seem to recall that Wikibooks specifically reported they did not want mathematical proofs? Am I mistaken? Did this change? — Arthur Rubin | (talk) 22:35, 9 September 2007 (UTC)

I have specifically asked a wikibook administrator this question and he has given me the go ahead for the book.[11] Also certainly saying that mathematics is all about proofs was a mistake. Cheers--Shahab 06:57, 13 September 2007 (UTC)

I don't remember that, but if it's true, it seems like a huge mistake. You can't possibly write a textbook on any interesting mathematics without including proofs. Perhaps some people were creating 'books' containing just one proof, and that was being objected to? That would make more sense. --Sopoforic 22:40, 9 September 2007 (UTC)

Thinking it over, it may have been wikisource, rather than wikibooks. But it might be a good idea to check, first. I seem to recall that some of "our" (Wikipedia's) mathematical proofs were moved to another Wiki, and then deleted from there. — Arthur Rubin | (talk) 23:21, 9 September 2007 (UTC)

Wikibooks are supposed to be textbooks. Does it make sense to aim at a textbook of all major theorems (and some not so major ones) in Algebra, Analysis, Applied Mathematics, Discrete Mathematics, Geometry, Logic, Mathematical Physics, Number Theory, Probability, Set Theory, Statistics, and Topology? Mathematics is not all about proofs, but at least as much about concepts and interconnections, which are needed to give the theorems context – so in the end you are creating an encyclopedia of Mathematics.  --Lambiam 23:22, 9 September 2007 (UTC)

To be fair, mathematics is also about proofs, and there have been published math books similar in some ways to the concept behind this one: Aigner and Ziegler's Proofs from the Book, notably. —David Eppstein 03:02, 10 September 2007 (UTC)

I wrote "Mathematics is not all about proofs" – and not "Mathematics is not at all about proofs". Since the thread originator wrote: "Mathematics is all about proofs!", I think it's fair to point out that there are other important aspects. Proofs from THE BOOK definitely did not aim to present all major theorems in mathematics, something that makes the envisaged endeavour rather incomparable.  --Lambiam 10:42, 10 September 2007 (UTC)

Not strictly on the topic of this discussion, but I disagree that Aigner and Ziegler, which is full of chefs d'œuvre that any mathematician will savour, is about proofs: it is about THE BOOK (orig. cap.) Arcfrk 03:31, 10 September 2007 (UTC)

Explaining R^n notation

I made an {{anchor}} so that Real number#R for Rⁿ notation can be linked directly from its use, e.g. in Leech lattice. The notation really needs some kind of explanation for those unfamiliar with it, but is this the best way to do it? ←Ben^B4 15:46, 12 September 2007 (UTC)

Possibly a footnote at the end of the sentence would be better: something like "Here Rⁿ denotes n-dimensional Euclidean space." I've been using a similar approach for set-builder notation. Jim 16:11, 12 September 2007 (UTC)

As I was going to do that, I thought, that would be a pretty small footnote. I'm happy with using a few words to explain it inline and linking.[12] ←Ben^B4 20:42, 12 September 2007 (UTC)

Null space introduction

I'm having a disagreement with another editor on Talk:Null space about the introduction to the null space article, and we could use an outside perspective. Thanks in advance, Jim 01:19, 14 September 2007 (UTC)

List of references

Hi everyone. I've recently been working to improve the articles related to introductory linear algebra, and as part of that project I've been expanding the list of references for articles that I work on. The standard list started to become unwieldy, so I came up with the following solution:

1. Create a separate article for the list of linear algebra references.
2. Create a Template:Linear algebra references with a few selected references, and a link the list.

Does this seem like a good idea? I haven't found any other articles with lists of print references, so I'm a little worried that the list of references will be perceived as listcruft. I also don't know anything about templates, and whether this sort of use is allowed. (You can see the result of the template at Linear algebra#References or System of linear equations#References.)

Any thoughts? Jim 02:02, 13 September 2007 (UTC)

It seems like a good idea to me. It probably can be used for other math topics as well, such as calculus, or theory of probability or wherever we have a significant number of well established textbooks. (Igny 05:07, 13 September 2007 (UTC))

I do fear the list of references might be considered listcruft, but if you make it as a subpage of Wikipedia:WikiProject Mathematics then it should be OK, because it serves a purpose in helping to build the encyclopedia, but the criteria for inclusion can be more arbitrary. Unfortunately, you would not be able to link to that location directly from articles. But just copying to each article two or three sources that cover its subject well would be a help to the reader. — Carl (CBM · talk) 05:25, 13 September 2007 (UTC)

Another article with a list of print references. But it only barely survived an AfD, so I don't know how strong a precedent it is. —David Eppstein 05:29, 13 September 2007 (UTC)

Interesting—the article you mention is part of a category called Category:Lists of books. Among other things, this contains Books on cryptography, List of notable textbooks in statistical mechanics (which survived a deletion nomination), and a subcategory called Bibliographies by subject to which I've added the linear algebra references article. Jim 06:46, 13 September 2007 (UTC)

It's really a catch-22 situation. It isn't acceptable to simply plan to list every book in the area, even if that was desirable, because WP:NOT states Wikipedia is not for indiscriminate lists. So these articles have to make up some arbitrary criteria for inclusion, which are then criticized as arbitrary, or as original research. — Carl (CBM · talk) 05:46, 13 September 2007 (UTC)

As I said on my talk page, I think it's a cute solution and my guess is that the list will survive AfD (this is most definitely a guess though). Another precedent is List of important publications in mathematics; see Wikipedia:Articles for deletion/List of publications in biology and Wikipedia:Articles for deletion/List of publications in biology (2nd nomination) for prior discussions (I seem to remember that these lists were all part of WikiProject Science pearls). I think the case for list of linear algebra references is stronger because it helps to achieve Wikipedia:Verifiability.

By the way, the References section should only contain books and articles which were actually used when writing the article (Wikipedia:Guide to layout). This rule is not really adhered to, but it does probably imply that the template shouldn't be used in a References section. -- Jitse Niesen (talk) 06:18, 13 September 2007 (UTC)

I don't see this working in its present form. One problem is, articles will need distinct references of their own in addition to any on a common list. The accepted way to order references is alphabetically by author, then by date, which this does not accommodate. Also, I have my own opinion of books, and I don't want to have to argue about what should and shouldn't be on a common list. Maybe we'll all agree, but I'd be surprised. To get a taste of the kind of disagreements I expect, look at Amazon.com reviews of calculus books and note the divergence of opinions and the broad spectrum of book styles. What an chemical engineering undergraduate wants or appreciates in a linear algebra book and what a mathematics post-doc wants will surely differ. Is linear algebra Numerical computing with MATLAB by Moler; or Introduction to Linear Algebra by Strang; or Linear Algebra by Greub; or Invitation to Linear Operators by Furuta; or Linear Operators Part I, General Theory by Dunford and Schwartz? Of course, linear algebra encompasses all these options, from finite-dimensional real vector spaces to abstract vector spaces to Hilbert spaces and functional analysis. And even an undergraduate studying quantum mechanics needs to learn more than matrix manipulation.

Maybe my concerns can be met; I'm open to ideas. Certainly it would be nice to have a common pool of good texts to consider adding to an article. One option for editors might be to have subpages of our reference resources; that doesn't directly help the general Wikipedia reader, but it should dodge AfD and one-size-fits-all concerns. --KSmrq^T 06:38, 13 September 2007 (UTC)

It seems to me that we ought to be able to agree on short set of canonical linear algebra references in roughly the same way that we agree on the content of major articles (slowly, with lots of heated arguments on talk pages, eventually settling on a compromise). I don't think that every linear algebra article should have the same set of references, but I do think that it helps to have a standard list available for articles covering topics that can be found in any introductory text. (I imagine that the editors of a print encyclopedia must use a similar strategy.)

Your point about the reviews of calculus books is well taken, but remember that there's a difference between "standard" and "good". If we tried to do this with calculus books, we wouldn't have to agree that Stewart is good introductory calculus text—we'd only have to agree that it's pretty much standard, and should therefore be included in the short list of references for an article on the product rule. Jim 07:29, 13 September 2007 (UTC)

so long as we're on the subject of linear algebra, the article linear algebra begs for improvement, something clearly worthy of the attention of the many knowledgable folks here. the scope of the article is rather narrow. linear algebra proper by itself is a huge subject with still open questions. the present article does not really convey this. related areas of mathematics and application are mentioned only haphazardly. maybe it can be made the collaboration article of the month. Mct mht 00:33, 14 September 2007 (UTC)

I think adding a list of general linear algebra references to any article which is on something in l.a. does not serve the purpose of a reference section well. I guess every l.a. reference will say something about a null space or kernel, but there are many articles which are much more specialised and there scatter some 8-10 references which possibly say nothing or very little about the article in question distracts the reader from the one refernence among the 10 he/she really needs to look at to learn more. Probably, whenever a article has one of these no-reference-templates, people will start adding the above mentionned reference-template. But on the long run, I guess it will actually do more harm than good. Last but not least: if a number of references is given, all of which have similar title, it is very helpful to the reader to give a short indication of the level of the book etc. (i.e. point out that it is an introduction for undergrad, or specialised text or whatever).

For example, consider normal matrix (which has no reference at all right now). Which of the books in the reference template even mention this notion? (I can only guess, but I'm fairly sure that at least one doesn't).Jakob.scholbach 10:32, 14 September 2007 (UTC)

Instead of the template, one could just add a number of standard references to the article linear algebra (or what the respective top-level category is) and put a link in the reference section of kernel: "see also the references of linear algebra#References." Jakob.scholbach 10:28, 14 September 2007 (UTC)

Spectral analysis

I could not find a Wikipedia article about the spectral analysis in this sense. I think spectral analysis is not quite what I have in mind. (Igny 16:28, 14 September 2007 (UTC))

There's an article on spectral theory, which I found by going to the spectrum of an operator article. (There's also an article on harmonic analysis, though I don't know how related the two fields are.) I've changed spectral analysis to a disambiguation page. Jim 17:04, 14 September 2007 (UTC)

the content of the spectral theory article probably is something closer to what Igny had in mind than what an operator theorist would now call spectral theory, which begins with models and decompositions of operators. Mct mht 00:11, 15 September 2007 (UTC)

FA reconsidered

I requested a copyedit for an FA article a while back; I am somewhat pleased to report it has been promoted. I have commented on the process on the FAC talk page; I would like to thank Jim and Ksmrq particularly.

I am only somewhat pleased because FA seems to have worsened significantly in the last few months; my evidence is also in the link above. For here, some regulars insist on enforcing every tittle of MOS, even when it is merely stating general good advice, not literally applicable to every article. (They do not seem to understand that it's a {{guideline}}.) Other articles pass with no substantive attention at all.

What should we do?

• Withdraw from FA and GA and ignore them to death; deprecate efforts to promote math articles, as giving recognition to a broken process?
• Try to fix it, as Geometry Guy is trying to fix GA?
• Institute dispute resolution against the worst reviewers? (While personally satisfying, this will have limited long-term effect; FA and GA have a fatal attraction to the sort of mind who would like to contribute to Wikipedia by demanding changes in other people's work, without doing anything themselves.
• Other. What?

Regards. Septentrionalis PMAnderson 20:51, 6 September 2007 (UTC)

I had to click on the link you gave above to be reminded what the abbreviation "FA" stands for, and if anyone's not a newbie here, it's certainly me. Michael Hardy 21:38, 6 September 2007 (UTC)

My contribution was trivial, changing hyphens to dashes (which I do routinely). Contrast that with the massive amount of work Septentrionalis put in to research, write, and otherwise improve the article.

Geometry Guy was called away by real life, so don't expect him to influence GA any further.

My impression is that featured article candidates are not intended to rally editors, but to focus reviewers. By way of analogy, nobody expects film critic Roger Ebert to direct a film himself. Wikipedia culture has a habit of distorting language for the worse, and one example is the term "editor". In traditional publishing, whether novels or textbooks or newspapers, we see clear distinctions between the writer and the editor and the reviewer and the critic. Such distinctions would be helpful here as well.

I do not aspire to having articles "featured", given the taste of the "selection committee" (and the general public?), but I do appreciate good reviewers, even if they only comment. For example, I recently did a complete rewrite of an article but with no critical input. It might benefit from the feedback of well-informed reviewers, and from a good "editor" in the traditional sense, but those who comment on featured article candidates seem a poor pool for such purposes.

Anyone who has graded homework or reviewed journal submissions knows there's usually more bad stuff than good, and even the good often needs work — whether for content or grammar or organization. It does not surprise me that we do not have enough volunteers who are good reviewers; it's hard work! And for those whose writing has been marked down all their lives, the lure of being able to attack others must be irresistible. --KSmrq^T 14:32, 7 September 2007 (UTC)

Without commenting on the FA process itself, I'd like to mention something that might be of interest. There have been a couple of discussions of the Stub/Start/B/GA/A/FA hierarchy, and it has been pointed out that reviewing for content accuracy and correctness of references is best done by a WikiProject, since you need multiple experts working together to be able to cross-check each other's work. This is particularly true in a field such as mathematics, where the great majority of the articles can't be verified by anyone not familiar with the topic. However, non-experts can review for form errors -- even if you don't agree with all of the MOS, some of it is uncontroversial, such as the requirement for a summarizing lead. So there were suggestions to separate the two hierarchies, with GA in one branch and B/A in the other; the latter branch would review for content and would be done by WikiProjects.

I understand that one of the points of contention is how to cite: in humanities articles, it is common to have multiple citations in a paragraph so that every fact can be traced to a reliable source. Although this might be beneficial for a biographical article on a mathematician, I can see why it might be unnecessary for an article such as Gamma function, for example. I think if you have clear and useful citation standards that comply with current guidelines (which don't require inline citations) you should be able to get the articles through FA, if you want to; if not, just bring things up to A-class and leave them there. Mike Christie (talk) 04:05, 9 September 2007 (UTC)

And one of the problems is that it is infinitely easier to check that emdashes conform to a mechanical standard than to check writing, let alone content; so reviewers, being human, gravitate to trivialities. Septentrionalis PMAnderson 22:42, 15 September 2007 (UTC)

The multiple-cites-per-paragraph style can also be useful for some mathematics articles, when different facts about some mathematical object are gathered from different research papers; as an example of this style, see complete coloring. But I agree that there's little point in peppering paragraphs with citations that are all going to point to the same source. —David Eppstein 23:04, 15 September 2007 (UTC)

A while ago I went for a compromise in Homotopy groups of spheres by adding a single footnote to each of the first three sections to explain to the reader that the material could be found in many textbooks on algebraic topology, with a reference to the source actually used. For articles which use footnotes, this is a way to satisfy, more or less, the scientific citation guidelines, and help the reader, without peppering the article with cites to the same source. For instance, this could be done at Hilbert space, which is still potentially in the GA/R firing line. Geometry guy 17:22, 16 September 2007 (UTC)

Another ultimatum from our friends in GA

Not that I like beating on the dead horse (if that's the expression), but a recent ultimatum note posted on Talk:Hilbert space marks a new low for self-appointed sheriffs of Wikipedia. Very reminiscent of an earlier condescending/ultimative requirements expressed by a certain user in connection with the GA review of Georg Cantor, which many people might remember well. I strongly object to the tone and the content of their messages, whether it is a product of a specific doctrine or just a sign of immaturity of the individuals posting these messages on behalf of their project, and urge mathematics community to speak out against this practice. Arcfrk 22:25, 8 September 2007 (UTC)

These inline cite issues seem to have also reared its head on Talk:Derivative which is being put on hold for GA. --Salix alba (talk) 22:45, 8 September 2007 (UTC)

And Talk:Exponentiation and Talk:Homotopy groups of spheres‎. He's been busy. --Salix alba (talk) 22:52, 8 September 2007 (UTC)

There is little tangible benefit to meeting the requests of GA reviewers; I don't see that working to get articles to GA status makes much difference in terms of their actual quality. — Carl (CBM · talk) 02:41, 9 September 2007 (UTC)

And there is a risk that attempts to meet their demands result in actual deterioration – see, for example, Wikipedia:Good article review#Derivative.  --Lambiam 04:26, 9 September 2007 (UTC)

Euclidean geometry was recently "bold delisted" (meaning without ultimatum).  --Lambiam 04:47, 9 September 2007 (UTC)

I've restored that listing. LaraLove♥ 16:33, 9 September 2007 (UTC)

Greetings from GA. I was directed here from a message on my talk page. I'd like to clarify some issues from the above posts and those elsewhere. First let me speak on the comment "Geometry Guy was called away by real life, so don't expect him to influence GA any further." G'guy may be on wikibreak, but his influence remains in GA. I have immense respect for him. His input has proven to be invaluable to GA and has triggered much of the changes being made in the project now to improve it. He is sorely missed; surely not forgotten.

Alright, to address concerns. As far as the message on Talk:Hilbert space, I've replied here. Please take the time to look over that reply. Concerning GA as a whole, the situation is sort of a catch-22; We're damned if we do, damned if we don't. Working in the GA project guarantees that one will encounter hostility regularly. We get criticism for lack of quality, our procedures, our criteria and enforcement thereof... pretty much everything. And now we're catching flack for trying to improve it. But that's fine. We need all the perspectives.
[Que seque]

Please attempt to see the situation from our point of view. It's all volunteer, of course, so there's always going to be reviewers that don't necessarily know what's up with the differences in reviewing certain articles. Scientific, wrestling, current events, etc. I try to stay on top of it, and I have addressed the issue. Math articles are exempt from sweeps for now. Hopefully G'guy will return and he can do those. But I've talked with the reviewer, explained the difference in needed citation, directed him to read the Wikipedia:Scientific citation guidelines, and asked that he not review any other math articles. All math, physics and chemistry articles will be reviewed by someone who is knowledgeable in the respective field at a later time. Considering the quality generally maintained within your projects, I'm sure there won't be issues. Additionally, any messages you've received on your article talk pages regarding sweeps, disregard those at this time. The GA/R for the above article has also been archived.

We're trying to improve the project. We can only do so much at a time and everything takes time. There are over 3,000 GAs right now, and more added every day. Sweeps will eventually get them all... that's the plan anyway. It's simply to ensure that GA is of respectable quality. All GAs should be maintained after being listed. We're simply reviewing listed articles to ensure this. The template is not meant to read as an ultimatum, but as a notification that the article has been reviewed and changes are necessary. The seven days is for abandoned articles. Those that show no changes, no response in seven days will be delisted (providing the changes are more than we can fix ourselves). Addressing issues isn't necessarily making the requested changes. If the reviewer is misunderstanding or otherwise requesting unnecessary changes, explain that to them. Acknowledging the review and talking with the reviewer about the possible issues is all we're looking for as far as the "If issues are not addressed..." part.

Past that, we have proposals coming out of our ears. I'm sure you all know how long consensus can take for various things. We're trying. Seriously. It's a learn as you go process. We're adapting as we encounter issues. Please feel free to voice your concerns with us. Like I said, we need all perspectives. The template message is a perfect example. I wrote it, I didn't realize it came off as harsh, but I wrote it to the point. This is who we are, this is what we're doing, this is what we need from you, these are the possible results. Let me know if you need help. Suggestions to improve it are certainly welcome. I acknowledge that I view things such as these differently that others, so I apologize for the misunderstanding with that. LaraLove♥ 05:51, 9 September 2007 (UTC)

It really does sound harsh and threatening. Instead of saying "This article has been reviewed" you could say: "We are in the process of reviewing". And "I found some issues that I think may need to be addressed" sounds a lot more friendly than "I have found there are some issues that need to be addressed". After all, the GA reviewers are not the ultimate authority; they can be wrong for a variety of reasons. And what about "I hope these issue can be addressed in a satisfactory way, so that the article can remain listed as a Good article. Otherwise, it may need to be delisted" instead of "If issues are addressed, the article will remain listed as a Good article. Otherwise, it will be delisted". It would really help if the reviewer can be somewhat specific about the perceived problems; a message like "References are not specific and not enough" is not actually helpful. Finally, the meaning of "GA on hold" is not immediately and readily clear, and could perhaps be replaced by something less cryptic, such as "Good Article review opened", and if a review is started, wouldn't it be nice to include a link to it? Not everyone is familiar with the process, and it is not immediately apparent where to look.  --Lambiam 08:10, 9 September 2007 (UTC)

Thank you. I will alter the template to read more like what you have here. Although, not the "Good article review opened part" as on hold is separate from the review process. Ugh. I have a proposal in to make this more clear as well. IvoShandor doesn't think this is an issue. Okay, GA/R is where articles are taken to be nominated for delistment. In the sweeps process, we're putting articles on hold with the above message to be improved. If they are, fantastic. If not, then they are either delisted on sent to GA/R, depending on quality. So there is no link to an article review. It's just the one editor listing possible issues and asking that they be addressed. Hopefully I worded that clearly. LaraLove♥ 16:24, 9 September 2007 (UTC)

The tone of your comments differs greatly from those found here. I would point specifically to "There are not enough references to cover all the keypoints. I am giving seven days for improvements to be made. If issues are addressed, the article will remain listed as a Good article. Otherwise, it will be delisted." That sounds like a demand, and a threat, with a strict timetable (he has given seven days). Furthermore, the "problem" he describes is terribly vague (his post is obviously cut-and-pasted). But "not enough references"? Did he do a quick count (seven in-line citations, five general references) and apparently 12 is not enough? If he wants to identify some number of "keypoints" that are not properly cited, he's free to do so, but "there are not enough references" seems absurd to me. If a "reviewer" wants to be taken seriously, s/he should put a little more into communicating what they think is wrong with the article.

I'm sure it's hard, since "it's all volunteer" - but it's not like we're getting paid either. Writing highly-technical and high-specialized articles like those on higher mathematics is difficult and qualified work, and if the reward for writing a "good" article is feeling like one is constantly being told what to do by these "reviewers," I expect that "good article" status will come to mean less and less. I would like to see "good" articles continue to truly be measured by how good they are, and I completely understand that this requires upkeep, but if the reviewers make the editors (who actually took the time to write a good article) feel like they're being condescendingly bossed-around or told what to do by someone who is otherwise uninvolved in the editing process, they are going to stop writing "good" articles. If that means that "good articles" will no longer be a concern to mathematics editors on Wikipedia, so be it, but it could even drive people to stop contributing altogether. I assume that all of this "reviewing" is done with the best of intentions, and I'd surmise that this sort of an outcome may not have been anticipated, but if I contributed for months, or even years, on an article that was recognized as "good" I would not be pleased to have a reviewer come by once a month and say "I don't think you have enough references, you fail the criteria, I will give you seven days to fix it." It's frustrating, and it makes people feel like they're being bossed around and like their work is being devalued. (Note that I am repeating alot of what Lambiam said, due to an edit conflict - we both wrote our responses at about the same time.) --Cheeser1 08:24, 9 September 2007 (UTC)

I think it's a matter of time management. Which is an issue that will be addressed. We're at the start of sweeps and it's a little overwhelming. A few thousand articles and a hand full of reviewers. So I think there is a feeling of pressure to hurry up with it which is apparently resulting it vague reviews. I'll make some adjustments to the instructions for sweeps to correct this. I understand your frustration and it is certainly not our intention to devalue your work. Not at all. We want all the the listed articles to be of good quality, obviously. We want consistency within the project, and we want people to know that if they're reading a Good article that it's been reviewed and is maintained at that quality. But it's counter-productive to attempt to improve our reputation in the community if we kill our reputation in the process. LaraLove♥ 16:24, 9 September 2007 (UTC)

I've only just noticed this thread: I saw that Derivative was up for good article review and was intrigued (well, it is hard to stay away!). I was delighted to see how differently it was handled to Georg Cantor: check the archive and you will see the fast consensus to keep, and also knowledge and respect of the scientific citation guidelines.

GA/R is not a monolith: the regular reviewers there have many different shades of opinion. There are still one or two who count inlines and say "delist" if there are not enough, but they are a minority, and there are currently discussions to clarify the criteria so that the verifiability issue is considered more thoughtfully. It is better to engage than be enraged.

I appreciate the kind comments about me above, but I have noticed also that GA/R has been enriched recently by a number of thoughtful new reviewers, such as Drewcifer, who add to the ranks of those such as LaraLove and Jayron who have realistic expectations and bend over backwards to be helpful. For example, reviewers such as LingNut, whose own personal standards for citation are very high, argue quite strongly that GA should not impose such high standards. The GA/R process is not perfect, but no WP process is, and GA/R seems to be one of the better ones right now, with growing aspirations to be one of the best, despite the demands that places on reviewers.

As for the math GAs, I actually went through these checking them several months ago: some people will remember I brought up several cases, and several articles (mostly game theory) were delisted as a result. I think the current math GAs can withstand any sweep. In particular, for Homotopy groups of spheres, I did quite a bit of work myself improving the verifiability of this article, and so I say "bring it on"! Take the article to GA/R. I am confident it will remain listed, just like Derivative. Geometry guy 23:24, 11 September 2007 (UTC)

Addendum: actually, Hilbert space, Exponentiation and Euclidean geometry do seem to lack sources to me. Euclidean geometry is perhaps the weakest, because the material isn't purely mathematical: there is history and applications. I agree that a "bold delist" was inappropriate in this case, but I think the article is in desperate need of sources right now. Geometry guy 23:10, 13 September 2007 (UTC)

If you have time, can you list the issues for each article on their respective talk pages with a note of which articles you have reviewed and their status here? LaraLove 23:33, 13 September 2007 (UTC)

In addition Order theory and Ordinal number are poorly sourced (and the former does not meet WP:LEAD). With these five articles (the three above, and these two), the issue is not purely about inline citations, but about making it clear to the reader where to verify the article's claims. It isn't so much to ask: the other GAs in mathematics seem to me to meet this criterion. If you care for these articles, please fix them. I have told the GA sweeps project that all of the other math GAs are fine. Geometry guy 23:27, 14 September 2007 (UTC)

A few remarks on in-line citations

I'd like to add a few remarks to this regularly appearing discussion on in-line citations in maths (and other science) articles. While I've had my share in arguing against (and occasionally adapting to) what I've perceived as overly strict application and/or interpretation of WP:CITE, I do however perceive a need to improve the standards of citing references in maths articles.

First of all, let's remember that in writing professional maths papers or books, we do indeed constantly use lots of in-line citations. This is particularly true in the types of articles whose nature comes fairly close to that of an encyclopaedia article: namely survey articles (such as regularly appear in, e.g., the Bulletin of the American Mathematical Society) or (in longer format) monographs. It is not unusual to have well above 100 sources listed in the bibliography of such articles and monographs, normally with each one of them cited in the text of the article.

What differs, though, is the format of citations used: instead of footnotes mainly advocated by the WP:GA and WP:FA reviewers (due to the prevalence of footnotes in social science articles?), we quite typically use our own variant of Harvard citations, where:

• cited works are numbered in the order in which they appear in the bibliography, sorted in the alphabetical order of the authors, with publication time as the secondary ordering criterion; and
• citations in the text are made by bracketing the work's number, with or without the author's name, and possibly giving a more precise pointer within the cited work. Examples of the usage could be like:
  • "For more details on derived categories, see [12], [17], [23] or [36]."
  • "The concept of a fibred category was introduced in Grothendieck [16]."
  • "This is an immediate consequence of Serre [7], Theorem 3."

We are quite adept at using this citation framework in a way which is unintrusive and does not negatively affect the flow of text or hamper reading (in ways we feel footnotes do). So perhaps it would make sense to adopt this common maths referencing scheme on Wikipedia as well. Or use the Harvard scheme as a substitute, they are not that different, even though the latter takes more space in cases such as the first example above.

Second, some reluctance to cite sources as requested by GA and FA reviewers probably stems from a combination of the stated reason to include the citations and the nature of articles under discussion. In my mind citations serve three primary purposes:

1. Providing the reader with a list of the best sources for further reading on the topic, compiled by someone with a good overall perspective on the existing literature; this is what I consider the most important purpose, the importance of which is growing while the number of articles and books keeps growing together with increasing specialisation of subfields of maths;
2. Pointing out the original sources for the purposes of keeping track of the history of the subject and establishing the priority of work and results; and
3. Making it possible to verify the statements made in the text.

Now I think that few would argue that purpose 1 above is something that a good Wikipedia article could ignore. This is however an area where work remains to be done in maths articles. To a lesser extent I believe the same applies to purpose 2.

However, it is purpose 3 that is (explicitly or implicitly) pointed out by FA and GA reviewers on Wikipedia. Again, if the article was on a topic at the forefront of current research, the need to be able to verify recent results used to prove new ones would be clear to every mathematician, and proper (in-line) citing of sources would be required automatically. But since most of the maths articles for which FA or GA status has been attempted / obtained are in certain sense of "general interest", they contain mostly results that are so "obvious" or "common knowledge" to the authors that the citation requirement seems like arbitrary harrassment. And of course it would not be good style to either reference every statement about derivatives to a standard textbook with its own in-line citation, nor would it make sense to source every theorem and proposition to the original (17th, 18th or 19th century) source. A typical solution has been then to list a few textbooks at the end of the article and leave it at that.

Now this approach is easy enough for other maths article authors to recognise and accept. However, this implicit way of saying that all in the article is "well-known" and "standard" may not be clear to a non-mathematician reading the article. And it clearly is not to several reviewers on Wikipedia. Therefore it would make sense to make the situation explicit in the article. One way to do it is a footnote like the one in Derivative (note number 1). I would advocate an approach more in line with our normal practice in writing introductions to original reserch articles and in writing survery articles and monographs: write the content of such a footnote in the text of the article, and provide the references to best sources in it. This way the purpose 1 of the list above is also better served, in particular when the article's list of references grow beyond the list of good general references to the topic. And this way the verifiability issue, which is indeed secondary for "standard, established" results, would get resolved as a side effect. Stca74 09:40, 9 September 2007 (UTC)

Your recommendation to put a list of general references early in the article is mirrored by the scientific citation guidelines, which were developed to express our ideals for inline citation in Wikipedia articles about math and some science areas. — Carl (CBM · talk) 12:00, 9 September 2007 (UTC)

Mathematics editors should also visit Wikipedia:WikiProject Mathematics/Editor resources, and especially Wikipedia:WikiProject Mathematics/Reference resources. We provide guidance for choices about article design, citation templates (some, like {{Harv}}, with automatically generated links), helpful citation generating tools, and where to find sources. We realize that neither readers nor editors always have close at hand a great personal library, university library, or bookstore, so we list over 40 Web sources such as the Internet Archive. --KSmrq^T 19:26, 9 September 2007 (UTC)

• Purpose 1 is, I think, one reason that some things don't get sourced. If an article can be sourced, as many can, out of "any decent algebra textbook", choosing Jacobson or Lang or Herstein seems unfair to the rest of them. Septentrionalis PMAnderson 22:36, 15 September 2007 (UTC)

As I mentioned above, a footnote (if footnotes are used) can say explicitly that "the material in this section can be found in any decent algebra textbook, such as ..." followed by one or more representative texts. It is dead easy to do (cf. Homotopy groups of spheres) and keeps the GA crowd happy by essentially satisfying WP:SCG.

As for which text(s) to use, since the purpose is to help the reader, not the authors of textbooks, the cited text(s) should be the ones used by the editors (if these are known), since they will probably bear the closest resemblence to the material in the article. Geometry guy 17:32, 16 September 2007 (UTC)

today's featured article...

...is worth a look. --Trovatore 00:17, 15 September 2007 (UTC)

For future visitors of this page: Wikipedia:Today's featured article/September 15, 2007.  --Lambiam 01:13, 15 September 2007 (UTC)

Congratulations! Well deserved, I think... Geometry guy 15:52, 15 September 2007 (UTC)

That's great! Thanks to everybody involved. Oleg Alexandrov (talk) 03:05, 18 September 2007 (UTC)

graphism

There's a proposition (graphism thesis) that hard sciences use more graphs than soft sciences. A paper once found that mathematics use less graphs than natural sciences. If the proposition is correct, then that paper's results could be useful in a discussion of whether mathematics is a science. Perhaps the topic should be discussed in Philosophy of mathematics or maybe in the Mathematics article itself. I added a sentence to Philosophy of mathematics, but I think more input is required. Ideally we should find someone who has expertise in graphism... Could you help me find out more about the implications of graphism in the question of whether maths is a science and locate more references in academic papers and the Web? NerdyNSK 17:04, 16 September 2007 (UTC)

The actual conclusion of the second source is that hard sciences use graphs, soft sciences use tables. They reject the most obvious explanation, that the hard sciences have more quantitative information (although this is likely to explain at least part of the result on mathematics); but they say nothing about the other obvious reason: independent variables in physics are often quantitative; in sociology they often are not. (Another factor in mathematics is that mathematics is not limited to a small number of dimensions. Even Edward Tufte has trouble putting more than six dimensions in a graph.) Septentrionalis PMAnderson 18:43, 16 September 2007 (UTC)

• Also, the definition of graph being used includes a numerical scale, apparently on both axes. This would exclude diagrams, such as the subjects of graph theory, which are not uncommon in mathematical papers. As far as I can see, the proposed mechanisms for graphism should include these, however. Septentrionalis PMAnderson 17:29, 17 September 2007 (UTC)
  • To be pedantic: graph theory is about a family of combinatorial objects that are not any more closely related to visualization of data than linear algebra is related to tables. The diagrams you describe are more properly the subject of graph drawing. —David Eppstein 17:35, 17 September 2007 (UTC)
    • No, I was merely not being pedantic; I meant, and mean, graphs, in the full mathematical sense; but a different term may be clearer to a philosopher. Septentrionalis PMAnderson 18:04, 17 September 2007 (UTC)

Graphs are more useful when the data is uncertain, such as in physical observations, than when it is known and described exactly, such as mathematical functions. I doubt this hypothesis very strongly. ←Ben^B4 21:57, 16 September 2007 (UTC)

mbox Template

Hi everyone! I'd like to offer a tool for editing math articles.

Many math articles have HTML mathematics in them, e.g.

''f'': ''X'' × ''Y'' → [0, 1]

which displays as f: X × Y → [0, 1]. One of the disadvantages of this practice is that the HTML math might get broken between lines. A common solution is to use non-breaking spaces:

''f'': ''X'' × ''Y'' → [0, 1]

However, this makes the code a bit unreadable, and it seems like there ought to be a better solution. I looked into it, and eventually found a suggestion by grubber on the talk for the mathematics manual of style:

<span style="white-space: nowrap;">''f'': ''X'' × ''Y'' → [0, 1]</span>

This is somewhat nicer, but still a bit lengthy, and hard to remember. So I thought, why not make it a template?

{{mbox| ''f'': ''X'' × ''Y'' → [0, 1]}}

I tried this (Template:Mbox), and it works nicely. See User:Jim.belk/Sandbox for a demonstration. The only disadvantage is that equals signs work strangely in templates, so you need to use |= instead of | for equations. (It probably also gets confused if you try to use a vertical bar for absolute value.) By the way, the name "mbox" comes from a TeX command that can be used to prevent line breaks in formulas. It's possible that some other name (e.g. "hmath") would be better. Jim 03:59, 18 September 2007 (UTC)

Looks good! Michael Hardy 04:42, 18 September 2007 (UTC)

Nice, but addresses the wrong problem. The real problem is that the rendering of LaTeX by wikimedia is bad. If Planetmath can do right, why not here. After wikimedia gets around to fixing the problem, we'll be all busy putting the LaTeX back where it belongs to get rid of the incorrect fonts and codes that were never meant to be used for math... Jmath666 06:17, 18 September 2007 (UTC)

"Morera's theorem" needs work

Two quite different statement on the same topic, both purporting to be Morera's theorem, have appeared in that article. Reverted to the first version, which says if the integral of a continuous complex-valued function of a complex variable along any closed path is zero, then the function is holomorphic. The second version said that if the function has an antiderivative, then it's holomorphic. I've commented out the proof of the second version since that's not what's now in the article. At talk:Morera's theorem there is some discussion. I'm posting here in order to ask others to either join that discussion or work further on the page itself. (So maybe that talk page (rather than this one) is the best place to say something about pros and cons of the two versions.) Michael Hardy 02:35, 14 September 2007 (UTC)

actually the difference is the following. version 1: if the the function has an antiderivative everywhere on its domain (this is equivalent to integral along any closed path being zero), then it's holomorphic. version 2: if the the function has antiderivatives locally on its domain, then it's holomorphic. version 1 is trivial and there's nothing complex-analytic about it. version 2 actually characterizes holomorphy. Mct mht 00:01, 15 September 2007 (UTC)

Could we actually, like, approach this in a Wikipedia fashion? You know, citing sources that use Morera's theorem in some or other fashion? Abstractly discussing what the theorem might say is not what we should do, at least in the first instance. I have just added the EoM page as external link. Charles Matthews 15:16, 18 September 2007 (UTC)

Journal of the Royal Statistical Society

Hi, I noticed that Journal of the Royal Statistical Society was on the User:Mathbot/Most wanted redlinks list. This article is the first "collaboration of the week" for the Academic Journals project. If you have any suggestions regarding this topic, or know of notable articles that have appeared in this journal, please join in the discussions on the project talk, or the talk page of the article. John Vandenberg 03:18, 21 September 2007 (UTC)

WP:SACU

A heads-up about (I fear) the latest example of the WP box-ticking fetish. Wikipedia:WikiProject Short article clean-up is devoted to making sure every article of five sentences or less has an explicit "References" section. (An "external links" section doesn't count, but may get renamed).

Editors are also encouraged to ask themselves whether the articles should be PROD'd / AFD'd / speedy deleted.

Yes, they could do some good; but the whole gung-ho project mission makes me feel uneasy, and I just hope somebody will keep an eye on them. Projects like this can get out of control and start doing real damage, if people start instituting formalised requirements ahead of actual usefulness. Jheald 22:00, 19 September 2007 (UTC)

I agree this is worrisome. Articles should grow and also add references naturally. Just mass slapping {{unreferenced}} indiscriminately or even complaining that an article has citations, but not inline, can be a bad idea. Hopefully those guys will use common sense and focus more on articles with disputed content rather than math stubs. Oleg Alexandrov (talk) 14:43, 21 September 2007 (UTC)

Animated image of sphere packing

An anon contributor has repeatedly removed this animated image from the sphere packing article. He/she believes that it is distracting. I believe it is informative and improves the article. Does anyone else have views on this ? (Raising this here rather than on article talk page to get more attention and hopefully establish a quick consensus) Gandalf61 16:25, 21 September 2007 (UTC)

I've commented there - I think the issue should be raised there too (to avoid what might be considered forum shopping - I know that's not what you're doing, but the discussion should be taking place over there). But I believe forcing the removal of that image without consensus is definitely not in the spirit of consensus-building. I also think there's nothing wrong with the image. --Cheeser1 16:58, 21 September 2007 (UTC)

Update: the user has violated WP:3RR, and I've issued a warning. The next reversion, if one comes, will prompt me to throw this one to the WP:AN/3RR. --Cheeser1 17:54, 21 September 2007 (UTC)

In a previous discussion, though I don't remember when, concerning animated images on pages, we decided that the best compromise was to replace the animation with a still shot and a link to the full animation, so that the motion was no longer intrusive on a casual reader. Surely this will work here as well? Ryan Reich 19:06, 21 September 2007 (UTC)

You're probably referring to Wikipedia talk:WikiProject Mathematics/Archive 24#Moving pictures and the Tesseract on our Project page.  --Lambiam 21:34, 21 September 2007 (UTC)

That would, of course, be a reasonable alternative. Of course, nobody provided a replacement image, nor did the user in question take the time to discuss the matter on the talk page. I think you'd do well directing your comments to the (seemingly more active) discussion at Talk: Sphere packing. Thanks for your input. --Cheeser1 20:36, 21 September 2007 (UTC)

I've now done that here. While I like the idea of animations as a useful device, I often find them infuriating. Take the spiral on spheres in spiral. If you actually want to see whats happening the animations move too quick. The still frames in Rhumb line actually convay more information. --Salix alba (talk) 20:57, 21 September 2007 (UTC)

"Planning statistical research" nominated for deletion

Planning statistical research has been nominated for deletion. Post opinions at Wikipedia:Articles for deletion/Planning statistical research. Michael Hardy 15:44, 22 September 2007 (UTC)

New Harvard citation template

I drafted a citation template {{Harvard citations}} (or {{harvs}}) as an upgrade of {{Harvard citation}} that can handle multiple links, and has some other extra features. For example, it can do things like

"The theory was developed by F.J. Murray and J. von Neumann (1936, 1937, 1943)",

with links to multiple papers and to the authors. Feedback and bug reports can go to the talk page. R.e.b. 19:27, 23 September 2007 (UTC)

Smooth coarea formula

This short orphan article was nominated for deletion. I don't understand any of it, but I'm hoping that the members of this wikiproject will know enough to tell if the article is legitimate and the topic notable enough. Cheers,Itub 13:01, 24 September 2007 (UTC)

There was a problem with the article, but it was fixed by an edit that occurred after the nomination for deletion. However, I am not familiar with the subject matter, so I can't comment on whether the content would qualify it for deletion. If someone familiar with Riemannian geometry could look at it, maybe that will be enough to settle the question. Michael Hardy 23:20, 24 September 2007 (UTC)

I had a look and commented on the article talk page. Should be a "keep" as far as validity of content. I'm not an expert on coarea formulae, but I suspect we'd eventually prefer to expand the article to cover a selection of them. --KSmrq^T 05:10, 25 September 2007 (UTC)

Multiplicative inverse

Can someone help keep an eye on multiplicative inverse, please? For some reason it's being persistently vandalized today (I already warned one anon and blocked another) and if I revert again it'll be my third. —David Eppstein 03:58, 25 September 2007 (UTC)

Reverting obvious vandalism is exempt from 3RR. (But help is always nice; vandal fighting is tedious.) --KSmrq^T 05:13, 25 September 2007 (UTC)

fnof character

I recently noticed an HTML entity called "ƒ" that makes an italic f with a tail. Compare:

ƒ(x) = 2x + 1    vs.    f(x) = 2x + 1

The ƒ looks nicer than a standard italic f (shown on the right above), and KSmrq has observed that it makes derivatives more visible:

ƒ′(x)    vs.    f′(x)

This character is standard HTML 4.0 (the same as greek letters and many other math characters we use, see list of XML and HTML character entity references). It appears to work in older browsers like IE5/Win, and is included in common fonts such as Arial and Times New Roman. The unicode description of the character suggests "function symbol" as one of the intended uses. (Many thanks to KSmrq for compiling this information.)

I've added ƒ to Wikipedia:Mathematical symbols. If there are no objections, I'd like to suggest the widespread use of this symbol in articles involving functions. Jim 02:51, 25 September 2007 (UTC)

Actually, that does not look nice in my setup (Firefox 2, Win32). I wonder if that's my taste or my fonts. It seems to kern poorly at small sizes (here I default to TNR 16pt, but I don't know what it's in). CRGreathouse (t | c) 03:18, 25 September 2007 (UTC)

That's interesting. I think the difference is more pronounced in a sans-serif font, which is the most common default setting on Wikipedia for both Firefox and IE. Here are some comparisons:

• Sans-Serif (e.g. Arial): fnof gives ƒ′(x), italic f gives f′(x)
• Serif (e.g. TNR): fnof gives ƒ′(x), italic f gives f′(x)
• Monospace (e.g. Courier): fnof gives ƒ′(x), italic f gives f′(x)

Jim 04:04, 25 September 2007 (UTC)

The "fnof" character also matches the TeX "f" better than does the ordinary plain-text italic "f", at least on the browser I'm using. Michael Hardy 06:10, 25 September 2007 (UTC)

Oddly on this computer all of the examples look good. I guess it goes to show that font differences are a pretty significant part of it. On my other setup the fonts are set smaller and there was distracting hinting (or lack of same) going on. CRGreathouse (t | c) 13:03, 25 September 2007 (UTC)

In other words: Using this charactre for any reason other than wanting the long tail (which is why it exists) is pointless, as everything else depends on the font. JPD (talk) 13:55, 25 September 2007 (UTC)

Oh well. Jim 16:25, 25 September 2007 (UTC)

Eh? Of course the appearance depends on the font, but that hardly makes the choice pointless. The prime stroke is so close to the f with a typical monobook.css appearance that I've seen editors deliberately add a space before it, "f ′(x)", not a desirable dodge. Many readers will benefit from use of ƒ, some will see little difference, and perhaps a handful will see something slightly worse.

We all know mathematics is the redheaded stepchild of Web typography, and is largely ignored by the MediaWiki developers. The latest word from the STIX Fonts project is "We are still looking for a beta release by the end of September." All the benefits of MathML and blahtex lie untapped. As we limp on, if ƒ helps us remove a pebble from the shoe on our good foot while doing nothing about the peg leg on the other side, it's still an improvement. :-) --KSmrq^T 18:22, 25 September 2007 (UTC)

Why don't we do it the way it should be logically - with <math> tags - and just live with the fact that it looks miserably for now, until MediaWiki gets its act together? PlanetMath and many other sites - for example fulltext html display at some major journal publishers - get it right, why Wikipedia cannot? The alternative preferred by many editors here is so put illogical kludges like this over the place, which may look well in one browser but not another, may change with browser versions, an so on. Let's just take the long term view. It would take a lot of work to fix all this and put the proper <math> tags where they belong after the rendering gets fixed, why keep adding to it. Jmath666 02:07, 29 September 2007 (UTC)

I'm relatively new here, but I haven't seen any evidence that Wikipedia will be fixing the LaTeX display any time soon. Is there some reason to think that this is likely to happen? If the math rendering is getting fixed in the next few months, then it's probably a good idea to start using the <math> tags. But if this is something that's years in the future, then the goal for now should be to make the existing pages as readable as possible. Jim 02:41, 29 September 2007 (UTC)

I have been wondering how long it will take, too. Perhaps the problem lies in the fact that the little I heard goes in the noble MathML direction instead of getting the pedestrian image-based rendering work right now, like most everyone else has done. Someone knows what is going on here? My view is that Wikipedia is here for the long haul. So even if it may take years that is not a reason to use illogical code for a temporary benefit that is quite minor, it looks bad anyway (displayed formulas are different than in text), just not as bad. I understand opinions differ. Jmath666 04:44, 29 September 2007 (UTC)

For it to happen it will take some agitation of our behalf. Its fine discussing this here but the people who get to make the decisions do not read it, WP:VPT is good, voting on bugs 672 and 6383 will help, currently they oly have 4 votes each which does not show strong community interest in the bugs. Maybe we need to contact the foundation board members to help raise the issue. The upcoming realease of STIX Fonts could be a good time to mobilise effort, but we do need to show that sufficient number of people care about the issue to convince devs to do anything about it. --Salix alba (talk) 07:53, 29 September 2007 (UTC)

Just a note: the developers have said that voting on bugs has absolutely no effect on when they fix them. The voting feature is just used as a watchlist for bugs. --Sopoforic 02:09, 30 September 2007 (UTC)

I haven't the foggiest idea who maintains the TeX software on Wikipedia. I am grateful that they exist. I have only an argument from design to tell me that they do. However, if they don't read THIS page, one wonders if they know what they're doing. Michael Hardy 18:25, 29 September 2007 (UTC)

Mathematical quilts

Mathematical quilts is a clumsily written article. Maybe there's something to it and it can evolve. Someone's proposing to delete it (not yet on AfD, but with a "prod" template. Does anyone here know anything about this? Michael Hardy 04:49, 26 September 2007 (UTC)

Whether the book reaches normal wikipedia notability standards is questionable. However there should be a place for this sort of thing in wikipedia, maybe with an broader article, whos name escapes me at the moment. There do seem to be others doing mathematical quilting as well[13]. --Salix alba (talk) 08:32, 26 September 2007 (UTC)

I was going to say that although there are quite a lot of people doing mathematical quilting, I wondered if the subject was sufficiently notable for inclusion. But then I thought it was likely that there had been at least one Mathematical Intelligencer article about it over the years, and shortly thereafter I remembered that quilting was the cover story in some journal I've read. So the topic is almost certainly verifiable, and meets the notability requirements. -- Dominus 12:32, 26 September 2007 (UTC)

The cover was not on a math journal. It was the May 2005 issue of the journal The Physics Teacher. Complete article. -- Dominus 15:54, 27 September 2007 (UTC)

Two Penrose tile quilts: [14] [15]. Not exactly published in reliable third-party print publications, but at least they show that there's more than the book to this. —David Eppstein 15:15, 26 September 2007 (UTC)

AMS Special Session in Mathematics and Mathematics Education in Fiber Arts (at the January 2005 Joint Mathematics Meetings in Atlanta, GA) Reference List: Mathematical Articles on Fiber Arts -- Dominus 15:33, 26 September 2007 (UTC)

See also Irena Swanson's page for some spectacular examples of mathematical quilts. Arcfrk 18:01, 26 September 2007 (UTC)

And this sublink of the AMS page above. Septentrionalis PMAnderson 19:11, 26 September 2007 (UTC)

Notability definitly seems to be there, quite a few of Ellison works are in the London Science Museum [16]. IEEE Spectrum have run at least 5 block design contests.[17]

The main question seems to be do we want a page on a book by one author, or a general mathematical quilts page. Or maybe a Mathematics and Fiber Arts page which could include quilting and kitted mobius bands and hyperbolic planes[18] etc.? I'd probably go for the latter. --Salix alba (talk) 20:11, 26 September 2007 (UTC)

An article on the general idea seems about right to me. Paul August ☎ 20:15, 26 September 2007 (UTC)

I've now moved it to Mathematics and fiber arts and reworked it somewhat. --Salix alba (talk) 22:16, 26 September 2007 (UTC)

If I recall correctly, there was an article (in one of those MAA periodicals or something) about someone who knits various shapes related to topology. I don't remember much of it. That might fit the article - I'll see if I can recall where/when I saw the thing. --Cheeser1 23:40, 26 September 2007 (UTC)

I have a Klein bottle ski cap that a fellow grad student knitted for me. (She may have gotten the idea from such an article). VectorPosse 00:46, 27 September 2007 (UTC)

If you could get a picture of this that would be wonderful. Currently the article has no illustrations and I guess most of the pictures on the interweb are not under open licences. --Salix alba (talk) 08:18, 27 September 2007 (UTC)

I am willing to release the photographs and illustrations here under a suitable license if you think it will be helpful. There is a more detailed explanation here. Also, my wife made a quilt that tabulates the values of the GCD function and we can put these on commons if you want to include them. Somewhere I have pictures of a quilt that incorporates the decimal digits of e, but the design itself is not mathematical. -- Dominus 14:07, 27 September 2007 (UTC)

May I suggest to move the discussion to the talk page of Mathematical quilts? Jakob.scholbach 17:47, 27 September 2007 (UTC)

Boolean logic

User:StuRat has replaced the redirect from Boolean logic to Boolean algebra (logic) by an article that is an older version of his hand and is unwilling to discuss this.[19] He appears miffed about not having been consulted.[20][21]  --Lambiam 05:28, 25 September 2007 (UTC)

I take exception to it being called "an older version of his hand". I, and many other, have worked to make the article useful for it's intended audience (general) and applications (computer science and electronics). I am completely willing to let others edit the article to improve it, as has happened all along. What I won't allow is for the article to be scrapped and replaced by yet another article aimed at a different audience (PhDs) and application (higher mathematics), in addition to the current article (Boolean algebra (structure)) aimed at that audience and target. StuRat 02:40, 26 September 2007 (UTC)

Actually he probably should have been consulted. I take some responsibility for that; I considered notifying him but decided he'd probably see the discussion and could decide for himself what to do about it.

However Boolean logic is not a good name for what he has in mind, and there's a lot wrong with the current content. I have reluctantly come to the conclusion that an article called introduction to Boolean algebra might be defensible as a "trampoline" article, an expedient I don't much like in general but which might be necessary in some special cases, this being one of them.

As regards the content, I think that at the moment the article overemphasizes the algebra of sets. There's already an algebra of sets article. --Trovatore 05:38, 25 September 2007 (UTC)

How about if we name it Boolean logic (computer science), to show that this article is aimed at those who use Boolean logic for computer science and electronics applications, not for mathematicians. I'm willing to do just about whatever it takes to maintain an article that won't be taken over by mathematicians and made completely incomprehensible to the general audience for which it is intended. If you read my comments on the discussion page, you will get some background on the article history: Talk: Boolean logic. StuRat 02:34, 26 September 2007 (UTC)

Boolean logic in computer science. If that's the content, more or less, it is a good title. Charles Matthews 08:40, 30 September 2007 (UTC)

Math Girl

A cute animation, which has in fact gotten some press coverage. Article written, fairly neutrally, by the principal author. Is this notable? Septentrionalis PMAnderson 17:19, 28 September 2007 (UTC)

It seems to have passed WP:N. The information, as far as I can tell, is correct/verifiable. Any other opinions? --Cheeser1 17:23, 28 September 2007 (UTC)

This strikes me as outside the scope of this project. CRGreathouse (t | c) 02:45, 30 September 2007 (UTC)

It's about math education at introductory calculus level, which should be in scope. Regardless of the fact that the animation style and the story seem to be at the kindergarten level. Jmath666 04:10, 30 September 2007 (UTC)

I like the article and the idea, but I'm skeptical about the notability. Here are some interesting Google results:

• Jungic "math girl": 57 hits
• Crockett "math girl": 43 hits

By contrast, a search for my last name plus the title of my thesis (which I'm certain is not notable enough for Wikipedia) garnered 191 hits, and a search for "mathematical quilts" generated a whopping 1530 hits. Though I hate to discourage anything called "Math Girl", right now this seems like too small a project to have an article on Wikipedia. Jim 05:53, 30 September 2007 (UTC)

I strongly disagree with the previous writer on the use of google as the measure of notability: "blah-blah-blah" generated about 2,330,000 hits. So what? It's not ten thousand times more notable than your thesis! On the other hand, "Riemann uniformization theorem" generated 178 hits, but, with all due respect, your thesis is epsilon notable compared to that. No one I know has written a mathematics Ph.D. thesis of comparable notability.

The article itself is all right, and the idea behind the show is very valuable. Granted, it's hardly notable enough for Encyclopaedia Britannica, but the content of the article is more suitable for an encyclopaedia than the content of many (most?) of the pop culture articles on wikipedia: take a look at Pokémon and the links therein to get an idea of what I am talking about. Arcfrk 08:08, 30 September 2007 (UTC)

I agree that Google isn't necessarily the best judge of notability. Its primary advantage is that it gives some numerical data, though this may be of dubious value. However, I think that Google is probably a reasonably good indicator on matters of pop culture, and is certainly fairly efficient at indexing media articles on a topic. Some important math topics may not get many hits on Google, but we both know that such topics are covered in many advanced textbooks on mathematics. Where do you suspect references to "Math Girl" are hiding?

(Incidentally, "Riemann uniformization theorem" doesn't get many hits on Google, but "uniformization theorem" gets 36,600. The former name doesn't seem to be too common: currently, "Riemann uniformization theorem" isn't even a redirect page here on Wikipedia.)

Again, I would really like to say that I'm completely in favor of a cartoon called "Math Girl", and I hope the creators are successful in popularizing their work. However, the fact remains that this seems to be a relatively small project stemming from cooperation between two departments at Simon Fraser University. The "press coverage" referenced in the article consists of a single article in the Simon Frasier student newspaper, and a single paragraph in the middle of an article on the CTV website. Unless there is some evidence that this is being shown to students in more than a handful of calculus classes, I can't see how it possibly qualifies as notable. Jim 09:09, 30 September 2007 (UTC)

That article on the CTV website is reproduced from The Globe and Mail, which is a high quality news source. That seems to be about it as far as news coverage goes. Generally to satisfy WP:N we ask for more than one brief mention, either multiple sources or a more in-depth source, but of course in practice people will fudge it if they think the article should be kept. Consider for example, an article on a self-published author who got only one small mention in the Globe and an article in a neighborhood newsletter. Let's also assume the subject created the article. I wonder if you would argue so strongly in favor of keeping it.

Arcfrk, I think you know all this which is why you end by making a value judgment. Sure, some people may say that we should have more material on Jean-Pierre Serre and his work instead of the latest exploits of Paris Hilton, Britney Spears, and Lindsay Lohan, but to a large degree that's not how Wikipedia functions. There are times when one can lobby people to fight for inclusion of what one considers worthwhile educational material, so maybe that is what will happen here too. --Horoball 09:49, 30 September 2007 (UTC)

(double edit conflict)

Indeed, we have learned to be skeptical of arguments for or against notability based on a Google search. I like the "blah" example; I'll try to remember it for future discussions!

Newcomers like Jim naturally take Wikipedia at its word, and also begin by assuming words have their common language meaning. But words get twisted and interpreted in strange ways here. The idea of "notability" is, logically, that a topic should be important enough to be worth an encyclopedia article. By itself, that's already tough to decide without the constraints of a printed work; in practice, that's often not how it's used here.

We argue about notability because some kid with a garage band adds a page for that name, or his friend does. Maybe our gut says that's a misuse of Wikipedia, but on what grounds do we argue our case? Notability! Fine; but every town, however small, is considered notable. (No misuse there!) The tiny school the kid attends passes review. It wouldn't surprise me if every band that recorded a "one-hit wonder" is notable. But an article on the American Institute of Mathematics can get nominated for deletion by the same kid, and we have to prove it's notable by Wikipedia's idiosyncratic standards. The general public knows few mathematicians, few theorems, and no higher mathematics. Thus we, the mathematics community, are a critical influence, and my impression is we are far more conservative than, say, anime fans. --KSmrq^T 09:51, 30 September 2007 (UTC)