Wikipedia talk:WikiProject Mathematics/Archive/2007/Apr

Upright d in math notation
An anon has been going through articles replacing italic d with upright d in math articles, for example


 * $$\frac{d f}{d x}$$

to


 * $$\frac{{\mathrm d} f}{{\mathrm d} x}.$$

There is a small discussion about this at talk:Derivative.

As pointed out by Geometry guy, the previous discussion about this at Wikipedia talk:WikiProject Mathematics/Archive 4 did not achieve consensus on what to use.

However, I would argue that while people should be allowed to use whatever notation they choose, I believe it is not a good idea to do mass changes to articles which used one type of notation for a long time.

That is to say, the vast majority of Wikipedia articles (all articles that I am aware of) use italic d notation. I vote to revert the anon conversions and to go back to status quo italic d notation at derivative and differential form. And if somebody starts a new article, and want to use roman d, they should be allowed. Comments? Oleg Alexandrov (talk) 15:00, 23 March 2007 (UTC)


 * I agree. Paul August &#9742; 15:40, 23 March 2007 (UTC)


 * As do I. (Of course, I also think that the non-italicized d is wrong, but that's another story.) there are a lot of areas in mathematics where there are variations in notational style, and it seems a lot of energy gets spent on discussions about which one is right. However, within a single article, I think we should try to be consistent, if possible. Greg Woodhouse 15:46, 23 March 2007 (UTC)


 * The archived discussion on d vs. d is not so good, but there was a really intense argument over i vs. i for the imaginary unit linked there. I will paste their summary of the resulting accord:
 * {| border="1" style="margin-left:20px; border-collapse:collapse" cellpadding="2"

! Reasons for pro italic usage of imaginary unit in Wikipedia: i ! Reasons for pro non-italic usage of imaginary unit in Wikipedia: i ! Reasons for usage of \imath
 * + Summary of reasons
 * - bgcolor="#DDDDDD"
 * most mathematics books/papers use italic notation of imag. unit
 * italic notation of imag. unit looks better Oleg Alexandrov
 * is a conceptual case of definition, italic i is needed Septentrionalis
 * - bgcolor="#DDDDDD"
 * is a conceptual case of definition, italic i is needed Septentrionalis
 * - bgcolor="#DDDDDD"
 * - bgcolor="#DDDDDD"
 * Better semantics. This has several beneficial implications. PizzaMargherita
 * prevents confusion with running index i, electr. current, etc. Wurzel
 * offers electrical engineering technicians an imaginary unit notation which has no interference with neither (Maxwell's) current density j nor with electr. current i Wurzel
 * allows parallel usage with running indexes i,j Wurzel
 * improves readability of formulas containing the imag. unit i because of no overlapping definitions Wurzel
 * i is easily acessible on many computers/text systems / fonts Wurzel
 * - bgcolor="#DDDDDD"
 * allows parallel usage with running indexes i,j Wurzel
 * improves readability of formulas containing the imag. unit i because of no overlapping definitions Wurzel
 * i is easily acessible on many computers/text systems / fonts Wurzel
 * - bgcolor="#DDDDDD"
 * i is easily acessible on many computers/text systems / fonts Wurzel
 * - bgcolor="#DDDDDD"
 * Is an alternative offered by TeX Michael Hardy
 * }
 * Looks to me like the arguments for italics are: "it's convention" and "I like it" (the third one is not generally applicable). The arguments against are "better semantics" and numerous practical advantages, though anything mentioning current is irrelevant to mathematics and the one about being more accessible for many systems is irrelevant on Wikipedia.  The decision was then made to keep i because it was the existing practice on Wikipedia, although there was no consensus.
 * Basically, italics won because we have a commitment to: (a) following widespread standards in the non-Wikipedia literature, and (b) when this is ambiguous, giving preference to existing standards on Wikipedia. Basically the same sorts of arguments work for d vs. d and the outcome is that the former is more common and the latter is better in every way except for being less convenient in LaTeX, so we stick with d since we already use it.  I support the decision, though there's a good chance that one of these days I'll write something of my own when I'm in a mood to be pedantic and Romanize all my operators and symbolic constants, because that's how I am sometimes.
 * However, I think the single most influential reason people really hate these notational crusades is that everyone resents the use of a notation they don't personally endorse but they learn to tolerate it with a little rolling of the eyes, until someone rubs salt in the wound by unilaterally imposing their notation on Wikipedia. For this particular reason, I would say that even though we should use really clever, unambiguous notation as mathematicians, the use of this particular, slightly ambiguous notation has been suffered for decades if not centuries and we've learned to work around it.  This can be done with a minimum of effort, such as paying attention to good choice of notation (which we should already be doing) and therefore the "practical advantages" of changing the italics to roman are pretty irrelevant, and far less than the practical disadvantages resulting from various bad feelings and revert wars that might ensue from doing so.
 * So I'd go further than agreeing to keep d. I would say that any time someone decides to make any sort of minor but widespread notational "improvements" they should be reverted with no more than a comment here to let people know what happened.  The discussion itself is foregone and, as I've argued, more trouble than it's worth.  Hell, by now it's standard practice :) Ryan Reich 16:07, 23 March 2007 (UTC)
 * Indeed. Both d and d are acceptable, both have advantages and disadvantages. We aren't going to ever have consensus to use just one of these, so we should follow Manual_of_Style and allow the first major contributor to decide these style variants (which should then be consistent within each article). Changing all "d"'s to "d"'s in articles where one is not editing actively otherwise is like changing all instances of "colour" to "color" or vice versa, which is a blockable offense. Kusma (talk) 16:21, 23 March 2007 (UTC)
 * So I'd go further than agreeing to keep d. I would say that any time someone decides to make any sort of minor but widespread notational "improvements" they should be reverted with no more than a comment here to let people know what happened.  The discussion itself is foregone and, as I've argued, more trouble than it's worth.  Hell, by now it's standard practice :) Ryan Reich 16:07, 23 March 2007 (UTC)
 * Indeed. Both d and d are acceptable, both have advantages and disadvantages. We aren't going to ever have consensus to use just one of these, so we should follow Manual_of_Style and allow the first major contributor to decide these style variants (which should then be consistent within each article). Changing all "d"'s to "d"'s in articles where one is not editing actively otherwise is like changing all instances of "colour" to "color" or vice versa, which is a blockable offense. Kusma (talk) 16:21, 23 March 2007 (UTC)

As an apology for causing trouble by partially supporting the anon, I promised to collect some links to previous discussions, to avoid (if possible) going over the same old ground. This is what I found so far: please add to this list if you find others. I tend to agree with User:Toby Bartels (although I am from the UK and he is from the US, we both personally prefer upright d's, but oppose the math project having a policy on this - see my comments after the list). Geometry guy 16:41, 23 March 2007 (UTC)


 * Wikipedia talk:WikiProject Mathematics/Archive 20 (User:KSmrq makes a particularly thorough comment.)
 * Wikipedia_talk:WikiProject_Mathematics/Conventions
 * Wikipedia talk:WikiProject Mathematics/Archive 4
 * Wikipedia_talk:WikiProject_Mathematics/Archive_3
 * Wikipedia_talk:WikiProject_Mathematics/Archive_1 (Only User:Toby Bartels comments here.)


 * I am not convinced we need a policy on this. Has there been an edit war over this issue? Clearly if (as discussed before) one user makes 75 large edits in one day changing italic to roman or vice versa in many articles, these edits will be reverted and few will argue. I like diversity in wikipedia and such an edit damages this. However, sometimes a mass change in one article is not a bad thing and can increase diversity. I don't think it is a good thing if "the vast majority of Wikipedia articles... use italic d notation". Also, one of the great things about wikipedia is that it is dynamic. I don't like the idea of setting the original notation in stone, although there are of course cases where this is entirely justified, so that (for example) related articles evolve with similar notation. However, no one is going to get confused if one article with an italic d links to another with roman one, are they?


 * I, for one, frequently make minor edits of a repetetive nature when I contribute to an article: for instance I often italicize Greek letters in wiki-text so that they look more like their TeX counterparts. Am I right? There are pros and cons, but I would be sorry to see a policy on this, or to find my edits systematically reverted because the original contributor didn't use italic Greek letters.


 * As regards this particular issue, I sometimes replace an italic d by a roman d when it is clearly the exterior derivative (an operator) and not part of a "diphthong". I have no problem with another user reverting this, but I would not like it to happen systematically as a matter of policy. This is maths, not bureaucracy! In terms of the recent reverts, I would therefore like differential form to retain an upright d, and may one day give exterior derivative the same treatment. There is possibly a case for a policy for this particular usage, and I'm not sure it has been discussed separately before. Comments anyone?


 * I also think there is no harm in keeping the change to derivative, as this article needs a shake-up. Maybe the next editor who substantially improves this article should decide? If I were proposing a policy (and I'm not), that's the kind of suggestion I would make!


 * Finally, appearance (just so you know all the prejudices which inform my comments!): my view is that dx looks better as wiki-text, but for display math, the roman font renders so poorly that the case is not clear. Geometry guy 16:41, 23 March 2007 (UTC)


 * It is kind of unclear to me how notation change to dx from dx would cause a beneficial shake-up at derivative. Anyhow, I guess we all agree (including Geometry guy) that mass notation changes are not a good idea. And I do agree that officially codifying dx over dx or vise versa is not necessary. Oleg Alexandrov (talk) 02:56, 24 March 2007 (UTC)


 * Without falling into the loathsome practice of codifying, we could point out that (a) Wikipedia follows common conventions; (b) in mathematics (but not necessarily physics) italic d (and i and e) are the usual convention, at least for dy/dx and ∫y dx, in spite of ISO 31/XI and the recognized advantages of upright boldness; and (c) making mass changes of notation without discussion and consensus is not the way to go about it. The Manual of Style for mathematics already contains formulations like "Which method you choose is entirely up to you, but in order to keep with convention, we recommend ..." and "Either form is acceptable, but do not change one form to the other in other people's writing". I think that also here shedding some light on the issues for editors who are seeking guidance in this matter is only beneficial. --Lambiam Talk  09:38, 24 March 2007 (UTC)
 * Where did you see this in ISO 31/XI ? I only found this paper which states that, according to ISO 31/XI, the operator of differentiation should be set in roman type (as well as constants, i and e). pom 11:39, 24 March 2007 (UTC)
 * I didn't mean to suggest anything different. --Lambiam Talk  14:34, 24 March 2007 (UTC)
 * A non-mathematician's take. This has some similarities to the discussions of changing standard English spelling ('thru' for 'through' etc.).  Let's face it:  Convention matters, in life, dictionaries, grammar, & on Wiki.  The law is made up of such conventions (driving on one side of the road, rather the other in a given country, etc.).  What matters is not necessarily what convention is selected but that there is a convention for reaching agreement, so that expectations are satisfied, reducing the cost of communication.  Of course conventions can change, which Wiki can reflect as necessary.  So far as the math community is concerned, it would be misleading to portray notation that is non-standard as standard, which is what one expects in an encyclopedia, akin to no original research.  But I like the flexibility shown above by the conventionalists in existing vs. new articles.  --Thomasmeeks 12:46, 24 March 2007 (UTC)


 * Conventions operate in a context. What is convenient and clear in one context may be awkward and confusing in another. We are flexible with mathematical notation because we must be. As mentioned above, I have previously posted at length detailing a wide assortment of mathematical contexts for dx and friends. If, as I perceive it, we do have a consensus against any mass conversion or "standardization" compaign, then: (1) we may wish to put a note on our conventions page, and (2) revert without debate and leave a note on the editor's talk page informing them of said consensus.


 * I need to ask about two other notations mentioned, italic Greek letters and the square root of −1.
 * Long ago (in wiki time) I saw an admonition somewhere to not italicize Greek variables inline, or so I remember. Thus I have used, say, "&theta;" (&amp;theta;) instead of "&theta;" ( &amp;theta; ). Maybe it was an issue of font availability, maybe times have changed, or maybe I remember wrong. What say ye?
 * Perhaps because of long exposure to quaternions and Clifford algebras and the like, I use upright bold for our little friend i, and when only a few instances exist in an article I may inflict my preference on existing material. (See Cayley transform, for example, which I recently gave a major facelift.) I have been unaware of any consensus, and strongly prefer the bold convention for most things I create. However, in topics of complex analysis, say, it could be a serious pain to use anything other than TeX's default italic, and I would not dream of changing or complaining about i there. Except, as a matter of reader friendliness, I do object to using i as both an index and a special constant. Never would I use upright "i" without boldface. Any strong feelings?


 * As always, I see our role as being a bridge between a diversity of common mathematical practice and the needs of our readers, while making editing less onerous. In that regard, I note that despite our thousands of mathematics pages, the developers have not yet switched to blahtex instead of texvc, not for its much broader TeX compliance in producing PNGs, and certainly not for MathML output. I suppose I have been biding my time until the STIX fonts release (currently anticipated for April), but come that day I would like us to begin serious lobbying. To produce attractive, consistent mathematics notation by using &lt;math&gt; markup everywhere — that would be a great benefit for editors and readers alike. --KSmrqT 15:26, 24 March 2007 (UTC)


 * I agree with KSmrq. Imposing even a "usual" usage for imaginary units is even worse that doing it for d, since it affects a much broader range of contexts. In some contexts it is entirely familiar to use i for the square root of -1, even when it is simultaneously an index. In other contexts, much greater clarity is obtained by using a different notation, and there are plenty of options: i, i, i, $$\imath$$, $$\boldsymbol i$$ and $$\mathrm i\,$$ being a few that I have seen. Geometry guy 19:15, 24 March 2007 (UTC)


 * It is generally the best idea to use the italics one when dealing with the imaginary unit because that is the way most textbooks do it. It is the same with the ƒ(x) vs. f(x) argument, most textbooks use italics for it. The engineering books I've seen have all used non-italics for their variables so we can either: switch all the 'i's in the context of imaginary units to italics and all the ones representing resistance with regular face 'i's; or just use an italic 'i' as the standard, unless it would cause confusion in context with electrical engineering. The Roc 1217 21:43, 2 April 2007 (UTC)

Wikipedia:Mathematics Collaboration of the Week
Mathematics Collaboration of the Week has just been marked as inactive, its not received much activity since November. Should anything be done to revive the collaboration? --Salix alba (talk) 07:34, 28 March 2007 (UTC)


 * The knack would be to nominate things people actually want to work on ... this idea has not ever really got off the ground. It's not really adequate to say "I decide, you write, that's what I mean by collaboration". On the whole a more elitist approach might be more welcome. (I mean it might work better, not that it is more desirable.) Charles Matthews 19:16, 30 March 2007 (UTC)


 * I agree. I also have another suggestion. I think it would be more productive to place the emphasis on producing A-class articles rather than FA-standard. I realise the distinction is slight, but it seems that FAC can be a rather dispiriting and bureaucratic experience of dotting every i and crossing every t in the definition of a "perfect wikipedia article" (it will be interesting to see what happens to mathematics), not to mention adding inline citations for everything, which do more to damage the readability of an article than enhance its authority. Instead with A-class, WikiProject Mathematics sets the standard and places the goals, and we can enthusiastically concentrate on the things we really care about in a mathematics article: its readability, liveliness, accessibility, depth and breadth and interest of its mathematical content. I also think that the number of A-class articles is an excellent way to judge the project's success. Geometry guy 11:11, 31 March 2007 (UTC)


 * I don't know whether it can be (re)vived, but if it can, it needs someone who is on top of this to coordinate the process, and it would help if that coordinator is one of our several well-known and respected contributors. As far as I'm concerned the coordinated effort could also be to raise the quality of a painfully embarrassing article to above the embarrassment threshold, or in general anything, as long as there is a potential and promise of a real improvement. Personally I'd really like to see the "entry level" articles improved, the ones must likely to be consulted by mathematically relatively unsophisticated readers, but others may be more inspired by advanced topics. A crucial aspect is the process by which each time the next "Collaboration of the  " is selected. The coordinator could perhaps each time select a few improvable articles as candidates for something like a run-off vote, and announce the selected candidates on this page. "Voting" for a candidate would signify: "I'm willing to work on this". Sometimes it also makes sense to deal with a group of related articles together.
 * For the process, I think the collaboration can be facilitated if we proceed by a few distinguishable phases. Phase 1 would be similar to a peer review, in which everyone can propose improvements, but also identify what the problems with the current version are, whether they have ideas or suggestions for fixing them or not. This first phase is not discussion-oriented. In phase 2, we attempt to reach consensus on the target: what are (and what are not) problems and what is the approach to be taken for improvement. Phase 3 is the implementation of the consensus reached. Each phase could last like a week. Once this process is going well, we could consider a three-article pipeline with staggered phases, so while article n is in the implementation phase, article n+2 is in the peer-review phase. --Lambiam Talk  12:46, 31 March 2007 (UTC)


 * I think a good approach would be to start with the Requested Articles page and try to get some impression of priorities there. I'd be looking for articles such as Lefschetz duality, for example (associated with one of the greats of topology, builds on Poincaré duality). In terms of grand strategy, I say it is more important to bring coverage up to date (i.e. fill in late twentieth century mathematics, at least in outline) than to worry about 'featured article' style requirements right now. (I can't imagine what group theory would look like as a FA; a comprehensive article on groups? What does that mean? It is long ago that there could be a comprehensive book on groups.) Putting it another way, there is plenty of online material on current developments, but WP should be putting in place the essential stuff which means it becomes more easily readable.Charles Matthews 11:44, 1 April 2007 (UTC)

Sines and cosines of sums of infinitely many terms
I just added this paragraph to the list of trigonometric identities, and maybe someone can add some information in response to the questions below.


 * $$ \sin\left(\sum_{i=1}^\infty \theta_i\right)

=\sum_{\mathrm{odd}\ k \ge 1} (-1)^{(k-1)/2} \sum_{ |A| = k } \prod_{i \in A} \sin\theta_i \prod_{i \not \in A} \cos\theta_i $$


 * $$ \cos\left(\sum_{i=1}^\infty \theta_i\right)

=\sum_{\mathrm{even}\ k \ge 0} (-1)^{k/2} \sum_{ |A| = k } \prod_{i \in A} \sin\theta_i \prod_{i \not \in A} \cos\theta_i $$

where "|A| = k" means the index A runs through the set of all subsets of size k of the set { 1, 2, 3, ... }.

In these two identities an asymmetry appears that is not seen in the case of sums of finitely many terms: in each product, there are only finitely many sine factors and cofinitely many cosine factors.


 * Is absolute convergence strong enough to entail the identities, and is convergence on the right side also then absolute?
 * How much can such a hypothesis be weakened, and at how much cost in weakened conclusions?

I derived the identities from scratch without careful attention to these questions (that's quite easy, as you'll see if you try it), then I found them in a couple of old books (19th-century trigonometry texts were quite detailed and thorough), but again without careful attention to what the books said about these questions. --- Lazily... Michael Hardy 02:41, 2 April 2007 (UTC)


 * Should not the exponent of -1 in the first equation be (k-1)/2 rather than (k+1)/2? So that the first term is positive? JRSpriggs 10:38, 2 April 2007 (UTC)

Oh. Yes. Clumsy of me.... Michael Hardy 19:56, 2 April 2007 (UTC)

...and now I've fixed it in what I wrote above, lest rely on it. Michael Hardy 19:57, 2 April 2007 (UTC)

In mathematics
We quite often begin our articles with "In mathematics..." or something similar. This has been discussed previously at Wikipedia_talk:WikiProject_Mathematics and I'm happy to report that the consensus was to support a variety of styles. :)

Anyway, I just wanted to draw attention to a variant which I have started using, which I think might be quite useful to other editors (who also might have their own variants which they would like to share here). Quite often it is necessary and appropriate to set a more precise context in the first sentence, such as "In analysis...". Sometimes it is safe to do this without the risk that the reader will assume the article needs psychotherapy, sometimes not. In the latter case, mathematics needs to be mentioned. In this particular example the problem is solved easily by linking directly as "In mathematical analysis...", but it is not always so easy to find an elegant solution. A common approach (which I have used), is to begin "In mathematics, more specifically in widget theory...", but I am increasingly finding this rather awkward, so I came up with an alternative


 * In the mathematical field of widget theory...

This is a bit shorter and seems less forbidding to me, so I thought I would share it here. Comments and suggestions very welcome! Geometry guy 18:29, 30 March 2007 (UTC)


 * I think the main issue is the very quick orientation of the uninformed reader. "In mathematics ..." is very good for that; "In geometry ..." is OK. Anything else risks drawing people further into the article than they personally need. The lead section of articles is rightly subject to a tighter and (in a sense) more tabloid style. Charles Matthews 19:03, 30 March 2007 (UTC)

I agree, but I think "In the mathematical field of..." does this. Do you agree? (This is one reason that I posted.) Geometry guy 19:25, 30 March 2007 (UTC)


 * When I first saw the article titled schismatic temperament, I was the uninformed reader. I knew the usual meaning of the word schismatic and the usual meaning of the word temperament, so it sounded like an article about a psychiatric disorder.  It's actually about musical tuning.  That's why quick orientation of the uninformed reader is needed.  In some math articles, the non-mathematician reader could go through the first paragraph before finding out it's not about about chemistry, theology, archeology, etc.
 * I think that saying "In the mathematical field of ..." suffices. "In algebra..." or "In geometry..." probably also suffices.  "In analysis..." often would not suffice given the plethora of different meanings of that word. Michael Hardy 19:42, 30 March 2007 (UTC)


 * I usually prefer something more explicit than "in mathematics". I expect analysis is more likely to cause confusion than, say, matrix theory. A helpful test is whether the broad topic requires disambiguation. If it does, then I have no problem with Geometry guy's concise phrasing. --KSmrqT 04:00, 31 March 2007 (UTC)

Here is an example similar to user:Geometry guy's from a page I've edited recently:
 * Hilbert space is a mathematical construct (...)

In this case starting
 * In mathematics, Hilbert space is (...)

creates more problems than it solves: does it not imply that in physics Hilbert space means something else? I find it annoying, by the way, when the lead to an article with an elegant or not so elegant, but, at any rate, well thought-out solution is edited down to "In mathematics" with a happy note "boilerplate" in the edit summary. Arcfrk 08:26, 31 March 2007 (UTC)


 * I couldn't resist; I tried my hand at revising the intro to Hilbert space with completely different wording. (No doubt my hand will be bitten off.) I worry that the opening sentence is a long and winding road, but I hope it's one that the average reader can follow. Of course, it's impossible to make anything "idiot-proof"; idiots are so clever at finding new ways to misunderstand. ;-) --KSmrqT 21:19, 31 March 2007 (UTC)


 * Certainly, writing
 * Hilbert space is a mathematical construct...
 * sufficiently notifies the lay reader that mathematics is what the article is about. Michael Hardy 23:30, 31 March 2007 (UTC)


 * I would just comment here that I don't think it's very helpful to link the word "mathematical" to the mathematics article. There are three good reasons to link a phrase: Because a reader might not be sure what it means, because it provides information helpful to understanding the article it's linked from, or because readers of the linking article are especially likely to want to see it.
 * The first reason will never apply; the second very rarely, the mathematics article being much too general to be helpful in the specialized articles we're discussing. The third reason also strikes me as a bit unlikely -- if a reader has wandered in by mistake he really just needs to know that this isn't what he's looking for, and if he's there on purpose he's probably more interested in the article on the specific field of mathematics.
 * Summing up, I prefer "in the mathematical field of foo theory...", with no link on the word "mathematical". --Trovatore 18:48, 3 April 2007 (UTC)


 * This is a really helpful comment. I'm tempted to go right away to the articles I have edited and remove the link. Do others agree? Geometry guy 19:01, 3 April 2007 (UTC)


 * I really like Trovatore's suggestion. Seems very reasonable. VectorPosse 23:36, 3 April 2007 (UTC)


 * I agree with not linking "mathematics". However, I'd like to question whether it's necessary to mention the field within mathematics. Of course, it depends on the circumstances, but I think it's done too much. For instance, here are the first sentences of two articles that came up on my watchlist:
 * "In mathematics, especially in elementary arithmetic, division is an arithmetic operation which is the inverse of multiplication."
 * "In mathematics (specifically linear algebra), the Woodbury matrix identity says that the inverse of a rank-k correction of some matrix can be computed by doing a rank-k correction to the inverse of the original matrix."
 * I'd argue that in both cases it's of little use. In the first case, I wouldn't even mention mathematics, and just start "Division is an arithmetic operation which is the inverse of multiplication": if the reader doesn't know what division is, does it help to mention mathematics? In the second case, the word matrix tells the reader that it's a topic in matrix theory / linear algebra, unless of course the reader doesn't know what a matrix is, in which case linear algebra probably won't mean much either. -- Jitse Niesen (talk) 02:19, 4 April 2007 (UTC)
 * Don't look now, mathematics is going to be an orphan at this rate. I think that Build the web is a little relevant here. I kinda disagree with Jitse, in that if someone happens on an article like term algebra, the immediate link to universal algebra is useful. Smmurphy(Talk) 02:50, 4 April 2007 (UTC)


 * If WP:BTW is relevant, so is WP:CONTEXT. Here's how I see it:  Think of a tree going from general at the root to specific at the leaves. It's very natural for more general articles to have links to more specific ones, but links the other direction are usually somewhat less useful. Extremely general articles, like mathematics, don't need many links to them, because they're likely to be primary search terms.
 * In general I think a link one step "up" the tree (my trees have the root at the top and the leaves at the bottom, don't ask me why) is useful, but going all the way up to the root is not. So pointclass should say it's in the context of descriptive set theory, which in turn should say that it's part of mathematical logic, and that last article might link to mathematics. In the other direction, from the root to the leaves, leap by as many stepas as you feel like.
 * As to Jitse's point, I pretty much agree on division (though a link somewhere to arithmetic wouldn't hurt; that's just one step "up"). On Woodbury matrix identity I'd probably throw in linear algebra, unless there's a narrower subfield that would be useful to mention instead. --Trovatore 03:04, 4 April 2007 (UTC)


 * I think I agree with Trovatore here. When the lead sentence links to mathematics, it makes the reader recreate the context all the way back to the current article. If the lead links to a subfield, it makes the user back up only one level. Presumably that article allows the user to back up another level, and so on until the reader finds a comfortable article.  So in most cases I think it is better to link to the general subfield of mathematics, when possible. But as long as there is a link in the opening paragraph it doesn't matter so much whether it's in the first sentence or not. CMummert · talk 03:18, 4 April 2007 (UTC)


 * I'm already on record as preferring to name a context more specific than mathematics. Now I shall attempt to illustrate why it's a good idea to provide context even when the context seems obvious. I'll use the examples given, "division" and "Woodbury matrix identity". The American Heritage Dictionary definition of division includes
 * 1.f. A group of several ships of similar type forming a tactical unit under a single command in the U.S. Navy.
 * 1.g. A unit of the U.S. Air Force larger than a wing and smaller than an air force.
 * 1.h. Variance of opinion; disagreement.
 * 2. Mathematics The operation of determining how many times one quantity is contained in another; the inverse of multiplication.
 * My point is that what seems "obvious" when we're thinking about mathematics may not be what the reader had in mind; our definition is secondary to these others! Likewise, our meaning for "matrix" comes eighth; a geologist naturally assumes we're talking about "The solid matter in which a fossil or crystal is embedded." Unless we tell otherwise. Which is what we should do. --KSmrqT 08:03, 4 April 2007 (UTC)
 * That sounds right to me. Smmurphy(Talk) 12:51, 4 April 2007 (UTC)


 * I believe I was not very clear. I agree that there should be a link to some specific context. My point was that we should be more creative than the standard template, because the first sentence of an article is often rather convoluted. It is easy to end up with beasts like "In mathematics (more specifically topology), the Poincaré conjecture (pronounced pweh-cah-ray, IPA [pwɛ̃kaˈʁe]), which is named after Henri Poincaré who formulated it in 19xx, states that &hellip;"
 * Here is what I'd prefer in the examples I gave:
 * "Division is an arithmetical operation which is the inverse of multiplication." The idea is that arithmetical signals which meaning of division we're talking about (though I'm willing to concede that this might be a bit too subtle and that we should start with "In mathematics, division &hellip;").
 * "In mathematics, the Woodbury matrix identity says that the inverse of a rank-k correction of some matrix can be computed by doing a rank-k correction to the inverse of the original matrix." Here, matrix (mathematics) gives better context than linear algebra, and mathematics is mentioned at the start so it's clear we don't mean the geologist's matrix.
 * I think these formulations make the context sufficiently clear. -- Jitse Niesen (talk) 15:11, 4 April 2007 (UTC)


 * If the article title is a dab title like Division (foology), where foology is a field with its own terminology, then – assuming the article title is aptly chosen – it is entirely appropriate to start with a sentence such as "In foology, the separation of an entirety into several parts is called a division." --Lambiam Talk  15:46, 4 April 2007 (UTC)

Wikipedia poll on Attribution now open
Further to the discussion above (Wikipedia talk:WikiProject Mathematics), the poll on the new WP:ATT summary of the verifiability and no original research guidelines is now open, and closes on April 6th. So far it looks close, so please vote to give the mathematicians' point of view.

It is my strong belief (and I am not alone, see above), that the proposed change is definitely a good thing for mathematics, since it reemphasises "attributable" (i.e., the fact that an article has sources) rather than "attribution" (choosing a particular source and citing it). Also, "no original research" would follow as consequence of being attributable, rather than a directive in itself. In some sense, every wikipedia article is "original research", because it gathers information from a variety of sources and presents it in a new (and hopefully fresh, lively and interesting) way. This is particularly true in mathematics, where an encyclopedic explanation of a fact requires careful writing in a novel way, even though there are hundreds of sources. This change, although clarifying existing policy, should help us (I'm an optimist) against the bureaucrats who insist on an inline citation to source every nontrivial sentence.

I also happen to think the change is good for wikipedia, because in other fields, the concept of "truth" is less clear-cut than it is in mathematics, and by replacing "verifiable" by "attributable", the role of an encyclopedia to describe knowledge (which is partly what people believe rather than what is true) is greatly clarified. Geometry guy 19:27, 31 March 2007 (UTC)


 * The poll is actually here: Attribution/Poll. I voted against, because I didn't see how the new thing fixed the problem of bureaucratic insistence on citations, while at the same time, it seemed loosen structure enough to allow propaganda and misinformation to be included in a back-handed manner, simple because its "attributable". Its precisely because "truth" is less clear-cut in other areas that it needs to be more carefully defended. linas 05:15, 3 April 2007 (UTC)

I was probably being a bit polite when I said "less clear-cut", as people can disagree quite vehemently on what the truth is, or even what "truth" means, which is why I find it not a terribly helpful term. I'm not actually sure what it means in general, and the wikipedia article on truth, while an interesting and amusing read, left me none-the-wiser! :) More seriously, I share your concerns, but hope that WP:ATT, backed up by a clear idea of what a reliable source is, will actually tighten the structure through its clarity. Geometry guy 13:14, 3 April 2007 (UTC)

Moving pictures and the Tesseract on our Project page
I really do not like moving illustrations in articles (unless they only move when you click on them or some such). They are distracting and the use up the band-width of my modem connection. In some cases, the article never settles down enough for me to see what is there and edit it. In this particular case, I would like to ask whomever is responsible for the moving picture of a tesseract on our project page to please fix it so that it does not move or only moves when the reader clicks on it. Thank you. JRSpriggs 07:04, 1 April 2007 (UTC)


 * It was added in this edit. If you don't like it, you can replace it yourself by another image you prefer. I don't think the wikimedia software currently offers the ability to start and stop animations; this is just a gif and the animation is that of your browser's renderer. In this case the file size is 491 KB, which is not excessive, and once loaded, it should not use up further bandwidth. Perhaps blanking an imagine could be realized with a JavaScript function in a user's monobook.js. --Lambiam Talk  09:49, 1 April 2007 (UTC)


 * I replaced it with a still image, but I view the current image as a temporary choice. It would be nice to get a sense of what people think about having an image on the project page. If there will be an image, suggestions of images would be welcome. I am inclined to agree that moving images are distracting. CMummert · talk 12:42, 1 April 2007 (UTC)


 * There's always the solution realized on the tesseract page itself (see the "net"): a still picture with a link in the caption to "view animation". This partly addresses JRSpriggs's request, though of course it would be nice not to have to leave the page to see the animation. Ryan Reich 13:59, 1 April 2007 (UTC)


 * I looked for a still version of the previous image but didn't see one. I think it's a good idea.  If you can make a still image, feel free to do so and then edit the project page accordingly.  CMummert · talk 00:49, 2 April 2007 (UTC)


 * To CMummert: Thanks for replacing it.
 * To Lambiam: I see it in your diff, but I do not see it in the source of the page itself. Is it in an included file? JRSpriggs 10:27, 2 April 2007 (UTC)
 * The source of which page are you referring to, and what is the it you don't see there? --Lambiam Talk  17:31, 2 April 2007 (UTC)


 * The image is included from WikiProject Mathematics/Nav, which is available from the "edit" link in the resources column of the main project page. This isn't obvious from the way that the page is structured, but there aren't a lot of good places to put the edit link for the sidebar. CMummert · talk 17:49, 2 April 2007 (UTC)


 * I've replaced the image by a still version of a newer animated image with "glass" faces, and provided a link to the animation as suggested above by Ryan Reich. --Lambiam Talk  18:01, 2 April 2007 (UTC)
 * My thanks again to CMummert, Ryan Reich, and Lambiam. It is much better now. JRSpriggs 08:06, 3 April 2007 (UTC)

Rao integral equation
I wonder what people think of the recent additions to integral equation. To me at appears that it is not yet well-enough established research (see the references section) to show up on Wikipedia. Comments? Oleg Alexandrov (talk) 04:25, 2 April 2007 (UTC)


 * "Not well-enough established" is too polite. Kill it. Jmath666 07:50, 2 April 2007 (UTC)
 * I tend to agree. Policy wise it seems to fail WP:OR, WP:RS. This stuff may be true, but it could be duplication of other work under a new name. However I think I can vouch that Rao is actually a distinguished computer vision worker. I met him one at the Newton institute in Cambridge, so it not a straight forward crank case. --Salix alba (talk) 10:15, 2 April 2007 (UTC)
 * There is a photo here, if you remember the face. --KSmrqT 17:22, 2 April 2007 (UTC)
 * This is Dr. Murali(dhara) (Subba)Rao from Stony Brook. Is this the same Rao you met, or was that this Murali Rao? The mathematics seems genuine to me, but nevertheless it is reasonable to wait until we have a reference from a peer-reviewed source. --Lambiam Talk  13:05, 2 April 2007 (UTC)
 * I looked at the professional publications listed at his university page, and found not one about "Rao transforms". Charitably, perhaps he is trying to secure international patents before he publishes, and we know journals can have long delays between submission and publication. Uncharitably, most researchers consider it bad form to name something after themselves; such names are best bestowed by others. If he wishes to promote his ideas for financial gain, that is his choice; but if he wishes to use Wikipedia to help him do so, without peer-reviewed publications to back it up, we must decline. --KSmrqT 17:22, 2 April 2007 (UTC)


 * OK, I removed the Rao transform from Integral equation (and also Fredholm integral equation, and Volterra integral equation which were edited later). I mentioned to Dr. Rao (more precisely, to the anon address who edited) about WP:NOR and asked him to comment here if he has objections. Oleg Alexandrov (talk) 15:04, 2 April 2007 (UTC)


 * (The comment below is from my talk page. Oleg Alexandrov (talk) 01:57, 3 April 2007 (UTC))

I am new to Wickipedia contribution and I was not aware of the policy that material should have been published in well recognized journals. My contribution was made in good faith with a view to benefit the readers and enhance the value of Wickipedia. Now I know the policy of Wickipedia, and in this case it is likely to be a limitation of Wickipedia. You can read an expert review of my work here: http://www.integralresearch.net/#Expert_Review. Please see page 4 here which establishes the link between standard integral equations and Rao Integral Equations: http://www.integralresearch.net/wps.pdf. It is easily verified. As for naming my result, if my result is very important, then it is appropriate, and I believe it to be so. I am confident that my results will become "well established" in time. I am not in any hurry. I have protected my idea by applying for US patents. In public interest, in order to evaluate the correctness of my idea, if any qualified person volunteers to verify my results on behalf of Wickipedia, I can send one copy of my book free. But you can get a lot of information on my approach here: http://www.integralresearch.net/RTslides.pdf http://www.integralresearch.net/wps.pdf http://www.integralresearch.net/apex.pdf

Lastly, if someone shows that my new method of solution to solve the Fredholm Equation of the First Kind when applied to shift-variant image deblurring is not better than current methods, I will give them $250.

If you volunteers are serious about enhancing Wickipedia, you should restrict your comments to technical merits of my idea and point out technical weaknesses. Keeping the interests of Wickipedia and the users in mind, you can decide whether your want to post my method or not, is upto you.

I appreciate the voluntary service your are rendering to the public. Dr. Muralidhara SubbaRao (Rao) rao@integralresearch.net —The preceding unsigned comment was added by 24.47.184.148 (talk) 23:33, 2 April 2007 (UTC).


 * Greetings, Professor, and welcome to Wikipedia. We very much appreciate your wish to contribute. Our decision not to accept this particular item is not meant as a reflection on the merits, pro or con. We must ask you to accept the limitations of Wikipedia. Let me briefly explain. We have many thousands of articles on mathematics alone, and we simply do not have a sufficient pool of qualified editors, nor a procedural equivalent of peer review, to weigh the merits of novel unpublished material. No doubt this sometimes means we must forgo the inclusion of valuable new ideas from the cutting edge of research. But the fact is, we are hard pressed to supervise the accuracy and readability of material on topics already published in reputable journals. So we have decided, as a matter of policy, to not accept unpublished material. Perhaps in the future this will change; but for now, we have the awkward task of explaining our limits to enthusiastic new editors such as yourself who are unaware of our restrictions. Please feel free to contribute in other ways, within our limits. We could use the help! --KSmrqT 03:43, 3 April 2007 (UTC)

requesting help at randomness
A new user calling himself MrMiami has been adding some bizarre nonsense to randomness that claims that, if a probability can be assigned to something, then that thing is "ordered" and therefore "not random". I don't think he's likely to see reason. Please come help. --Trovatore 18:03, 2 April 2007 (UTC)
 * It's not bizarre nonsense - it's theology. (Err... perhaps I should think that through again ;->) But the sources he cites are clear, for instance C. S. Lewis. He may have misunderstood them; but it is more likely that they are appealling to a pre-Laplacian vague concept of "randomness". Septentrionalis PMAnderson 03:21, 4 April 2007 (UTC)
 * Did Lewis claim that randomness cannot be mathematically modeled? I'm not aware that Lewis ever wrote on mathematics. Anyway the material I was referring to was (mostly) in the section called "In the physical sciences"; Lewis can hardly be taken as a source there. --Trovatore 03:24, 4 April 2007 (UTC)
 * There he seems to be trying to say that no single sequence is random, which is more or less true ("randomness" is a property of an ensemble of sequences). Saying so might help. Septentrionalis PMAnderson 03:35, 4 April 2007 (UTC)
 * Look back in the history a little further; you'll see what I'm talking about. OK, I'll dig up the diff for you: this diff. --Trovatore 03:37, 4 April 2007 (UTC)
 * Yes, that's what I meant. This is trying to say, strained thtough a lay understanding, that "a random sequence" is uncomputable. It is true that "almost all random sequences" are incomputable; there are only a countable number of algorithms. Septentrionalis PMAnderson 03:42, 4 April 2007 (UTC)
 * No, that's not what he means. Look at the talk page. He says quite clearly that if an event has a probability, then it's not random. --Trovatore 03:46, 4 April 2007 (UTC)

Rose (mathematics)
Can someone please check over the area section I added here to make sure it's alright? Thanks :-) — M ETS 501 (talk) 20:39, 2 April 2007 (UTC)
 * The areas are correct. Those b's in the second equation should be k's yes?  I don't know what you mean by "therefore integrals of the form..": do you mean that this applies also to rose curves defined with sine?  That would make sense.  Also, since the integrands are the same, the antiderivatives are identical, so the intermediate step shown between the integral and the area should (or at least could) look much more similar.  Cheers, Doctormatt 21:26, 2 April 2007 (UTC)


 * After a couple of stabs at it, I think I've fixed the ambiguity. VectorPosse 22:12, 2 April 2007 (UTC)
 * The areas are not given by the same integral. For odd parameter, the limits need to be 0 to pi, and for even paramater, 0 to 2pi.  The earlier version was correct. Cheers, Doctormatt 22:15, 2 April 2007 (UTC)
 * Yup, I got confused for a moment. It's correct now, I think.  VectorPosse 22:27, 2 April 2007 (UTC)
 * Looks good. Nice improvement.  Cheers, Doctormatt 23:22, 2 April 2007 (UTC)

Thanks everyone! :-) — M ETS 501 (talk) 02:47, 3 April 2007 (UTC)

Math portal
Portal:Mathematics looks kind of odd these days, with redlinks for the article of the week and picture of the week. From the history it appears that this state of affairs was present for a while. Anybody knows why things are like this? Oleg Alexandrov (talk) 02:08, 3 April 2007 (UTC)
 * Fropuff fixed it (that was quick :) Oleg Alexandrov (talk) 03:31, 3 April 2007 (UTC)


 * These things will happen from time to time if the portal maintainers (presently myself and User:Tompw) are remiss in their duties. Somewhere I'll put up a set of instructions on how to fix things should this happen again (which I'm sure it will). In the meantime, feel free to pester Tompw or myself if something is wrong with the portal. -- Fropuff 23:54, 3 April 2007 (UTC)

Equilateral triangle
Equilateral triangle redirects to triangle. Shouldn't there be an own article for it? -- 212.149.217.110 18:59, 3 April 2007 (UTC)
 * Well, if there is enough material concerning equilateral triangles, it should be separated. But, what (besides the definition and perhaps how to construct them) would be in this separate article? Jakob.scholbach 20:32, 3 April 2007 (UTC)
 * There are plenty of topics that already link to equilateral triangle. Perhaps some of them would be relevant for an article on equilateral triangles themselves? —David Eppstein 21:16, 3 April 2007 (UTC)
 * I think there is plenty which is specific to the Equilateral triangle so I've reverted back to before it was a redirect and added a few things. --Salix alba (talk) 08:19, 4 April 2007 (UTC)

Lie groups
I've noticed that User:Smylei, with the only contributions in the areas of Television and Uncyclopedia under his belt, has removed a paragraph from the history of Lie groups, and I reverted it as vandalism. However, since he insists that he is right, and I am an interested party, having written the section, I would like to ask other mathematics editors to take a look. Arcfrk 23:07, 3 April 2007 (UTC)


 * Septentrionalis put it back in and I agree that it's fine. Arcfrk, keep in mind that vandalism is quite an accusation (it means that somebody is deliberately trying to compromise the integrity, per Vandalism). While I struggle to understand Smylei's comments, I try to reserve the vandalism accusations for the most clear-cut cases. -- Jitse Niesen (talk) 01:15, 4 April 2007 (UTC)
 * Thanks for the reference. Indeed, according to the policy page you've mentioned, removing parts of the text, even if it looks random or is accompanied with disparaging comments, does not constitute vandalism. So what is it called, then (e.g. for the purpose of motivating a revert)? By the way, I reverted and put a note here before I saw his comments on the talk page. Arcfrk 01:37, 4 April 2007 (UTC)
 * I think what you may be looking for is "Don't be dense", which refers to Hanlon's razor.
 * There is another consideration, as well. You're new to this wiki, so may not have seen the test message left on someone's user page when they have done something untoward to an article. If you visit the template page, you will see that we have a very mild version at first, becoming progressively more firm. One reason is because vandalism is often a "Hey, look at me!" message, intended to provoke a strong reaction; we don't want to give them that reward. Another is because kids experiment. Yet another is because people make mistakes. If we call someone a vandal, they may choose to identify themselves that way and act that way, whereas if we take this approach, perhaps they will become good contributors. (And if not, we gave them every chance.)
 * We still have to deal with bad judgment, bad knowledge, and strong differences of opinion. Most of our editors make consistently helpful or harmless contributions; a very few are distressingly consistent the other way. We do the best we can with an encyclopedia that "anyone can edit". --KSmrqT 03:56, 4 April 2007 (UTC)

Computational mathematics
I wrote computational mathematics (last version edited by me), which User:JJL proposes to replace by redirect to computational science with some additions there. Please see Talk:Computational mathematics and Talk:Applied mathematics for details. Jmath666 04:16, 4 April 2007 (UTC)


 * I can tell how I use the terms:
 * numerical analysis = the development, analysis and application of algorithms for problems in continuous mathematics.
 * scientific computing = same, but with more stress on the applications and less on the analysis
 * computational science = even more stress on applications
 * computational mathematics = as numerical analysis, but not restricted to continuous mathematics. Includes for instance computational number theory, complexity theory and symbolic computation.
 * For me, the first three terms are almost synonymous and they can perhaps best been described in one article. Computational mathematics is quite different, but also a collection of disciplines with little overarching concepts so it will be hard to write an article about it. Furthermore, all these concepts are rather amorphous and discussions about their proper definition tend to not very fruitful so I usually stay away from them. -- Jitse Niesen (talk) 05:45, 4 April 2007 (UTC)

JJL, I see you have reverted my edits to scientific computing that made a distinction between computer science and computational mathematics, consistently with your POV about computational mathematics. I suggest to wait for the outcome of the discussion here. Jmath666 05:54, 4 April 2007 (UTC)

I apologize if this historic view may be incorrect in details. Computer science a.k.a. Informatics split off from mathematics and electrical engineering sometime in the 1950s and separate CS departments started to appear in the 1960s. The term computational mathematics (CM) was used for numerical analysis as well as other computing things done on the math side, see the 1985 Rheinboldt report Future Directions in Computational Mathematics, Algorithms, and Scientific Software commissioned by NSF. A DOD commisioned position paper (page 45) refers to CM as "The Mathematics of Scientific Computation". This report was followed by the NSF program in CM which currently defines its scope by: "Supports mathematical research in areas of science where computing plays a central and essential role, emphasizing algorithms, numerical methods, and symbolic methods. The prominence of computation in the research is a hallmark of the program." Meanwhile, the 1982 Lax report Large Scale Computing in Science and Engineering commissioned by NSF and DOE was the beginning of substantial funding for (and thus the very existence of) what was at different times called Large scale computing, High performance computing, Supercomputing, and Computational science. (The reason for evolution of names was mostly political, to justify funding for something new, buzzwords get tired.) [Added: Scientific computing seems to be a more general term for everything here, used already before 1980.] To date, computational mathematics kept its identity as the theoretical side of things (most ly of us do proofs) distinct from Computer science (only the complexity people still do proofs) [Added: and Computational science (what proofs?)]. So that is my take, which includes sources. Maybe too much original research for an article unless I am lucky enough to find an independent history source. Jmath666 06:32, 4 April 2007 (UTC)

Added: SIAM Timeline says the Rheiboldt report "led to the High Performance Computing initiative.". Jmath666 15:44, 4 April 2007 (UTC)


 * With terms that overlap in meaning, and that are used in different ways by different people, it is always difficult to draw boundaries. Unlike Jitse, for example, I perceive a clear distinction in focus between numerical analysis and scientific computing/computational science. To me, numerical analysis is a field within mathematics, practiced by mathematicians. It is traditionally more about methods or algorithms than about programs. The abstractions addressed by these algorithms are still clearly within the realm of mathematics; they offer methods for solving, for example, a certain type of pde. Scientific computing, on the other hand, is typically an activity that takes place in the context of a class of problems specific to some science, like protein-protein interaction prediction, even when it draws heavily on methods of numerical analysis. But then, of course, the numerical challenges of the Navier-Stokes equations are such that for example in computational fluid dynamics you can't tell where the researcher stops being a numerical mathematician and becomes a computational scientist. Now I think of computational mathematics as a generalization of numerical analysis: computing in the context of a mathematical class of problems, and as such being within mathematics and practiced by mathematicians. To the extent that we do not consider mathematics a science, it would not classify as one of the computational sciences, just like computational linguistics does not. A final remark: in general it does not make for pleasant reading if the introduction to a topic starts with an enumeration of everything it should be distinguished from or not be confused with, something that is hardly helpful to the typical reader who has only a vague idea of what all these things are. Better to focus on what it is and push such distinctions to a section at the end like Related fields. --Lambiam Talk  07:45, 4 April 2007 (UTC)


 * There is certainly scope for historical writing on half a century of computation. It should probably be clearly flagged as history, though. OR is really a concern when there is some spin being put on the material; I wouldn't worry about it for a timeline article. Charles Matthews 10:13, 4 April 2007 (UTC)


 * I feel we're really splitting hairs between num. anal., num. methods, computer/computational sci./eng., computational math., sci. computing/computation., etc. There's a clear line between computer science on the one end and computer eng. on the other, but the rest are muddled. If we have an article for each then I think OR/POV is inescapable. We'll all have different views on this; I think redirecting to a computational science article is most sensible. What meaningful content goes in computational math. but not in computational science and/or sci. computing? I would agree that computational math. is more aligned with numerical analysis--I don't draw the distinction, personally--and feel that sci. comp. is the next step toward computer science, and that computational science is a more interdisciplinary field. Looking at the two SIAM journals in this area and their CS&E conference (which I just attended in the OC), I feel the distinctions are very minor and far from agreed-upon. It's more a matter of how one prefers to be identified, for the most part. I've known some great numerical analysts who knew very little about actually using a computer themselves, but that was back in the 80s--no one is in that boat nowadays (thank you, MATLAB). I think num. anal. deserves a page and computational sci. does, but comp. math. is just multiplying entities beyond necessity. That some groups use a different term doesn't move me. JJL 13:20, 4 April 2007 (UTC)

Thank you. There are still great numerical analysts and even engineers who do not program, not even MATLAB, and will be. I am familiar with the the SIAM CS&E conference though I sent there someone with the paper instead of going myself. Yes, the distinctions are blurred. Some are along the line proofs/no proofs. Some are what is done in what department / academic program / center.

Perhaps someone could take an integrative view on the whole business of applied mathematics / computational mathematics / computer science / computational engineering / computational simulation / large scale simulation / large scale computing / supercomputing / high performance computing / scientific computing / scientific computation / numerical analysis / numerical mathematics [Added: / computational science] etc. and make some sense out of it. Maybe start with disambiguation page and/or a category. What we have is an inconsistent hodgepodge of separate articles, written at very different levels of expertise. The separate pages or sections should give attribution for whatever they say instead of POV. I'd think the role of major conferences, commissioned panels, editorial policies of journals, funding agencies programs, and funding initiatives in naming things and distinguishing fields is notable and also provides the required citations not otherwise available and will help to avoid editor's POV.

Right now, the question is: should computational mathematics be remain a separate page or not? Jmath666 16:23, 4 April 2007 (UTC)

Tagging math articles
It appears that User:Parker007 has asked User:Snowbot to tag a lot of math article talk pages with maths rating. Has anyone heard anything about this? CMummert · talk 00:50, 30 March 2007 (UTC)


 * Nothing beyond the obvious activity on my watchlist. But I don't see any particular harm in tagging them all. —David Eppstein 03:10, 30 March 2007 (UTC)


 * The consensus the last time it was discussed was not to do it. The bot doesn't actually rate the articles, it just puts the tag on them, so a real person still has to do the work. In which case there is no real reason to add the tag.  Also, thousands of articles will only slow down User:WP 1.0 bot more than it is currently slow.  About half articles in the master table are in the "no grade/no importance" entry of the table. CMummert · talk 12:49, 30 March 2007 (UTC)

Snowbot has now been stopped. Thanks to all of you for addressing the issue. The bot has tagged about 1.6K talk pages, what I need to know if the project want them to be detagged or not.  Snowolf (talk) CON COI  -  13:31, 30 March 2007 (UTC)


 * Unfortunately it stopped at the letter 'S'. There was also a bug which meant it tagged articles which already had a "math rating" (spelt without an s). If we want for all these pages to be detagged, we need to be sure Snowbot will only detag the pages it tagged, and not previously tagged pages (which may reflect editor priorities). Otherwise, I suggest we let it finish the job: there are only a few letters remaining. Geometry guy 14:04, 30 March 2007 (UTC)


 * I'm looking at WikiProject Mathematics/Table. I would support removing the math rating tag from all the articles that are both "Stub-Class" and "Unassessed importance". It is OK when someone manually adds a tag with these settings, because then someone else can go through and assess the importance and field, but there are currently 2300 such articles, more than we can expect anyone to go through by hand.  We might as well clear them out and then pick up with the manual rating again. This is just my opinion. The user who requested the tagging has not yet commented on his/her goals for doing it. CMummert · talk 14:10, 30 March 2007 (UTC)


 * Would anyone object to the removal of certain tags as described in my previous comment? CMummert · talk 03:10, 4 April 2007 (UTC)


 * I support this, especially if Snowolf can persuade Snowbot to remove only the tags of this form which were previously placed by Snowbot: if the bot can read the recent edit history it should be possible. Geometry guy 19:58, 4 April 2007 (UTC)
 * I support removing auto-tagged articles - it serves no use. (Btw, if the bot can't read the edit history, it could just remove stub-class unassessed-importance articles in Category:Unassessed-field mathematics articles). Tom pw (talk) 20:13, 4 April 2007 (UTC)


 * In my opinion, it would be fine to make a list of the intersection of those categories and ask SnowBot to untag those articles. I have been manually assigning importance ratings to articles periodically, so not very many of the articles with "unassessed" importance were manually added. CMummert · talk 20:25, 4 April 2007 (UTC)


 * It should be unassessed importance plus unassessed field. I went through and assigned a field to some of those articles, but not always an importance; those tags could maybe be left in place. --Trovatore 20:50, 4 April 2007 (UTC)


 * I would generate the list fresh, so articles that have been rated by hand ought to be left alone. There are currently 2153 articles in both Category:Unassessed-importance mathematics articles and Category:Unassessed-field mathematics articles. That seems like a reasonable set from which to remove the tags. I would make sure that every article with a quality class other than stub has an importance rating before making the list, so those would not be untagged. CMummert · talk 21:35, 4 April 2007 (UTC)


 * User:SnowBot is making the changes now; I estimate it will take about 8 hours from now to go through the entire list. CMummert · talk 15:48, 5 April 2007 (UTC)

Request tags be put back. It's useful information for users that the articles fall within the scope of Project Mathematics, and this should be indicated on their page.

It is a good thing for articles to be tagged with a Project, so people know where to come to for experts. And so 'bots can easily ID the subject matter, just from a links dump. Jheald 18:13, 5 April 2007 (UTC)


 * There is already a system for identifying which area the articles are in: categories. They allow User:Mathbot to generate the list of mathematics articles. The project banner is for rating the quality of articles in the WP 1.0 scheme. It does no good to put on the banner if every article ends up "unrated". This was discussed last fall and the (unanimous) consensus then was not to put the banner on every math article. User:Parker007 asked for the articles to be tagged without bringing it up here; the untagging is essentially just restoring things to the way they were before then. CMummert · talk 18:34, 5 April 2007 (UTC)


 * There is a great opportunity here for editors to go through Snowbot's recent edit history and tag any articles they think deserve a tag (with rating, importance and field of course). I did a small handful recently. It doesn't take long to scan through the list, particularly if you just want to tag a small number of articles. Geometry guy 12:04, 6 April 2007 (UTC)

Use of quantifiers ('for all', 'there exists')
I don't seem to find any previous discussions concerning the use of quantifiers (e.g. $$\forall, \exists$$, as in the definition section of Inner product space) in articles (I might be bad at searching, though). I personally think these should not be used in a 'textbook' setting, and they greatly reduce readabillity. Granted, they provide a clean and very clear exposition for readers who are comfortable with them, but since articles are usually directed to a general audience, I think they should be avoided. Comments? JohanK 07:51, 4 April 2007 (UTC)


 * They should be avoided, except in logic articles where they have a legitimate reason to be used. We 'expand abbreviations' routinely, and if quantifiers are just used to avoid writing words in plain English, they should be expanded.Charles Matthews 10:10, 4 April 2007 (UTC)


 * Our Manual of Style says:
 * In particular, the English words "for all", "exists", and "in" should be preferred to the ∀, ∃, and ∈ symbols.
 * I think there hasn't been any discussion about it because everybody agrees that quantifiers should be avoided (though I have far less problems with using ∈ instead of "is an element of"). -- Jitse Niesen (talk) 11:02, 4 April 2007 (UTC)

Let $$Q=\mbox{quantifier},$$ $$W=\mbox{Wikipedia}$$, and $$M=\mbox{math style manual}$$. Then,


 * $$\forall Q\in W: M\Rightarrow \mbox{expand to plain English}(Q).$$

Oleg Alexandrov (talk) 15:00, 4 April 2007 (UTC)


 * What about a page like peano's axioms. As its dealing particular with set theoretic concepts it seems somewhat appropriate to use this notation. Expanded it could become quite unwieldy. --Salix alba (talk) 18:39, 4 April 2007 (UTC)
 * Peano's axioms uses quantifiers only in quoting a set of equivalent axioms (from Kaye) which introduce addition and multiplication explicitly (and are therefore much longer). I wouldn't take the quantifiers out; but then I wouldn't have included Kaye's axioms either. Do you think quantifiers are needed anywhere else? Septentrionalis PMAnderson 19:16, 4 April 2007 (UTC)
 * Probably a case of ignore all rules there will always be exceptional cases, where the style can be broken. --Salix alba (talk) 19:28, 4 April 2007 (UTC)
 * The symbolic quantifiers in Peano's axioms are only used for a particular set of formal axioms; the rest of the article, which is not about the formalization, uses natural language. Other examples of this practice include Zermelo–Fraenkel set theory and Prenex normal form. It is only when we are discussing axioms qua symbolic expressions in a formal language that they should be written symbolically. In other contexts, when the actual formalization is not being discussed, symbolic quantifiers are avoided in published mathematical writing,and they should be avoided here too. CMummert · talk 19:46, 4 April 2007 (UTC)

There is a reason that formal languages were invented. Natural languages are very confusing and clumsy at expressing complicated concepts, e.g. when there are quantifiers in both the hypothesis and conclusion of an implication. See Transfinite induction for an example of a concept which is expressed very unclearly in ordinary English, but would be trivial in formal logic notation. JRSpriggs 10:21, 5 April 2007 (UTC)
 * Which is why our mathematical style guidelines advise moderation, not abstinence:
 * Careful thought should be given to each formula included, and words should be used instead if possible. In particular, the English words "for all", "exists", and "in" should be preferred to the ∀, ∃, and ∈ symbols.
 * (Emphasis added.) The intent, as always, is clear, correct, and compelling communication with the reader. It would be lovely if we could just say, "Write well." :-) --KSmrqT 11:28, 5 April 2007 (UTC)

Category:Applied Mathematics
Should there be such category? There was only Image analysis in it and the category page was blank. So have have changed Image analysis to Category:Mathematics. Jmath666 20:48, 5 April 2007 (UTC)


 * That's a upper-case problem; see Category:Applied mathematics. Charles Matthews 21:05, 5 April 2007 (UTC)

I see. So I have corrected the page pointing there. Jmath666 21:22, 5 April 2007 (UTC)
 * I've deleted the now-empty category. — M ETS 501 (talk) 04:59, 6 April 2007 (UTC)

Need help with computational mathematics
I need some help there. From User:JJL on Talk:Computational mathematics: "I don't believe this is a separate field but rather a different name for what is now usually called computational science. I suggest redirecting it there and editing that article to mention that, indeed, computational science as a named academic program often has an emphasis on the applications to science." This [sentence] makes no sense. Some of you made few helpful edits to computational mathematics but that does not seem to be enough to make User:JJL go away and stop claiming that computational mathematics does not exist and pushing to delete the article. Thanks. Jmath666 00:40, 6 April 2007 (UTC)


 * The real problem is that no-one is wrong or right here. Some groups use the term in a way that is almost indistinguishable from computational science, except for the pretension that, since they are mathematicians, they must be doing it right. Others prefer to use the term differently. I think the first may have more clout. --Lambiam Talk  01:00, 6 April 2007 (UTC)


 * Of course. But the issue is that the article should exist and explain that. The different uses of the term can be easily supported by sources and I am not asking for help with that. I need support for the existence of the article. Jmath666 01:34, 6 April 2007 (UTC)

Calling for a straw vote
Do you think that computational mathematics should be a separate article or, as JJL suggests, redirected to computational science and that article edited by JJL "to mention that, indeed, computational science as a named academic program often has an emphasis on the applications to science" (sic)? Please vote separate article or redirect. Thank you. Jmath666 04:08, 6 April 2007 (UTC)


 * With the lame discussion we have had (for which I'm more than a little bit responsible, but which is most likely due to there being no persuasive arguments either way) I don't see how this will lead anywhere. Either approach (separate article, or redirect + treatment in Sci. computing) seems fine to me. What I find much more of an issue is that the current article Computational mathematics is a painful read, and when you manage to read through you emerge at the other end with the feeling "huh? what happened?". Merging it with Scientific computing will not improve the readability of the latter article – which itself offers considerable opportunity for improvement on the scale for reading pleasure. --Lambiam Talk  07:25, 6 April 2007 (UTC)


 * In the spirit of the previous comment, I'd like to concentrate more on the content than on the nomenclature. Neither article seems to mention any work on scientific computing (computation ?) in algebra. Yet, pioneering work on computer algebra systems, such as PARI and MAGMA aside, at the moment, number-therertic applications such as primality testing and factorization are arguably among the most important in the whole field of scientific computation (think RSA, cybersecurity, etc). Is there an article that discusses it? Which article would be closest in spirit to Richard Crandall and Carl Pomerance, Prime numbers. A computational perspective (ISBN: 978-0-387-25282-7)? Arcfrk 07:46, 6 April 2007 (UTC)

If the information in computational mathematics (not form) is preserved after such move that's OK with me too but given the attitude of JJL that is highly unlikely, please do read the quoted sentence above that shows it clearly. The current painful state of computational mathematics (yes I fully agree) emerged after me trying to respond to objections of JJL but that was a mistake obviously nothing will satisfy JJL given his POV that computational mathematics does not exist and it just messed it up.

I would not mind taking look at the whole scientific computing area and I suggested just that above and even wrote a rought draft of a the top article but there was no response. and I cannot afford the kind of struggle like with JJL reverting my edits every step of the way - too much waste. Can you look? Sorry I do not know anything about computing in algebra etc.

I can give it one last shot and rewrite computational mathematics completely reporting on documented facts not any editor's wish what is should or should not be. But I would like to see some interest/support before I take on that task. Thanks. Jmath666 08:37, 6 April 2007 (UTC)


 * Carve up between scientific computing and numerical analysis. Reduce current article to no more than a dismabiguation stub.  Jheald 10:25, 6 April 2007 (UTC)


 * Redirect (and by the way, see WP:OWN). JJL 12:40, 6 April 2007 (UTC)


 * Looking at the various orginisations, confrences and journals with computational mathematics in the title it seems the most of these would fit numerical analysis/scientific computation. There are others which seem to have a distinctive mathematical flavour The Complexity of Computing the Hilbert Polynomial of Smooth Equidimensional Complex Projective Varieties and tend to fall under the Automated theorem proving and Symbolic computation. Following Arcfrk, I think there is a need for a general article on using computers in mathematics, detailing its history and some of the key moments say the Four color theorem, E8 (mathematics) and calculation of primes, but its not this article. This article should be kept short as it mostly mirrors other articles. --Salix alba (talk) 13:54, 6 April 2007 (UTC)

Thank you very much for the feedback. I have replaced the page by disambiguation. Jmath666 00:31, 7 April 2007 (UTC)

Lifting (mathematics)
I hastily created a stub article titled lifting (mathematics), and at this moment I'm not even sure I've got the definition right. Could others here add more? Michael Hardy 21:04, 6 April 2007 (UTC)
 * ... looks like haste doesn't always make waste: several people have edited this thing that I put there in a few seconds without yet proving I included something horribly dumb in it. We'll see.... Michael Hardy 01:23, 7 April 2007 (UTC)
 * Yes, perhaps we should start a category "Collaboration of the last 30 minutes" :-) Jakob.scholbach 01:54, 7 April 2007 (UTC)
 * The most important thing is to include the diagram illustrating it. I think, User:Fropuff was offering his help with things like that. Arcfrk 02:43, 7 April 2007 (UTC)


 * I've regularly seen the term "lifting" with a quite different meaning; expressed categorically it would be the action of a functor on a morphism in the category Set. Examples are the "lifting" of a function from A to B to one operating on sets in the usual covariant way: from Set(A) to Set(B). An other example is how the addition operation of a field carries over to vector addition in a vector space over the field. Is this notion of lifting sufficiently standard and entrenched that it should be mentioned? --Lambiam Talk  07:59, 7 April 2007 (UTC)
 * I have not yet heard the word "lifting" in this context; rather extension (of the field operations to the vector space) or that a functor Fun(A, B) category inherits the properties of B. Jakob.scholbach 19:24, 7 April 2007 (UTC)
 * There are many uses of the term lifting outside of category theory. In Sobolev spaces, the extension operator (to extend a function from the boundary of a domain in Rn inside, say) such that its norm is bounded in certain Sobolev norms, is sometimes called lifting (it is the opposite of the trace operator, which is basically a restriction). There are many papers on lifting of some operator in some functional spaces. Perhaps lifting is a general term that refers to a commutative diagram in a certain context. Jmath666 01:09, 8 April 2007 (UTC)
 * There is a (slightly) different usage within category theory - or at least within homological algebra, which at least formally is not the same thing:-)
 * Another common usage in `practical' homological algebra is within diagram chasing, where you lift elements (ultimately often in order to construct homomorphisms). Element-wise, a lifting of an element is simply a preimage of it; possibly involving an arbitrary choice of one out of many. In favourable cases, you may emply element-wise liftings in order to lift homomorphisms. Of course, this is not `the proper way' to do it, since not all caregories have forgetfulness functors to the category Set (even if small abelian categories may be represented thus). I note that Saunders Mac Lane in Categories for the Working Mathematician does not have a reference to either 'lift' or `lifting'; he does refer to digram chase, but spende the time with explaining how to do it 'using "members" (in A) instead of elements (in Ab)' (from section VIII.4, 'Diagram lemmas'). He very clearly assumes that the reader already is used to the concrete 'chasing elements around diagrams' technique from homological algebra.
 * See the enlightening discussion in Five lemma. There, too, 'liftings of elements' are not mentioned; but the following instance from the first proof:
 * Since p is surjective, there exists an element d in D with p(d) = t(c&prime;)
 * may be referred to thus: 'Since p is surjective, t(c&prime;) may be lifted to an element d in D.' JoergenB 05:46, 8 April 2007 (UTC)
 * I believe this has gotten enough publicity here. Further discussion should move to the article talk page, where it will be found by future editors who may want to know what motivated our choices. --KSmrqT 06:03, 8 April 2007 (UTC)

PanelWhiz
Cross-posted from Requests for comment/Maths, science, and technology:


 * Talk:Stata - Should external and wiki links to PanelWhiz be included in the Stata article? Is PanelWhiz notable? - 16:14, 7 April 2007 (UTC)

Tearlach 12:56, 8 April 2007 (UTC)

AfD nomination
Articles for deletion/Curve transformation. The author of the page describes it as original research. Charles Matthews 19:16, 8 April 2007 (UTC)

AFD nomination
Articles for deletion/Internet shorthand notation seems to be misnamed for what I've always known as "ASCII math notation". Note, there may not be one standard method, but I believe the overall concept is notable for an article, at least in terms of the history of computing. Yea or nay, I think this nomination needs more knowledgeable input. --Dhartung | Talk 21:55, 8 April 2007 (UTC)

Differential equations
I've come across the article Examples of differential equations which has been tagged as having insufficient context (in my opinion, rightly so), and following the links, found an article ordinary differential equation. Now, the interesting thing about it is that there is also an article differential equation which is not linked from the ODE page. Was there a schism a while back, with one school of editors leaving and founding their own page? What do other editors think about the current distribution of the basic material between different articles? (I'll mention two more: partial differential equation and phase portrait, which is just a stub with a picture.) Arcfrk 03:51, 6 April 2007 (UTC)
 * I don't know about the distribution, but an ordinary differential equation is a specific case of a differential equation, so they (rightfully) should have two separate articles. — M ETS 501 (talk) 04:58, 6 April 2007 (UTC)


 * I've visited these pages a few times for various reasons, and am also somewhat unhappy with the current arrangement. It is mainly partial differential equation that fills me with despair, although I have been trying to keep quiet about it because a complete cure may be hard to find! The main problem, in my view, is that the article implicitly assumes (without stating it) that a PDE is just one equation (i.e. real-valued, rather than an equation in Rn or some other space). This means in particular that the most fundamental trichotomy of PDEs - into the underdetermined, the determined, and the overdetermined - is not properly discussed. Instead aspects of this trichotomy (such as the h-principle) are mentioned in passing, and these discussions are consequently completely unilluminating.


 * I like the idea that differential equation provides an elementary introduction to the subject, but partial differential equation should then be a really comprehensive survey of the scope of this immense and hugely important field, rather than just a list of a few one variable examples (I am being a bit harsh here, I admit).


 * As for the sociological concerns about previous editorial groups, my experience so far at wikipedia suggests that the best approach is to read the edit history and then be bold! The quality of the article is what matters, not how it got there. If a complete rewrite is needed, go for it! Geometry guy 16:31, 9 April 2007 (UTC)


 * I could't find h-principle even after your tip, which is just as well, because there are even more fundamental concerns about that article, namely, that it somehow conceals the fact that there are (many!) important nonlinear equations, and we do have some theory (integrable systems, Hamilton-Jacobi, etc). The distinction between linear and nonlinear appears only casually in the middle of one section, and not given any serious treatment. Systems are mentioned, but only linear systems of first order! On the other hand, the article is fairly long as is, meaning, in particular unfocused. It the same story as with ODE, only worse (since the material is more complex). Should we propose them as collaborations of the week? Arcfrk 03:07, 10 April 2007 (UTC)

Rose (mathematics)
Should a formula be included for defining 2-, 6-, 10-, etc-petalled roses such as
 * $$r = \cos(k\theta)(-1)^{\left \lfloor\displaystyle\frac{\theta-\pi}{\pi}\right \rfloor}$$

or something? — M ETS 501 (talk) 05:11, 6 April 2007 (UTC)


 * See WP:NOR. --Lambiam Talk  06:40, 6 April 2007 (UTC)
 * This is not original research. It may not need to be included, but original research is defined as "a term used in Wikipedia to refer to unpublished facts, arguments, concepts, statements, or theories, or any unpublished analysis or synthesis of published material, which appears to advance a position".  This is clearly not to advance a position.  I'm not getting rich off this :-) — M ETS 501 (talk) 06:47, 6 April 2007 (UTC)
 * Also, I didn't say that this formula should be included. What I meant was should we put a note on the article that a (4k+2)-petalled rose can be created in other ways such as using the floor function or plus/minus sign on a (2k+1)-petalled rose? — M ETS 501 (talk) 06:53, 6 April 2007 (UTC)
 * I think you may be interpreting WP:NOR a bit too literally, Mets. The real issue is not about "advancing a position" but rather about the "unpublished" part.  (See WP:ATT instead of WP:NOR perhaps.)  At any rate, this would be a better issue for the talk page for the article itself.  VectorPosse 16:52, 6 April 2007 (UTC)
 * It would be very silly to call such a thing "original research". Michael Hardy 23:13, 7 April 2007 (UTC)
 * Then Wikipedia is very silly. What about versions that miss the even petals for originally "odd-petalled" roses (which might be called "she-loves-me roses"), or roses with only every third petal, or only the petals whose index is prime? Or roses with petals of different sizes? The variations on functions that can be invented have no limits. Fortunately, the prohibition on OR is a clear criterion by which we can keep such products of unbridled inventivity out of Wikipedia. --Lambiam Talk  00:25, 8 April 2007 (UTC)
 * I think you hit the nail on the head when you said Wikipedia is silly.  However, I think we have to do what we can to keep Wikipedia usable for those of us interested in the non-silly and not-so-silly portions.  I could write a (set of) article(s) on the multiplication table of of numbers in base 2007 (Well, I couldn't physically type it, but theoretically a computer could do it.)  The resulting article wouldn't necessarily be original research excepting creating of a notation for numbers in that base, but it wouldn't be notable either, because the results are trivial.
 * Hehe, it looks like I've just created new policy for Wikipedia. Wikipedia is not a multiplication table.  It is not a Rainbow Table either.Root4(one) 03:51, 9 April 2007 (UTC)

The formula above is yet another excellent reason why vertically stacked fractions should not appear in superscripts. Consider: Michael Hardy 21:08, 6 April 2007 (UTC)
 * $$ r = \cos(k\theta)(-1)^{\displaystyle{\left \lfloor(\theta-\pi) / \pi \right \rfloor}}. \, $$
 * $$ r = \cos(k\theta)(-1)^{\displaystyle{\left \lfloor(\theta-\pi) / \pi \right \rfloor}}. \, $$
 * Thanks, but either way, there's objections, and I don't want to add it if people don't want it. — M ETS 501 (talk) 23:33, 6 April 2007 (UTC)

Double integral and Multiple integral
Anyone think that the double integral article should be merged into the multiple integral article? The triple integral article already redirects to multiple integral.--Jersey Devil 02:52, 9 April 2007 (UTC)


 * To be sure, multiple integral is a lot more detailed, but very general + many details &rArr; intimidating (frequently). I think that the double integral article can be considerably improved, with more geometric treatment and little or no measure theory (at the moment, it reads more like misconceptions about double integral). Arcfrk 04:27, 9 April 2007 (UTC)


 * Currently the double integral covers basically the same as Fubini's theorem but expressed in such a way as to cause uncertainty and doubt in any freshman. Multiple integral has similar problems, this long page could be shorterened by moving the majority of content for polar, spherical and cylindrical coordinates into their respective articles. I've now made double integral a redirect, but there is scope for expanding it. --Salix alba (talk) 10:40, 9 April 2007 (UTC)


 * I don't think that was appropriate. Multiple integral is not very user-friendly. Double integral needed some better introductory material, but if you are just in need of a basic introduction to the calculus of multiple integrals, you don't want generality, you want some idea of the picture and the reasons why two definite integral operations should commute. Charles Matthews 07:31, 10 April 2007 (UTC)

History alteration
Something bizarre and troubling has come to my attention. I was looking at the article on classical modular curves and saw that two superscripts were closed with &lt;sup&gt; instead of &lt;/sup&gt;. I fixed the problem, but was puzzled that something with such dramatically visible consequences had previously escaped notice. So I checked the edit history, which showed the problem had been there for a long time, flying under the radar of a number of seasoned editors. This struck me as odd, but I moved on. But just now, I found the same phenomenon with a subscript tag in an article, area of a disk, I have worked on and scrutinized repeatedly. Here, another editor did the fixing. The history claims a subscript tag has been missing for some time. I don't believe it; I'm sure the presented history is false.
 * False histories:
 * classical modular curve
 * area of a disk

I don't know what the source of this may be, whether a software glitch or deliberate foul play; but we really need to be able to trust histories. At the moment, I do not. --KSmrqT 03:44, 9 April 2007 (UTC)


 * Essentially the same "disclosures" occur in Talk:Area of a disk, where they are at least as visible. It is hard to imagine David Eppstein not noticing the diminished appearance of this contribution of his. --Lambiam Talk  04:11, 9 April 2007 (UTC)


 * Huh. Weird.  The same thing happened on the Hilbert matrix article.  I wondered how that could've escaped so many people's attention for so long...  Lunch 04:16, 9 April 2007 (UTC)


 * Must be some bizarre server problem. I just saw another example today, at crystalline cohomology. It has already been corrected, but I am absolutely certain that the problem (unclosed sub) wasn't there whenever I had looked at it since the article was created, and the last time was less than a week ago. Arcfrk 04:44, 9 April 2007 (UTC)


 * So the server is taking the slashes out of the close tags? Weird. I wonder if it's happening with any tags other than sub and sup...ref could be particularly disasterous since if the references tag is inside a ref it won't ever get expanded. And yes, I imagine I would have noticed the shrunken text in that comment, had it been like that before. —David Eppstein 04:54, 9 April 2007 (UTC)


 * (edit conflict) This is probably related to a recent update of HTML Tidy, announced at Wikipedia:Village pump (technical)#Tidy upgraded. It seems like the old version of HTML Tidy repaired stuff like  by replacing the second   tag by a close tag, while the new version does not do this. -- Jitse Niesen (talk) 04:58, 9 April 2007 (UTC)


 * So does this mean that xy z nested superscripts will start working? Good news if so! —David Eppstein 05:12, 9 April 2007 (UTC)
 * We have had a workaround for the nesting, using a &lt;span&gt; around the inner script. But if Jitse is right, the errors have been in the source the whole time, silently corrected. --KSmrqT 06:39, 9 April 2007 (UTC)


 * Should we ask a bot owner to list pages with  or   that were previously automagically repaired, so that we can fix problems now no longer covered-up?  --Lambiam Talk  07:20, 9 April 2007 (UTC)

I think that Jitse is correct. The errors have been there all along, but were hidden by bugs in the processing of html. Notice also that where previously tags terminated automatically at the end of a paragraph, now they do not. A user had trouble with the "small" tag on the signature of Signpost not terminating any more and affecting the rest of his user talk page. JRSpriggs 11:11, 9 April 2007 (UTC)


 * Yes, I agree. I often make the mistake of missing out the slash in a closing sup or sub, but was previously forgiven by the generous HTML tidying code. Now it is not so generous, we should all go back to articles we have edited and check them for mistakes! I made this mistake several weeks ago at affine connection for example, but only discovered and fixed it today. Geometry guy 16:50, 9 April 2007 (UTC)


 * I can recommend the editor wikEd, which helps saving time by inserting sup and sub and much more. For those (like me) who do not crave for the command line text editor atmosphere, this nice tool creates some comfortable editing environment. Jakob.scholbach 17:37, 9 April 2007 (UTC)

Tidying up after the change to HTML Tidy

 * I'm going to do a database dump search and find all instances of ......, ......, ......, and ...... . Note that some of these might actually not be problems (like if the user intended xy z ) but if they're listed by a bot then we can go through them and check. — M ETS 501 (talk) 17:59, 9 April 2007 (UTC)


 * For completeness, include ... and ... . --Lambiam Talk|undefined  18:34, 9 April 2007 (UTC)
 * OK. I'm doing each case separate, because it's easier that way, and I'll include those two.  It looks like it's very successful, almost all the articles it's catching have the problem, which is thousands and thousands of articles, unfortunately.  I'll post when I'm done making the list. — M ETS 501 (talk) 18:49, 9 April 2007 (UTC)
 * Also, the newest dump is 7 days old, so it's catching many that've already been fixed (it caught the versions before these edits, for example) so I'm going to have to parse the live versions of pages before presenting the list. — M ETS 501 (talk) 18:52, 9 April 2007 (UTC)


 * What a hero M ETS 501 is! Things like this are what make WP so great! Geometry guy 18:59, 9 April 2007 (UTC)
 * LOL. First set of data done at User:Mets501/Pages that need to be fixed. — M ETS 501 (talk) 19:31, 9 April 2007 (UTC)
 * OK, two sets done; that's about 900 pages to work with so far. I'm supposed to be on Wikibreak now, so I can't spend any more time, but that should be enough for the time being. — M ETS 501 (talk) 19:48, 9 April 2007 (UTC)

Since I see that a lot of the subscript problems involve chemical formulae, I left a note over on Wikipedia talk:WikiProject Chemistry describing the problem and pointing to Mets501's page. Perhaps some of the people there can help as well. —David Eppstein 20:23, 9 April 2007 (UTC)


 * Perhaps we should make a separate list of problematic maths articles (the intersection of the list produced by M ETS 501 and the list of mathematics articles). --Lambiam Talk  20:28, 9 April 2007 (UTC)


 * I fixed a very small handful of /sup errors (often mine) in mathematical articles. It would be good to update this problem from time to time, even if it requires us to drag the magnificent user Mets501 out of wikibreak. Geometry guy 20:47, 9 April 2007 (UTC)
 * I've updated the list. I can do this once or twice a day if that's the kind of update that you mean. — M ETS 501 (talk) 21:43, 9 April 2007 (UTC)


 * Wow, fantastic! Once a day (or every few days) would suit me fine! Lets hope the chemists can sort out their stall too. Geometry guy 21:54, 9 April 2007 (UTC)
 * OK, I've rewritten the program, so all I have to do is press one button and the bot updates the list :-) I can run it often, because it's no work for me!  — M ETS 501 (talk) 22:12, 9 April 2007 (UTC)
 * Good work so far. Some questions: Does the consolidated list include all combinations of subscript and superscript? Can we have a consolidated list intersected with mathematics articles to let the chemists fend for themselves? How do I prevent an article with properly nested script tags ( xy z ) from being listed every time? The numbers do not seem to indicate the number of errors; can that be changed? Can we get a display of a little text surrounding each error, so we don't have to search through twenty tags to find one? --KSmrqT 23:21, 9 April 2007 (UTC)
 * Thanks. Unfortunately, I have very little time now; not nearly enough to invest in writing code much more complex than it is now.  It looks like the list is being weeded through quite fast now by Beetstra with AWB, and the rest we'll have to be done by hand.  I'll look into putting in some surrounding text, but I'm not sure how quick it would be do that.  If you remove a page from the list, the bot won't add it back, so if it's on the list in error, just remove it.  The bot has two functions (at the moment): it can go through the database and check for  ... and .... , and it can go through each page listed at User:Mets501/Pages that need to be fixed and check if it still has a syntax error.  I'm going to expand the first function soon to include other syntax errors as listed above. — M ETS 501 (talk) 00:10, 10 April 2007 (UTC)

Citizendium content
Just to bring back an old discussion about content from Citizendium, the Special:Export page works on Citizendium to get the wikicode of any page. For example, to get the code of their Mathematics page, visit http://en.citizendium.org/wiki/Special:Export/Mathematics. Not in ideal form, but it's a way. — M ETS 501 (talk) 00:01, 10 April 2007 (UTC)
 * The copyright of these pages is, unfortunately, still unclear. It would be prudent to avoid copying material from CZ to here until CZ gets their act together. CMummert · talk 00:52, 10 April 2007 (UTC)
 * Definitely. I was just pointing out that we actually don't need registered users there to get their source code. — M ETS 501 (talk) 03:11, 10 April 2007 (UTC)

Institute for Mathematics and its Applications
In one of the most idiotic edits I've seen in a long time, User:The Kinslayer, who seems to spend most of his efforts on topics of no importance, marked Institute for Mathematics and its Applications for speedy deletion on the grounds that it is not important, and did not notify anyone who had edited that page. It was recreated recently. Someone else then deleted it. Michael Hardy 02:40, 29 March 2007 (UTC)


 * This user seems to have a problem with articles on institutes. --Lambiam Talk  07:03, 29 March 2007 (UTC)


 * This user is not alone. I wrote a stub on Fachinformationszentrum Karlsruhe which got marked for speedy deletion by Realkyhick the next minute. Now that's itchy fingers. Jmath666 07:47, 29 March 2007 (UTC)


 * It is an unfortunate consequence of the large number of nonsense pages created each day that, when creating a stub on a person, company, institute, etc., one must immediately add references to assert notability or the article is likely to be tagged for deletion almost immediately. The number of math related articles created each day is far less than the number of nonsense or non-notable pages, so we can't really hope for some sort of special treatment. One solution is to write such articles in a subpage of your user page and move them into place once they are basically done, even though this violates the basic idea of collaboration on the wiki. CMummert · talk 11:27, 29 March 2007 (UTC)


 * Sounds like these human spam filters need their Bayesian statistics adjusted. The odd thing is, ever since I began bringing up random articles frequently, I've been struck by the fact that almost all are a paragraph or two on an obscure topic. So this "shoot first" approach, while an understandable reaction to garbage, is tainted by ignorance. A town of 300 people: no problem; a minor sports figure: no problem; a major mathematics institute: kill it. The garbage is a cancer, and these people want to feel useful. But, as so often happens, the "cure" has damaging side-effects, sometimes worse than the disease. --KSmrqT 14:32, 29 March 2007 (UTC)


 * I recently exchanged a few pleasantries with Realkyhick about his shoot first and ask questions later approach to new articles. I notice that he's a member of the New Pages Patrol. He also ignores the guidelines for that process with some regularity. I suggested that he at least look at the page history before tagging an article with a speedy delete tag, but he doesn't seem inclined to accept that suggestion. Type A personality, I guess. DavidCBryant 15:22, 29 March 2007 (UTC)


 * Fachinformationszentrum Karlsruhe happens to be marginal (not a major mathematics institute like IMA) but the original article did indicate why it was notable, even if the article was very short. It took some speedy expansion (which I really did not plan on doing right at the time) to keep the trigger happy patroller off. Apparently it is not clear what it takes for an article to be a valid stub. Well, the lesson is just keep the red tags in and do not go the next step unless one is ready to invest some work right at the moment. Jmath666 15:41, 29 March 2007 (UTC)


 * It seems that the root of the the problems discussed here is that the speedy deletion is too speedy - both the tagging and the admin action. Per CSD: The word "speedy" in this context refers to the simple decision-making process, not the length of time since the article was created. There should be a reasonable time required for both. There is something vague about that in CSD but I could not find mention of some mandatory wait period anywhere. There should be one. Jmath666 23:14, 29 March 2007 (UTC)


 * There are, I think, good reasons for speedy deletions to be speedy. The problem is not the existence of speedy deletions, but rather someone abusing the speedy deletion process. The only criterion I can see in WP:CSD that might fit is A7, "does not assert the importance or significance of its subject". That is, even when you write a stub, you have to explain what is notable about the subject; why is it important enough to be in WP? E.g. in the current article "largest single mathematics grant the NSF has ever awarded" seems to be enough to counter that criterion. I don't know how to access the previous versions in order to tell whether similar language was present in the deleted versions, but if it was, they shouldn't have been deleted. I don't know what the process is when one feels that an admin has been consistently abusing speedy deletion, but I assume there is some appeal procedure available. —David Eppstein 23:52, 29 March 2007 (UTC)


 * I added that fact because I thought it was interesting, not really to establish notability. It was not present in previous version, which also had no references except the external link.  I added the reference to the SIAM article to assert notability.  WP:notability doesn't mean notability - it means "discussed in multiple, independent reliable sources", so the only way I see to assert notability is to include such sources.
 * As for appeal, you could I suppose go to WP:DRV, or you can ask a friendly admin to give you a copy of the deleted version, add some content, and recreate the article. There are about 20 admins associated with WP:WPM, which is about 1 per 800 articles, a better ratio than WP as a whole. CMummert · talk 00:35, 30 March 2007 (UTC)

You can't know what should be considered notable and what should not without familiarity with the field. It seems to me those who were involved in the present case disregarded that fact, which I would think would be obvious. Michael Hardy 00:00, 30 March 2007 (UTC)


 * Of course you can't tell what is notable and what is not without some knowledge, but it's not too hard to determine whether an article asserts notability (in this case, importance or significance), which is the speedy deletion criterion. Articles about even obviously notable subjects will get deleted if they do not include anything telling the reader what their importance is. The only thing that could possibly be interpreted as an assertion of significance in the original article was that the IMA was a body with academic aims associated with a particular university. That description does indeed suggest that the organisation could be quite significant and I wouldn't list it for deletion, but I can't really blame anyone for thinking there was no claim to importance. JPD (talk) 10:01, 30 March 2007 (UTC)

Married in haste, we may repent at leisure." — William Congreve, The Old Bachelor, Act V scene 1.
 * Now that the history has been restored, we can look at what was there before deletion. Two things strike me. One, behind the scenes, is that the article has a long history. Two, the article has a link to the organization's web site. It would take less than thirty seconds to check both of those, which should give more than enough information to see that a "speedy" tag was wrong. Note I am assuming a tagger and an admin who know nothing about the standing of the editors who touched the article, and who know nothing about the mathematics world. Was it a stubby, lackluster article? No doubt. Was "speedy" appropriate? No way.
 * Here's what the WP:SPEEDY page says, at the top:
 * Before nominating an article for speedy deletion, consider whether an article could be improved or reduced to a stub; speedy deletion is for cases where an article does not contain useful content. Note that some Wikipedians create articles in multiple saves, so try to avoid deleting a page too soon after its initial creation. Users nominating a page for speedy deletion should specify which criteria the page meets; it would also be considerate to notify the original author.
 * From this lead paragraph of three sentences, not one was properly followed. I frequently look at random articles, and the one that I just hit is HaShevet. Compare it, especially what it explicitly asserts (or fails to assert) about notability, with the IMA article at the time of tagging and deletion. So forget the CYA, which is not very convincing. The typical catastrophic failure involves a sequence of things going wrong; here, article, tagging, deletion, and response.
 * From Barlett's Familiar Quotations:
 * "Thus grief still treads upon the heels of pleasure;
 * Sum the time spent by multiple editors in response to the hasty deletion and compare to the time necessary to do the right thing in the first place. Need I say more? --KSmrqT 17:28, 31 March 2007 (UTC)
 * Unfortunately this IMA article *did* look like a typical Speedy candidate when it was nominated. It said in effect, 'There is this institute at University of Minnesota, it exists, and it's wonderful'. Its only reference was its own web site, and it had no links to other WP articles except to the Univ. of Minnesota and another institute with a similar name. There was no mention of NSF or SIAM in the version nominated; certainly nothing about the largest NSF grant ever.


 * A mathematics editor described the tagging as vandalism; I think that was unfair. (See User_talk:The_Kinslayer and User_talk:Coredesat/Archive_7). Speedy deletion does not poll the universe to see if the deletion is wise, it just goes and does it. Such deletions are easily reversed if someone complains. The clue that the speedy-tagger missed was 'mathematics of the highest caliber'; I guess that should have registered as the claim for notability. I suggest that the phrase 'idiotic edits' (which sounds like a personal attack) be removed from the above comment. EdJohnston 01:19, 5 April 2007 (UTC)


 * I agree that there is an overreaction here. Vandalism is a serious charge, alleging more than just simple ignorance.  I think it's best to put the best front possible when engaging people on their talk pages as a representative of a WikiProject.  There is a problem of ignorance here though.  It ought to be addressed in a forum for people like the new article and recent changes patrollers.  A lot of these anti-cruft policies like CSD A7 were created for expediency in deleting non-notable people and groups.  One thing failed to be understood by these people is that an institute at a major American research university is already far more notable by its affiliation than an institute in say, Elbonia, just like someone who wins an AMS prize is automatically far more notable than somebody who wins some Elbonian prize. (Or even a full professor at an American research university is probably already far more notable than a winner of said Elbonian prize, even though the latter may look more like an assertion of notability to the ignorant.)


 * where the heck's Elbonia and why we hatin' on the Elbonian folks? Mct mht 12:52, 11 April 2007 (UTC)
 * Elbonia is a very backward country (6th or 7th world) in the cartoon universe. :) JRSpriggs 08:11, 12 April 2007 (UTC)
 * In addition, a fact which may look like an assertion of notability to someone unfamiliar with the mathematical world may in fact be completely irrelevant to what makes a well-written, informative article; hence people used to writing good articles will not include such assertions. Rather than rail about this, it is more productive to let the patrolling community know.  In fact, if you go to these pages, you will find they do have guidelines such as don't speedy an article too quickly after creation, don't be too hasty in prod'ing, etc.  People who do a bad job of speedying articles are not representative of the best of the patrolling communities, and we should not declare all-out war on these people.  Instead, let's work with them and straighten this out pleasantly, by say, discussion at RC_patrol and New_pages_patrol.


 * The real problem in this instance seems to be that the wording of CSD A7 ("An article about a real person, group of people, band, club, company, or web content that does not assert the importance or significance of its subject.") is problematic. In the past, I've had to argue with people that an article stating "This guy is really, really important.  He lives on 3213 Wakefield Street, Ohio." does not assert notability.  It seems that fewer people think of this as such an assertion nowadays, most having come to the realization that said assertion should satisfy a condition such as plausibility.  Actually, it's funny, but at one point I'm certain CSD A7 said it should be "plausibly assert", now it doesn't! (Some of these points have been raised at Wikipedia_talk:Criteria_for_speedy_deletion; I think it's a good idea to join this discussion quickly and make our views known in order to avoid this kind of situation in the future)  --C S (Talk) 11:13, 11 April 2007 (UTC)

Proof sketch for Gödel's first incompleteness theorem
This article was created a few months ago and has sat mostly untouched for two months. As it stands, the article has problems visible from even a cursory reading. I was about to start working on it, but first I want to get a sense of whether this article belongs on WP at all. On the one hand this is a famous and important result that is covered in practically every text on mathematical logic. On the other hand, the proof can be found in practically every text on mathematical logic, so sketching the proof here only duplicates what is already available (and probably better) in other locations. CMummert · talk 22:42, 4 April 2007 (UTC)


 * Would it be much work to fix it up? It's kind of entertaining that the proof is here, and a person can jump into it and at least pretend that they understand it. Adding a list of books that explain the proof more thoroughly would certainly be useful as well. I see on the Talk page that you left some suggestions for the creator of the article, and he hasn't followed up. EdJohnston 22:57, 4 April 2007 (UTC)


 * It looks like a lot of work to me! For one thing, it is dangerously close to WP:OR. The use of three digit codons follows Hofstadter, but a different coding is used and is unsourced. It is also too technical, and the English is not great. At the moment, I find the proof sketch in the main article more helpful. In particular, I think it is better not to emphasise a particular choice of coding. Geometry guy 11:43, 5 April 2007 (UTC)


 * To fix this article would require a great deal of effort. The correct proof has some technical, subtle parts that take care to explain. I think the article is too vague, not too technical. For examples of problems in the current article, look at the paragraphs numbered 1, 6, and 7. CMummert · talk 11:53, 5 April 2007 (UTC)


 * I have seen this discussion only now, indeed English is not my first language, and I'll see what I can do regarding your comments. Anyway if the article is too vague for some and too technical for others, maybe that means that it's just in the corect level for others? no original research was done here, and the 3 digit codon I used is borrowed from the "Goedel Esher Bach" book though I used a different coding exactly because the choice is not important - surely you wouldn't call this WP:OR. There's also a link to the formal proof. Since this theorem atracts a lot of attention from people with only limited mathematical background, it seems to me nice that there's something they can read which would hopefully give them a clue about how the theorem is proven.Dan Gluck 19:51, 5 April 2007 (UTC)
 * Editors here will be uncomfortable if they don't think your article is correct. Do you have the background to fix the subtleties mentioned by CMummert? I suppose you could skip the hard parts if you left a pointer to a book that supplied the missing material. But then the fact that your coding is different might confuse people who had to switch to the book. EdJohnston 01:35, 6 April 2007 (UTC)
 * I thought the problem wasn't that it's incorrect, but that I skipped some technical parts. Since there's a pointer to the full proof on the web, and (I think) I mentioned all the places where I have done so, I don't see a problem. The proof on the web doesn't use a coding, since it does not need to specify one explicitly; the coding I have given is only an example for pedagogical purpose, as in  Hofstadter's "Goedel Esher Bach" book. I would love to use the coding in that book, but I don't have it, and anyway I'm not sure it's permitted by copyright law.

Anyway perhaps I missed it, but all of CMummert's specific comments (in the article's talk page) are related to the form of the article - he suggests deviding it to sections, adding more links etc. I saw no content-related specific comment. Dan Gluck 19:54, 6 April 2007 (UTC) Anyway mathematical logic is not my field of expertise, so if it is yours, and you think the article is incorrect and cannot be repaired, feel free to erase it :( Dan Gluck 20:00, 6 April 2007 (UTC)
 * I added some comments on the article's talk page about the content. It's not the content of this particular article that led me to ask a question here; I am certain that this particular article can be fixed up, and i was about to do so myself. The real question is whether an article entirely devoted to a proof is acceptable. I didn't want to spend a while making changes only to have the article is nominated for deletion. Personally, the more I read this sketch the more I like it. CMummert · talk 20:16, 6 April 2007 (UTC)


 * I'm not a lawyer, but I believe it is legitimate to use Hofstadter's coding as long as his book is clearly cited (and this has to be done anyway to source the approach taken by the article). This would require some work, however, because Hofstadter's coding uses the successor function to define numbers, instead of the (rather nice) base 10 coding of the article. Perhaps a compromise coding would work better (although I still worry a bit about OR - unusually for me, since I think explaining things in new, interesting and engaging ways is exactly what an encyclopedia should do). Anyway, I have GEB and would be happy to contribute the relevant information. Geometry guy 15:58, 9 April 2007 (UTC)
 * Well, the criterion for OR is this WP:OR. I don't think the article satisfies any of the conditions there. If everybody agrees that there are no mistakes, and it is a rephrasing of the proof (hopefully in a more understandable way) with some technicalities overlooked, I don't think it's OR. Dan Gluck 20:52, 12 April 2007 (UTC)

Redundant articles
Hi. I noticed when adding a link somewhere to Bijection that we have individual articles for Bijection, injective function and surjective function, as well as an article called Bijection, injection and surjection. Is this the optimal way to cover these topics? It seems redundant to me. At the least, should bijection be renamed to bijective function, for consistency? -GTBacchus(talk) 18:52, 9 April 2007 (UTC)


 * I think your suggestion to rename is spot on to make the titles at least parallel. I would say that the Bijection, injection and surjection is redundant, except it's actually very nicely written. For now, I've linked the three concepts in that article to their respective pages, and I've added some main templates in each section. I'm interested to hear other opinions on this. - grubber 19:04, 9 April 2007 (UTC)


 * A quick review of the talk pages shows that this has been discussed several times. I made some comments here. CMummert · talk 21:13, 9 April 2007 (UTC)


 * It appears that, while opinions differ on what to do with the joint article vis-a-vis the separate ones, there is at least a broad agreement that bijection should be moved to bijective function, for consistency with injective function and surjective function. I see that this point has been raised at Talk:Bijection, but not much came of the discussion.  I've commented there; perhaps it would be appropriate to do an "official" move request through WP:RM, or we could just agree to do it because it makes sense. -GTBacchus(talk) 21:27, 9 April 2007 (UTC)


 * I don't think that you will be able to do the move yourself, because of edit history at the redirect page (I'd be curious to know if you can, in fact). In any case there are quite a few admins in the math project, so you won't need to go to WP:RM to get this done.  Let's wait a little while to give people a chance to object, though. CMummert · talk 21:42, 9 April 2007 (UTC)


 * (edit conflict) While opinions have differed in the past, perhaps it is time to revisit the idea. I can't imagine anything doing a search and lading on Bijection, injection and surjection.  I agree, though, that the material there is good.  Is there any reason this good stuff can't be moved to each of the three separate articles?  VectorPosse 21:45, 9 April 2007 (UTC)


 * (reply to CMummert) I'm one of the admins who monitors WP:RM and moves pages daily, so that part's not a problem. The only reason I suggested going through RM is that it sets up a somewhat official discussion area, and that's often a good way to bring people with opinions out of the woodwork. It's no problem at all for me to set up a discussion, but at the same time, it's really not a formal necessity. -GTBacchus(talk) 21:48, 9 April 2007 (UTC)


 * Sorry about that - you weren't on my list of admins who contribute to the math project. I think that a formal process isn't warranted if there is clear consensus for the move. Renaming pages isn't supposed to be a big deal. I went through the page history of the destination and it has only ever been a redirect. CMummert · talk 22:06, 9 April 2007 (UTC)
 * No worries; most of my contributions here aren't math related. You're certainly welcome to contact me anytime you need an admin who's mathematics-literate.  For this page, I think we'd be fine moving it after allowing a day or so for anybody to raise an objection.  (It would probably be fine to do it right now, but it can't hurt to sleep on it.)  Because I may not be online tomorrow, I've gone ahead and deleted all but the top redirect at bijective function, so now anyone can carry out the move. -GTBacchus(talk) 22:16, 9 April 2007 (UTC)


 * I think redundancy can be discussed, but I object to changing bijection, which is definitely used, although, perhaps, not as much as bijective map, to bijective function, which is very contrived sounding and not at all common. See my comments here. Arcfrk 04:02, 10 April 2007 (UTC)


 * Huh, now that you say that, I guess I have heard "bijective map" more than "bijective function". Should we be talking about moving the other two articles to injective map and surjective map?  I think all three should be the same, and injection isn't available. -GTBacchus(talk) 04:48, 10 April 2007 (UTC)


 * I don't really care, because there will be redirects, but I think that "map" is a poor choice because it has too many real world meanings. "Function" is much more clear, and the bare word "Bijection" is pure jargon and thus unconfusing. CMummert · talk 00:45, 11 April 2007 (UTC)
 * I see nothing wrong with injection (mathematics) and surjection. Septentrionalis PMAnderson 23:42, 11 April 2007 (UTC)
 * Yeah, that mirrors common usage the best. -GTBacchus(talk) 23:44, 11 April 2007 (UTC)

A question about plagiarism
I recently ran across some copyright violations, and read the policy. I think I understand how to go about cleaning those up. This article is different. It's an outright copy of this newsletter article. Interestingly, there's no copyright notice on the isi web site – or at least, I couldn't find it. Also of interest, this article was contributed by User:Pdagum, who one might reasonably suppose to be a relative (son Paul?) of Prof. Dagum.

Anyway, I'm unsure what to do about this one. It's an "in memoriam" article, quoted verbatim. As such it certainly doesn't sound very encyclopedic. Suggestions? DavidCBryant 14:40, 12 April 2007 (UTC)
 * It sounds like a violation to me- copyright is assumed unless specifically released. So "no copyright notice" isn't good enough, we need a specific statement that it is free content. Staecker 14:48, 12 April 2007 (UTC)


 * I would remove the copyrighted text, leaving a stub, and leave an explanatory message for the article's creator. You could try uw-copyvio along with an explanation of the specific situation. CMummert · talk 14:56, 12 April 2007 (UTC)

Projective space - painting of Dürer
Does someone know the title of the painting by Albrecht Dürer (I guess, not absolutely sure though), which shows a projection a smaller shape onto a bigger canvas? I know, the description is very vague... The image has some geometrical interest, I'd like to put it to projective space (if it is available somewhere). Jakob.scholbach 04:51, 13 April 2007 (UTC)


 * The one most often used is this picture of a lute, from Dürer's Underweysung der Messung mit dem Zirkel und Richtscheyt, Nuremberg, 1525. However, another choice is the reclining nude from the same work. I would expect that an image scanned from the book is free of copyright restrictions, but that's a legal question, and the scans I have seen on the web are not ideal. --KSmrqT 07:41, 13 April 2007 (UTC)

Strange article
Any hope of evolving Mathematical landscape into something encyclopedic or shall it go straight to AfD? --Pjacobi 13:15, 8 April 2007 (UTC)


 * I think the point of the article is something called the "Mathematical landscape conjecture". The author of the article is new here, so maybe asking him/her for a reference is a good first step. CMummert · talk 14:14, 8 April 2007 (UTC)

Just saw this before coming here. Tagged several statements that seemed speculative and unsourced with cn. But if it doesn't improve, I'd likely vote Delete in an AfD. —David Eppstein 16:38, 8 April 2007 (UTC)


 * Strange indeed. Although there is a game some theoretical physicists play with speculating about why certain numbers appear, we can say things about any number. For example, 1728 (which is 123) is a very special number that is not listed. As written, this article is hardly more than numerology. --KSmrqT 21:21, 8 April 2007 (UTC)


 * I posted a message on User talk:Qloop. Let's see what the author has to say. -- Jitse Niesen (talk) 02:59, 9 April 2007 (UTC)


 * This one has been struggling me somewhat. Although it is not very far from my research area, I have not heard of the "mathematical landscape", and the "mathematical landscape conjecture" seems to be not only unsourced and very vague, but also complete nonsense. I tried to source this article myself, but the best I have found so far is one of John Baez's nice pages. I think I know what the article is trying to say (and there is something interesting to say here), but at present it is clearly in the AfD firing line, and is a magnet for a whole load of numerological speculation. For example, the "26 dimensions" section has already attracted the comment that there are also 26 sporadic groups, which is a completely unrelated fact (as far as I know from sources to date). Geometry guy 17:07, 9 April 2007 (UTC)


 * I've never heard it given the explicit name of "mathematical landscape conjecture", but I'll affirm that it is the defacto deeply held tenet held by those theoretical physicists who have wandered off to string theory and beyond. After the fantastic successes of General Relativity, Standard Model and QCD, and the near-misses of Kaluza-Klein theory, supersymmetry and etc., a large part of the theoretical physics community decided that surely, the Theory of Everything would be immediately obvious, if they just knew only a tiny little bit more math, or that, at least, if nothing else, they'd be the second person on the planet to figure it out after Ed Witten did. The near-miss of Monstrous Moonshine only affirmed this belief of the congruence of mathematical landmarks and physical reality: there was 3 or 6 or 12 months or so, in the mid-1980's, where a lot of physicists honestly, truly believed that the monster group really was the group that described the known universe and everything in it. A lot of ink was spilled. Well, here we are 20 years later, and we've got doodly-squat to show for it.  The experimental physicists are all pissed off and are saying that the theoreticians are shirking their work and have abandoned them, and string theory is theological mumbo-jumbo.  But the core belief remains: if we can only find that one magic mathematical expression, it will be everything, and of course it'll be a Lie group, and of course it'll be p-adic, and of course it will have a j-invariant, and a moduli space, and etc. woven into one beautiful whole. Why, in fact, the ratio of the strong/weak/electric/gravitational forces are in the same proportion as the first four Mersenne primes! Didn't know that? Well, there are string theory papers out on ArXiv that explore this numerology... linas 00:04, 10 April 2007 (UTC)


 * To be clear, I'd probably vote to delete this article, since it does seem to be OR. linas 00:30, 10 April 2007 (UTC)


 * Perhaps the article could be saved by changing the focus to the sociological phenomenon described by Linas in his message here. JRSpriggs 07:05, 10 April 2007 (UTC)

Call me a pessimist, but writing about this kind of sociological stuff is difficult and will no doubt be easily prey to OR, lack of citations, and just generally crap. We already have great difficulty with articles about mathematical education (which can be readily sourced and is written quite frequently about). Who's ready and qualified to change this article anyway? --C S (Talk) 11:59, 15 April 2007 (UTC)

Help the Physics Project
The level of activity at the physics project has fallen way off. See Wikipedia talk:WikiProject Physics. If any of you have an interest in physics, but have not been paying attention to it lately, now would be a good time to get involved. JRSpriggs 07:15, 10 April 2007 (UTC)


 * I don't think emphasising high-profile departures is very helpful. Charles Matthews 12:10, 11 April 2007 (UTC)


 * I do not think that killing the messenger is very helpful. JRSpriggs 08:13, 12 April 2007 (UTC)


 * I wasn't intending to. I was commenting on the content of that discussion. Charles Matthews 11:44, 15 April 2007 (UTC)

Requested articles
Many of the requests currently listed on Articles requested for more than two years seem to be mathematical ones. The requests on this page are those which have been unfulfilled for the longest time, and we therefore tend to treat them with a higher priority than those requested only for one year and the other request pages. If any members of this WikiProject have sufficient knowledge and access to sources to write a good stub on one or more of these topics, it would be much appreciated. Thanks – Gurch 10:31, 11 April 2007 (UTC)


 * Those are almost all logic articles; more specifically set theory. Charles Matthews 12:54, 11 April 2007 (UTC)


 * All but one of the titles listed on Articles requested for more than a year are also on the &gt; 2-year list, and are moreover conveniently selected on being maths topics. (The extra title looks more like a physics topic to me.) --Lambiam Talk  15:16, 11 April 2007 (UTC)

Many of them can just be redirects. Take the topmost example from the 2 year list, amenable set, for instance. I was, just the other day, going to write an article on the Gödel, Jensen, Lévy etc. constructions (none of which exists as an obviously named article), but then I soon found that an article already exists on the constructible universe. It is pretty unlikely that we will be able to write a feature length article on amenable sets (without abandoning all pretence to being a general encyclopedia, that is), but we should certainly be able to do a better job of presenting the various independence results&mdash; if only to redirect key concepts to more comprehensive articles. — Kaustuv Chaudhuri 16:50, 11 April 2007 (UTC)


 * These are lazy answers :) If e.g. amenable set should be a redirect then why not make a redirect rather than just chatting about it? But are we talking about admissible or amenable sets? The first is mentioned in constructible universe, the second is not. After an hour searching, I eventually found a reference discussing the distinction between admissible and amenable sets, but did not understand it. Surely we can do better than this :( Geometry guy 22:16, 13 April 2007 (UTC)


 * Cheer up geometry guy! I think I finally understood the difference between amenable sets and admissible sets, so I fixed that. I also redirected the redlinks that seem to have something to do with Determinacy to that article. Geometry guy 14:11, 14 April 2007 (UTC)


 * I'm glad you took some action, but I must say that the end result is not as perspicuous as one wishes. Someone will eventually have to sit down and write some text. I myself must continue to plead laziness for the time being. — Kaustuv Chaudhuri 09:05, 16 April 2007 (UTC)

Category:Mathematicians by religion
At the moment, mathematicians are divided by religion in Category:Mathematicians by religion. Analogous categories for many other professions (including in many scientific fields) have generally been deleted, mainly because religion is largely irrelevant to the given profession. Some people have defended these categories as being used specifically for clergy or devoutly religious, but they are rarely used that way in practice, as the category names leave open the possibility that someone will use these categories to identify anyone who is a mathematician of a given religion.

I was wondering whether these categories had the support of WikiProject Mathematics or not. If people here generally disapprove of these categories, then I can nominate them for deletion. On the other hand, if people really want these categories, then I will leave them alone.

Could other people here please comment? Thank you, Dr. Submillimeter 15:28, 13 April 2007 (UTC)
 * I think this is an example of categorization gone astray. It is not that important what a religion a mathematician had. One's got to pick and choose which are the most relevant categories to add a person too, and I doubt this qualifies. Oleg Alexandrov (talk) 15:38, 13 April 2007 (UTC)
 * Classifying G. H. Hardy as an atheist is probably harmless; it is one of the few notable non-mathematical features of his character. Pity the article doesn't mention cricket. Extending this, even to Erdős or Russell, is very doubtful; and since the whole variety of category is normally a form of nationalism, bad for Wikipedia. Septentrionalis PMAnderson 16:39, 13 April 2007 (UTC)


 * This category certainly doesn't have my support: it looks like a case of overcategorization. However, there is a potential minefield with the subcategories here, and I wish Dr. Submillimeter good luck in finding a path through it. There is probably little problem with the deletion of the atheist, Buddhist and Christian subcategories, and it seems reasonable to keep Category:Pythagoreans. However, the other three subcategories are a bit more tricky...
 * Category:Hindu mathematicians. This is a subcategory of Category:Hindus by occupation. The name of this category suggests the whole tree should be deleted as overcategorization, but in fact most of the occupations listed are relevant to the religion (perhaps the category needs to be renamed). Is mathematics relevant to Hinduism or vice-versa? Are we already well enough served by Category:Ancient Indian mathematicians and Category:Kerala school, or not? I have no idea.
 * Category:Jewish mathematicians. This is even more difficult as there is an argument that being Jewish (especially in the 20th century) does affect ones career and it is therefore much more legitimate to catagorize Jews by occupation. On the other hand, such special treatment for this religion over others has the potential to fuel antisemitism. I would not like to be an arbitrator over this issue!
 * Category:Muslim mathematicians. This is probably also an overcategorization, but it belongs to the Category:Muslim scholars tree, which contains several similar subcategories. Should they all be deleted?
 * Something similar is happening with Category:Women writers. If the overall category is legitimate, is it not legitimate to create intersection subcategories when the category becomes too large? In this project Category:Women mathematicians seems well accepted as is Category:Mathematicians by nationality (or geographical location) and its subcategories. I am not necessarily questioning the legitimacy of this, but where do we draw the line when it comes to closely related issues of ethnicity and religion? Geometry guy 19:04, 13 April 2007 (UTC)


 * Most other occupation by religion categories are about careers that involve religion (missionary work, religious leadership positions, philosophy, etc.). Some of the remaining inappropriate categories for Muslims (Category:Muslim astronomers) and Hindus (Category:Hindu physicians) are already nominated for deletion.  I think some of the other occupation by ethnicity categories have also been deleted; I certainly have not seen any for astronomers.  The categories for Jewish people, however, have been treated a bit differently.  If I nominate Category:Mathematicians by religion for deletion, I will include all of its subcategories except Category:Pythagoreans anyway (just to be unbiased towards all currently-practiced religions, if nothing else).  Dr. Submillimeter 22:30, 13 April 2007 (UTC)


 * Thanks for this answer. The case for some of these subcategories of mathematicians (especially Jewish and Muslim) is a bit delicate, but maybe we can just see what happens at the CFD. Geometry guy 22:50, 13 April 2007 (UTC)

Spiritual beliefs are not necessarily irrelevant to how a mathematician thinks about the status of mathematical objects and of the infinite -- and this can affect his research as well; not the content of his conclusions, perhaps, but the way he interprets them, and what questions he chooses to study. --Trovatore 01:07, 14 April 2007 (UTC)


 * I agree that for some people that religion is an important factor in their careers. However, these categories, as they are currently named, will be used for any mathematician who can be described as belonging to a specific religion, regardless of whether that religion had any influence on the person's career.  For example,  Carl Friedrich Gauss is currently categorized under Category:Christian mathematicians.  However, it is unclear from his article as to whether Christianity was at all influential in his career.  I am certain that I can identify others.


 * This is the general problem with these religion/career categories; despite the intentions of a few editors of wanting to restrict these categories to monks or scholar-priests who worked in mathematics, the categories will be used by other people for any articles that appear to describe mathematicians who were a certain religion, even if the religious beliefs had no influence on their careers or thinking.


 * I will hold off on nominating these categories for deletion a little longer. I would like to see what happens with some currnet discussions, and I would like to see if anyone else comments here.  Dr. Submillimeter 22:43, 16 April 2007 (UTC)

Scholarpedia
I stumbled across Scholarpedia today, its a wikipedia type thing written by academics. The coverage seems patchy coving computational neuroscience, dynamical systems, and computational intelligence. But the articles that exists seem to be more in depth than here. They have quite a destinguised list of authors Milnor, possible Lonez Conway and Mandelbrot. --Salix alba (talk) 16:43, 14 April 2007 (UTC)
 * Their articles are under full copyright, by the way .  Oleg Alexandrov (talk) 16:54, 14 April 2007 (UTC)
 * The calibre of authors they're approaching -- and getting to sign up -- for the dynamical systems articles is impressive. Many of the defining people in the field.  Their advertising model is interesting too -- matching on-topic ads from Amazon and Google to articles by people of the very first rank in the field.  With luck they may generate enough cash-flow that they may actually be able to pay their contributors and their server bills.  Which would be a good thing, because Scholarpedia is generating reliable and available online material written by very impressive people, including coverage even for its version 0.1 of topics we still haven't reached.  (Dynamical systems has always been a somewhat weak area for WP; but even in other areas they're covering subjects that we don't).


 * On the other hand, SP is a very different beastie to Wikipedia. Even if their article's were available GFDL (which they aren't ), the interesting thing is that I think (and I hope and trust) our eds would want to heavily re-write them -- to make them much more integrated, more accessible for typical readers coming in from other WP articles, more geared to answering typical readers questions, and generally just differently pitched somehow.  And it makes me appreciate the real freedom of the anyone-can-edit ethos at WP, because I wouldn't dare touch any of those contributior's articles at SP, even though I'm not sure they all play particularly well together (or are even very well laid out).


 * So while I'm glad that there's now an online article on Fuzzy Logic by Lotfi Zadeh himeself up at SP, an article on K-S entropy by Y. Sinai, etc., they are useful resources, but they are not actually what I would like to see as articles for WP. Jheald 21:05, 14 April 2007 (UTC)


 * Link and learn. Charles Matthews 11:47, 15 April 2007 (UTC)


 * (via edit conflict) One of the things which intrigue me about Scholarpedia, and possibly one of the reasons why they get these top-notch authors, is that they use a traditional peer review system. After the article is written, it's sent to two referees who comment on it. The article is only accepted after the referees give the go-ahead. This is called "initial peer review" in their instructions for reviewers, which say that it allows "authors to list their papers as peer-reviewed in their CVs and resumes". -- Jitse Niesen (talk) 11:52, 15 April 2007 (UTC)


 * I doubt John Milnor needs to pad his CV! Perhaps it's simpler: survey articles that can be updated and are on the Web probably seem sensible to experts. Charles Matthews 19:37, 15 April 2007 (UTC)


 * Of course it wouldn't add to John Milnor's CV. I see now that I chose my quote poorly, giving the impression that I think that Scholarpedia's authors contribute to fill their CV. The whole paragraph is "Scholarpedia enforces the same rigorous anonymous peer review process as most printed journals. This is done primarily to insure the accuracy and quality of information, and to allow authors to list their papers as peer-reviewed in their CVs and resumes."
 * I agree that "survey articles that can be updated and are on the Web probably seem sensible to experts". The question is: why does Scholarpedia attract experts that do not write for Wikipedia (as far as I know)? I suspect that this is partly because Scholarpedia's model (in particular, peer review) is closer to what they are familiar with, while Wikipedia is just a wacky idea. Another factor may be that Scholarpedia actually asks people to write for them. It's harder to decline when you get an email saying "The contributors to the Scholarpedia website have decided that you are the best guy to write this article. Would you please do so?" -- Jitse Niesen (talk) 03:26, 16 April 2007 (UTC)
 * I think "lack of familiarity" is too kind to Wikipedia. We can look at examples of experts who came here, contributed, then gave up and departed. Suppose Milnor devoted two weeks to writing a survey of exotic spheres, complete with expert insights and a well-chosen bibliography. What would happen to it in the next six months here? Some possibilities:
 * A vandal adds an illustration; you know the kind.
 * Another vandal blanks the article and replaces it with homophobic hate.
 * Some crank inserts a pet theory.
 * The inline citation squad splatters it with fact tags, ignoring the bibliography.
 * Any new insights not previously published are removed as violations of WP:NOR.
 * Several irate readers demand that the introduction be made more "accessible".
 * Some of them, who know nothing of the topic, and little of English, take turns rewriting it.
 * A revert war ensues, Milnor is lambasted for WP:OWN, and blocked for WP:3RR.
 * Need I go on? Do I exaggerate? Wikipedia is not kind to expert scholars. As recently mentioned, topics like general relativity theory fare worse than mathematics, and topics of interest to a wider audience fare worse still.
 * I wonder if Wikipedia may be like peer-to-peer music; it fills a temporary void. The quality is not great, and we would prefer better offerings, but nothing else was adequately addressing our needs. Where, freely available on the web, do I read about exotic 7-spheres? If nowhere else, perhaps Wikipedia. --KSmrqT 04:43, 16 April 2007 (UTC)

I agree with the general sentiment, but let's not get carried away. Since you've chosen the exotic sphere article as an example, let's look and consider if any of these things has happened to it. Vandalism? Not really. Crank theories? No. Inline citation problems? No. Accessibility complaints? No. Illiterate ignoramuses revising the page? No (except me perhaps). As for the problems with say, Milnor putting in his new cutting edge research (which has somehow not made it into the literature after all these years...), I doubt he would have any trouble understanding the problems with that; in my experience, mathematicians tend to be good at realizing these things, even those that haven't edited Wikipedia much. Certainly after one of the people who normally edit the page pointed it out to him, I can't imagine Milnor unreasonably insisting on a revert war and being eventually blocked (he seems calm enough in person).

So, in summary, I would imagine if some famous mathematician were to edit Wikipedia (and somehow I can imagine this extremely well even with a lack of imagination...) I would expect little problems. The math portion of Wikipedia functions quite differently than some other parts though. So I would agree there is a kind of expert problem. But I don't know how much of a problem it is to us. And let's not disparate the exotic sphere article too much, eh? It's not bad; it's informative, gives some good references. Certainly one can find a fairly elementary introduction by Milnor (in one of those MAA lecture series from the 70s) that is wonderful, but that's a very high standard to try and match.

In your hypothetical example, I imagine some people would read Milnor's article and it would get promoted to A-class :-). --C S (Talk) 07:16, 16 April 2007 (UTC)


 * I was certainly thrilled by what I read above concerning Scholarpedia, but before making your far-reaching judgements about its advantages and disadvantages viz. Wikipedia, take a look at the product. It touts itself as the new, wonderful model,


 * The approach of Scholarpedia does not compete with, but rather complements that of Wikipedia: instead of covering a broad range of topics, Scholarpedia covers a few narrow fields, but does that exhaustively.


 * Well, the first 3 links in the Differential Equations part of (supposedly, exhaustive) Encyclopedia of Dynamical Systems are Ordinary Differential Equations, Boundary Value Problems and Initial Value Problems. The first is worse than a stub: it's an announcement of the authorship; the second is a bit better, having a potential author and potential table of contents (with typos); the third (finally) is an article, but it jumps straight off into numerical solutions, with no explanation or motivation. I don't want to imply that the authors or curators lack qualifications, but let's put it this way: there is no Milnor (or anyone of comparable stature) among them. For the time being, I'd regard the site as a half-baked imatation of Wikipedia, with ambition, but uncertain future.


 * I have to differ with you about the authors. I have some history in nonlinear systems, and I was blown away by the calibre of people they have signed up - these really are the first team, people who defined and interpreted and are leading the subject.  And Scholarpedia haven't signed up just a few: they've signed up dozens and dozens and dozens of them, and matched them to the most appropriate topics.


 * Kicking SP because most of the articles are only stubs seems premature: most of the topics in dynamical systems appear only to have been commissioned last month, or to be still going through the commissioning process. But most of the authors who have been commissioned are aiming to deliver in the next few weeks.  If you look again in June/July it may be quite a transformation.  (In contrast, how may WP articles now stubs will reach Good Article status in that time ?)


 * Where I think you might have more of a point is perspective. WP can't call on this calibre (or quantity) of talent, would take years to evolve to even mark out with stubs a survey this detailed, constantly faces articles losing their shape as more details are added that don't quite fit the original plan, and frankly, well, it's a rare article in a technical subject here that couldn't be substantially improved.  But what WP does have, like a sandstorm smoothing a stone, is continual pressure to make things more accessible, and to make different articles fit together better, and to make articles appropriate for the pages linking into them.  That matching of individual articles to a multiplicity of entry perspectives, and making them play together at a category-wide level, is what I would currently see as the most significant weakness in SP (version 0.1), and its authored approach.  Lots and lots of old-fashioned editorial smoothing required - casting and re-casting of leads and intros to give more perspective and accessibility.  Jheald 12:28, 16 April 2007 (UTC).


 * The larger point that was touched upon by User:Jitse Niesen is more interesting: why would the best of the experts write something for a web encyclopedia? I think that this question is incorrectly posed: given the spread of electronic publishing, TeX, arXiv and all that, it's remarkable how little of this is occurring so far! Now, I don't believe for a second that a world expert in a certain field needs any of this "peer review" hullabaloo to somehow validate their survey papers. In fact, it seems that there is a certain aura of purity about "peer reviewed publications" in public sentiment outside of academia, bordering on worship. Very few serious mathematicians rely on peer review to justify their work (there are some notable exceptions). Contributing surveys by invitation, on the other hand, has a long and distinguished tradition. I have little doubt that it can be extremely successfully transferred into electronic medium. If Wikipedia (or mathematics project) wanted to toy with the idea, it's quite feasible to invite the very same people to make a contribution, and quite possibly, many will agree.


 * On the other hand, I think that User:KSmrq's comments are very relevant. In my short (about one month) time on English Wikipedia I've seen pretty much the whole spectrum of his bullet points, some of which unfolded (or were mentioned) in this very discussion page! By the way, note that he was talking about perceived problems with original research, lack of clarity, undocumented statements, and so on, and being an expert in the field by no means shields someone for those kinds of issues! If anything,
 * a famous author may attract unwanted attention from cranks, vandals, and possibly, scientific foes;
 * an article on a popular topic will get scrutiny from legions of people who demand it to be comprehensible to everyone ("if it's in the news, why don't I get it? I graduated from high school and consider myself brilliant"), and I can easily imagine how having a high profile author alone can produce this effect.
 * Arcfrk 09:01, 16 April 2007 (UTC)


 * Warts and all, Wikipedia fills a need. I would not want that point to be lost. But to respond to C S: suppose I were to invite Marcel Berger to write on geometry, as he has done with great appeal in his books; the article history forebodes ill, and I think his time would be better spent elsewhere. --KSmrqT 10:18, 16 April 2007 (UTC)


 * This has generated an interesting discussion! I tend to agree with C S that things are not so bad in the mathematical oasis of WP. Vandalism is not intensive enough to be a problem: just hit that undo button and it is gone. And I've noticed that for every anonymous IP user who blanks a page or adds something scatological, there will be another who fixes a typo while browsing.
 * As for credit and peer review, one of the reasons that I contribute anonymously is that I don't want credit for any of my contributions. Also, if I were to be a named or invited author in article, I would want to maintain control over the article. So I'm not sure the wikipedia model is compatible with soliciting surveys. I like the WP model, but I think you need to have the right attitude/character to enjoy contributing here as an expert. I was recently browsing through the talk page at Lorentz group and I can see why Chris Hillman left. You have to be very flexible to contribute to WP. It also helps to be not just civil, but to make an extra effort to be friendly with other editors, especially when you disagree with them.
 * One area where I think WP lets mathematics down is in its desire to be a "general encyclopedia" in the Encyclopaedia Brittanica model. However, WP is so much larger than EB that it is really a whole new thing: a union of specialized and general encyclopedias. The mathematics coverage here is already becoming comparable with the Springer Encyclopedia. Unfortunately the generalist model has resulted in policies and guidelines which are not relevant or appropriate for mathematics. The whole concept of a featured article is totally unsuited to most mathematics articles, and it is not surprising to me that the few FAs we have are mostly on elementary subjects or are biographies. Having recently witnessed an FAC, and the inline citation crowd adding tags to any sentence which is not utterly bland, I have no desire for any article to which I have contributed significantly to become a featured article.


 * Perhaps we need our to define our own standard for the ultimate article, the FMA perhaps? Geometry guy 11:26, 16 April 2007 (UTC)


 * In general I don't think wikipedia is the right place for most academics. However I do think that they can help in someways. I've had good experience with emailing various academics and other experts on a number of topics, mainly for clarification on specific points, they have all been happy to help. Perhaps this a model, with us as a buffer between the academics ans wikipedia, work works best. --Salix alba (talk) 12:49, 16 April 2007 (UTC)

One point that I think is particularly relevant to mathematics here. The aim is not so much to have 'great articles' but a great piece of hypertext. So that (for example by the end of this decade) it would not be an empty boast that cutting-edge research mathematics can be referred back to definitions by an unbroken chain of blue links. Deligne said the proof of the Ramanujan conjecture would write out as 1000 pages of graduate level mathematics. We have a new model for making that less scary. Charles Matthews 15:51, 16 April 2007 (UTC)


 * Absolutely. By a quirk of fate I ended up doing interdisciplinary research, and ofter need to study small parts of new fields in a hurry - and fan out to the basic concepts needed, but no wider than that. Wikipedia with its blue word links is priceless, warts and all. Scholarpedia has more classical survey articles. There is a place for both. All these wikis are social experiments, and only time will show which equilibrium each one will hit. One aspect that Wikipedia is missing - when a professional mathematician spends a significant time on something he/she needs to take credit for it. Time is not free and even if the CV is thick the yearly increments do count. I write it off to paying back, public service and education, Wikipedia will make one a better writer over time. But from my short time here, many of us seem to be in the stage of their careers where they should seriously consider spending more time on writing journal papers than on Wikipedia in order to get that tenure or promotion. [added] A credit taking mechanism like Scholarpedia has would help and may attract more expert contributors. Jmath666 01:18, 17 April 2007 (UTC)

What do you mean by a credit-taking mechanism? Scholarpedia offers you an option: although you need to use your real name to register and you need to log in to edit, you can nonetheless edit either anonymously (but while logged in) or with your name on your edits in the edit history. Michael Hardy 01:37, 17 April 2007 (UTC)


 * Something to show on the CV for the time and effort spent that the tenure committee would consider. I have seen more than one case when someone was giving time and effort to noble things but the papers were not there so the ax fell. Authorship is more important than edits of course. I do not need that personally, but my guess is many of the best contributors here might. That may be a factor in favor of a model like Scholarpedia. Time will show. Jmath666 01:56, 17 April 2007 (UTC)

I for one don't think any tenure committee is going to be impressed by time and effort spent on Wikipedia or Wikipedia-like systems. Giving some way to give people credit on their vitae for those efforts is just going to lead to the committee asking why they aren't spending more effort on research. I think, as Jmath666 wrote earlier, that effort spent on WP is not only valuable community service but also that it is valuable practice in writing readably. But I don't see any kind of credit system as much of a draw, and I worry that such a system would lead junior into spending more time than they should in WP and hurting their careers. The way it is now, there's a clearer picture of what you're getting: a chance to help others understand the world better, but not really a step on the career ladder. —David Eppstein 02:17, 17 April 2007 (UTC)


 * There is an item for "service to the community" on our merit forms but it will not save the day. Maybe the whole business how we collaborate and publish will evolve towards some form of wiki. I see a pressure here towards original research all the time. Maybe the right way to parcel credit and control is would be the ingredient to spark the revolution. My group is already using CVS for pretty much everything, and I might set up a wiki instead for the next bunch; then it is only a step from a private server to online publishing. Jmath666 02:45, 17 April 2007 (UTC)


 * Tenure committees have different priorities at different schools, and in different departments within schools. Some give little weight to proven excellence in teaching; some give little weight to books authored. If the focus is on research, only peer-reviewed research publications count for much, and even then volume may not matter as much as perceived significance. Some journals garner more respect than others, and astute faculty judge accordingly. Like being kind to children and pets, contributing to Wikipedia can be a good thing; but don't expect it to help win academic advancement. --KSmrqT 04:32, 17 April 2007 (UTC)
 * This brings to mind the ideal of a gift economy which is frequently applied to open source projects and online communities. Both academia and wikipedia opperate (information) gift economies, reputation in each is largely measured by the amount of information produced. In academia this is measured by the CV in wikipedia its the list of articles i have contributed to on your user page and things like edit count. As seperate economies there is a poor rate of exchange between the two, academic credentials matter little to WP and edit counts matter little to academia.
 * I'm definitly with David on my motivation for editing WP in that my motivation is to help people understand mathematics and wikpedia (despight all its problems) to me seems the most effective way of acheiving this. --Salix alba (talk) 07:41, 17 April 2007 (UTC)

Mean information
The article Mean information has been nominated for deletion. Comment as you see fit. Anville 19:49, 15 April 2007 (UTC)

Variable Shape Geometry
I guess this does not qualify as encyclopedic content per WP:NOR. Comments? Oleg Alexandrov (talk) 22:22, 15 April 2007 (UTC)


 * I received the pamphlet I referenced not too long ago in school. The front cover said Variable Shape Geometry by Val Bess and that's all. I asked one math professor about it and he said that it (the geometry) is still being developed and the pamphlet is what the "creator" has 'discovered' so far. I really don't know if it was published or not but it seemed worthy enough for an article so I made one. Burned thru  22:33, 15 April 2007 (UTC)


 * What makes it worthy? I'd like to know.  There's not enough in the article for me to really understand what this geometry is about.  It seems kind of strange to include something you found in a pamphlet here in Wikipedia before many other, more standard notions of geometry are covered. Rybu 22:57, 15 April 2007 (UTC)


 * Before what others? Burned  thru  23:00, 15 April 2007 (UTC)


 * If more information is required, I'd be happy to add some, it's just that I have no one to ask at this moment and am pretty busy. Burned  thru  23:04, 15 April 2007 (UTC)


 * Non-commutative geometry is a big branch of modern mathematics. Variable shape geometry on the other hand, I doubt many people have heard of.  My point is, just because somebody produces a pamphlet and includes the word geometry in the title, does that mean it should be linked to the geometry Wikipedia page?  I don't understand Wikipedia's mission well enough to answer the question.  I don't think any reasonable encyclopedia would include variable shape geometry in a listing of geometries, since it's obscure. But maybe Wikipedia is happy to promote anything, regardless of how obscure it is.  Take your article on variable shape geometry for example.  Does isometry group make sense for this geometry, and if so, what is the isometry group of the geometry?  That would help me understand what this geometry is about. As it is, the article tells me very little about anything, other than some guys name. Rybu 23:14, 15 April 2007 (UTC)


 * This should go straight to AfD. It is not a question of priority: deletion is clearly justified under WP:NOR and the article seems to me to be palpable nonsense. For a start, semicircles (oblate or otherwise) are not parabolas (and how are they defined anyway?). Then we have "Later on, triangles can be proven to be squares because of the Congruency Postulate, which is very complicated." The congruency postulate is, of course, not defined. The word "indiminent" does not exist, and one of the diagrams has a "parabola" drawn between lines which are not parallel, contrary to the unexplained hypothesis. Lets not waste any more time on this! Geometry guy 01:45, 16 April 2007 (UTC)
 * I agree. I nominated it for deletion, see Articles for deletion/Variable-shape geometry. Oleg Alexandrov (talk) 01:56, 16 April 2007 (UTC)

Euler on infinite series
Someone has proposed that Euler on infinite series be transwikied to Wikiquote. I think instead the article should be expanded to be more than just a quote. Perhaps some of this pages public can contribute. Michael Hardy 23:58, 15 April 2007 (UTC)


 * Can't this be merged into Divergent series? --Lambiam Talk  06:50, 16 April 2007 (UTC)

Category:Complex systems
Category:Complex systems is on CfD []. --Salix alba (talk) 15:01, 16 April 2007 (UTC)


 * Since I nominated the category for deletion, the category has been cleaned up, and a solid justification has been given for its use. It no longer looks like a category to collect anything that is just "complex" and a "system" according to the casual reader, and I would certainly not attempt to delete this category if it remained this way.  However, it would be good to avoid having people put things like Category:Role-playing game systems back into the category because "they are called systems, and they look complex" (which describes why some articles and subcategories were in this category before clean-up).  What are people's thoughts on renaming this as Category:Complex systems studies so that the category stays focused on the field of complex systems and does not become a category for anything named a "system"?  Dr. Submillimeter 15:16, 18 April 2007 (UTC)


 * This may be a way forward, although I think something like Category:Complex systems science or Category:Complex systems (science) would be better. Geometry guy 17:16, 18 April 2007 (UTC)

I think the name as it is is fine, it is what people in the area use and it is not ambiguous with any other meaning for the same phrase. If the problem is people not knowing the technical meaning of the phrase and guessing that things belong when they don't, wouldn't it suffice to add appropriate text to the category page? —David Eppstein 17:22, 18 April 2007 (UTC)


 * The approach that some people take to categorization is similar to jamming a square peg through a round hole. If the category name vaguely seems to describe an article, then some people will stuff that article in the category even if it is inappropriate.  Explanatory text may limit the problem, but some people will ignore the text anyway.  Renaming the category with a less ambiguous name is a more robust solution.  Dr. Submillimeter 17:46, 18 April 2007 (UTC)


 * If it must be moved Category:Complex systems (science) sounds good to me, Complex systems studies does not work for me as no one uses that term. BTW the category seems only weekly linked in the mathematical should it be included in Category:Chaos theory or Category:Non-linear systems. --Salix alba (talk) 18:55, 18 April 2007 (UTC)

Warel?
Please consider this list of edits which hits all of WAREL's pet topics. I didn;t see any that looked harmful; but I can't tell where the ja: links are going. Would people keep an eye on this? Septentrionalis PMAnderson 23:20, 16 April 2007 (UTC)
 * Of the first three ja: links I saw, one was valid, one was a change to the literal translation of the English title, which was a redirect at ja:, and one was a translation that didn't seem to have an article at ja:. In other words, the links seem to be made without any reference to existence of Japanese articles. JPD (talk) 10:58, 17 April 2007 (UTC)

12988816 (number) is nominated for deletion
12988816 (number) is nominated for deletion. The AFD discussion is here. CMummert · talk 15:55, 17 April 2007 (UTC)

Arabic/Islamic mathematics
Nationalists are trying to move this article to a silly title again. —Ruud 23:24, 17 April 2007 (UTC)


 * The article used to have the name "Islamic mathematics" for most of its existence. This is a rather common name for the subject (I get 19300 Google hits for "Islamic mathematics" versus 16200 for "Arabic mathematics"), and hardly a silly title. You yourself moved History of mathematics in Islamic culture to Islamic mathematics on April 28, 2006. Then, on March 10 you moved Islamic mathematics to Arabic mathematics. Why? Was there any discussion on this, and was a consensus reached? When User:Jahangard moved it back, complaining that the move was unilateral, you immediately reverted this, as far as I see without discussion why "Islamic mathematics" is a worse title than "Arabic mathematics". Was that wise? It looks like your March 10 unilateral move triggered the proposal to rename the article from "Arabic Mathematics" to "Mathematics in Medieval Muslim World". On the talk page, the supporters of the (ill-advised) proposed renaming appear to favour "Islamic Mathematics" over that long-winded title. My suggestion to you is to undo your last move, and try to obtain consensus that "Islamic mathematics" is to be preferred over "Mathematics in Medieval Muslim World". If such consensus has been reached, emotions have come down, and you still think "Arabic mathematics" is the best title, open a discussion about it. --Lambiam Talk  06:49, 18 April 2007 (UTC)
 * I think both "Arabic mathematics" and "Islamic mathematics" are fine (I personally even prefer the term Islamic mathematics, my move last month was motivated by more pragmatic reasons and, given the lack of dedicated editors of this article, more a case of WP:BOLD than a "unilateral action".) The proposed move is to "Mathematics in Medieval Muslim World", however. —Ruud 08:21, 18 April 2007 (UTC)


 * I was probably too frustrated yesterday and should have approached this in a more calmly and with more tact. I would appreciatie it, if some looked at it with a fresh look. —Ruud 09:09, 18 April 2007 (UTC)

Question on Relation notation
I've always seen a relation shown as something like x~y on some set S. And that this relation is an "equivalence relation" iff it is reflexive, symmetric and transitive. So that for instance you could have x~y if x<y on the integer set Z. So this relation would not be an "equivalence relation" since x is not less than x so x~x is not true and so ~ is not reflexive. So ~ is a relation but not necessarily an "equivalence relation". But the article on the following articles seem contradictory to this saying that ~ alone denotes an equivalence relation:


 * Equivalence relation: An equivalence relation between a and b is often denoted as "a ~ b" or "a ≡ b".


 * Tilde: In mathematics, the tilde, sometimes pronounced "twiddle," is often used to denote an equivalence relation between two objects. Thus "x ~ y" means "x is equivalent to y". (Note that this is quite different from stating that x equals y.)

Is there something I am missing here?--Jersey Devil 01:23, 20 April 2007 (UTC)
 * It's fairly rare to use the tilde for a relation that's not an equivalence relation, but you won't go to jail for it. Not sure I really understand your question, though. --Trovatore 02:08, 20 April 2007 (UTC)


 * You may not go to jail for it, but the physical world is not what one ought truly be concerned with. Rather one's soul.  There is a very special section of Hades reserved for those that use a tilde for a non-equivalence relation.  And it's the section that is not filled with interesting people.  --C S (Talk) 21:52, 21 April 2007 (UTC)


 * I agree, x~y almost always means an equivalence relation. But suppose a broad survey finds that one third of the uses are for relations not satisfying equivalence (a generous estimate); the two "often" statements quoted are no less correct. (Note the word is "often", not "always".) As a practical matter, we are unlikely to be confused by seeing a "~" used with uncertain meaning, because authors habitually tell us the meaning they intend. --KSmrqT 03:44, 20 April 2007 (UTC)


 * When I've seen ~ used, there really is an idea of equivalence involved -- not just a generic relationship. Two states in a communicating class in a Markov chain are "equivalent"... or whenever you have equivalence classes, for example, in the field of fractions (which, even though 1/2 and 2/4 are fractions of different integers, they are equivalent). In your example, x<y, this also fails because of symmetry (that if x~y then y~x, that is, x<y and y<x, which is clearly not true). However, even though < is not an equivalence relation, the relation &le; over the real line is. I don't see the contradiction in the examples you give. - grubber 03:19, 20 April 2007 (UTC)


 * No, "&le;" is not an equivalence relation, since it also obviously fails symmetry. Paul August &#9742; 03:57, 20 April 2007 (UTC)


 * Grr, of course. My bad :) - grubber 17:17, 20 April 2007 (UTC)
 * Don't worry we all have mental lapses - mine come more and more lately. The distinguishing property between "<" and "&le;" that you were probably thinking of is antisymmetry, which "<" fails, but "&le;" satisfies, thus making the latter, but not the former a partial order. In fact, of course, "&le;" is the archetypical partial order, hence the symbol "&le;" is often used to represent any partial order, much like "~" is used to represent any equivalence relation. So there is that connection as well. Paul August &#9742; 15:39, 21 April 2007 (UTC)

The ~ may be very often used to express equivalence relations, as the article says, but this does not contradict the fact that it is sometimes used for any relation, equivalence or otherwise. The confusion is simply that mathematical notation varies in different places and contexts. Perhaps the Tilde article could be updated to reflect the more general use of the symbol, but apart from that I don't see any issue at all. JPD (talk) 11:24, 20 April 2007 (UTC)

Strange edit at Relativity of simultaneity

 * "Recent research in the history of mathematics has shed light on the mathematical suppositions Einstein brought to the formulation of the relativity of simultaneity ..." etc. -- it doesn't look completely incoherent and gives sort of some sources, nevertheless I'm sceptical. Comments? --Pjacobi 18:39, 22 April 2007 (UTC)


 * Someone had made word-for-word the same edit in general relativity, which I promptly removed. Its applicability to GR was even more dubious than its presence in the relativity of simultaneity.  Silly rabbit 14:48, 23 April 2007 (UTC)


 * I've now seen it at History of special relativity too -- that just didn't look right. So I'll take it out and direct further discussion to Talk:History of special relativity. --Pjacobi 16:21, 23 April 2007 (UTC)

An invitation to categorize uncategorized math stubs
Hello. The categorization taskforce is trying to find WikiProjects interested in using the bot of Alai to identify mathematics stub articles which do not currently have a category (besides the stub category of course). If the project is interested, we could create something like Category:Uncategorized mathematics stubs which could then be categorized by people knowledgeable in the subject, thus reducing the risk of improper categorization. Please let us know on the taskforce's talk page if you're interested. Cheers, Pascal.Tesson 00:23, 23 April 2007 (UTC)

Category:Complex systems nominated for renaming
I have proposed renaming Category:Complex systems as Category:Complex systems (science). Here is my justification for the rename:
 * I previously nominated this for deletion on 13 April 2007, mainly because it was being used to categorize anything that could be described as "complex" and a "system" by the average Wikipedia user (such as "role-playing game systems"). This actually seemed to categorize things by name rather than categorize things that were related to each other.  Following the nomination, several people familiar with the scientific field of complex systems explained that the field deserved a category and cleaned out the category.  However, the category is still at risk for being used to list anything that could be described as "complex" and a "system", and it would be good to have the category focus speficially on the field of complex systems itself rather than gathering together everything that could be called a system (like the deleted Category:Systems, which was deleted following a 12 April 2007 discussion; see User:Jpbowen/Back up - Category Systems).  After a discussion at Wikipedia Talk:WikiProject Mathematics, a couple of people suggested renaming this as Complex systems (science), which I now recommend as the new name for this category. Dr. Submillimeter 18:06, 19 April 2007 (UTC)


 * Although I don't have a strong view on the rename, I originally suggested Complex systems (science) as an alternative to Complex systems studies. A better suggestion has now been made: Complex systems theory. I think this choice avoids most of the drawbacks of the rename (it seems to me to be as harmless as the distinction between relativity and relativity theory), while retaining the benefits. Comments can be added here. Geometry guy 11:11, 25 April 2007 (UTC)

Karel de Leeuw
I came across this article on a mathematician while working on categories. (Don't ask how.) The article says little about why this person was notable, although I suspect that part of the reason why the person was notable was because he was a murder victim. I may nominate it for deletion, although first I have asked the article creator to improve the article.

Is anyone here familiar with de Leeuw's research? Was he notable in mathematics? From this standpoint of this person as a mathematician, should the article be kept, or should it be deleted? Dr. Submillimeter 20:59, 19 April 2007 (UTC)


 * Tenured mathematicians at Stanford (even back then) are very notable. I also spy a publication in the Annals and several in lesser (but still prestigious) journals.   --C S (Talk) 21:39, 19 April 2007 (UTC)


 * I do not concur that "tenured mathematicians at Stanford ... are very notable." However, if C S is correct is correct as to publications, that should be adequate for notability.  I'm just not sure.  &mdash; Arthur Rubin |  (talk) 21:55, 19 April 2007 (UTC)


 * Please note that WP mathematician notability guidelines appear to be far more strict than those in other areas. This is particularly the case for anyone who's made a name for themselves on the net, e.g. various luminaries in the Linux/open source world, or in the computer gaming industry. The list of accomplishments are often of the form "so-n-so wrote this-n-such piece of software", yet the software is not particularly deep, original or complex. Point-for-point, they'd be completely outclassed by thousands of utterly anonymous engineers. Or, to compare to academia, the accomplishments seem at best comparable to those of junior math professors at state universities, the kind of which get AfD'ed. Yet, internet fame seems to be a deciding factor. It seems to be all so very unfair... linas 00:05, 20 April 2007 (UTC)


 * Dr. Submillimeter is right: many more people are familiar with the story of his murder (linked in the article) than with his research. It was a very high profile case, in my opinion, deserving coverage in Wikipedia. Arcfrk 00:10, 20 April 2007 (UTC)
 * I think the murder does contribute to his notability (hard way to get it!) but he would deserve an article without it. --Trovatore 00:14, 20 April 2007 (UTC)


 * Questions of notability would be easier to address if the name were spelled properly: "Karel de Leeuw" should be "Karel deLeeuw", without the space. Then a web search would find this description of his life and contributions. --KSmrqT 03:55, 20 April 2007 (UTC)
 * He seems to have published under both spellings, so I don't know how you're picking which is the "proper" one. --Trovatore 05:48, 20 April 2007 (UTC)
 * Two factors: number of search hits (unreliable), and Stanford department remembrance. The article should mention both spellings, and should definitely draw on the information I linked above. I don't care enough to move it, but my leaning is it should be moved. --KSmrqT 05:57, 20 April 2007 (UTC)
 * Better than search hits is MathSciNet. Have you looked there? It was a long time ago, but when I looked there seemed to be more listings with the space than without. --Trovatore 06:00, 20 April 2007 (UTC)

By the way, does anyone know if he's any relation to Jan de Leeuw, who definitely should not have a redlink? --Trovatore 06:02, 20 April 2007 (UTC)

Could someone please expand the article to indicate how the person was notable? Dr. Submillimeter 11:12, 20 April 2007 (UTC)
 * Just to conclude this for the archives, deLeeuw is notable for two reasons: his research and his murder. I suspect that after a few decades the murder will evolve into urban legend, since it seems to express two emotions common to graduate students. He was well liked by most students, and his murderer was frustrated by 19 years of failure. As to his research, the article's list of selected publications occupies as much space as the text, which surely is enough. The biography is brief, but a Stanford faculty member whose advisor was Emil Artin at Princeton, who was a Fulbright Fellow at Cambridge, who spent time at IAS, and who was a coauthor with Walter Rudin — clearly this man had something to contribute. As for the spelling, both Stanford and the mathematics genealogy project omit the space, while his publications do not.
 * Violence and murder are too common a tragedy in the U.S., and the victims are too often the generous and gentle souls. Perhaps Moez Alimohamed deserves a page as well, especially as Penn established an award in his name. Sadly, he has only this major publication left to represent his contribution to mathematics. Contrast the worth of these two individuals with the 65 names on the list of Formula One drivers who never qualified for a race. From the pages of Spider-Man, who himself saw his beloved uncle murdered, "'Nuff said". --KSmrqT 21:21, 25 April 2007 (UTC)

I was bold, and moved it to deLeeuw, which most of the sources seem to use. Does anybody mind? If so, let's do WP:RM. Septentrionalis PMAnderson 05:29, 26 April 2007 (UTC)

Systolic geometry -- legitimate topic or link spam?
Please check out Systolic geometry. I noticed it when twice put a link to it into Hyperbolic geometry. Both times I removed the link after looking at the article and not seeing any relevance to hyperbolic geometry. Apparently, Katzmik is Mikhail G. Katz, author of a book linked to by the article. He did not create the article, but he has done most of the edits to it. Do you think that this is a legitimate topic or is it just link spam? JRSpriggs 07:54, 25 April 2007 (UTC)


 * i think there is no reason to assume it's spam. requesting for clarifications from seems to be a sensible thing to do. there's a section Systolic geometry that hints at some kinda relationship with hyperbolic geometry. perhaps  would be willing to expand it. Mct mht 09:29, 25 April 2007 (UTC)


 * This is an entirely legitimate topic, as far as I can tell. It's probably worth drawing the author's attention to WP:NOR. I think the AMS monographs are peer reviewed, but it might be best if the book is only used as a secondary source for the article.


 * As for the relevance to hyperbolic geometry, for one thing, the length of the systole will be a hyperbolic invariant. Geometry guy 12:08, 25 April 2007 (UTC)


 * If the AMS is willing to publish something, it ought to count as a reliable source for Wikipedia. The article is actually pretty reasonable, and lists numerous sources (there are some minor OR issues that would come up in an FA review, but this is a common problem). I agree with R.e.b.'s comment below that when experts want to improve articles in areas they are familiar with, and they write quality articles like this one, there is no reason to discourage them. CMummert · talk 18:39, 25 April 2007 (UTC)


 * Systolic geometry is a legitimate topic, it is indeed related to hyperbolic geometry, and is an expert on it, who should be thanked rather than insulted for his efforts. R.e.b. 16:42, 25 April 2007 (UTC)

We have assume good faith for a reason. This case illustrates why. Charles Matthews 16:50, 25 April 2007 (UTC)

It looks perfectly legitimate to me. The basic definition is stated in such a way that it should be instantly comprehensible to everyone (except non-mathematicians, maybe) and that's more than can be said for some math articles. This seems like a good reason why an author should not be forbidden to put in an external link to his own book. Michael Hardy 20:53, 25 April 2007 (UTC)

Capitalization question
I was reading ordered field and noticed that the "Archimedean property" was capitalized, whereas in my Lang Algebra book, it is not. So, I noticed: I had always learned that when the property is modified (like with -ian), that it loses its "properness" and should be written lowercase -- and I have never seen abelian capitalized. But, on the other hand, I have always seen Gaussian written capitalized. I would prefer to never capitalize such adjectives (but that could look very strange with Gauss, Hamiltonian, Hilbert space, etc) Any ideas or has this been discussed before? - grubber 16:02, 18 April 2007 (UTC)
 * abelian group: abelian never capitalized
 * Archimedean property: Archimedean always capitalized
 * Euclidean domain: Euclidean capitalized
 * Gaussian distribution: Gaussian capitalized


 * Capitalization of proper names in mathematical terms is sort of inconsistent. You say, for example, that "Archimedean property" is always capitalized.  Nonetheless, number theorists routinely refer to non-archimedean valuations (actually, that article uses both capitalizations).  I've seen both "Gaussian" and "gaussian", and in rather old books, also "Abelian" rather than "abelian".  I can't remember who in particular I heard this from, but it's said that a lower-cased name in this context is a sort of honor, that the term has become so standard that it's not a matter of professional courtesy anymore, but simply a fact of life that, e.g., commutative groups are "abelian".  In light of this, it's not really something that Wikipedia should regulate, in my opinion. Ryan Reich 16:32, 18 April 2007 (UTC)


 * This is to confirm what Ryan Reich said. I have two old group theory books (Herstein, and Hall). The first author refers to abelian groups, and the second uses Abelian groups. I don't think there's a standard in the literature, so Wikipedia ought not impose one. Oh -- I'm pretty sure I've seen both "Gaussian noise" and "gaussian noise" in physics/electrical engineering contexts. DavidCBryant 16:44, 18 April 2007 (UTC)


 * I would disagree here. As far as I know, not capitalizing "abelian" far outweighs capitalizing, so there is a pretty widespread convention; as indicated by Ryan's comments and your example, some old books may use capitalization.  In fact, as pointed out by Dave Rusin in a nice sci.math post it is the only name in the MSC that is not capitalized; I took a quick look and certainly abelian is not capitalized except when beginning the name of a category.


 * I'm not sure about other terms. If indeed it is a matter of preference, then Wikipedia ought not to impose as David said, but it would be silly to take this ambivalence to the extent of overturning a nearly unanimous convention.  --C S (Talk) 17:01, 18 April 2007 (UTC)


 * (edit conflict) Once upon a time I was told such a rule too. But it's a rule many people don't use!  As for "archimedean", I've  seen it both ways.  "Hilbertian field" can be either way too.   It's interesting that indeed Gaussian (even in other cases as in "Gaussian integers") and Euclidean (in "Euclidean domain") is always capitalized.  The usual way to find out if something is capitalized is to put everything in lowercase, submit it to a journal, and see if someone busts a gasket.  I was told an interesting philosophy behind this lowercasing: that putting a name in lowercase is bestowing even greater honor upon the person; unfortunately an editor did not see it that way.  So it goes.


 * Interestingly enough, I found this entertaining blog post which has a comment which led me to Abelian group, where the note on typography states, "Among mathematical adjectives derived from the proper name of a mathematician, the word "abelian" is rare in being expressed with a lowercase a, rather than A (cf. Riemannian). Contrary to what one might expect, naming a concept in this way is considered one of the highest honours in mathematics for the namesake."  Huh, no citation needed tag? ^.^ --C S (Talk) 16:52, 18 April 2007 (UTC)


 * (edit conflict) I was once told that it is the great honor of Abel that his was one of the few non-capitalized math names turned adjective.   This is discussed outside of WP, but I don't think it is in WP:MSM, but maybe should be. Smmurphy(Talk) 16:56, 18 April 2007 (UTC)


 * I'm not suggesting that WP should be the body that determines the standard. However, it would be nice to have a standard that we use consistently in all mathematics articles on WP (and alternate usage notes where appropriate); for example, a few lines in WikiProject Mathematics/Conventions to discuss which one is "preferred". I hope that it would not degenerate into a "British vs. American spelling"-esque debate, although you never know :) - grubber 17:04, 18 April 2007 (UTC)


 * I agree with Ryan Reich that this is not something that we should regulate. On the other hand, I would not like to see the use of unmodified proper names without capitals such as hilbert space. I quite like the convention to decapitalize words derived from proper names, but not the proper name itself (even when used in a noun phrase like Hilbert space). Hence: Hermite polynomial, hermitian form, Gauss map, gaussian distribution, Abel-Jacobi map, abelian group, Klein geometry, kleinian group. (This convention is systematically used in French.) Geometry guy 17:11, 18 April 2007 (UTC)


 * PS. More importantly, though, Abel-Jacobi map is still a redlink! Can someone write a stub, or find a suitable redirect?


 * I'm writing an article right now. If anyone else is interested, wait a bit until I'm finished to avoid a conflict and then join in. Ryan Reich 17:40, 18 April 2007 (UTC)


 * The following, found in the Style manual of the U.S. Government Printing Office, corresponds to what I believe to be the general rule for English:


 * 3.3. Derivatives of proper names used with a proper meaning are capitalized.
 * Roman (of Rome) • Johannean • Italian


 * 3.4. Derivatives of proper names used with acquired independent common meaning, or no longer identified with such names, are set lowercased. Since this depends upon general and long-continued usage, a more definite and all-inclusive rule cannot be formulated in advance.
 * roman (type) • brussels sprouts • venetian blinds • macadam (crushed rock) • watt (electric unit) • plaster of paris • italicize • anglicize • pasteurize


 * As far as I know this is also the prevailing rule in the U.K. Unfortunately the application of the rule is not authot-independent. While I write "angora wool", "benzine" and "cardigan sweater" without thinking of Ankara, Benz or Cardigan, I can't see "abelian" or "Abelian" without thinking of Abel. But in any case, I think we should preferentially follow general rules as laid down in authoritative style guides (rather than the possibly haphazard choices made by authors of mathematical texts) unless there is a compelling reason not to do so. (And note that in view of the completely different rule for French, texts by semi-French authors like Lang are perhaps somewhat suspect.) --Lambiam Talk 19:26, 18 April 2007 (UTC)


 * Is there an authoritative style guide in mathematics? I remember looking at some Oxford general science writing guides, but they were not exactly helpful in this regard. While it's true that authors could be inconsistent (even between different texts of the same author!), but usage and grammar are not frozen in time, either: for example, capitalization of nouns was a lot more common even 100 years ago. Thus it may be preferable to consult current literature, with all the attendant faults, to using outdated manuals. Also, the last comment about Lang is a bit perplexing: it's true that he had peculiar accent, yet I never heard any implications that his English had been influenced by French or any other language, certainly, not in the context of his prolific mathematical writing (only small Bourbaki part of which was done in French, as far as I know). Arcfrk 23:25, 18 April 2007 (UTC)


 * Surely the rules are not topic-dependent? If it is Archimedean point with a capital A in philosophy, then also Archimedean property in mathematics. If it is Hamiltonian economic program in economic history, then also Hamiltonian path in maths. Rather than consulting outdated manuals, what about contemporary manuals that are kept up-to-date? The GPO Style Manual referred to above is from 2000. --Lambiam Talk  03:07, 19 April 2007 (UTC)


 * The problem is that different contemporary style guides almost certainly disagree. My opinion is that the best we can do - and this is not sarcasm - is to make each article internally consistent, and not worry about global consistency. CMummert · talk 03:14, 19 April 2007 (UTC)


 * I dont think we'll have a universal solution, but for each word I think we should make it consistent throughout WP. - grubber 03:16, 19 April 2007 (UTC)
 * I must agree with Arcfrk below; any given word is subject- (and author-) dependent. For example, I believe Frank Harary uses hamiltonian, and the usage in graph theory is clearly divided. We represent the state of mathematics best by being inconsistent.  Septentrionalis PMAnderson 15:09, 26 April 2007 (UTC)
 * My impression is exactly that the rules are both subject-dependent and time-dependent (different editions of the same book may use different conventions), and mathematical usage tends to diverge from more general science usage. Arcfrk 03:44, 19 April 2007 (UTC)


 * Replying to your question on style guides in mathematics: The only one I know is Nick Higham, Handbook of Mathematical Writing, SIAM, 1993. Not a style guide, but a good reference for the finer points of typography, is Donald Knuth, The TeXbook. I also use P.R. Halmos, "How to write mathematics", Enseignements Mathématiques, 16:123–152. Perhaps we should start a list at Manual of Style (mathematics)? -- Jitse Niesen (talk) 14:23, 19 April 2007 (UTC)


 * Yes please! Paul August &#9742; 14:50, 19 April 2007 (UTC)

AfD: Mathematical landscape
Mathematical landscape has been nominated for deletion. Comment as you see fit! Anville 15:34, 26 April 2007 (UTC)

Calculi versus algebras
What is the difference between caluculi (e.g. propositional calculus,predicate calculus,proof calculus, and the various comp-sci calculi) vs. algebras (Boolean, heyting, etc)?

I was hoping to maybe find something on wikipedia explaining the difference, but I couldn't find anything. Brentt 03:03, 26 April 2007 (UTC)
 * Well, there's an ambiguity here that's caused us lots of grief in the past. "Boolean algebra", taken as a mass noun, means pretty much the same as the propositional calculus. Our article on Boolean algebra in that sense is (supposed to be) at Boolean logic, though the criteria for including information in that article are not particularly clear.
 * On the other hand, a Boolean algebra, count noun, is a mathematical object. It's not a calculus at all; it's an algebraic structure, like a group or ring. Not sure if I've helped, but it's a start; maybe we can narrow down what your question means. --Trovatore 03:17, 26 April 2007 (UTC)

I have always associated the word "calculus" with a method to manipulate strings of symbols in a meaningful way. There are "transformation rules" that tell you how to convert one string into another string. Not every transformation may be suitable for every string - you have to follow some rules. This matches the way that Newtonian calculus, lambda calculus, proof calculi, propositional calculus, etc. work. An algebra, on the other hand, is a set with operations such that any two elements can have the operation applied to them. In some cases, you can make up a semantics for the calculus that show that when the strings represent elements of a certain type of algebra then the transformation rules preserve some algebraic properties. But this is a very loose connection. Is that your question? CMummert · talk 03:35, 26 April 2007 (UTC)
 * Yeah, actually your first three sentences here are pretty much what I was getting at. The mass-noun sense of "Boolean algebra" is really a calculus; the count-noun sense is an algebra. (We should perhaps reconsider making Boolean algebra a disambiguation page, because this keeps biting us from time to time.) --Trovatore 03:52, 26 April 2007 (UTC)


 * Isn't an operation sort of a inference rule though? Brentt 00:57, 27 April 2007 (UTC)
 * Not sure what you mean by that. Can you elaborate? Note that the objects on which you're performing the operation may not be things you can write down in a finite amount of space. --Trovatore 02:32, 28 April 2007 (UTC)


 * Most of elementary algebra (where "algebra" is the mass noun) consists of rules for meaning-preserving manipulations on formulas (strings of symbols). In that sense, the algebra rules are much like (for example) the rules of the differential calculus, and one might call it a calculus. The soundness of these rules corresponds to algebraic properties enjoyed by the mathematical structure (possibly an algebra) whose elements are denoted by the formulas. The mathematical structures (such as rings) and the calculi (such as high-school algebra) live in different universes. For a Platonist, the mathematical structure has an existence independent of human knowledge; it was discovered. A calculus, on the other hand, is a human construction that was invented.
 * To complicate things, strings of symbols are themselves also elements of carrier sets of various algebras (for example, regular languages form Kleene algebras), and so we can also "discover" and examine the algebraic structure of some calculi. Given the somewhat frayed ends of most interesting calculi, this is usually not a rewarding undertaking, but for the Risch algorithm it definitely was. --Lambiam Talk  04:15, 28 April 2007 (UTC)

FAC of Equipartition theorem?
Umm, hi,

This is surely off-topic here, but I'm hoping that people here would be kind enough to evaluate Equipartition theorem, which is a Featured Article candidate now. It's at the level of basic multivariable calculus, although there is a multidimensional integration by parts at one step of the derivation. Does the article read OK to you all? Any suggestions for improvements? Thanks ever so much for your time and trouble, Willow 21:58, 27 April 2007 (UTC)

Mathematics ratings and tables
It seems to be regarded as a positive thing to put templates on talk pages as a way of tracking progress in our task to provide a good range of high quality articles, with emphasis on the most important ones. However, I am confused by the current organization and would appreciate some links/clarification/discussion. The whole process seems to be intertwined with the separate but related selection of articles for the CD-ROM Wikipedia1.0, whose classification we use (modulo our additional B+ class).

At first this seems fine: I follow the "mathematics grading" link on a template to WikiProject_Mathematics/Wikipedia_1.0 and find explanations of the grading, with helpful tables of how many articles there are in each class, analysed both by importance and field, automatically generated each day by a bot - great! When I follow links in this table to a particular class (A,B etc.), or importance level (low, mid etc.), I find myself on a category pages which automatically list the articles in the given category - also useful!

Finally I follow the link to geometry and topology and I see something which looks even more useful: a list of articles in my field, ordered first by decreasing importance and then by increasing quality (class), together with comments, presumably from the template on the talk page. Wow, this is the most useful page of all!

But then I notice that the page is incomplete (articles with ratings that I know are not there), and has an extra "Has template" column. The page appears not to be automated (indeed it is months out-of-date). Then I remember WP1.0 and guess this is some list of articles chosen to go on a CD-ROM. Is this right, or am I just confused? Wouldn't it be really useful to have pages like these which were updated automatically from templates on article pages? Wouldn't it be better to have an extra column "Selected for WP1.0" (which could be, and is usually, indicated on the article page)?

Forgive these mumblings if I have completely missed the point, but the current structure has left me very bewildered about what is going on. Geometry guy 21:01, 12 April 2007 (UTC)


 * These per-field pages are not related to WP 1.0; they are just for in-project use. They aren't (yet) automatically updated, mostly because I haven't gotten around to writing the code. There are no technical obstacles to doing the updates automatically except that the automatic versions would not include the "comments" that the current tables do. It would not be too hard to write the code to update them at the same time the main table is updated. CMummert · talk 22:29, 12 April 2007 (UTC)


 * This definitely seems a good thing to do when you get the time. We are now getting to the stage where the number of rated articles has reached the level where further sub divisions is necessary. --Salix alba (talk) 23:42, 12 April 2007 (UTC)

Many thanks for the clarifications. It would be great if these pages were automated (although the "algebra" one would be a bit long unless stub class or low importance articles were somehow omitted).

However, one of the most useful things about these pages is that comments are there. Now there is a "comment" field on the template. This does not separate the comment from the author, but that is not important. At least in principle, couldn't the "comments/updated/has template" columns be replaced by a single column with the "comment" field of the article template? I know it requires someone to do the work, but it would be very useful and might encourage editors to use the "comment" field more often, and keep it up-to-date. Geometry guy 09:44, 13 April 2007 (UTC)


 * The table is created by using a database query interface to get lists of articles with various properties and then cross-referencing those lists to generate the table. These queries doesn't put much strain on the database and are very fast.  In order to get the comments, it would be necessary to download the actual source of the talk pages and parse out the comments, which is a much slower and technically more difficult operation.  At least the first version of automatic update wouldn't do it. CMummert · talk 11:33, 13 April 2007 (UTC)

I see, that's a pity. Does that mean that if we moved to an automatic update regime we would lose the comments completely on the field pages, or would it be possible to write the code so that it checks the current field page for existing comments and writes them to the new page? (I guess that would be more work to do, though, because it involves parsing the field page.)

Would it be feasible to update comments from templates as a separate operation that happens less frequently? Weekly would surely be enough, or the bot could cycle through the list of fields to reduce the daily load (so each page would be updated every 11 days). Well, I know this is work, and work which I am not able to do, but I think it might add some energy to the project to have such a system for monitoring progress. I would at least be willing to go through (some of) the existing field pages and merge the existing comments into the templates on the article pages.

Also, I wonder if this is something that Snowolf and Snowbot would be able to do for us. Even a one-off update of the pages would be great. Geometry guy 15:54, 13 April 2007 (UTC)


 * There are two ways comments could be be done. It can be specified as an inline field in the template, its can also be done as a sub page. This has been done on Talk:Blaise Pascal, with Talk:Blaise Pascal/Comments as the sub page. If this structure is used then it becomes a trivial matter to transclude the appropriate comment page in the listings of articles, using etc.
 * It should not be too problematic to migrate articles to this latter fashion, a bot could do this. It would only need to run once so the server load would not be problematic. I've changed to put articles with inline comments into Category:Mathematics articles with inline comments. That category could be checked periodically to complete the migration. --Salix alba (talk) 19:25, 13 April 2007 (UTC)

This is a clever work-around &mdash; nice one Salix alba! I've changed to put articles with a comments page into Category:Mathematics articles with comments page. There are not so many of them yet, so I was able to go through them by hand, fixing them if necessary so that they are not also in Category:Mathematics articles with inline comments. It would be nice to have a third category with Category:Mathematics articles with no comments, but that required me to pluck up the courage to make a more substantial edit to the template, or for someone with more Template experience to do it before letting me the chance to mess it up ;) Geometry guy 15:50, 14 April 2007 (UTC)

I wasn't courageous enough so I just added a Category:Mathematics articles with no comments page. I also wasn't careful with the includeonly/noinclude issue so the template itself is an example. This is just a temporary fix to see where we are now. I will not be at all offended if someone reverts my edit! Geometry guy 20:08, 14 April 2007 (UTC)

I plucked up a bit more courage, and my ugly code appears to work, although it has the side effect of raising the "field" field slightly. I expect this is due to my lack of understanding of spaces and newlines in template code, and I hope an expert can fix it quickly (and improve my ugly nested code!): explanations on my talk page most welcome (and don't be afraid to be patronizing!). Meanwhile the temporary category Category:Mathematics articles with no comments page needs deleting (I don't think this needs to go through CFD since it contains no articles!). Geometry guy 21:23, 14 April 2007 (UTC)

Proof of concept
Salix Alba suggests migrating from inline comments to only using comment subpages. I have no objection to that. I don't know the historical circumstances that led to the current redundant system.

If the comments are all moved to subpages then there is no technical problem with making the field summary pages. I wrote some proof-of-concept code whose results are available at User:CMummert/Sandbox4. That page is entirely automatically generated. Please let me know what can be improved. CMummert · talk 15:06, 14 April 2007 (UTC)


 * This is fantastic! Thank you CMummert! The only suggestions I have are cosmetic.
 * A fixed width column for the article name would mean the coloured class boxes would line up nicely :)
 * I can see that splitting the page up into "importance" sections is a logical thing to do, but I wonder if a left-hand coloured "Importance" column would look better. This also has the advantage that when browsing half-way down a long list, the importance level is still clear.
 * I will check out some of the existing per-field pages as promised above and report back. Geometry guy 15:50, 14 April 2007 (UTC)


 * I added a color key for the importance and tried to fix the table width. CMummert · talk 16:42, 14 April 2007 (UTC)


 * I like it! Meanwhile I've made a first pass through the current G&T page, copying the comments to /Comments pages. Now the second pass, to eliminate any duplication. Geometry guy 18:07, 14 April 2007 (UTC)


 * Second pass now done... From my point of view the current G&T page could be replaced by the proof of concept. Geometry guy 19:39, 14 April 2007 (UTC)

The plan is for all the field pages to be generated this way, unless there are objections. The argument in favor is that the old pages are horribly out of date and unlikely to be kept in sync manually as the number of rated pages increases. I just used the geometry field as a proof of concept to illustrate what can be done manually. CMummert · talk 21:39, 14 April 2007 (UTC)


 * I am completely in favour of all field pages being generated this way! However, there may be useful information on the current field pages which is not duplicated in the articles (In the geometry and topology test case, most of the information was duplicated, but I was able to incorporate some additional information into the article talk page in some cases). I think, however, editors interested in other fields should be given the chance to do what I did with G&T for a few days before the new automated scheme is introduced. Do you agree? Geometry guy 22:26, 14 April 2007 (UTC)


 * The proof of concept looks fine and generating all field pages automatically is an excellent idea. Do I understand it correctly that the comments column in pages like WikiProject Mathematics/Wikipedia 1.0/Analysis will be overwritten and that we thus lose the entries currently there? If yes, then that worries me. Is that necessary? A few days seems a rather short period. Can't the comments there be transferred to the /comments subpages (e.g., Talk:Manifold/Comments)?
 * I read an essay some time ago which I can't find anymore. Its main point was that we spend to much time on classifying and rating articles, while we should just be writing and improving them. I don't want to tell others what to do and rating articles is certainly useful, but I do want to raise the issue for you to consider because the essay struck a chord with me. -- Jitse Niesen (talk) 11:37, 15 April 2007 (UTC)


 * Most of these entries are essentially copies of comments in article talk maths rating templates. I believe a bot should be able to move these to /Comments pages: Salix Alba has created a category to facilitate this process. For the remaining entries, I don't believe moving them to /Comments is easy to automate, so I'm doing it by hand: I've done the work on the Analysis and Geometry and Topology fields so far. (Maybe User:CMummert would be willing to run the code on the Analysis field in another sandbox so we can see where we are so far.) Putting these comments on article talk pages is a useful thing to do anyway, I believe.


 * I agree entirely with your second point: classifications and comments are a means to an end - improving the quality and coverage of the artices - not an end in themselves: they should make it easy to identify where there is work to be done and what improvements are needed. To this end, they should be easy to update, easy to refer to, and easy to maintain. The present situation, with comments on an article placed by hand in two separate places, wastes editors' time that could be spent writing articles. CMummert's code would solve this. It also contains several other improvements: for instance, both article page and talk page are linked, and there are links to add comments to articles without them. Geometry guy 16:55, 15 April 2007 (UTC)


 * A proof-of-concept for analysis is at User:CMummert/Sandbox5. CMummert · talk 18:06, 15 April 2007 (UTC)


 * Thanks! Geometry guy 18:23, 15 April 2007 (UTC)


 * This is wonderful :-) I've always wanted the rating summary pages to be automatically updated and sorted. The most important thing is that they allow editors to prioritise their efforts (either on important or low-quality articles, accordingy to taste). Migrating the comments to the subpage is a sensible thing. (To be honest, the only reason why I used the "comment" parameter was laziness on my part - it was quicker than creating a subpage, esp. when rating lots of articles). (What would be really nice would be an AWB extension which allows one to rate/categorise an article and post a comment in one click.) So, I think the concept is definately proven. :-) Tom pw (talk) 00:28, 17 April 2007 (UTC)


 * I agree, the concept is certainly proven to me, and there appears to be no disagreement on this now, so I suggest we implement it. I'll start a new subsection below in case further discussion is needed. Geometry guy 12:15, 19 April 2007 (UTC)


 * I'm planning to work on it this evening. CMummert · talk 13:08, 19 April 2007 (UTC)


 * Great! I've made a couple of suggestions below. I'll be glad to help out, anyway. Geometry guy 13:44, 19 April 2007 (UTC)

Progress on checking comments on field pages
I believe that all comments on the current Geometry and Topology page are now covered by /Comments pages. If someone wants to check what I have done, that would be very welcome, but from my point of view the current G&T page could be replaced by the proof of concept. Geometry guy 19:39, 14 April 2007 (UTC)

Note that not all comments at Analysis can be found in this proof-of-concept. This is because (I hope) they already exist in article talk page templates, and I've not migrated them to /Comments. As mentioned above, I really hope a bot will be able to do this. Geometry guy 18:23, 15 April 2007 (UTC)

I've now been through Algebra. This was relatively easy, as most of the comments are by Tompw (who almost always makes consistent comments on article and field pages). Geometry guy 19:34, 15 April 2007 (UTC)

I've now also checked Mathematical physics, which was relatively easy for the same reason. Geometry guy 20:21, 15 April 2007 (UTC)

I think we should not include the Mathematicians field in this process until an interested editor comments, as the format there is rather different (mathematicians are organized by year). Anyway, this means the checking is half-done, but I won't be able to do much more for the rest of the week. Geometry guy 22:17, 15 April 2007 (UTC)

In addition to the Mathematicians field, I also notice that there is a Theorems and Conjectures field. In the long term it would be nice to automate both of these, by adding e.g. a "mathematician=year" tag and a "theorem=Y" tag (cf. "vital=Y") to the template. Geometry guy 19:44, 16 April 2007 (UTC)

I've now checked the Basics field page. After automation, I think it makes sense to split Applied into Applied Mathematics and Probability and Statistics. These are, after all, completely different fields. It would also be convenient, I think, to split General into General Mathematics and History. The split pages could be given different names to save checking them for now. Geometry guy 19:44, 16 April 2007 (UTC)

I've now also checked Number theory, Discrete mathematics and Foundations. Subject to the above comments, this completes the checking process. Geometry guy 22:24, 16 April 2007 (UTC)


 * Thanks for migrating all those comments.


 * Probability/Statistics is already split from Applied Math in the rating template, it just doesn't have a ratings page like the others. I can easily add the ratings page. Similarly, General and History have different template parameters already.


 * So the only page that I think needs to be left alone is the mathematicians page. I think the best thing would be to move the old pages to subpages of their respective talk pages, so that they are still available online, and then delete them after a month or so if they are no longer needed.


 * Does anyone else have any thoughts about this? CMummert · talk 22:55, 16 April 2007 (UTC)

This is a good idea. I would also suggest moving the current "Applied mathematics" page to a (probably temporary) location ("Applied" or "Applied and statistics"). Similarly, the current "General" page could be moved to "General and history". The "Mathematicians" and "Theorems and Corollaries" need further discussion. Geometry guy 23:23, 16 April 2007 (UTC)

I'm sure there was a reason why I put Probability and Statistics on the same page as Applied, but I can't remember it... I think the history category was added as an option after the general page got created. Anyway, seperate rating pages for P&S and history are a good thing. I'll sort it out properly tommorrow morning (it's late, and I'll make silly errors if I do it now) Tom pw (talk) 00:31, 17 April 2007 (UTC)


 * In response to "the best thing would be to move the old pages to subpages of their respective talk pages, so that they are still available online, and then delete them after a month or so if they are no longer needed": Perhaps it's better to have the bot overwrite the pages (I know I'm contradicting my earlier comment). The old versions are forever accessible on the talk page. You can add a link on the talk page to the old revision just to be sure. I don't see much point in moving the old pages to subpages, and very little point deleting them after a month. -- Jitse Niesen (talk) 01:54, 17 April 2007 (UTC)


 * I went through "General" and "Applied". In all cases but one, the comments were already included in the comment field of the maths rating template, so I didn't do anything. In the remaining case (Theorem, if I remember correctly), I migrated the comments.
 * So, now somebody has to move the comments from the template to the /Comments subpage. I could write a bot to do that, but I doubt I'll have time to do that soon. One thing I thought of: what happens for pages which fall in several WikiProjects? Should we instead make a /MathsComments subpage? Or should we assume that the comments are shared by all projects? The latter option has my preference (simpler is better), but there might be situations in which that doesn't work well.
 * Geometry guy, I deleted Category:Mathematics articles with no comments page. For future reference, the easiest way to get a page deleted is to put db-author on the page. -- Jitse Niesen (talk) 13:48, 17 April 2007 (UTC)

Thanks Jitse! I agree simpler is better. During my edits, the only places I noticed potential conflict with other WikiProjects were two or three articles such as Quantum mechanics which are joint with the Physics project. Even so, I think it is better (as well as simpler) to share comments between projects.

As for the migration of comments to /Comments, there are only about 130 pages to go, and there is no rush, so if we all move a comment from time to time (and I've done quite a few already), it will get done even without the help of a bot. See the categories created above for further information. Geometry guy 19:25, 17 April 2007 (UTC)

Implementation and theorems
There appears to be general agreement that the automatic of per-field ratings pages is a good thing for all fields except "Mathematicians" and "Theorems and Conjectures" (which is not really a field), and that "Probability and statistics" should have its own page, separate from "Applied mathematics". I therefore suggest we begin to implement it. I have a few other ideas as well. Here is a possible list of actions to strike out when done!
 * 1) Move Geometry and Topology to Geometry and topology.
 * 2) Similarly move Theorems and Conjectures to Theorems and conjectures.
 * 3) Choose a consistent title for the field Foundations and mathematical logic aka "Foundations and set theory".  Suggestion: Foundations, logic, and set theory. Move the old page to the new name . Fix WikiProject Mathematics/Wikipedia 1.0/Table accordingly.
 * 4) Create new page Probability and statistics.
 * 5) Replace existing field pages by pages automatically generated by CMummert's code :
 * Algebra
 * Analysis
 * Applied mathematics
 * Basics
 * Discrete mathematics
 * Foundations, logic, and set theory &mdash; VeblenBot needs to be told the new field name
 * Geometry and topology
 * General
 * Mathematical physics
 * Number theory
 * Probability and statistics
 * 1) Write code and run a bot to migrate current inline comments to /Comments page.

I wonder also if it would be useful to add a "theorem" tag the template to place articles into a category Category:Mathematics articles about theorems or conjectures (I'm happy to do this and tag the relevant articles). Could this then be used to automatically generate Theorems and conjectures? Geometry guy 13:35, 19 April 2007 (UTC)


 * I hope you don't mind that I numbered the bullets for easier reference.


 * Re bullet 3, I disagree - mathematical logic includes more than just set theory.


 * Re bullet 6, if these have all been done by hand then it is only necessary to remove the inline comment option from the math rating template. I don't think a bot is needed.


 * Re tagging the articles as theorems, this is already done via Category:Mathematical theorems. It would not be hard to cross reference the contents of that cateogry with the list of rated articles to find the intersection. CMummert · talk 13:53, 19 April 2007 (UTC)


 * Thanks for adding numbers.


 * Re 3, you cannot disagree with me :) as I did not express an opinion! I don't mind what the field is called, as long as the same name is used in the title of the page, in links to the page, and in . However, if I were forced to express an opinion, I would say that, at least to a non-specialist, neither of (mathematical) logic and set theory is a subset of the other, so a more inclusive title would be be Foundations, logic and set theory. I think the word mathematical is unneeded, just as it is unneeded in (mathematical) analysis, because this is a WikiProject Mathematics category, and "mathematical" is understood.


 * Re 6, alas it has not been done by hand: there are still 131 articles with inline comments. Furthermore, editors continue to add comments using the comment method. Geometry guy 19:19, 19 April 2007 (UTC)

Could you tell again what exactly is to be done with these 131 articles? I would volunteer to handle some of these, if help is needed. Jakob.scholbach 19:37, 19 April 2007 (UTC)


 * Here is the process. Take an article talk page from Category:Mathematics articles with inline comments (for example, Talk:Abacus) . Remove the comment from the maths rating template and put it in the Comments subpage (for example, Talk:Abacus/Comments). If you can do a few of them, it would certainly help. CMummert · talk 20:49, 19 April 2007 (UTC)
 * AutoWikiBrowser can make this process a lot easier. I've created a page at User:Salix alba/Sandbox which lists each page in the category followed by its comments page. You can make a list in AWB with it. Turn off all auto edit functions and make sure skip empty pages is off. The step through the list, cut the comment from the maths rating tag and save, go to the next page paste in the text and save. I manage to do about 30 pages in half an hour with this method. --Salix alba (talk) 23:00, 19 April 2007 (UTC)


 * The comments appear to be dealt with; check again in an hour or two to make sure the category doesn't repopulate. I edited the template to show an error when inline comments are left. The next thing to do is to make changes 1 through 3 and then wait a while for the job queue to update the articles. CMummert · talk 00:02, 20 April 2007 (UTC)


 * Very good! Clearly, I should learn to use AWB! Geometry guy 10:28, 20 April 2007 (UTC)

I have now checked that all comments on WikiProject Mathematics/Wikipedia 1.0/Theorems and conjectures are reflected in the article Comments pages. I've asked the marvellous VeblenBot if it can generate this page for us using Category:Mathematical theorems and Category:Conjectures. Geometry guy 11:16, 26 April 2007 (UTC)

History and mathematicians
I am no longer convinced that a separate WikiProject Mathematics/Wikipedia 1.0/History page is helpful. At the moment it only contains two articles, and only one is really about history. There is already a WikiProject on the History of Science which rates articles, there is already the Mathematicians field, and I think it is more helpful to place articles like History of manifolds and varieties in the relevant mathematical field ("Geometry and topology" in this case).

One possibility would be to treat these as suggested above for "Theorems and conjecture" using a "history" tag to generate a category called Category:Articles about the development of mathematical concepts or something similar. Comments? Geometry guy 13:41, 19 April 2007 (UTC)


 * I've now added an experimental "historical" tag to the template to put articles in a Category:History of subject mathematics articles (not a great name, I admit). This could be used to autogenerate WikiProject Mathematics/Wikipedia 1.0/History, while at the same time allowing historical articles to belong to another field. Geometry guy 10:32, 26 April 2007 (UTC)


 * This has now been used to autogenerate WikiProject Mathematics/Wikipedia 1.0/History. I find the results very satisfactory! Many thanks again to CMummert. Geometry guy 01:05, 27 April 2007 (UTC)

Finally, does anyone have any ideas for the Mathematicians field? Sorting these by date is the hardest part. Would we be able to do it if every mathematician had a subpage containing their date of birth (or death)? Geometry guy 13:41, 19 April 2007 (UTC)


 * Although it's a couple of days late: do you mean that you want to be able to sort the mathematicians by date of birth/death? In that case, we actually have a very nice way to do that: persondata. It's special metadata for articles on people which includes (among other things) date of birth/death. Ideally all articles on people should have this, and it ought to be machine readable, so this should solve that issue. Of course, if that's not the issue you meant, then please ignore this. --Sopoforic 00:17, 22 April 2007 (UTC)


 * The difficulty with that is the bot doesn't download the articles or their talk pages, it only looks at category links and some 'what links here' links. The persondata info is machine readable, but only at the high cost of downloading the article. Putting the info into another subpage, as Geometry guy suggested, wouldn't help. If there is a way to do some crude sorting via categories, that could be made put into the bot. CMummert · talk 02:16, 22 April 2007 (UTC)


 * I see. It would, of course, be unnecessary to download the article for each update; once to retrieve the info initially and then perhaps monthly (or less often) thereafter would probably be sufficient. It should be possible to do something like this with categories, but that would be at the (very high) expense of manually inserting every article into the category. One ought to be able to put the articles into categories like this:, where 1512 is the DOB of the mathematician. Presumably one would be able to extract this ordering from the category page, though not (apparently) the actual sort key used. This may be insufficient for the task at hand--I'm not totally sure what the goal is. --Sopoforic 02:42, 22 April 2007 (UTC)


 * The best way I see is to use the categories in sequence with Category:19th century mathematicians.  This would at least allow sorting by century, so the sort order would be century, then importance, then quality. This is not technically difficult to implement, but the code to do it isn't written. CMummert · talk 02:58, 22 April 2007 (UTC)

Yes, that is what I meant, simply because the current page sorts the mathematicians that way. Sorting by century is one option, but there is the problem that many mathematicians straddle two centuries (e.g. Gaston Darboux and Elie Cartan). There are also Category:1917 deaths style categories, which I suppose in principle could be used, but this may be too complicated to be worth the trouble! Geometry guy 14:12, 22 April 2007 (UTC)

Mathematicians by year: a proposed way forward and some questions
There is now only the mathematicians page left to automate, so I have been looking at it again. There is in fact an obvious way to sort mathematicians by year: just make the table(s) "sortable". This can be automated using /Dates subpages giving the lifetime of the mathematician. Together with a couple of categories, such as Category:BCE mathematians and Category:First millenium mathematicians, sortable tables can easily be produced by transcluding these subpages. Such tables have the additional advantage that the tables can be sorted by importance or quality by clicking on the links provided by the sortable format.

It would be nice to be able to sort the table(s) also by name. One way to do this would be to introduce additional /SortName subpages, which give a sortable form of the mathematicians name (e.g. "Fermat, Pierre de" for Pierre de Fermat).

The creation of /Dates and /SortName subpages from the current page can probably be done using AWB (alas, I still don't have the expertise for this). However, there is another more subtle issue: at present the mathematicians page provides an additional "field" column, which lists the expertise and contributions of a mathematician, in addition to any comments on the quality of the article. This can partially be covered by the /Comments subpage. However, I think it might be useful to transclude another page containing the expertise and contribution information.

Many of these list entries describe the field(s) in which the mathematician contributed. This makes me wonder whether we should replace the "field = mathematicians" tag by a "mathematician = yes" tag, so the field is still free (for the primary field in which the mathematician contributed). One advantage of this would be the possibility to sort mathematicians by field. The mathematicians could then also be listed on the relevant field pages, although it is not obvious that this is necessarily a good thing, so please add comments! Geometry guy 19:47, 27 April 2007 (UTC)

Proof of concept for mathematicians
To show how this can be done, User:Geometry guy/Mathematicians WikiProject Mathematics/Wikipedia 1.0/Mathematicians now contains a sortable list of mathematicians with some dates. Most of the content of this page was automatically generated by VeblenBot with no change to the code. There are links for adding /Dates, /SortName, and /Contribs subpages to mathematician talk pages. At the moment, the sort order for the "class" and "importance" columns is not the natural one. Also, BCE and first millenium dates do not sort correctly. Feel free to make comments, changes, or add dates etc. Geometry guy 12:32, 28 April 2007 (UTC)

I've now added almost all the dates, and many sortable names to the table. I have also managed to make the class and importance columns sort naturally, as well as BCE and first millenium dates, although my solutions are not particularly elegant. Geometry guy 18:17, 29 April 2007 (UTC)
 * Apologies for not getting back sooner. The /Mathematicians table is nice, its certainly fun to be able to sort by dates class etc. I'm a little unsure about the /Dates, /SortName, and /Contribs sub pages, a perfusion of sub pages seems a bit excessive.
 * An alternatives is to use a database type structure to store the data. I've done a sort of database for the properties of polynomial at Template talk:Polyhedra DB, quite esoteric as its pre-parser function and could be much nice using a switch.
 * There are two ways this could be done, as one big page with the data for all mathematicians. Or with a sub-page per mathematician say /DB, this latter could be like
 * this would save in creating too many extra pages and could be extensible.
 * One thing I've been doing on a separate wiki is treating each wikipage as a object which has properties which are stored in a seperate table in the database. In the page, say in Pascal you can use  to set a property, an extracted anywhere using  . This is of course just wishful thinking as getting this implemented in wikipedia would be a task.
 * This might be a little too late to change things as I see you done a lot of good work creating the sub-pages, so we might as well stick to that style. --Salix alba (talk) 08:34, 30 April 2007 (UTC)
 * This might be a little too late to change things as I see you done a lot of good work creating the sub-pages, so we might as well stick to that style. --Salix alba (talk) 08:34, 30 April 2007 (UTC)

I agree that the profusion of subpages is not ideal and your ideas are interesting indeed. I think the data should be stored with each mathematician, but I am not wedded to the current (experimental) approach. The /DB subpage per mathematician is a nice idea, and it might not be too much work to migrate the data I have created to produce entries like

(This above was generated by a template User:Geometry guy/MakeDB which might make the migration easier.) [Note added: I am now substituting for this template, so I can delete it from my user space.]

As I understand, I then just have to modify WikiProject Mathematics/Wikipedia 1.0/Mathematician row format template to replace transclusions like by , which is easy to do. Geometry guy 09:48, 30 April 2007 (UTC)

The mathematicians page has now been fully automated by CMummert and VeblenBot. However, in view of the above discussion, it may be subject to some change for a short period. Geometry guy 16:38, 30 April 2007 (UTC)

I have now copied the /SortName, /Dates and /Contribs subpages into /Data subpages of the form

and modified the templates to use these /Data subpages in the Mathematicians table. The only disadvantage of this approach is that the /Data subpages appear to be empty. This could be fixed by adding instructions inside a "noinclude". I could do that, but it will take a while, even using AWB. Geometry guy 16:48, 3 May 2007 (UTC)


 * I've had a play about with Talk:René Descartes/Data and a template in my sandbox User:Salix alba/Sandbox:
 * gives:
 * gives:
 * gives:
 * gives:
 * Basically whats happening is I've changed Talk:René Descartes/Data to be


 * with a template name instead of the switch. The key parameter to Descartes/Data is passed through to the second template. This is just a switch, but with a default argument which prints out the data in human readable form. if a parameter is given it just behaves as if the switch was there, no parameter, ie when viewing the /Data page you get the human readable form. --Salix alba (talk) 23:46, 3 May 2 007 (UTC)
 * with a template name instead of the switch. The key parameter to Descartes/Data is passed through to the second template. This is just a switch, but with a default argument which prints out the data in human readable form. if a parameter is given it just behaves as if the switch was there, no parameter, ie when viewing the /Data page you get the human readable form. --Salix alba (talk) 23:46, 3 May 2 007 (UTC)

Nice! Can you implement this? I know it is boring going through 147 articles with AWB, but I think it is worthwhile. Using a template as you suggest has the huge advantage (which I didn't appreciate when I created these /Data pages) that the format can be changed without having to change all the pages. For the same reason, it is probably worth being flexible about any "noinclude": I was going to transclude a template explaining how to edit these pages, but it might be possible to absorb this into your idea. Geometry guy 00:13, 4 May 2007 (UTC)

Geometry and topology is now live
Please look at WikiProject Mathematics/Wikipedia 1.0/Geometry and topology, which is a test example of how to use the automated tables. Before implementing the other fields, I want to collect feedback on this one. It's worth looking at the source of that page to see how the implementation works. CMummert · talk 15:21, 20 April 2007 (UTC)
 * Perfect! I might slightly adjusts the column width. Links to the other field pages at the top could be helpful. --Salix alba (talk) 16:37, 20 April 2007 (UTC)
 * Still looking good! Maybe it is worth explaining in the lead how to Add or update Comments by following the handy links. Geometry guy 18:11, 20 April 2007 (UTC)
 * Looks good to me :-) Could we have WP MATH 1.0 at the bottom? Tom pw (talk) 18:30, 20 April 2007 (UTC)

It isn't obvious without looking at the page source code, but that page can be edited by anyone and the changes will not be lost. The tables themselves are included from another page, and changes to them will get overwritten. So it's possible to edit the lead section or add stuff to the bottom without having to ask me to edit the bot code. Tompw has suggested a way of formatting the tables so that the formatting will also be able to be tweaked from the wiki without changing the bot. CMummert · talk 20:06, 20 April 2007 (UTC)
 * (For the record, I suggested a template... it's at User:Tompw/sandbox12 currently, and is being trialed at User:CMummert/Sandbox5. Tom pw (talk) 12:29, 21 April 2007 (UTC))


 * This looks really nice, but I have a small request. Having every comment field starting with 'Comments:' is a bit distracting, so is there any way we can do something like:
 * Consider re-nominating for Featured Article status. Tompw 12:25, 6 October 2006 (UTC) (edit comments)
 * I think the same link code could be used for articles with and without comments too. Cheers, darkliight[&pi;alk] 09:50, 21 April 2007 (UTC)
 * Like this? Tom pw (talk) 12:29, 21 April 2007 (UTC)
 * Yeah I think thats much better. Is there any reason not to use the url that sends people straight to the edit page though? darkliight[&pi;alk] 13:42, 21 April 2007 (UTC)
 * Yes, there is... if you have an article with spaces in the name, you end up with an edit link like: coordinate system/Comments&action=edit (see source). Tom pw (talk) 14:07, 21 April 2007 (UTC)
 * Ahh I see, no problem then, and thanks. darkliight[&pi;alk] 14:25, 21 April 2007 (UTC)
 * Can we find a work-around for this? Is there a method for replacing spaces by underscores? Geometry guy 22:13, 21 April 2007 (UTC)
 * Since anyone can edit the format I see no reason not to roll out the entire list. However, I have one more suggestion: these pages are probably all going to have essentially the same format, so why not have a single page or template describing this format, which can be included in each of the field pages. I've done this with WikiProject Mathematics/Wikipedia 1.0/General so you can see what I mean. The format is defined at WikiProject Mathematics/Wikipedia 1.0/Field page format. This makes it trivial to roll out the remaining pages. Geometry guy 13:55, 21 April 2007 (UTC)
 * That's a good idea. Tom pw (talk) 14:07, 21 April 2007 (UTC)
 * I changed the bot to use a template to format the table rows and hacked with the template to make them get along. You can see the results at WikiProject Mathematics/Wikipedia 1.0/General. CMummert · talk 20:38, 21 April 2007 (UTC)
 * I bashed the template some more, and it all seems to be working OK :-). Are we ready to roll this out across the board?Tom pw (talk) 21:12, 21 April 2007 (UTC)
 * The template I made was just a primitive first attempt: feel free to alter it radically! Anyway, this should be easy to do, and everything else seems to be well, so I will roll out the other fields. Geometry guy 22:13, 21 April 2007 (UTC)
 * I made the changes necessary because of the new name for the logic field, and updated the tables. That takes care of all the bullets above. There is still the remaining issue of a new "Theorems" field page, which could be automatically generated as well. That will take some time to implement. I want to say that the transition process went much more smoothly than I had anticipated because of Geometry guy, Salix Alba, Tompw, and others taking care of important parts of the work.  CMummert · talk 23:59, 21 April 2007 (UTC)


 * That's the beauty of wikipedia! The response to my puzzled query has been fantastic, and with a great outcome. If I were in to barnstars and such things, I'd be awarding them to everyone involved, especially CMummert! Geometry guy 14:25, 22 April 2007 (UTC)

Category:Mathematicians by religion nominated for deletion
Category:Mathematicians by religion nominated for deletion before I could get to doing it. (Ask me about Slovenia!) Please go comment at Categories for discussion/Log/2007 April 29. Dr. Submillimeter 22:49, 29 April 2007 (UTC)

Automated archiving
I have edited the source of this page to ask User:MiszaBot II to automatically archive it. If any errors occur, please remove the template from the very top of this page's source code. The cutoff is currently set for 7 days, and the bot appears to run at 6:35 and 18:35 UTC each day. CMummert · talk 13:33, 30 April 2007 (UTC)

Cultural comment
I added this comment to FOIL rule:


 * This mnemonic is not used by mathematicians and perhaps not known by most mathematicians, but is taught by secondary school teachers.

I took every math course my high school offered; the curriculum was better than at most high schools and I not only mastered calculus when I took the course but absorbed a lot of material beyond what was required in every course, and I NEVER heard of "FOIL" until as an undergraduate I was tutoring other undergraduates. It seems this mnemonic is completely universally known to everyone who hates algebra but is required to take an algebra course. Am I right to think that perhaps most mathematicians don't know it? Michael Hardy 17:39, 30 April 2007 (UTC)


 * Pardon me while I retch. An apt commentary is
 * Stapel, Elizabeth. "'FOIL': A Special (and Misleading) Case." Purplemath. Available from http://www.purplemath.com/modules/polymult2.htm. Accessed 30 April 2007
 * Rote has its place, as in learning to add and multiply single digits. But in the case of this mnemonic, whoever is teaching the teachers should be taught better. --KSmrqT 19:02, 30 April 2007 (UTC)


 * (edit conflict) Perhaps this is a generational thing (although I make no supposition about how old Michael Hardy is). :)  I grew up with this terminology as did most of my peers (currently or recently in grad school).  It is often used as a verb, and in fact, its use has extended beyond even multipling two binomials together.  Although technically incorrect, one might say, "Foil these terms together" when referring to multiplication of any polynomials, the idea being that you extend the idea of "FOIL" in a natural way to more terms.  I'm not saying this should be mentioned in Wikipedia, but it is widely known among people I know. VectorPosse 19:06, 30 April 2007 (UTC)


 * I would definitely say it should be exposed in Wikipedia, along with other silly rotes that every recent high school graduate seems to know. I would even go as far as suggest a category: silly rotes (mathematics). As for generational thing, VectorPosse is probably right. It would be actually be interesting to pinpoint the origin of some of these widely known rules. On a side note, I pity the generation that had to grow up with things like that, as opposed to, let's say, Martin Gardner's amazing books and columns. Arcfrk 19:23, 30 April 2007 (UTC)


 * Teaching students this age is a challenge, and we should support — not ridicule — the teachers. Not all mnemonics are bad, and different students respond better to different means. Should we lampoon "SOH CAH TOA"? How about "xyzzy"? Pity the poor astronomy student who must remember star classifications. --KSmrqT 19:55, 30 April 2007 (UTC)


 * Teaching in any age is a challenge, and one may choose whether to rise to the challenge, to supplement teaching with entertainment, or to dispose of the substance entirely. Far from ridiculing teachers, categorizing silly rotes is a first step to correcting the incalulable damage that they inflict on students of mathematics (yes, I said damage, and I mean it). If you want an amusing 'advanced' rhyme (which no one takes seriously, by the way), how about
 * Гомоморфный образ группы,
 * будь, во имя коммунизма,
 * изоморфен факторгруппе
 * по ядру гомоморфизма!
 * This is a poeticized form of the first isomorphism theorem widely known in Russia Arcfrk 20:16, 30 April 2007 (UTC)
 * Hmm; I think that second line will have to change for the English-speaking world. ;-) --KSmrqT 20:59, 30 April 2007 (UTC)


 * I'm confused. What is this rule even about? Some weird kind of algebra with a non-commutative addition where one needs to remember the order in which the terms of the folded-out product comes? I fail to imagine a distributive law that could feel halfway natural and still dictate an internal order between the O and I terms. –Henning Makholm 20:25, 30 April 2007 (UTC)
 * Hypnotize yourself and regress back to a time in your life when words like "commutative" and "distributive" meant nothing to you. The order of the terms in FOIL is irrelevant; the point of the mnemonic is to help students remember the four terms of the result. Apparently their brains would explode if asked to understand and use the distributive law twice, especially since the pieces are not numbers but symbolic monomials. --KSmrqT 20:54, 30 April 2007 (UTC)

I don't understand all the fuss.

The statement, "This mnemonic is not used by mathematicians and perhaps not known by most mathematicians, but is taught by secondary school teachers" seems a little strong. Why not just say "This mnemonic is often taught in secondary schools" and leave it at that? In the article it also says, "The FOIL mnemonic is commonly taught but is sometimes frowned upon because the method does not work, without modification, for higher order polynomials (the double distributive method, by contrast, is easily extended to the latter case). Foiling can thus be seen as an example of learning by rote memorization of rules rather than by understanding underlying concepts." But I always have to memorize base cases, and I can combine FOIL with recursion to expand the multiplication of sums of arbitrary terms.

I disliked the article on purplemath.com referenced above. It's primarily because the author adopts the tone of "this is the way I do it so this is correct and it is the way you should do it." I'd agree more with KSmq's second comment that "Not all mnemonics are bad, and different students respond better to different means." (I also have a strong distaste for the use of "simplify" to describe the process of expanding the multiplication. Yuck.)

IMHO, the "Factorization" section of the article should either be removed or modified. It's not at all clear how one FOILs in reverse.

My two cents, Lunch 20:37, 30 April 2007 (UTC)


 * Everything should be taken off the page that is not reliably sourced. I suspect this is US-oriented remedial material. Charles Matthews 21:12, 30 April 2007 (UTC)
 * On the talk page, Taxman commented in 2005 that "the phrase could certainly use some attribution to a reputable source, and I haven't been able to locate one." A web search for 'FOIL polynomial' brings up a plethora of supporting hits, so there is no question that this name is used for this topic. I suspect it is not just remedial, as educational standards for mathematics lack the Cold War motivation of Sputnik and the beneficial influence of an influx of educated immigrants fleeing Nazi purges. --KSmrqT 21:30, 30 April 2007 (UTC)


 * (edit conflict, again.) About removing unsourced material, yes. About being US-oriented, I suppose so if someone from Europe says so.  About remedial, no.  It is, as I mentioned above, at least for people in my generation, common parlance.  The points about rote learning are well-taken, but ultimately irrelevant to the article in question.  In my calculus classes, I refer to it simply because "foiling" is easy to say and understood by everyone present.  The article probably can be sourced, but I just want to point out that this term has widespread usage even apart from any pedagogical considerations.  VectorPosse 21:33, 30 April 2007 (UTC)


 * My impression is that many textbooks discuss FOIL, but I don't have a stack of them handy. However, Steege &amp; Bailey, Schaum's Outline of Intermediate Algebra, 1997, ISBN 978-0-07-060839-9, is searchable at Amazon.com; perhaps it will suffice for now. One can also find several discussions like this and this critique at "The Math Forum". --KSmrqT 22:53, 30 April 2007 (UTC)
 * PS: Further discussion specific to FOIL should probably move to the article talk page. --KSmrqT 22:55, 30 April 2007 (UTC)

Unital ring poll
There is an ongoing opinion poll at Talk:Ring (mathematics) as to whether or not the rings on Wikipedia should be defined to unital or not. Opinions welcome. -- Fropuff 04:22, 1 May 2007 (UTC)