Wikipedia talk:WikiProject Mathematics/Archive 22

From Wikipedia, the free encyclopedia

This is the archive file "Wikipedia talk:WikiProject Mathematics/Archive 22". It is for February 2007.

Archive This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Contents

Articles listed at Articles for deletion

Uncle G 10:54, 29 January 2007 (UTC)

It wouldn't hurt to have some intelligent people comment on the above AfD, since the first several comments were written by silly gullible people. Michael Hardy 23:40, 29 January 2007 (UTC)

The AfD was closed with a resolution of keep. --KSmrqT 20:21, 1 February 2007 (UTC)

Ancient pages

Out of curiosity, I made a version of Special:Ancientpages just for math articles: User:CMummert/Oldpages. It lists articles on Mathbot's list whose last edit was in 2005. There are no articles on Mathbot's list older than that (except for one redirect page, but I think I fixed that). CMummert · talk 04:53, 1 February 2007 (UTC)

List of mathematics articles are not Mathbot's pages. They were there long before mathbot or me were around. Oleg Alexandrov (talk) 05:13, 1 February 2007 (UTC)
Point taken. CMummert · talk 05:18, 1 February 2007 (UTC)
So the list shows that the vast majority of articles are edited quite a bit, although of course those edits could be trivial or vandalism. Interesting. Oleg Alexandrov (talk) 15:44, 1 February 2007 (UTC)
I was pleasantly surprised. The median age, by the way, is Dec. 21, 2006, and over 70% have been edited since Nov 1, 2006. CMummert · talk 16:49, 1 February 2007 (UTC)

Modularity theorem

For some time the article modularity theorem had the incorrect title Taniyama–Shimura theorem, a name invented by an editor who wrongly thought that was what the Taniyama-Shimura conjecture was called when it was proved, and never used by mathematicians. This has been fixed on the English wikipedia, but unfortunately this mistake was copied to wikipedias in many other languages. So the corresponding page in the following languages needs to be fixed: Català Deutsch Español Français Italiano עברית 日本語 Português Русский Suomi Tiếng Việt 中文 (There are links to the pages in these languages at modularity theorem.) R.e.b. 19:45, 1 February 2007 (UTC)

Infinite monkey theorem FAR

Infinite monkey theorem has been nominated for a featured article review. Articles are typically reviewed for two weeks. Please leave your comments and help us to return the article to featured quality. If concerns are not addressed during the review period, articles are moved onto the Featured Article Removal Candidates list for a further period, where editors may declare "Keep" or "Remove" the article from featured status. The instructions for the review process are here. Reviewers' concerns are here. LuciferMorgan 04:54, 2 February 2007 (UTC)

User:Farever, who apparently created his/her account for the sole purpose of nominating this article for FAR, has stated a kind of vendetta against the article's existence. This would appear to be a misuse of the FAR process. Is there a way it can be "speedy" closed? --C S (Talk) 06:13, 2 February 2007 (UTC)

Recreational mathematician article for deletion.

See Wikipedia:Articles for deletion/Ed Pegg, Jr.. Oleg Alexandrov (talk) 16:23, 31 January 2007 (UTC)

The AfD was closed with a resolution of keep. (After the stub was expanded there was no dissent.) --KSmrqT 18:21, 4 February 2007 (UTC)

Nearly orphaned article

I found number spiral a complete orphan---no other pages linked to it at all. I did a few small copy-edits and put a link to it into another article, and added the "number theory" category. Perhpas others here can figure out which other articles should link to it or other categories it should be in. It would also benefit from an illustration and perhaps other additional work. Michael Hardy 23:11, 4 February 2007 (UTC)

Somebody linked Ulam spiral to it.--70.231.149.0 01:23, 5 February 2007 (UTC)
I believe number spiral fails the notability guideline's requirement of multiple independent nontrivial references, so I nominated it for deletion. Discuss at Wikipedia:Articles for deletion/Number spiral. -- Jitse Niesen (talk) 01:38, 5 February 2007 (UTC)

If it's not kept, I think it should be merged into Ulam spiral, perhaps with comments comparing and contrasting the two, and probably in terser and more efficient language than what is now in this article. Michael Hardy 02:06, 5 February 2007 (UTC)

Factorization

We really need to merge polynomial factorization and factorization; most of the article factorization is in fact on polynomial factorization, but it is treated more basically. Should we merge polynomial factorization into factorization, or merge most of the stuff in factorization on polynomial factorization into polynomial factorization? —Mets501 (talk) 04:11, 5 February 2007 (UTC)

Although polynomial factorization is certainly a kind of factorization, a merge in either direction would be unhelpful. Each topic has a great deal to be said about it, and neither should be burdened with the baggage of the other. If anything, factorization should be expanded in other ways. --KSmrqT 05:23, 5 February 2007 (UTC)

Problem with HTML entities

Yesterday I was poking around through some W3C tables when I ran across a couple of HTML entities (&thetasym = #x03D1 and &weierp = #x2118) that were new to me. So I added them to the table of symbols in Fropuff's user space, and they showed up fine: ϑ and ℘.

Today the Weierstrass "p" symbol ℘ still works OK for me, but I get an ugly little hook for the "\vartheta" symbol that I should be able to produce by coding ϑ (this ought to look like {\scriptstyle\vartheta} – yesterday it did, and today it doesn't). I figure it has to be in my browser somewhere. I'm running SuSE Linux 9.3, and I'm using Firefox 1.06 (yeah, I should upgrade Firefox, but it's kind of a pain to do, and I haven't gotten around to it). I did shut the browser down and restart it in the interim, so that's probably how I lost the glyph for ϑ, but I don't understand how that could happen. Oh – I also have my Wiki-preferences set to render all math expressions as PNG's.

Anyway, I'm curious if other people have any insight into this phenomenon. I also think I'm starting to understand the problem with in-line <math></math> expressions a little better. Please take a look at the following line in this message.

{\displaystyle\vartheta} \quad {\textstyle\vartheta} \quad {\scriptstyle\vartheta} \quad {\scriptscriptstyle\vartheta} \quad ϑ {\displaystyle\wp} \quad {\textstyle\wp} \quad {\scriptstyle\wp} \quad {\scriptscriptstyle\wp} \quad ℘

Anyway, if everything were working right, there should be five copies of each of the symbols I'm talking about on the preceding line, in four or five different sizes. I only see three different sizes through my browser, but one of them (the {\scriptstyle} size) looks about right for rendering in-line symbols. Does it look that way to you? I'm especially curious what it looks like to Windows users. Should there be something more about different ways of rendering math symbols in the style manual for math articles? Thanks! DavidCBryant 17:36, 4 February 2007 (UTC)

Both symbols look ok to me, in Camino. They are bolder and more upright than the png math symbols, and each of the html symbols is larger than two of the corresponding png math symbols and smaller than the other two. —David Eppstein 17:56, 4 February 2007 (UTC)
All work for me. Although Fropuff's table includes TeX invocations, it's a tiny subset of all the symbols one might wish to use. Compare to my table, which includes HTML entity names (underlined) and MathML/Unicode names. --KSmrqT 18:29, 4 February 2007 (UTC)
I am using the latest version of Firefox now; and the last of the five thetas looks very crudely drawn compared to the other four. JRSpriggs 06:23, 5 February 2007 (UTC)
Looking at them with my Internet Explorer, the last of the five in each group just look like rectangles. JRSpriggs 06:27, 5 February 2007 (UTC)
Thanks for the feedback. Are you using Windows XP, JR? "Crudely drawn" on the fifth one makes sense, since that's a font rendering, as compared to a graphic image (PNG) for the first four. I've done enough digging around on my own machine to convince myself that the problem is not really in my browser. It's buried deeper, in the font engine ("xft") for Linux. Oh, joy. Something else to learn about.  ;^>
Thanks for the big table, KSmrq. My system renders most of the symbols OK, up to about 0x'2900'. Past that I mostly get little square symbols that contain the four hex digits buried in the Unicode string. The "vartheta" string is anomalous (for me) because it's just 0x'03D1', a fairly low value, but it still renders badly. I guess I have to locate my LaTeX/Amsmath fonts and get them configured so that "xft" can find them. DavidCBryant 12:39, 5 February 2007 (UTC)
Squares appearing instead of the characters has nothing to do with which browser is being used - it is just a matter of whether an appropriate font including those unicode characters is installed. JPD (talk) 13:07, 5 February 2007 (UTC)
Not to put too fine a point on it, JPD, but it's not only a matter of installing the fonts. With most browsers, there's also a font configuration tool. So if you have two different fonts installed that can both render a particular Unicode glyph, your browser is going to select one of those ... and it may not be the one you'd rather see.
Anyway, I've got my ϑ problem straightened out now. I found fairly simple instructions on this web page, which may be of help to other Linux users. I installed about 50 new fonts (~ 7 MB) by following the instructions ... KSmrq's table looks better now, although I still have a significant hole between roughly 0x'2900' and 0x'02A00'. DavidCBryant 04:59, 6 February 2007 (UTC)
We are expecting the beta release of the STIX fonts "Real Soon Now", but meanwhile I find that one font, Code2000 fills almost every hole. --KSmrqT 08:56, 6 February 2007 (UTC)

To DavidCBryant: I am using: Windows 2000, version 5.0 (Build 2195: Service Pack 4). I do not understand fonts, so all I can say about that is that I use: Verdana, style regular, size 10, script western. JRSpriggs 05:38, 6 February 2007 (UTC)

Hoaxing

One zero one (talk · contribs · deleted contribs · logs · block user · block log) is editing math and physics related articles and (I think deliberately) introducing errors. They are not so easy to spot, see this edit for example. Such things are much more worrisome than plain vandalism. Something to watch for. Oleg Alexandrov (talk) 04:23, 5 February 2007 (UTC)

84.92.224.122 (talk • contribs • deleted contribs • WHOIS • RDNS • trace • RBLs • http • block user • block log) did something similar at General relativity and elsewhere — changing the values of physical constants. JRSpriggs 05:46, 6 February 2007 (UTC)

Jacobi rotation

I think this should be merged with Givens rotation because it's the same. See e.g. Golub/van Loan "Matrix Computations" --Mathemaduenn 12:55, 7 February 2007 (UTC)

Yeah, they're the same thing. In the context of eigenvalue computations, they're usually known as Jacobi rotations, and they're otherwise usually known as Givens rotations. To quote Golub and van Loan, "Jacobi rotations are no different from Givens rotations, c.f. Section 5.1.8. We submit to the name change in this section [Jacobi methods for the Symmetric Eigenvalue Problem] to honor the inventor." You might add a {{merge}} tag to the Jacobi rotations article. Lunch 23:46, 7 February 2007 (UTC)
Perhaps you have not read the articles. Both are planar rotations (rotations of a two-dimensional linear subspace of a vector space), but the Givens rotation is a one-sided transform chosen to introduce a single off-diagonal zero in a matrix without regard to symmetry, while a Jacobi rotation is a similarity transform chosen to produce a pair of zeros in a symmetric matrix. The computations are completely different. Each article does refer to the other. Sometimes the names are used interchangeably, but the distinction is well worth keeping. I believe if you actually read the excellent Golub and Van Loan book, which is cited, you will find no contradiction.
One possible source of confusion is that when we have a real symmetric matrix like
M = \begin{bmatrix} 2.6&0.2&-0.7&-2.4\\ 0.2&-4.4&1.9&2.2\\ -0.7&1.9&4.2&-2.5\\ -2.4&2.2&-2.5&-1.7 \end{bmatrix} ,
then a Givens rotation of the last two rows that zeros the −2.4 entry in the last row, when applied from the left, will (by symmetry) act in transposed form from the right to zero the −2.4 entry in the last column.
Q = \begin{bmatrix} 1&0&0&0\\ 0&0&-0.28&0.96\\ 0&0&-0.96&-0.28 \end{bmatrix}
Q M = \begin{bmatrix} 2.6&0.2&-0.7&-2.4\\ 0.2&-4.4&1.9&2.2\\ 2.5&1.58&-3.576&-0.932\\ 0&-2.44&-3.332&2.876 \end{bmatrix}
Q M Q^T = \begin{bmatrix} 2.6&0.2&2.5&0\\ 0.2&-4.4&1.58&-2.44\\ 2.5&1.58&0.10656&3.69392\\ 0&-2.44&3.69392&2.39344 \end{bmatrix}
We can systematically work in this fashion to reduce M to tridiagonal form, but we cannot diagonalize it. If we zero an element on the subdiagonal with a Givens rotation, the zero is immediately destroyed when we complete the similarity transform. So we must choose our rotation as described in the Jacobi rotation article to actually get zeros. Of course, a side-effect of that is to create a "bulge", destroying our tridiagonalization.
\begin{align} M &{}= \begin{bmatrix} 2.6&0.2&0&0\\ 0.2&-4.4&1.9&0\\ 0&1.9&4.2&-1.2\\ 0&0&-1.2&2.4 \end{bmatrix} \\ Q &{}= \begin{bmatrix} 1&0&0&0\\ 0&1&0&0\\ 0&0&0.8944272&-0.4472136\\ 0&0&0.4472136&0.8944272 \end{bmatrix} \\ Q M Q^T &{}= \begin{bmatrix} 2.6&0.2&0&0\\ 0.2&-4.4&1.69941&0.84971\\ 0&1.69941&4.8&0\\ 0&0.84971&0&1.8 \end{bmatrix} \\ \end{align}
We are not surprised; otherwise we could find roots of polynomials of arbitrary degree using radicals, which is impossible. --KSmrqT 03:05, 8 February 2007 (UTC)
The transformation both use is the same, but you're right, they're applied for different effects. In context, Givens rotations are only applied one-sided whereas the Jacobi rotations are applied as a two-sided similarity transform. Sorry, I guess a merge isn't in order. (But I do wish you wouldn't just up and delete my comments.) Lunch 03:52, 8 February 2007 (UTC)
I have not knowingly deleted your comments, and would only delete any comments in extreme circumstances, such as severe spam or vandalism. However, this bug has bitten repeatedly, so I'm (sadly) getting used to the complaint. I never get the slightest warning in advance, only irate responses on the (so far) unpredictable occasions when it occurs. (Sigh.) Maybe one day the software will get fixed, or I will find a reliable workaround. Meanwhile, please, restore anything you wish that mysteriously disappeared. --KSmrqT 06:57, 8 February 2007 (UTC)
(No worries. It happens.) Lunch 18:53, 8 February 2007 (UTC)
Have you tried not hitting the "back" button after the preview, as you say you often do? I'd be curious if that solves the problem. There's not really a need to do that anyway. --C S (Talk) 10:06, 8 February 2007 (UTC)
I don't think that this is the common use of the term "Jacobi-Rotation". That's all. Maybe my first comment was to short. So a book source should be added. The content of Jacobi-Rotation can also be used to improve Jacobi eigenvalue algorithm. But it's only a suggestion. --Mathemaduenn 13:53, 8 February 2007 (UTC)
Do you mean this isn't a common use of the term "Jacobi rotation"? In that case, isn't Golub and van Loan good enough to establish that usage?
Or do you mean this isn't the common use of the term "Jacobi rotation"? If so, what else have you seen this term used for?
Thanks, Lunch 18:53, 8 February 2007 (UTC)
The use is "It is the same as Givens Rotation." and how KSmrq said these names are used interchangably. Also Golub van Loan defines Givens Rotations and uses Jacobi Rotation in the same manner. It's a Rotation in the plane spanned by e_k and e_l. --Mathemaduenn 09:02, 9 February 2007 (UTC)

Various issues

A few things I'd like opinions on:

First: are we (we, anybody) using the {{Maths rating}} template? I come across articles all the time without it on them (like Polygon, for example, which I guess is top-class or high-class importance, or whatever the word is we use). I could add this to articles (talk pages, rather) when I notice it's missing, but I don't want to do it if it'd just be a waste of time.

Second: should we have a category for math problems? I don't mean like 'integrate x^2', but things like Archimedes' cattle problem, Kirkman's schoolgirl problem, Doubling the cube, etc.

Third and fourth: It seems that Mathbot hasn't removed bluelinks from the math articles listed at WP:MST since august. It's not a big deal, but should I just do it by hand? Also, there are some things listed that I don't quite know what to do with: take, for example, Mud cracks (listed on Wikipedia:Missing_science_topics/Maths18. MathWorld says that cracks in mud tend to cross each other at right angles--the Mud Cracks article on mathworld redirects to Right Angle. Mathworld cites some sources to support this, too. Should wikipedia mention that in Angle? Or maybe in Mud? (Imagine: a project mathematics banner on Mud). It seems like it'd be hard to justify putting it in there. On the other hand, it's vaguely interesting (for some definition of interesting) and apparently verifiable. What should be done?

Suggestions, as always, will be appreciated. --Sopoforic 01:41, 8 February 2007 (UTC)

Second: I don't see why not. Article topics should be categorized by what they are in addition to the field of science into which they fall, and not just in math. Melchoir 00:23, 9 February 2007 (UTC)

Re the rating template: I can imagine them being used to prioritize effort, but I don't know that anyone does use them in that fashion. One other use is to show our recognition of well written articles or improvements to articles by giving them better quality ratings. I have at least added a few more of these templates (and filled out the fields in some uncategorized ones) since seeing your note here. —David Eppstein 02:53, 9 February 2007 (UTC)

Replies:
  1. According to Wikipedia:WikiProject_Mathematics/Wikipedia_1.0 anybody can grade articles with the {{maths rating}} template as they see fit. I'm sure the guys working on it would appreciate the help.
  2. I think something like Category:Famous mathematics problems would be a fine category.
  3. You have to ask User:Oleg Alexandrov about this one. He runs mathbot.
  4. If anywhere I think it should be mentioned in the Mud article. That article wouldn't warrant a math category because of it though. In short mud cracks shouldn't be listed at WP:MST.
-- Fropuff 04:56, 9 February 2007 (UTC)
All right, I'll put the rating template on articles when I notice them. Regarding the category, probably just Category:Mathematics problems or Category:Problems or something would be enough--if they weren't famous (or at least notable), we wouldn't have articles for them. I'll leave a note for User:Oleg Alexandrov if he doesn't chime in before I get a chance. And, finally, I'll see about adding a note to Mud (once I decide whether it's worth adding, anyway). Thanks for the replies, everyone. --Sopoforic 17:19, 9 February 2007 (UTC)

Template:Hilbert's problems

What do we think about: Template:Hilbert's problems. Given that we have Category:Hilbert's problems, do we think the template is useful? Paul August ☎ 19:27, 8 February 2007 (UTC)

Some people prefer template navigation. But if we keep it, we should do one of two things:
The ancient debate of navational boxes vs. categories. Navigational boxes do not add any information, but as PMAnderson notes some people find them user friendlier (less mouseclicks.) Now if all those boxes where properly marked up being navigational boxes, people who dislike them could hide them. (P.S. Maybe there should a way to automatically generate those boxes through the articles included in a certain category.) —Ruud 20:03, 8 February 2007 (UTC)
There is a class="messagebox standard-talk collapsible collapsed" command which gives a template the little show/hide box. See {{WikiProjectBanners}}. But is that enough to fix this? Septentrionalis PMAnderson 00:12, 9 February 2007 (UTC)

I agree with Septentrionalis. I would prefer an other article template that simply pointed to the main Hilbert problems article. These problems are so diverse and their numbers are so uninformative about the actual content of the articles that I don't see the navbox as useful. —David Eppstein 23:03, 8 February 2007 (UTC)

Agree. If I saw that box at the bottom of the page, I wouldn't know what to do with it. I know what the problems are (generally), but I wouldn't know them by number. There's no real benefit in providing easy navigation if the user can't tell where they're going. The table present in Hilbert's problems is much better for navigation. A notice like {{otherarticles}} should be fine (though it ought to point to Hilbert's problems, not the category, in my opinion). --Sopoforic 02:51, 9 February 2007 (UTC)
It does both; it has two arguments; see Template talk:otherarticles Septentrionalis PMAnderson 06:01, 9 February 2007 (UTC)

The new version of Template:Hilbert's problems is much more palatable. Paul August ☎ 16:22, 9 February 2007 (UTC)

Yeah, but I still question whether it adds any value. The only use cases I can think of are: you accidentally stumble upon an article for one of Hilbert's problems, and realize that you want to look at problem #17, but don't feel like clicking the link to Hilbert's problems; or, you want to navigate through all of them to read about them, but don't want to have an extra window/tab open to click them from Hilbert's problems. The first is unlikely; the second is possible, but I don't know whether it's going to be important enough to justify an otherwise-meaningless template. On the other hand, space at the bottom of the article is cheap: this isn't doing much harm, anyway. --Sopoforic 17:15, 9 February 2007 (UTC)
Possible uses would be quickly opening each article in a tab or reading the introduction of all the Hilbert problems by using popups. Some people find this useful, some don't. —Ruud 19:57, 9 February 2007 (UTC)
If you're opening them in tabs or reading with popups, you can do it from Hilbert's problems just as easily. But, like I said, it isn't really doing any harm, either. --Sopoforic 23:50, 10 February 2007 (UTC)

What can be done about "Exponential smoothing"?

Exponential smoothing has been around for a long time. Although it appears to be on a legitimate topic, it is one of the most poorly written articles. So badly written in fact, that the difficulty of fixing it probably deters people from even trying. As EconStat (talk · contribs) said, "I feel really sorry to see poor work like this on Wiki.". It is too far out of my field for me to fix it. If no statistician is willing to fix it, perhaps we should put it up for deletion. What do you think? JRSpriggs 06:46, 5 February 2007 (UTC)

Why not just rewrite it as a short stub, since the current content is quite odd. Eventually someone will decide to expand it from there. Deletion is usually reserved for topics that don't deserve an article even if it was written well; there is no deletion criteria for articles with bad writing. CMummert · talk 14:03, 5 February 2007 (UTC)
I'm willing to take a stab at a complete rewrite, since I have extensive experience with time series. I notice that moving average already exists as a financial topic. Although financial applications of time series attract a great deal of attention today, the statistical techniques for improving "signal to noise ratio" are more widely applicable than that. For instance, someone with limited daily temperature data for a particular location might base his "forecasts" for this year's daily highs and lows on a moving average of actual temperature data from the past two or three years. DavidCBryant 16:52, 5 February 2007 (UTC)

I've done some (very badly needed!) cleanup. More is needed. Michael Hardy 21:53, 5 February 2007 (UTC)

Thanks, Michael. I hope you and the others can stand to wade through it. JRSpriggs 05:54, 6 February 2007 (UTC)
OK, I just dumped a complete rewrite in there. It's not comprehensive, but at least it's a reasonable start. Please take a look and polish it up a bit. Thanks! DavidCBryant 00:50, 10 February 2007 (UTC)
Thanks, David. It looks much better. Now that you have fixed it up, I expect more people will jump in with changes. JRSpriggs 06:32, 10 February 2007 (UTC)
I put a comment on the weighting on the talk page Talk:Exponential smoothing#Problems with weighting. JRSpriggs 03:51, 12 February 2007 (UTC)
Many thanks to Michael and David for cleaning it up. --A bit iffy 07:53, 12 February 2007 (UTC)

question about definitions of objects

Some time ago I was looking at free algebra's for a paper I was doing in Universal Algebra and I looked at the wp page on the subject. The definition was a sort of "abstract algebraic" or "ring theory" definition, and it got me to wondering about definitions of mathematical objects on WP. A free algebra could be defined in the language of UA, of category theory, or probably in other ways. How does one know which way to go for a WP page on a mathematical object? If someone wanted, they could easily go to dozens of articles and add in UA or category theory or whatever definitions of everything from logic operators to who knows. I was thinking of adding this question to the Math MoS, but I figured it would be ok to add it here. I think that the answer should be incorporated into the MMos, or somewhere in a policy guideline.

To answer my own question a bit, in many cases the most naive definition is best, who wants to mess with this stuff in an arithmetic page. On the other hand there are some cases where almost all of the research in a subject is done by logicians or computer scientists or whatever, so the definition they use is best. But, this is a type of WP imperialism, as many pages are defined with the CS way of looking at it due to the high number of editors with CS backgrounds, frustrating other potential users. Sometimes the subject can be safely split into how different fields look at it (such as Combinatorial game theory and Game theory), which can also help. Smmurphy(Talk) 02:15, 13 February 2007 (UTC)

It depends on the level of the expected reader of the article. The article on natural numbers should be much more elementary than the article on groupoids. But there is no limitaton on the space that we can use to describe a topic, so any relevant and notable definition can be included, at least in a subsection. For example, Peano axioms gives a categorical definition as a subsection. CMummert · talk 02:30, 13 February 2007 (UTC)
I am having this daydream of future WP wars where category theorists approach every math article and add a category theory definition (or move that definition to the top), while putting new research into a new category, "math which is not yet sophisticated enough to warrant a category theory approach." Computer scientists counter marking articles as "math which doesn't and will never have real world applications," and universal algebraists add a reference to Burris & Sankappanavar to dozens of articles. Smmurphy(Talk) 02:46, 13 February 2007 (UTC)

Recently translated article, needs expertise

Could someone have a look at Gauss-Lucas theorem. It needs to be put into context and made understandable to a general audience. − Twas Now ( talk • contribs • e-mail ) 05:04, 13 February 2007 (UTC)

categories for deleteion

Category:Claude Shannon, Category:Norbert Wiener are up for deletion at WP:CFD 132.205.44.134 00:45, 15 February 2007 (UTC)

Some missing topics

I have a short list of missing mathematics topics. I have tried to find all relevant links to similar articles but I would appreciate if someone could have a look at it. Thank you. - Skysmith 13:44, 16 February 2007 (UTC)

There is a list of requested articles at Wikipedia:Requested articles/Mathematics. You should consider adding these to that list. CMummert · talk 19:25, 17 February 2007 (UTC)

New math articles

Just as a reminder to perhaps newer people, there are two pages where one can see what is going on with the math articles. First, Jitse's very versatile Wikipedia:WikiProject Mathematics/Current activity, and then my own User:Mathbot/Changes to mathlists which justs lists articles added and removed from math lists.

There is a lot going on in math on Wikipedia, with at least five (or more like ten) articles created daily (my unscientific guess is that we are creating articles at a much faster rate than either PlanetMath or MathWorld). There is a lot of work however in making sure that those articles have proper style, are correct, notable, and mathematical. So, if you have time, taking a look on those pages every now and then could be a good thing. Cheers, Oleg Alexandrov (talk) 22:28, 10 February 2007 (UTC)

I looked through several new articles this morning and I see two reasons why we may be getting so many of them.
  • People who speak English as a second language want to write articles about mathematicians with whom they're familiar. I noticed several articles about Russian mathematicians which (to judge by a conspicuous lack of the word "the", and by the names of the contributors) were probably written by Russian authors.
  • People who don't understand mathematics well enough to look for a topic related to the word they're curious about. For instance, there was a new article for "leptokurtotic", plus an older stub article for "leptokurtosis". We already had redirects for leptokurtic and platykurtic in Wikipedia, and an excellent discussion of fourth moments at kurtosis. So I was able to clean out a couple of useless articles by coding some new redirect pages (for leptokurtotic and leptokurtosis and platykurtotic). Those words look like poor English usage to me, but they're present on the web, so people look them up on Wikipedia. And if the words aren't here … Voila! A new article is born. DavidCBryant 17:07, 11 February 2007 (UTC)
That's why we need more people to look over recent additions. Bad articles are caught best when caught early. :) Oleg Alexandrov (talk) 23:48, 11 February 2007 (UTC)
I keep on noticing new patterns as I work through the lists of new articles. Here are a few I've been seeing recently.
  • Ideas that are "out there", but that accrete templates/stub notices/category tags without being deleted. Probability control is a "good" example. I count 28 words in the original article, and 107 words in the moss this rolling stone has gathered during its brief career.
  • Articles that discuss mathematical concepts from a non-mathematical point of view. Here are two (Self-referencing doomsday argument rebuttal and Negative probability) that I found particularly amusing, in case anyone else wants to have a look. I took the "probability" tags off both of these so they can go swim around as philosophy and/or "pseudophysics".
  • Redirect pages that have category tags stuck on them. The categorization crew is having a field day.
  • Real solid new math articles. Supersingular K3 surface is a good example. Thankfully there are quite a few of these. DavidCBryant 14:15, 17 February 2007 (UTC)
Yep, a good chunk of those articles should either not exist as written or are not properly categorized so they are often not math. Oleg Alexandrov (talk) 16:27, 17 February 2007 (UTC)
Here's another odd situation. 56 (number) has existed since 13 December, 2003. Because of a recent unfortunate edit, its entire revision history has been disconnected. That revision history is now associated with another article, 56 (game). The article about the card game was actually created on 16 February, 2007, but it now has 3+ years of history associated with it, even though it's really only 2 days old. How am I supposed to straighten something like that out? I guess it's not a huge deal, but it doesn't seem right, somehow, that an article should be divorced from its edit history. DavidCBryant 14:34, 18 February 2007 (UTC)

The story started with an anon blanking 56 (number). Then in a few days somebody else saw the blank page and created an article about the game 56 (note that the IP of the guy who blanked, 71.233.129.128, is very similar to the IP of the guy who created the game, 71.250.232.151, although that's probably a coincidence).

I moved back 56 (game) to 56 (number) and reverted to the pre-blank version. The game itself doesn't seem notable enough to have its article. Oleg Alexandrov (talk) 16:13, 18 February 2007 (UTC)

About images

I've already done some hunting through Help:Images and other uploaded files and some other associated pages, and I haven't found an answer yet, so I'm asking a question here. Is there a recommended maximum size for image files embedded in a Wikipedia article? If not, should there be?

Here's why I'm asking. I was reviewing the list of new math articles when I ran across Shallow water equations. It's an interesting article. And it has a very nice animated GIF embedded in it, which illustrates waves in a bathtub evolving over a period of time. The only problem is that the graphics file (Image:Shallow_water_waves.gif) is roughly 7 Mb in size. On my machine the animation takes about 30 seconds to run. So I'd need a data transfer rate of ~250 Kb per second to view this animation in real time, and I don't have that kind of a connection.

I'm just wondering if there's some sort of convention for a really big graphics file like this one. I think it's a nice animation, especially for people who have high-speed internet connections. But roughly half of all the internet users are still using dial-up, AFAIK. Wouldn't it be nice to link to an animation file like this one as a separate article, with a caption describing the file you're about to download? DavidCBryant 17:31, 14 February 2007 (UTC)

You might grab a single frame from the animation to use as a still image in the article. You could then put in a wikilink to the animation with a note about its size. As a further enhancement, if you're super-duper motivated, you could recompress the animated GIF into Ogg format.
Nice addition to the article, BTW. Lunch 19:50, 14 February 2007 (UTC)
That animation could certainly be clipped to the first 2 (maybe 3) splashes without losing any content. Tesseran 05:47, 17 February 2007 (UTC)
I think you should insist that images which load slowly be put in a separate file with a link to it. That way, if you know your connection is slow you can avoid clicking on it and just read the text of the article. But if your connection is fast or you are willing to wait, you can click on the link and see the image. JRSpriggs 07:45, 17 February 2007 (UTC)

Boy's surface kills my browser due to huge animations even with a high-bandwidth connection. I think the first animation on the page adds a lot to the understandability of the surface, but it should probably be significantly reduced in size, and one of the editors of that page seems to feel that if one animation is good more must be better. —David Eppstein 08:06, 17 February 2007 (UTC)

Same thing here David, it killed my browser too. I also have a high-bandwidth connection. Definitely needs to be reduced in size.--Jersey Devil 08:21, 17 February 2007 (UTC)
Thanks for all the responses. Apparently there is no convention about big images. Or at least, nobody has heard of it. I guess I'll try asking in a couple of other fora. Oh – I should clear up some apparent misconceptions. Lunch, I didn't make up any graphics files. The shallow waves animation is from Dan Copsey. I was wrong when I said half of all internet users are on dial-up – I checked some statistics on that, and it's more like one-fourth (23% in the U.S., currently). Interestingly, the total size of the graphics files I received on the Boy's surface article is just 2.0 Mb, or about 29% of the size of the splashing water animation. But the biggest picture there (1.2Mb) is served at the top of the article. At least Dan Copsey had the good sense to put his little movie at the bottom, so a user doesn't necessarily notice that the big grapic is loading until he scrolls down the page. DavidCBryant 12:29, 17 February 2007 (UTC)
I've just done a heavy prune of Boy's surface I've cut out two of the visulisations and two sections devoted to visulisation of the surface. --Salix alba (talk) 18:39, 18 February 2007 (UTC)

Apologies?

Saying that the Project consists of "only suggestions" sounds apologetic. I think that starting a Wikiproject with an apology or disclaimer is probably a bad idea. The overall tone of the project does not sound heavy-handed. Have others reacted badly to the existence of this project? Many of the other WikiProejcts do not start their page with such caution.

Also, there was briefly an attempt to provide some kind of realistic status. It read:

Perhaps the greatest sign of distress in this WikiProejct at the start of 2007 is that of the five articles that are both FA and Top importance, three of them are biographies (See also the chart in Wikipedia:WikiProject Mathematics/Wikipedia 1.0). The other two articles are Polar coordinate system and Game theory, the latter of which contains not a single equation or diagram of a spatial nature. Surely, more of the non-biography articles in Category:Top-importance mathematics articles can be brought to FA status this year. Let us aim for the goal that some of those new (or restored) FA's contain equations.

This assessment seems fair and points in an interesting direction. Should it be put back? For perspective, you might want to familiarize yourself with WP:100K.--199.33.32.40 22:08, 4 February 2007 (UTC)

I strongly support a non-prescriptive approach to Wikipedia; and one of the things that make this project worthwhile is that it begins with one. See WP:PRO.
My opinion of WP:FA is far lower: it will promote essentially worthless articles like Daniel Webster, which is written out of Henry Cabot Lodge's artificial and dated exercise in biography. FA cannot exercise better judgment than its members', and its members have none. The current review of infinite monkey theorem alone will show this. Septentrionalis PMAnderson 22:57, 4 February 2007 (UTC)
I do not mean to be contentious, but even if the FA process sux, it is all we got. I would like to roast alive the monkey in IMT as badly as you. In my imagination, I can smell that acrid burning hair even as we...but I digress. To start off with such weak-willed verbiage as we do betrays a lack direction in our WikiProject. Wouldn't it be more motivating to just have a rousing "Let us get some FA-quality content secured" and try to ignore the gauntlet of inline-obsessed editors we have to go through to get there? We can make it if we just do the content-building and the inline refs, and respond to the feedback, no matter how unfamiliar or uncomfortable the demanding FA reviewer is with, say, first-semester calculus. We should encourage a positive, constructive attitude and, you know, pretend like the FA process does not sux. It is just a matter of determination, judgement and productive focus on our part.--199.33.32.40 00:35, 5 February 2007 (UTC)

Before you start changing the WikiProject page to state your desires for what the WikiProject should focus on, you should discuss it first on this talk page. After all, it doesn't seem like you've even participated enough in this WikiProject to get a sense of what the community thinks. Also, is there a reason that you apparently created a new account (User:Farever) and stopped using it? Keeping track of the different IPs is kind of annoying.

Speaking for myself, non-prescriptiveness is a good thing. If it distinguishes us from other WikiProjects, that'll all for the good. I believe the main purpose of this WikiProject is to improve mathematical coverage on Wikipedia. We all have differing ideas of what this means. Some may argue that this entails more FA mathematics articles, while others disagree. I personally wonder if the our time is most effectively utilized tweaking articles to make it through the FA process. There are bigger issues, e.g. fixing the elementary mathematics articles. --C S (Talk) 04:22, 5 February 2007 (UTC)

The anon does raise some good points about lack of FA quality articles for our most important articles. I second the comments about getting a login name, you will be taken more seriously with one. Over the last year I think there has been improvment in the general quality of our more important articles. The situation looks better when you consider our A-Class articles, better still when GA and B+ class articles are taken into account.
Personally I'm not too fussed about FA-class articles, it takes a lot of work to acheive and maintain and their are some systematic dificulties we face trying to gain that status: anything sufficiently advanced is likely to fail as the reviewers will find it hard; continued debates over citations; brilliant prose is not something mathematicians are trained in. I'd much rather see our top and high importance article gain B+ or higher quality so we have a good all round coverage.
One thing I've been thinking about is having some sort of formal process for an article to obtain A-class status. We could set our own criteria which reflects the needs of mathematical articles and a brief review process. --Salix alba (talk) 09:46, 5 February 2007 (UTC)
I agree that developing our own criteria for A-class articles would be a good idea. A recent discussion in Wikipedia: namespace has highlighted the fact that the vast majority of rated articles in the whole encyclopedia are rated Stub, Start, or B, while very few are rated B+ or A. This is certainly the case with the math project. The criteria don't require that A-class articles are perfect, only that they are very good, and each project is free to develop customized criteria for A-class articles (although the only one I could easily find is the Military History project). CMummert · talk 01:35, 21 February 2007 (UTC)

Cleanup main project page

I would like to mark Wikipedia:WikiProject_Mathematics/Motivation and Wikipedia:WikiProject_Mathematics/Proofs as historical and direct additional conversation from them to this page instead. I looked at them today and realized that they are still gaining new comments, even though the main project page seems to describe them as historical discusions. Any objections?

Also, the 2006 update is beginning to look a little old in Feb. 2007. I would like to add a 2007 update. What are the main issues that require attention this year? There have been several comments in recent weeks about improving articles on elementary subjects, so that might be a candidate for one. CMummert · talk 13:53, 5 February 2007 (UTC)

I think a better idea is to move the historical discussion to an archive page, e.g. .../Proofs/archive1. Keep some of the more recent comments (from the last several months). That way when the conversation starts getting long and involved it will be localized in one obvious spot, as I predict there will be long gaps in the discussion; otherwise, Werdnabot will shelve parts of it in different spots. It should also decrease repetitionin discussion. --C S (Talk) 14:34, 5 February 2007 (UTC)
The look of the project page could be improved. Wikipedia:WikiProject Molecular and Cellular Biology has a fairly clean look to it, and a minimum of fancy graphics, which should please Oleg. Work on elementary, top and high class articles are my suggestions for attention. Also WP:MATHCOTW seems to have very little activity of late, the last article chosen was in mid November. --Salix alba (talk) 14:25, 5 February 2007 (UTC)
When did Oleg get the reputation for not liking fancy graphics? :-) I would miss the 2002 retro look, but the page is starting to get unwieldy. --C S (Talk) 14:49, 5 February 2007 (UTC)
I guess Salix alba means this. I actually kept the graphics the anon put in, but reverted anon's modifications to the text part. That because I'd rather have one of us look through the page and decide how to reorder things than allow an anon coming out of the blue to do things. And no, I have nothing against graphics, as long the pretty pictures help in usability rather than just distract. Oleg Alexandrov (talk) 16:44, 5 February 2007 (UTC)

I cleaned up the page some. I like to put links at the side so that it is easier to start reading at the top. I don't dislike images per se but I don't see how the images in the newly-added nav box actually helped the page any. And they were too big. CMummert · talk 02:27, 7 February 2007 (UTC)

The nav box is great. The box currently includes links to the discussion pages on proofs and motivations. So it seems like you agree with my prior comments. Should we archive those pages (to subpages) and create clean versions, which can hopefully be managed in a more organized way? At this point, I also recommend everyone watchlist those pages, since it appears that discussion on those issues may be centered there. --C S (Talk) 13:25, 7 February 2007 (UTC)
I don't feel strongly one way or another about keeping the other two talk pages. There have been no comments on Motivation since 2005, so my instinct is to mark it as historic and list in with the talk archives of this page. The Proofs talk page should definitely be archived and cleaned up. CMummert · talk 13:56, 7 February 2007 (UTC)
The proofs and motivations subpages should really be moved into the Wikipedia talk namespace since they are, in fact, discussion pages. Whether we should mark them as historical or keep them open is debatable. Should these pages ever produce some sort of recommendations/guidelines we can put those in the Wikipedia namespace. -- Fropuff 23:32, 7 February 2007 (UTC)
There is some discussion for having discussion pages in the Wikipedia namespace: the village pump and reference desk. But if you want to move them I doubt anyone will revert it. CMummert · talk 04:38, 8 February 2007 (UTC)

I've just redone the project page in a two column layout[1]. The Nav box by CMummert caused a few layout problems so I've cut and pasted it for now and and hacked it a bit to make it work. Feel free to revert. --Salix alba (talk) 21:18, 7 February 2007 (UTC)

I like the new layout :-) Interesting how the (superb) animation of tesseract has become the unofficial logo ... only (minor) problem is that the sidebar is bigger than the main content, and will (hopefully) get bigger as more maths articles become FAs and GAs. Hmmm.... I seem to be putting (occasional) words in brackets a lot.Tompw (talk) 00:12, 8 February 2007 (UTC)
Perhaps, the graded articles material (good/featured articles) should go on a subpage. That way, future growth won't overwhelm the main project page. Also, if we are going to have a logo for this project (official or otherwise) perhaps we should vote on one. The tesseract animation is superb indeed, but I find animated gifs somewhat distracting. -- Fropuff 02:10, 8 February 2007 (UTC)
The new layout is problematic of narrow screens. The actual text content is in a narrow column about 2 inches wide on the left while the sidebar on the right is about 4 inches wide. I'll tweak it some.
Is there a non-animated image that could be used instead of the animated one? I'm not so sure that an image belongs here, instead of on the Portal. In my mind, the function of the project page is to gather together resources that help editors. CMummert · talk 03:19, 8 February 2007 (UTC)
I tried, but I couldn't get the two-column layout to work with a narrow screen. The image here (warning: advertising) is a screenshot that demonstrates the problem in firefox; it's worse in opera. So I added the new content (image, MCOTW, graded articles table) to the old version and reverted. If there really is huge demand for a two-column layout I won't fight against it, but I think that as long as the rest of WP is in single-column format our page should be as well. I'm sure I'm not the only person with a narrow screen. CMummert · talk 04:23, 8 February 2007 (UTC)


Erdős number categories up for deletion yet again

Wikipedia:Categories_for_discussion/Log/2007_February_9#Category:Erd.C5.91s_number_1

The discussion has been going on since Feb.9, so speak quickly or you may not get a chance to speak at all. —David Eppstein 07:23, 19 February 2007 (UTC)

Update: discussion was closed early today with a result of no consensus. The closing admin doesn't seem to have paid attention to our requests for more time so that WPM members could respond, but we did get quite a few responses in the short time available. —David Eppstein 23:41, 20 February 2007 (UTC)
We don't want to give the appearance of group-think by mobbing these discussions; there were probably enough comments by mathematics editors at the end. A more useful way to convince people of the notability might be to add additional published references to the Erdos number article. I would guess that some of the papers named there as "External links" were actually published. Once Erdos numbers are established as clearly notable, the categories are harder to argue against. Personally, I am neutral about the categories - individual articles are free to state the Erdos number of mathematicians whether or not they are categorized as such. CMummert · talk 00:07, 21 February 2007 (UTC)

I've refrained from these discussions, because I don't have a leaning either way. One thing that disturbs me is that it seems arbitrary to have a category for say, Erdos number 6 (which I think was actually created in between these CFD debates). Is there, for example, a cutoff typically used by social network researchers? --C S (Talk) 00:41, 21 February 2007 (UTC)

Six degrees of separation says that everyone is connected by six steps, indeed we find that the modal Erdos number is 5 and virtually all mathematicians (97%-ish) will have EN of 7 or less.[2] (P.S. debate now closed as no consensus) --Salix alba (talk) 07:56, 21 February 2007 (UTC)
Having co-authored a paper with someone is a much closer relationship than just knowing him/her. JRSpriggs 08:33, 21 February 2007 (UTC)
Closer still is co-authoring a book. As with buying a certain brand of automobile, a revealing question is, "Would you do it again?". The Mathematics Genealogy Project is also fascinating. But most of one's professional life comes after graduate school, and it is hard to document influences and connections. A broad web of co-author connections tells more than the Erdős number, and even that cannot show how much a researcher was influenced by reading Lagrange or Atiyah or Grothendieck.
We can admit, without prejudice, that the Erdős number is partly of interest as entertainment, perhaps a step up from "What's purple and commutes?" (An abelian grape.) Beyond that, it also serves to remind us to collaborate, whether as co-authors or merely as colleagues. Such encouragement is especially important for mathematicians, because by inclination and by necessity, we often work in solitude.
How do we explain that to the rest of the world? Wikipedia devotes more pages to sports and films and games and other forms of entertainment than it does to mathematics, so it would be churlish to deny us our fun. And like Playboy's other content (short stories, interviews), the Erdős number has social value. --KSmrqT 11:00, 21 February 2007 (UTC)
At the risk of being obvious, one's Erdős number is a status symbol, an indication of how well-connected one is within the mathematical community. JRSpriggs 11:50, 21 February 2007 (UTC)

Numerical digit

I just rewrote this article; please visit, improve the article, and offer comments. It's a vital article, so I think we should work to improve it to at least good article standard. --N Shar 20:25, 20 February 2007 (UTC)

If you are interested in getting some sort of assessment for the article, I would recommend first improving it, then putting it up for peer review, and finally nominating it for Featured Article status. But you should be aware that the criteria for featured articles may differ from the criteria that we ordinarily use to assess mathematical writing. CMummert · talk 20:43, 20 February 2007 (UTC)
Actually, I'm just looking for reassurance that I didn't completely ruin the article. --N Shar 20:59, 20 February 2007 (UTC)
The new article is a significant improvement over the previous version. CMummert · talk 22:40, 20 February 2007 (UTC)
Thanks. I'm still working on it; maybe I will nominate for peer review/GA after it gets some references. --N Shar 22:45, 20 February 2007 (UTC)

special characters

The articles Naive set theory and Subset use special math characters that show up as squares in IE 7. There are probably other articles like that. Can someone substitute the TeX math characters? Bubba73 (talk), 05:09, 15 February 2007 (UTC)

Or, you could join the civilized world and install a font or two. The STIX Fonts project will be releasing theirs soon, and Code2000 has long been available. This is no different than if you were reading an article involving Chinese or Arabic; you'd need fonts for those notations as well. --KSmrqT 06:39, 15 February 2007 (UTC)
It works fine for me with IE 6, so it may be an issue of some font not being installed. Looking at Wikipedia:Mathematical symbols, can you identify which characters don't display properly?  --LambiamTalk 06:35, 15 February 2007 (UTC)
A large number of those don't display correctly, e.g. the last four of Analysis, the last six of Arrows, and the last four of Logic. None of Sets are correct except the next to last one. Bubba73 (talk), 20:03, 15 February 2007 (UTC)
With Lucida Unicide, all of those display except the fifth and sixth delimiters. Bubba73 (talk), 02:13, 16 February 2007 (UTC)
Or switch to Mozilla Firefox as I suggested at Naive set theory. JRSpriggs 09:04, 15 February 2007 (UTC)
Sometimes I use Firefox, and they are OK there. But I mainly use IE 7. Many other people use that too. Bubba73 (talk), 20:04, 15 February 2007 (UTC)
No, it's a matter of an appropriate font being installed. Unless it is, it won't matter which browser you use. JPD (talk) 10:24, 15 February 2007 (UTC)
I did some digging around, and found this very helpful web page. Should we try to get some links to this worked into a Wikipedia article about MathML / Unicode / UTF-8 somewhere? DavidCBryant 16:43, 15 February 2007 (UTC)
I changed my main browser font to Lucida Sans Unicode, and now they display correctly. Other people have had the same problem. I don't know if this is the best solution, but it fixes that. My thanks to the users who helped. Bubba73 (talk), 20:43, 15 February 2007 (UTC)
Lucida Sans Unicode can fill many character gaps if you have it. Unfortunately, Microsoft restricts distribution, and Wikipedia is viewed with many browsers and operating systems. If you want to see how good your character coverage is now, I keep a page with hundreds (!) of mathematics characters, here. --KSmrqT 01:02, 16 February 2007 (UTC)
I have Lucida Sans Unicode, it must have come with my system. It fills in many of the gaps in math symbols. Farther down the page, though, most are still missing. And using that font changes the font of a lot of other websites. Bubba73 (talk), 19:17, 16 February 2007 (UTC)
Ok this is the first place I've seen a contemporary discussion about the display problems that I've been

long plagued by in WikiPedia. I've tried to find and implement and analyze solutions, but I'm at a loss, so I thought that by sharing in some detail my own experiences / results / efforts that I can help others understand, solve, or become aware of the difficulties. I appreciate the efforts of those who've taken the time to author what appear to be many fine technical pages on WikiPedia, I simply wish that many of them were usefully accessible to me. I have to believe that my IT/PC/Browser setup is representative of or better than that of most casual Wikipedia users, so I imagine that if I (and others here) are having such difficulties that the problem must be widespread and that the solution(s) are poorly known. Yet given the substantial quantity of well authored and similarly encoded technical pages on WikiPedia, it stands to reason that the authors must be aware of the appropriate guidelines, tools, techniques, and encoding / authoring techniques to create their pages, view them correctly, and have some 'quality assurance' expectation that most of their audience should be able to view their works with reasonable success and facility. However, if that's so, I've certainly missed finding "the instructions" as to how to repeat their successes! I've been impatiently waiting for the STIX fonts and FireFox 2.x for nearly a year hoping that maybe it was just a local font and rendering engine deficiency that I could clear up with those tools, but alas it's appearing unlikely to me that it's "simply" explicable and soluble by those two issues. I'm hoping that someone who does have mathematical (et.al.) symbol display mostly or wholly working with their browser can speak up and help answer what configuration(s) are beneficial to achieve that result. Though I've often seen FireFox lauded for superior MathML et. al. display versus IE6, I'm presently (and miserably) having the opposite experience. Using a "full" default install of WinXP, fully updated recent versions of FireFox 2.0.0.1 and MS IE 6.0.x, I'm unable to correctly see a great number of Wikipedia's mathematical symbols under FireFox and I'm missing a lesser but significant number of them on MSIE 6. Following the tips of users' comments here and elsewhere I've tried both FF and IE, I've installed the recommended TeX, MTExtra, Mathematica, MathML, et. al. fonts, I've experimented with changing my default encoding's font and size to several various choices including the installed Lucida Sans Unicode, et. al. I've enabled JavaScript, Cookies, Style sheets, and web site overrides of my chosen default fonts. I've reloaded the pages containing inappropriately rendered characters many times to judge whether anything I've done has made an improvement. ...For example, with this representative page:

I notice that under MSIE, even before installing non-default fonts, automatically loaded many PNG images to display graphic representations of symbols / equations, and by so doing that it got many of them correctly displayed. Whereas on the same system under FireFox, even after correctly installing numerous additional fonts, most missing characters are still displaying TeX-like escape codes in plainly visible "text code" and the rendering is neither invoking "png image loading" to display graphic representations of the symbols/equations (like MSIE is doing), nor is FireFox apparently able to or trying to render the symbols / codes via any of the installed MathMl / Unicode fonts et. al. ...For another example:

In FireFox, the above page correctly shows a couple of dozen of the symbols; most of leftmost ones are all OK; the bottom rows as well as most of the symbols on the right sides of all rows do not display correctly; instead I see what looks like character entity references as one would type them "in code". In MSIE, the page display is just about equally broken as it is under Firefox. It's showing a combination of empty box characters and textual character entity reference codes instead of most of the characters on the right and bottom side of that page. ...I've followed the instructions / suggestions at the below couple of sites to install new Windows fonts and have verified that they're installed and recognized / available to the browser(s) and system in general:

...I'm a developer myself and am not unfamiliar with XML, MathML, fonts, UNICODE, HTML, TeX, et. al., but despite several hours worth of googling, experimentation, and looking for Wikipedia pages on tips for using its Mathematical Notations with browsers, I have yet to find a solution or even consistent diagnosis for this seemingly fundamental and oft encountered problem. Even in the context of this Talk Page there's contradictory information about whether it's the fault of the browser, missing fonts, user configurations of browser fonts, et. al. If it were just a browser issue, I would expect that in either MSIE6 or FFox2 I'd have "mostly successful" experiences; I do not. If it was a missing font issue, I'd expect that having installed all the platform default as well as dozens of commended added technical / symbol / mathematical fonts would have mostly solved the problem; it has not. If it was mostly a local problem, I'd expect that most Wikipedia pages would look uniformly good / bad and act consistently; they do not; the following page displays much more successfully than the second following one:

If it was "mostly just my errant setup" I wouldn't expect to see evidences of other users reporting similar problems, so many contradictory hypotheses about the problems, and I'd have expected to find some kind of FAQ / guide suggesting "the standard configurations" that'd be working for most people who had followed suit. Another example:

...looks mostly bad and unrendered in FireFox 2, whereas on the same system, IE6 renders it mostly fine though it's clearly resorting to WikiPedia server-side provided png graphic renderings of the equations / symbols instead of using any local font. Another example, this looks to be an excellent and useful test matrix of characters / symbols:

On both MSIE6 and FireFox2 (cum MANY additionally installed technical font sets) the page's table is mostly absent correctly rendered symbols. A great number of them (but still perhaps well less than half) are properly rendered in FireFox. A significantly lesser number of them correctly render in MSIE6; mainly the first 22 rows from the top are mostly OK in MSIE, whereas most of the rest of the table is not rendered. If it was a local font issue, I'd expect relatively uniform "pass/fail" between the two simultaneously active browsers. Clearly from the quaternion WikiPedia page there must be, at least in THAT case, a different stylesheet or something Wikipedia is using to tell MSIE to graphically load many of the page's equations, whereas doing something different and more unsuccessful for FireFox. Does ANYONE have these pages mostly / fully working, and, if so, what's the secret, please?!

Overall I'd say that most work very poorly in FireFox, and the ones that work 'well' in MSIE are exclusively ones where, somehow, a graphical versus textual representation is provided 'server side' via Style Sheet or some other mechanism to cause MSIE to not even try to display most of the symbols / equations via local font or rendering technologies. Obviously that .png bitmap graphic approach leaves a lot to be desired (scalability, searchability, accessability, et. al.), but at least it's a visual "on screen" step forward wrt. total gibberish. It must be a choice that's not consistently or fully implemented server side, though, since many of the pages don't try to do that in MSIE and in such cases the result is no better than in FireFox. I haven't extensively experimented with this issue in LINUX, though I've encountered similar problems sufficiently often with FireFox/LINUX to suggest to me that there's nothing uniquely beneficial about LINUX's browser / fonts that makes it work much better than I've experienced under MS Windows. If anything LINUX tends to lack some of the more 'common' TrueType / UNICODE fonts that are 'often' available on MS Windows platforms, so OS platform doesn't seem to be the essential problem. —The preceding unsigned comment was added by 216.99.216.242 (talk) 09:10, 22 February 2007 (UTC).

(You were use *foo* for emphasis, rather than the wiki ''foo'', which conflicted with the wiki interpretation of a line beginning with "*" as a bullet point. Fixed.)
I have seen no feedback suggesting your ordeal is typical. Perhaps that means widespread satisfaction, or perhaps it means most give up in quiet despair. Given the amount of effort you have invested, we can certainly invest some ourselves to help you achieve success. I suggest we first concentrate on using Windows XP, Firefox 2.0, and one or two pages. To begin, please install a copy of BabelMap. We will especially use its "Tools>Font Analysis Utility..." (shortcut:F7) menu option, to show which installed fonts cover a chosen Unicode block.
While you do that, I'll give some relevant background. Mathematics display within Wikipedia is far from ideal, mixing several different techniques. Complicated equations imitate TeX:
<math>\int_a^b f(x)\, dx = g(x) + C </math>
→
\int_a^b f(x)\, dx = g(x) + C
These go through a process that produces a PNG image. So as not to overload the servers, the images are cached. Lately I've found that images of all sorts, not just equations, often fail to load. Since I haven't changed anything on my end since six months ago when this rarely happened, I assume Wikipedia is having technical difficulties. Standard browser behavior for a missing image is to display the alt content, which in this case will be the TeX input for the equation. Simple equations may (partly depending on user preference) generate HTML instead of an image.
<math>g'(x) = f(x)</math>
→
g'(x) = f(x)
This should always work. Special characters and more complicated equations force the PNG.
<math>g'(x) = f(x) \,\!</math>
→
g'(x) = f(x) \,\!
<math>g'(x) = \tfrac12 f(x)</math>
→
g'(x) = \tfrac12 f(x)
At present Wikipedia has no MathML support, so our best alternative is Unicode and markup. We can illustrate with a line from Wikipedia:Mathematical symbols:
'''Logic:''' {{unicode|&not; &and; &or; &exist; &forall;}} &amp;not; &amp;and; &amp;or; &amp;exist; &amp;forall;
→
Logic: ¬ ∧ ∨ ∃ ∀ &not; &and; &or; &exist; &forall;
On this line, the characters displayed on the left are produced by the text shown on the right. (Notice that on the right the ampersand is entered as &amp;, so the ampersand character is displayed, rather than used as an escape character.)
Only a small number of characters have HTML names like this, but any Unicode character can be produced in one of two ways. The first is as a numeric entity, like &#x22A2;. The second takes advantage of WikiMedia's UTF-8 support, allowing any character to appear verbatim, like ⊢. This particular character, RightTee, is covered by Lucida Sans Unicode, so if you have that font installed and if your browser is configured to use it, in principal it should display without trouble. However, it is not covered by fonts like Arial or Times New Roman, so some users may see a "missing character" glyph, typically an empty box or question mark. The fonts in my table are basically ordered by Unicode code point (numeric value), which means the further down you go the more exotic the character, and the less likely it will be covered. For example, Unicode includes a dedicated symbol, Cross, for the cross product, ⨯, code point U+2A2F [VECTOR OR CROSS PRODUCT]. BabelMap tells me it is covered by Code2000, but by no other font I have installed, including Lucida Sans Unicode. (This is why I so firmly recommend Code2000!)
Maybe this information will help a little. --KSmrqT 16:12, 22 February 2007 (UTC)

Another AfD

The article Checking if a coin is fair is up for deletion. People who read this page might want to take a look at it. DavidCBryant 14:59, 15 February 2007 (UTC)

The decision was to keep the article. --KSmrqT 15:19, 24 February 2007 (UTC)

More deletions

Felix A. Keller and Keller's Expression

Keller keeps adding himself and his unimportant and obvious expression to the page about e. He has done this at least four times since 2004. Now he has promoted himself and his expression to a pair of articles.

See Wikipedia:Articles_for_deletion/Felix A. Keller to discuss.

-- Dominus 16:51, 15 February 2007 (UTC)

Obviously, given that the links have turned red, the result was delete both. --KSmrqT 09:59, 22 February 2007 (UTC)

Importance

Some articles about individual mathematicians who were important in the development of math are rated as top importance, e.g. Leonard Euler and Gottfried Liebniz, or mid importance, e.g. al-Khwarizmi. But shouldn't all of those articles be rated as low importance, because who developed the concepts is not relevant to math, it is historical trivia, wikiproject mathematics should really only concerned with articles that focus on technical aspects of mathematics. Prb4 19:55, 15 February 2007 (UTC)

So, you're saying that these articles you mentioned above might be top-importance in some biography project, but not in a Mathematics project? I think I see where you're coming from. I guess it all kind of depends on what this Wikiproject really encompasses. –King Bee (T • C) 20:02, 15 February 2007 (UTC)
The history of mathematics is an important component of mathematics. In many cases, the history says a lot about the mathematics itself. (The obvious example would be the Newton/Leibniz development of calculus.) Maybe you're saying that the history is fine, but the biography is less relevant to the math itself. Like the previous poster, I see what you're getting at, but I disagree. We should include key biographies in any list of important articles. VectorPosse 20:23, 15 February 2007 (UTC)
I think User:Pbr4 is referring to the "importance" field of the rating box appearing on some talk pages. This doesn't refer to the importance of the article to this project; it refers to the importance of including the article on a yet-to-be-made CD or DVD version of the encyclopedia. Presumably there would not be enough space for everything, so the importance field is supposed to help choose which articles to include. Interpreted this way, it makes sense for articles about Leibniz and Euler to be high importance — they are of high importance for making the DVD coverage similar to other encyclopedias. It is a historical coincidence that biographies of mathematicians are covered by the mathematics wikiproject. CMummert · talk 20:31, 15 February 2007 (UTC)
Historical coincidence? Well, I wouldn't quite put it like that. A lot of wikiproject members just find working on bios of mathematicians interesting, perhaps of relevance to the actual mathematics. I would say that editing mathematics and mathematical bios are often so complementary, e.g. the actual researching involved can be linked, that it makes sense things should be like this. --C S (Talk) 22:28, 15 February 2007 (UTC)
Technically importance does relate to importances within the specific project, and not for the CD or biographies as a whole. It is allowable to have difference imporance rating for different projects. coin flipping is one example with two different importances and theres others within the cross over with cyptography.
It is important to remember that we are writing an encyclopedia not a maths text book so historical information is a vital component. Actually I think haveing historical coverage is one of our unique aspects as history so so often neglected in mathematical texts. I learned must of my group theory not knowing in which century it was developed! Historical info can add context to the maths and helps show the development of the ideas.
We did spend some time trying to assess importance of the top 100 or so mathematicians and in some respect these are some sort of comparision with other mathematicains. Flawed yes, but a rough measure. There is also the field argument in the rating template and bios all have field=biography. At some point in the future, when the number of graded articles has grown, it may be possible to sperate out the mathematicians so we can have one list for strict maths articles and another for mathematicians. --Salix alba (talk) 00:26, 16 February 2007 (UTC)

I find this talk about 'historical trivia' to be actually offensive. What is more, the proposition that mathematics articles should be only about mathematics, narrowly defined, makes no sense in terms of the needs of the general reader; which is where WP aims, in principle. Further, I know from past comments that mathematicians themselves can find historical context helpful. Charles Matthews 19:31, 18 February 2007 (UTC)

I agree, although I would add that writing accurately about the history of mathematics is extraordinarily difficult compared to writing about mathematics. One reason is that writing about technical mathematics is more or less standardized, whereas I don't think that's true for history of mathematics.--CSTAR 19:54, 18 February 2007 (UTC)
I did some digging around to learn more about the Wikipedia:Version 1.0 Editorial Team (these are the guys trying to organize the contents of the CD version to which CMummert referred, above). You can see the list of "important" mathematicians the team has identified on this page. The list of "important" topics in mathematics is right here. DavidCBryant 16:17, 21 February 2007 (UTC)

A question about categories

I was reviewing Complementary sequences and noticed that the only math-related tag on the article was for the category "Elementary mathematics" (which seems strange, since Golay codes aren't all that elementary). Anyway, that got me curious about other topics associated with error correcting codes. I couldn't find any corresponding topics in the list of mathematics categories. So then I looked at several of the articles in list of algebraic coding theory topics and found that some of these are "math" articles (because they carry category tags like "finite fields") and some of them aren't (because they carry category tags like "error detection and correction").

So now I have a question. What's the preferred procedure in a situation like this? Is it OK to add another category to the list of mathematics categories? What if that category is likely to drag in a bunch of articles that aren't really math articles? Would it be better to add some more specific tag (like "finite fields") to some of the articles (in, say, "error detection and correction") that seem entirely mathematical? Frankly, I'm a little bit confused by the existing hodge-podge of "categories" on Wikipedia! DavidCBryant 17:58, 18 February 2007 (UTC)

There is no good answer to your question. Whether a given article belongs to a given category can be a subjective question, and whether a given article is a math article is also a subjective question.
I created the list of mathematics categories and I have mathbot add articles from those categories to list of mathematics articles so that we have a way of noticing and listing math articles. I try to be rather conservative about what categories to include in there, and categories which contain a lot of non-math articles usually don't make it in. I don't know if that's a good idea in the long run. Outside review of list of mathematics categories is very welcome.
If you feel strongly that a certain article should be in the list of mathematics articles you can either add it to a "mathematical" category as "defined" in the list of mathematics categories (then the bot will automatically list that article), or you could add that article directly to the list of mathematics articles (the bot does not remove articles from there, even if they are not in math categories, the bot just adds articles and removes redirects and redlinks). Oleg Alexandrov (talk) 18:12, 18 February 2007 (UTC)
OK, Oleg – thanks for the information. I ran across two categories that aren't on the control list for your 'bot, but that appear to have a great number of math articles in them. You might want to add "Category:Coding theory" and/or "Category:Information theory" to the list of mathematics categories. If you decide not to do that, then I'll try to make time to add the math articles in those two categories onto the list of mathematics articles. I looked through both categories for a little while, and found some articles that are already on the "math article" list, and others that are clearly math-oriented, but not on the list yet. DavidCBryant 00:43, 19 February 2007 (UTC)
I added Category:Coding theory to math categories. I don't feel comfortable adding Category:Information theory however. It has too much stuff which is not really math. And the category is generic enough that people may occasionally put odd stuff in it, and there's going to be more work to keep strange things from popping into the list of mathematics categories. Perhaps some of those articles could be categorized into "more mathematical" categories? Oleg Alexandrov (talk) 03:01, 20 February 2007 (UTC)

I've just done some cleanup on complementary sequences. Using an asterisk for ordinary multiplication, as if we were restricted to plain ASCII, is uncouth; we're not primitive troglodytes. Also, notice this difference:

N-1
N − 1

Michael Hardy 21:13, 19 February 2007 (UTC)

Thank you for working on it, Michael. My wife would not necessarily confirm your hypothesis: that I'm an actual example of Homo sapiens, and not a more primitive species of primate.  ;^> Sorry I didn't catch all the poorly coded stuff in complementary sequence the first time through, but after putting in about 75 or 80 of those non-breaking spaces (to make the big long vectors look better) I just got tired. There's only so much of that kind of stuff I can take at one sitting. DavidCBryant 23:47, 19 February 2007 (UTC)
No David, you ain't a troglodyte. Michael is a homo sapiens too, by the way, just one of grumpy subspecies. :) Oleg Alexandrov (talk) 03:01, 20 February 2007 (UTC)
The real troglodytes — Pan troglodytes. JRSpriggs 05:44, 20 February 2007 (UTC)

Oleg is mistaken as to which subspecies I represent. I'm a hothead, not a grump. Michael Hardy 20:39, 21 February 2007 (UTC)

Probability notations

Hi all! It would be a nice thing to try to set some rules concerning notation for the probabilty and expectation symbols. In the various probability/statistics articles I've seen at least three notations: P,E, \mathbb{P},\mathbb{E} and finally P,E. Personally, I prefer the latter, as it's the accepted notation of the scientific community (sometimes the letters are bold, i.e. P and E but always straight). I have not seen nevertheless any guide that woud explain such a thing. Some article even go as far as to right Pr... Let's make a public discussion about this resulting in some agreement and guide for wikipedia math community. Amir Aliev 21:05, 21 February 2007 (UTC)

Consistency would be nice, but I dunno that there's a problem. I've seen almost all of those notations in print in various places (\mathrm{E},\mathbb{E},E and P,P,Pr,Pr). I can't remember whether I've seen \mathbb{P}, though. Lunch 21:17, 21 February 2007 (UTC)
\mathbb{P} gets used alot. The most consistant place I have seen it used is in financial mathematics to distinguish risk-neutral probability measure from the physical probability measure.Thenub314 00:43, 22 February 2007 (UTC)
Could you please some decent sources with P,ProrPr ? Nevertheless, the fact is that we should pick only one notation. Amir Aliev 21:27, 21 February 2007 (UTC)
There are two extreme positions one can take with respect to notation conventions. One would be to have none. This means each article would be responsible for explaining the notation it uses (not a bad idea anyway). Another would be to standardize everything. The problem with the former is that it can confuse a reader browsing through related articles. The problem with the latter is mostly one of agreement: people use different conventions and many have strong feelings about one over another.
We should have one notation for all as wikipedia as a whole is one piece of work, not the collection of individual articles. Otherwise, it will be disastrous for the reader. Moreover the convenience of the reader certainly outweights the convenience of the writer. Amir Aliev 19:05, 22 February 2007 (UTC)
It's certainly true that five different probability theory books will use five different notations. So how do we declare one as being better than another? The few notation conventions that WikiProject mathematics has laid out in its style guide seems to be relatively uncontroversial since they agree with a majority of sources out there. (The glaring exception seems to be the use of "Dih" for dihedral groups. I don't have any idea about that one; the discussion for that decision must have taken place before I started hanging out around here.)
So this leads to a more general question. Is there some magic number that tells us what percentage of textbooks have to agree on a notation before we declare it here as a convention? Alas, for probability theory, the answer to that question may not be good news since there seems to be a lot more disagreement in that discipline, even for basic concepts like P and E. VectorPosse 22:49, 21 February 2007 (UTC)
Well, then we should organize some sort of election, for example every interested user citing some book on probability (conceivably the one they like most) and then we will count the notations in this books. It would be very interesting for me because I still don't know enough modern works from the respected authors with notation other than straight P and E. Amir Aliev 19:44, 22 February 2007 (UTC)
The universal convention I have seen for dihedral groups is Dn, and I recall no discussion ratifying the Dih notation. Sometimes things sneak in that should be removed; this is such.
Wikipedia as a whole depends more on collegiality and consensus than on conventions rigorously enforced. Sometimes we're lucky, and find almost universal agreement in current practice. For example, the ratio of circle circumference to diameter is denoted by π. Sometimes most of the English-speaking world agrees, even if others do not. For example, the unit interval including 0 but excluding 1 is denoted by [0,1). (Elsewhere, [0,1[ is used.) A great deal of mathematics notation seems to fall into these two categories. After that, things get more complicated. Sometimes an older convention has largely been displaced by a newer one, but it is important to know both to read the literature. For example, proofs once often ended with Q.E.D., but now commonly end with a block character. Sometimes different subfields or conceptual schools prefer their own convention. For example, algebraic topologists may write Z2 where number theorists would write Z/2Z. Finally, things get simple again, with "conventions" that are used by one editor alone; these we discard.
The editor of an article may not be aware that different conventions exist. When we know, our cardinal rule is to be kind and clear and helpful for the reader. A nice example of this can be found in Fulton & Harris, Representation Theory: A First Course, ISBN 978-0-387-97495-8. On page 100 we find "Remarks on Notation", which says:
A common convention is to use a notation without subscripts or mention of ground field to denote the real groups:
O(n),   SO(n),   SO(p,q),   U(n),   SU(n),   SU(p,q),   Sp(n)
and to use subscripts for the algebraic groups GLn, SLn, SOn, and Spn. This, of course, introduces some anomalies: for example, SOn R is SO(n), but Spn R is not Sp(n); but some violation of symmetry seems inevitable in any notation. The notations GL(n,R) or GL(n,C) are often used in place of our GLn R or GLn C, and similarly for SL, SO, and Sp.
Also, where we have written SP2n, some write Spn. In practice, it seems that those most interested in algebraic groups or Lie algebras use the former notation, and those interested in compact groups the latter. Other common notations are U∗(2n) in place of our GLn H, Sp(p,q) for our Up,q H, and O∗(2n) for our Un∗ H.
Very helpful; I may not know what to write, but at least I have a fighting chance of understanding what I read! My suggestion here is
  1. Write what you know.
  2. Help the reader.
  3. In case of unfamiliarity/conflict, use the talk page.
  4. For wider input, ask here.
  5. For decisions vitally affecting many articles, propose a convention.
Be prepared for both surprises and cranks. As Richard Feynman said, "Keep an open mind, but not so open that your brain falls out." --KSmrqT 00:28, 22 February 2007 (UTC)

I see that the notations listed above do not include this:

\Pr\,

produced in TeX by \Pr .

Fropuff made a good suggestion on the talk page for the Manual of Style during the flapdoodle with the physicists over the difference between := and ≡. Unfortunately, nobody responded to his suggestion then. Do you think this discussion should move over to Wikipedia talk:Manual of Style (mathematics)? DavidCBryant 20:18, 22 February 2007 (UTC)
Yes, I think it's reasonable to discuss style there. Amir Aliev 21:30, 22 February 2007 (UTC)

Mathematics Genealogy Project and an Afd

I come here by way of a mathematician Johann Christoph Wichmannshausen, who is up for deletion and is in the Mathematics Genealogy Project. I am unfamiliar with the importance of this genealogy tree, but I notice that a number of academics provide their "genealogy" on their personal website, so my initial assumption was that people higher up in the tree should be considered notable by default. Is that reasonable? Are there some rules of thumb that can be used to assist in determining WP:N using the "Mathematics Genealogy Project" data? John Vandenberg 00:36, 22 February 2007 (UTC)

Labyrinthine processes

While reviewing some of the "new" math articles I ran across this one. Someone hit it with a "cleanup" tag in November, 2005. The tag exhorts editors to "discuss this issue on the talk page". Interestingly, the talk page is still a red link, 15 months after the tag was placed. Apparently the person who tagged the article didn't even care enough to list any specific concerns right then and there, on that talk page. Like a graffiti artist, he tagged the article and moved on.

Last week I reviewed an article consisting of 28 words (plus 107 words of encrustations from various templates affixed by editors who apparently don't think others can judge a page unless there's a neon sign on it). I put a PROD tag on that one, but one of the author's sockpuppets deleted the tag, so I had to learn about the AfD process. Now the article itself (135 words, including the barnacles) is gone, and in its place we have a (roughly) 350-word archived discussion.

I understand why some of this happens. Maybe it's just that I haven't been here long enough to get used to it yet. But it seems that some of these processes aren't really helping to make Wikipedia any better. Can some of these labyrinthine processes be straightened out, or even eliminated, somehow? For example, could the code underlying a "cleanup" tag prevent its insertion while the associated talk page is a red link? If somebody really thinks an article needs cleanup, shouldn't that editor be encouraged to identify a specific problem before waving a big red flag at everybody else? DavidCBryant 16:25, 22 February 2007 (UTC)

Sounds like everything unfolded as it should. Your AfD was the right thing to do, and everyone who participated agreed with you. If you have a page on your watchlist, and someone adds a cleanup tag with no accompanying text, just write to them and complain. This nearly always gets a response. If there is no response, put something on the article's Talk page and then remove the tag. The topic of Wikipedia:Deletion is under constant review, and a large percent of the complaints are about inappropriate deletion. Sounds like you have the opposite complaint, 'inappropriate retention'. The other article you mentioned, Prefix codes, might be a candidate for deletion. It seems disorganized; any valid material would find a better home elsewhere. EdJohnston 16:48, 22 February 2007 (UTC)
I would vote very strongly against deleting Prefix codes. It is an important topic in coding theory. I did some work on cleaning it up this morning but it still needs more. —David Eppstein 17:52, 22 February 2007 (UTC)
I agree with D. Eppstein here. "Prefix codes", otherwise known as "prefix-free codes", are important in Kolmogorov complexity as well as in coding theory. The article in question just needs to be cleaned up.
In general, I give others about a week to comment on cleanup tags that are so vague as to be useless, and remove the tags if no comments are forthcoming. Look at the recent history of Turing degree for an example. But I think eliminating the tags from WP would be a losing battle to fight. CMummert · talk 20:19, 22 February 2007 (UTC)
I'm not sure precisely which processes you thing are troublesome, but I would like to defend the use of cleanup tags. I agree that it can be frustrating to see a cleanup tag on an article when you have no idea what was meant to be cleaned up, but I'd say that the person that added the tag probably thought it was clear. As for not 'think[ing] others can judge a page unless there's a neon sign on it,' you should realize that putting the cleanup tag on a page also puts the page in a category (Category:All pages needing cleanup) so that people who are looking for something to do can pick out an article that needs help.
As for the deletion process: it can be a little annoying to have to go through a week's debate for an article that you know needs to be deleted. But, then, "it's not what you know, or don't know, but what you know that isn't so that will hurt you." Better to go through a week's worth of tiresome debate than delete an article that should be kept, I think. --Sopoforic 20:55, 22 February 2007 (UTC)
The description of that category notes "This category exists primarily as an aid to bots and other automated processes." And it is correct - Jitse's bot will notice if a math article is added to the category and note it in the daily update. There is no legion of editors poring over the actual categories to fix all articles with maintenance tags. CMummert · talk 21:02, 22 February 2007 (UTC)
Eh, well, I mostly meant people doing that sort of thing--other wikiprojects also have listings of articles needing cleanup that fall within their scope. But, some people do, I think, look at the very old categories in Category:Cleanup by month to clean up things that have been sitting for a while. I try to, anyway. --Sopoforic 21:15, 22 February 2007 (UTC)
I meant to point out there is a list maintained by Jitse's bot that lists only math articles that have been tagged. It is much more tractable than the general categories. CMummert · talk 22:07, 22 February 2007 (UTC)

Request for Comment

Talk:Indian_mathematics#Request_for_comment:_Reliable_Sources_for_Indian_Mathematics Feedback is requested for a problem on the Indian mathematics page, where two users have a disagreement about what constitutes reliable sources for claims in the article. 05:29, 23 February 2007 (UTC)

Power sum and Power Sum

There needs to be some merging with the articles Faulhaber's formula, Power sum and Power Sum. I don't know that much about mathematics to clean those up. --Montchav 11:22, 24 February 2007 (UTC)

Thanks. I made Power Sum redirect to Power sum. No information was lost since the text at Power Sum is contained at Faulhaber's formula which is linked from Power sum. Oleg Alexandrov (talk) 16:18, 24 February 2007 (UTC)

Categories for number-theoretic properties

Many articles about special classes of natural numbers are categorized in Category:Integer sequences, apparently solely because the numbers in the class form a sequence. See for example Practical number, Vampire number, or Square-free integer. I think it would be better to have a Category:Number classes with subcategories Category:Base-dependent number classes and Category:Divisor-based number classes (and, of course, appropriate cross-references between "class" and "sequence" categories). Several articles currently in Category:Number theory would also be moved to the subcategory, leaving only very important or well-known classes such as Prime number or Perfect number as direct members of Category:Number theory.

Any reasons why this would be a bad idea? Any better suggestions for category names? Category names are cumbersome to change, so it would e best to get them right from the start. –Henning Makholm 20:39, 24 February 2007 (UTC)

As far as I can see, the articles concerned are about sequences of integers - so why not just leave them in Category:Integer sequences or the sub-category Category:Base-dependent integer sequences ? Any more complex or subjective classification scheme will be poorly understood (how do you distinguish between a "sequence" and a "class" ? how you define "important or well known" ?), and you will quickly get articles appearing in the "wrong" sub-category or being put in the top level category by default. Gandalf61 14:28, 27 February 2007 (UTC)
I don't see why the distinction should be complex subjective at all. If one can determine from the definition of the class whether a given number is a member of it without needing to know anything about what the other members are, then it is only a sequence in the trivial sense of being a subset of the natural numbers. This is true for all of the three examples I gave. Why should someone who wants to learn about square-free numbers have to look in a category about sequences to find it? Expecting users to be able to think "sequence" when what they have in mid is just a property with no inherent sequential features, now that is complex... –Henning Makholm 03:20, 28 February 2007 (UTC)

Enneper-Weierstrass parameterization formula

The article Enneper-Weierstrass parameterization needs some quick cleanup that I am unable to provide. It says:

Given complex-valued functions f(z) and g(z), parameterize minimal surfaces by taking the real part of
∫ (f(x)(1 − g(x)2), i*f(x)(1 + g(x))2, 2f(x)g(x)).

The formula here needs to be re-typeset, but it also does not appear to be strictly correct: where are the differentials? What are the limits of integration?

-- Dominus 08:09, 27 February 2007 (UTC)

Here it is typeset.
\int \left( f(x) \left( 1-g^2(x) \right), \bold{i} f(x) \left( 1+g(x) \right)^2, 2 f(x)g(x) \right)
The MathWorld version (here) takes f(z) and g(z), where r = reiφ, and writes
\begin{bmatrix}x(r,\phi) \\ y(r,\phi) \\ z(r,\phi)\end{bmatrix} = \Re \int{\begin{bmatrix} f (1-g^2) \\ \bold{i} f (1+g^2) \\ 2 f g\end{bmatrix} \, dz} .
It also has problems, including the use of z in different ways on the left side and the right. More reliable is
\begin{align} x_k(\zeta) &{}= \Re \left\{ \int_{0}^{\zeta} \phi_{k}(z) \, dz \right\} + c_k , \qquad k=1,2,3 \\ \phi_1 &{}= f(1-g^2)/2 \\ \phi_2 &{}= \bold{i} f(1+g^2)/2 \\ \phi_3 &{}= fg \end{align}
with ck constants, g(ζ) a meromorphic function on a domain D in the complex ζ-plane, f(ζ) an analytic function on D with the property that at each point ζ where g has a pole of order m then f has a zero of order 2m, and D either the unit disk or the entire plane. (Lifted from the Encyclopedic Dictionary of Mathematics, second edition, ISBN 978-0-262-59020-4.)
I claim no expertise in minimal surfaces, so someone may want to check this and fix the article. --KSmrqT 15:23, 27 February 2007 (UTC)
Many thanks. I should have thought to look at Mathworld myself. -- Dominus 15:53, 27 February 2007 (UTC)
I have changed the article; I would be grateful if someone would check my changes for errors. -- Dominus 16:28, 27 February 2007 (UTC)