Wikipedia talk:WikiProject Mathematics/Archive/2006/Mar

recategorizing recreational mathematics

I've been being WP:BOLD with the subcategories of Category:Recreational mathematics. In particular I've emptied its rather ill-defined subcategory Category:Mathematical recreations and puzzles; a lot of its articles have found much better homes, but those that really did want to be somewhere under both Category:Recreational mathematics and Category:Puzzles I've put in one of a few joint subcategories such as Category:Mechanical puzzles. (Putting "puzzles" as a subcat of "recreational mathematics", as suggested on one talk page, isn't really an option: there are a lot of puzzles there that really aren't mathematical.)

While I was at it I also emptied Category:Puzzle games, which had an identity crisis as some people thought it was Category:Puzzle computer and video games while others couldn't tell it from Category:Puzzles.

Anyway, I expect I've offended innumerable people one way or another. If I've put your favourite article somewhere you don't think it belongs, please don't hesitate to move it (hopefully not into the categories I've carefully emptied). If you dislike the entire new categorization, please don't hesitate to argue with me about it. Though I can't imagine I've made things worse, since everything was categorized more or less at random to begin with. —Blotwell 14:35, 1 March 2006 (UTC)

Category:Mathematicians by religion

Category:Mathematicians by religion has a single subcategory, Category:Jewish mathematicians. I would think that being Jewish does not necessarily mean being religious. And do we actually need to categorize mathematicians on whether they were relegious, and if yes, what relegion they were practicing? Oleg Alexandrov (talk) 23:48, 1 March 2006 (UTC)

Being a jew does not, of course, make one religious, any more than being a christian makes one religious. So the categories' names do not imply that the mathematicians in question are religious - They just state to which religion they belong. And I think such categories are useful, in the same way that categories of mathematicians by nationality are useful. But obviously, additional categories for other religions, not just judaism, are in order for it to be meaningful. -- Meni Rosenfeld (talk) 07:19, 2 March 2006 (UTC)

I note that Category:Christians in science is applied both to Blaise Pascal, a Christian writer, and Bernhard Riemann, where as far as I can see it does little. I didn't much like like classifying mathematicians by nationality, when it came in; but it was inevitable with the growth, and the issue of several nationalities has the solution of including all of them. There are problems with all such classifications, and I'm not keen on them. Charles Matthews 09:05, 2 March 2006 (UTC)

Hmm, I wonder if Voltaire belongs in the Category:Christians in science, as, like me, his parents were Christian? I don't like this kind of categorization either; I think its basically some subtle political POV-pushing. May I suggest one possible cure: IF the person preached a religion (other than math) at one point in thier life, or published articles on faith (in newspapers, as letters to the editor, etc), THEN they may be classified by faith. However, if they had the bad luck of having Christian, or Jewish parents, that alone is not a reason to classify. I would insist on proof of religious activity before allowing classification. linas 14:49, 6 March 2006 (UTC)

french spelling

Um, I don't actually know french, but I thought only the first "e" in "etale" had an acute accent. So is this edit incorrect? Dmharvey 03:11, 2 March 2006 (UTC)

I think in this context, it's correct: the term in Hartshorne is "éspace étalé". Ryan Reich 03:30, 2 March 2006 (UTC)

So how do you know when it's étale and when it's étalé? Dmharvey 03:36, 2 March 2006 (UTC)

As far as I can tell, it's étalé here, and étale for morphisms. "éspace étalé" means roughly "slackened space", or "stretched-out space", which is reasonable given what it is, while an "étale morphism" is simply a "slack morphism". The metaphor is roughly the same, in that the slackness refers to a space constructed from layers laid out flat, and the grammatical difference distinguishes the "slackened space" constructed from something which was not, of itself, slack, from the "slack morphism", which is inherently so. Of course, "éspace étalé" is not used much anyway. Ryan Reich 03:56, 2 March 2006 (UTC)

It is certainly espace (not *éspace) étalé in French, but this leaves open the question of what the English translation of this expression is. I had been under the impression that it was called the étale space nevertheless, but Google seems to support both usages. —Blotwell 05:12, 2 March 2006 (UTC)

Étaler being a verb, étalé is the past participle (has been spread out, roughly). My MicroRobert says étale, adjective, can be applied to the sea as 'calm', when the tide is about to turn. We have been using sheaf space for espace étalé, which is not so common in English. HTH. Charles Matthews 09:13, 2 March 2006 (UTC)

• Please continue with sheaf space. Septentrionalis 21:08, 6 March 2006 (UTC)

After a check in the "Annales de l'Ecole Normale Supérieure", the good term is "espace étalé". --pom 11:18, 2 March 2006 (UTC)

Location of "elementary function" article

I think Elementary function (differential algebra) should be moved to Elementary function, currently a disambiguation page with little value. Despite the title, said article covers the concept of elementary functions in the general sense. Fredrik Johansson 23:50, 2 March 2006 (UTC)

I think it would simplify a few links and a line could be added to the article pointing to the list of common functions. When Elementary function (differential algebra) was created what is currently List of mathematical functions was in an article called Elementary functions, so I had to create something else. XaosBits 02:10, 3 March 2006 (UTC)

Could someone execute the move? Fredrik Johansson 04:56, 6 March 2006 (UTC)

Definition of General Linear Group

Charles Matthews and I are having a discussion about the correct definition of general linear group. It might be useful to have more input. The question is whether it should be defined initially in terms of rings or fields. Talk:General_linear_group A5 22:19, 4 March 2006 (UTC)

LaTeX

I have created a template to tag articles in need of LaTeX formatting. My concern is that it uses the LaTeX logo, which may or may not be a problem. The image was created using LaTeX, and using LaTeX to create images like ${\frac {q}{2}}$ doesn't seem to be a problem; yet, the image is still a logo with questionable copyright status. I was wondering what everyone else thought? Isopropyl 00:04, 5 March 2006 (UTC)

I would like to note that per the math style manual html formulas are perfectly acceptable (unless they look awful, like Σ_i=1ⁿ). It is also advised that one not modify somebody else's formulas by converting them from HTML to LaTeX or viceversa.

In fact, formulas which become PNG images may actually be preferrable in HTML, as then they show up as text, and look better on the page, also per the math style manual.

All in all, I don't see any pressing need for putting the {{LaTeX}} template on articles which are properly formatted, but only in HTML. Of course, one may use this template for articles which have no formatting whatsoever, like people writiting x_2 or x2 without bothering to use proper markup or math tags. That's what I would think.Oleg Alexandrov (talk) 23:37, 5 March 2006 (UTC)

Thanks for your input! I'll keep it in mind in the future. What is your opinion on the logo used in the tag? Isopropyl 23:42, 5 March 2006 (UTC)

Should a page use a combination of LaTeX and HTML formatting, or should its use be consistent throughout an entire article? I have tagged sections with {{LaTeX}} when the section in question deviated from the precendent set by the rest of the article. Isopropyl 23:45, 5 March 2006 (UTC)

I don't quite know, and for myself I would be fine with a mix. But if you find it stylistically ugly to have html mixed with LaTeX, then a better solution would be maybe to just convert the html to LaTeX right away, rather than put a "work needed" template on it and hoping that a kind soul would do it some time. There is a huge amount of articles needing serious work, as listed at Wikipedia:Pages needing attention/Mathematics, and I think that labeling an article as needing work because of TeX/HTML inconsistency would be probably not good. Cheers, Oleg Alexandrov (talk) 23:57, 5 March 2006 (UTC)

I agree with Oleg. Paul August ☎ 01:48, 6 March 2006 (UTC)

Most linked to and least linked to maths articles

I've been playing around with the database dumps and extracted the most links and least linked mathematics articles.

• top linked maths articles
• orphaned maths articles
• maths redirect frequency

The top linked articles might be useful for directing our efforts as these are probably most visited pages. The orphaned articles and redirects could help with some housekeeping. For example there is Squircle which seems quite dubious, and there are several highly linked redirects which indicate a need for some topics to be expanded. --Salix alba (talk) 13:54, 6 March 2006 (UTC)

Heh. Pi has 314 links... Ryan Reich 14:15, 6 March 2006 (UTC)

No way.... Dmharvey 14:33, 6 March 2006 (UTC)

And it holds slot 77 which is almost pi/4. linas 15:31, 6 March 2006 (UTC)

I wonder about the correctness of these lists. I was browsing the "orphaned" list and I was very surprised to see Stone–Weierstrass theorem, which of course is linked to from many articles. Paul August ☎ 17:15, 6 March 2006 (UTC)

It is quite a tricky job, especially where redirects are concerned. For Stone–Weierstrass theorem the only pages which link directly to it are 6 redirect pages [1]. For some technical reason, I've not included redirects in the count of articles. So these lists are the bests my little scripts can produce at the moment. If people feel the need, I'll try to update them to get closer to a real number. In the case of Stone–Weierstrass, I'd actually say the appearence in the list is a good thing. Looking closely, the hyphen in the article name is an odd unicode character (0xE28093) rather than a regular ascii hyphen (0x2D). I'd say this would be a good case for the article to be moved to the name with the ascii hyphen. --Salix alba (talk) 18:31, 6 March 2006 (UTC)

Ok I see. Yes I noticed the odd name. I think I will move the article. Paul August ☎ 19:51, 6 March 2006 (UTC)

• JA: I thought we were standardizing the use of ndashes, not hyphens, for conjoining names of distinct people, as distinguished from hyphenated names of one person. Jon Awbrey 20:04, 6 March 2006 (UTC)

Were we? I missed that. Why would we want to do that? Paul August ☎ 20:12, 6 March 2006 (UTC)

• JA: I'm sure I was directed to do that by some WikiPundit or other -- I just assumed it was to mark an obvious logical distinction for the sake of better hyper-indexing or sumting. Jon Awbrey 20:25, 6 March 2006 (UTC)
  • Somebody likes m-dashes and n-dashes, hardcoded by use of — and – and goes through substituting them. I'm not sure why; portability, maybe? Septentrionalis 21:13, 6 March 2006 (UTC)
    • I think I would strongly oppose that policy, on ground of human nature. Most editors will use the ascii hyphen, never get to see the policy on ndashes, leading to the same redirecting problems we have seen on Stone–Weierstrass. --Salix alba (talk) 21:31, 6 March 2006 (UTC)
      • Well, I haven't seen these improvements in article names; only in text. But there does seem to be a tendency to avoid hyphenated article titles: loan word not loan-word. Septentrionalis 23:26, 6 March 2006 (UTC)

• JA: YARTIW (yet another reason to ignore wikipundits). Jon Awbrey 21:40, 6 March 2006 (UTC)
• dear reader, for a more complete view of the status of the discussion, please do have a look at User_talk:Jon_Awbrey#En-Dash_Protocol_Reigning_Over_Polynominal_Titles_.28EDPROPT.29

Endashes

I knew we'd have to discuss this one eventually. The arguments for the A-endash-B theorem if A and B are two people are (a) it parses uniquely if you don't happen to be able to recognise double-barrelled names, and (b) it is a more professional piece of format. I would, however, always recommend creating [[A-hyphen-B]]'' first, as a precaution, so as to pick up any hungry red links; and only then move to the endash version. Charles Matthews 21:58, 6 March 2006 (UTC)

Yeah, I didn't like it at first, but after thinking about it (and looking at typeset documents) I have to agree. Not so much for the unique parsing, which is a good argument in principle but not so much in practice (you can't reliably conclude that Burali-Forti is a single person just because the article is at Burali-Forti paradox, even assuming you do notice the difference in the length of the dash/hyphen, which I wouldn't have if it hadn't been pointed out). But the endashes really do make the title look more like typeset documents and less like Usenet.

Maybe someone could send a bot around to look for article names that are duplicates except for the hyphen-endash distinction (these should always redirect to the same place), and for articles with endashes with no corresponding hyphen redirects (redirects should be provided). --Trovatore 22:24, 6 March 2006 (UTC)

Agreed. Some folks care as much about typographical niceties as mathematicians care about proof validity, or musicians care about pitch correctness. Lack of personal interest or awareness of these subtleties is no good excuse for hostility toward the interests of those who do care. Accents and quotation marks are another common battleground. With redirection, there is no need to fight. The hypen-redirects-to-dash idea seems like a reasonable compromise. --KSmrq^T 22:26, 6 March 2006 (UTC)

• dear reader, for a more complete view of the status of the discussion, please do have a look at User_talk:Jon_Awbrey#En-Dash_Protocol_Reigning_Over_Polynominal_Titles_.28EDPROPT.29

Three forms of mathematical induction

This article was intended to be comprehensible to all mathematicians.

It was not intended to teach mathematical induction. It was not intended to explain what mathematical induction is, nor how to use it.

It was nominated for deletion by those who did not understand it. To some extent, they did not understand it because it was a stub and failed to explain what audience it was intended for and what its purpose was.

A bunch of (mostly) non-mathematicians looking at the stub form in which the article appeared when it was nominated from deletion saw that

• It was not comprehensible to ordinary non-mathematicians who know what mathematical induction is, and

• The article titled mathematical induction is comprehensible to ordinary non-mathematicians, even those who know --- say --- secondary-school algebra, but have never heard of mathematical induction,

...and voted to delete.

And so I have now expanded the article far beyond the stub stage, including

• Substantial expansion and organization of the introductory section.

• Two examples of part of the article that is probably hardest to understand to those who haven't seen these ideas.

• An prefatory statement right at the top, saying that this article is NOT the appropriate place to try to learn what mathematical induction is or how to use it, with a link to the appropriate article for that. It explains that you need to know mathematical induction before you can read this article.

Therefore, I have invited those who voted to delete before I did these recent de-stubbing edits, to reconsider their votes in light of the current form of the article.

I also ask others here to vote on it by clicking here.

(Nothing like nomination for deletion to get you to work on a long-neglected stub article!) Michael Hardy 23:42, 6 March 2006 (UTC)

WAREL

My assumption of good faith in User:WAREL (formerly User:DYLAN LENNON) is being sorely tested. I know I'm not the only one who has wasted a lot of time over the past few weeks dealing with him/her. I'm wondering whether anyone else here has any thoughts about how to deal with WAREL, short of deploying an automatic WAREL-edit-reverting-bot. Dmharvey 18:00, 7 March 2006 (UTC)

For context, see the following article histories Decimal representation, Real number, Twin prime conjecture, as well as User talk:WAREL (Link to today's version, as WAREL likes to delete things he does not like. See especially the bottom section.) Oleg Alexandrov (talk) 19:47, 7 March 2006 (UTC)

I left a comment at Wikipedia:Administrators' noticeboard/Incidents#Disruptive_contributor to_mathematics articles. Oleg Alexandrov (talk) 05:47, 9 March 2006 (UTC)

al-Khwarizmi

This isn't about mathematics, but it is about a mathematician. Anybody who has spare time and is willing to read a long talk page is kindly request to comment on the dispute regarding al-Khwarizmi's etnicity at Talk:al-Khwarizmi. Cheers, —Ruud 14:49, 9 March 2006 (UTC)

How about showing the whole lot of them the way to Wikinfo, which wants editors like that? ;-> Septentrionalis 19:24, 9 March 2006 (UTC)

Somehow I doubt that most persons involved are interested in updating his biography beyond the first two sentences. —Ruud 19:30, 9 March 2006 (UTC)

Articles for the Wikipedia 1.0 project

Discussion moved to Wikipedia:WikiProject_Mathematics/Wikipedia_1.0 Tompw 16:40, 13 April 2006 (UTC)

Wikipedia talk:Scientific peer review

Notice: interested contributors may wish to participate in the Wikipedia talk:Scientific peer reviews by working scientists.

--Ancheta Wis 17:10, 11 March 2006 (UTC)

Can you guys have a look

Gallagher Index is a Political Science article and subject. But currently it could probably do with a mathematicans eye (alongside a few more things as well). Essentially, is there a neater or nicer way of doing the table at the bottom as an example of how the index is generated? Cheers, --Midnighttonight 08:47, 13 March 2006 (UTC)

Categorizing articles

On my suggestion, Salix alba made a list of Wikipedia articles which are not categorized, but which are linked from a math article. That list has a bunch of false positives, but also articles which are math and are not categorized. I suggest we start a cat wiki-pet (short for a Categorizing Wikiproject), going through those articles and categorizing them.

I split the list into 47 sections of 50 articles each. One may choose a section to work on, and sign at the bottom when done. I did the first three, and found roughly 3-5 articles out of 50 which may need categorizing. See the list at User:Salix alba/maths/uncategorised maths. Oleg Alexandrov (talk) 20:14, 14 March 2006 (UTC)

I don't know much about the category system, but if I just tag relevant articles with Category:Mathematics, is that enough to get them on the radar? (i.e. should I mark a section as "done" if I do this?) Dmharvey 03:03, 15 March 2006 (UTC)

I'd shoot for at least one level more specific than Category:Mathematics. The names of the big categories are pretty intuitive: Category:Algebra, Category:Mathematical analysis, Category:Mathematical logic, Category:Geometry, Category:Topology, Category:Number theory. Just make sure to remember the "mathematical" before "analysis" or "logic". --Trovatore 03:16, 15 March 2006 (UTC)

Sometimes one can pick the right category by looking at the articles going from the current one. But yes, putting them in Category:Mathematics is a good first option. Then my bot will list them to the list of mathematics articles, so more people will see them and may refine the categorization further. So yes, marking a section as done if the articles there are listed in some category is good, thanks. Oleg Alexandrov (talk) 03:18, 15 March 2006 (UTC)

ok guys thanks Dmharvey 03:27, 15 March 2006 (UTC)

I made the sections be 20 items rather than 50, as those were too big I think. To continue with the note at the top of this section, the person who does most work will get a cat as a wiki-pet (the Wikipet which anybody can touch (and edit)). Oleg Alexandrov (talk) 05:08, 15 March 2006 (UTC)

Oleg, you are SO going to award it to yourself. That is, like, so totally not fair. Dmharvey 19:48, 15 March 2006 (UTC)

Not all is lost, the race is still fully open! By the way, if you look at my bot's changes page, you will see a good harvest of math articles for March 15. Awesome work! Oleg Alexandrov (talk) 03:37, 16 March 2006 (UTC)

Now, I eager to get the wiki-pet, reviewed a section, categorized around 10 of the 20 there, felt good of myself, and when I got to editing the section to say "done", I see the section was done already! Dmharvey, now that's unfair. :) Oleg Alexandrov (talk) 04:55, 16 March 2006 (UTC)

Perhaps people should mark their territory -- in a nice way -- at the top of the score of items when they start work on it? Jon Awbrey 05:00, 16 March 2006 (UTC)

I doubt it is worth it; I meant it to be a silly joke rather than a complaint. Oleg Alexandrov (talk) 05:01, 16 March 2006 (UTC)

$''\!$ Jon Awbrey 05:32, 16 March 2006 (UTC)

Cheaters!!!! Hey, I noticed that some of the "finished" sections are still contain uncategorized articles. Even if the article is not about math, please do make an effort to put it into some category, somewhere!!! linas 01:08, 18 March 2006 (UTC)

Be my guest, my friend. :) Oleg Alexandrov (talk) 02:10, 24 March 2006 (UTC)

History of human knowledge about pi

This is the new title of History of pi. Even I think this is pædantry, so it may be over the top. Can we discuss this here, away from the Pi day crowds? Septentrionalis 00:48, 15 March 2006 (UTC)

"History of pi" deserves an article. To think that a table of the history of numerical computation of pi is the same thing as a history of pi is very silly. I've moved the table to another article, and labeled this article a stub. Michael Hardy 01:40, 16 March 2006 (UTC)

Agree w/Michael. I remember reading, as a young student, of plenty of interesting snippets about Egyptians knotting strings, silly legislation in kansas about pi=3, and what not. It deserves an article. linas 22:26, 17 March 2006 (UTC)

LOL... I remember adding that to (what is now called) Chronology of computation of pi (see under 1897), except the reference I have is for Indiana not Kansas. Dmharvey 22:34, 17 March 2006 (UTC)

MathWorld

Hi guys,

I was wondering why I can find so many maths-related articles here that do not reference relevant pages from MathWorld. I'm not sure what their license model is, but I can only assume that this is the reason why it's not popular around here? Please let me know if you think including their articles as references is a desirable thing. I'm watching this page, so do reply here. - Samsara (talk • contribs) 13:18, 15 March 2006 (UTC)

Obviously, we can't include relevant sections of MathWorld articles, as that would be a copyright violation. The reason for not referencing MathWorld articles is probably the uneven quality (yes, even by our standards) and the presence of clear errors (possible copyright traps) and probable neologisms. (I don't think the neologism being published as part of Mathematica makes it any less a neologism.) — Arthur Rubin | (talk) 13:56, 15 March 2006 (UTC)

I agree with Arthur and the reasons he provides. A policy of providing links to mathworld just doesn't make sense for us. However, if you come across a particular article where they have a much stronger version, then certainly linking to theirs would be useful (even better: bring ours up to snuff). -lethe ^{talk +} 15:30, 15 March 2006 (UTC)

Yeah, making it a policy to link to mathworld does not make sense, but I would think we should be encouraged in making external links to mathwolrd on case-by-case basis when those links are relevant (not necessarily much stronger than ours :) Oleg Alexandrov (talk) 16:11, 15 March 2006 (UTC)

Just to clear up a possible misunderstanding: I was referring to the license model because Planet Math is more frequently linked to. Is quality really so divergent between the two? I'm not trained as a mathematician, so I admit my judgement is poor. - Samsara (talk • contribs) 16:17, 15 March 2006 (UTC)

We actually copy planetmath articles, see WP:PMEX, that's why we must refer to the original versions, per their site license. Oleg Alexandrov (talk) 16:21, 15 March 2006 (UTC)

I second Arthur's comments. Just in the past month or so, I've had to remove several external links to MathWorld because when I checked them out, I found out they contained major errors. Sometimes these MathWorld articles can be good, but other times, it looks like a real hack job. So it's definitely not good to just unilaterally add the MathWorld links. I think it best for editors working on particular articles in their area of knowledge to add the links they actually found the most useful. --C S (Talk) 10:20, 10 April 2006 (UTC)

Please vote on this proposed deletion

at Wikipedia:Articles_for_deletion/Proof_that_22_over_7_exceeds_π#.5B.5BProof_that_22_over_7_exceeds_.CF.80.5D.5D.

The delete votes seem to be from non-mathematicians who erroneously think they understand the article. The main idea is this:

$0<\int _{0}^{1}{\frac {x^{4}(1-x)^{4}}{1+x^{2}}}\,dx={\frac {22}{7}}-\pi .$

Therefore 22/7 > π.

But the article also includes exposition, discussion, and mention of the appearance of this problem in the Putnam Competition.

One "delete"-voter says this is no more significant than, for example, a proof that π > 3.14159 or the like. The fact that 22/7 is a convergent in the continued fraction expansion of π seems to mean nothing to that person or to escape his notice altogether. The fact that this particular integral is so simple and has a neat pattern also seems to escape them. Another shows signs of thinking that all articles on π-related topics should get merged into one article (see list of topics related to pi). Michael Hardy 02:22, 16 March 2006 (UTC)

arXiv

So what's the deal with linking to the arxiv? This has come up quite a number of times in the last little while. Someone has gone trigger-happy recently on some papers there by Diego Saá, and it took a lot of convincing to get User:WAREL to stop linking there. (Or maybe he/she is still at it.) I would think generally such papers do not qualify for linking from Wikipedia, unless there are very good reasons to the contrary. Somehow a link to the arXiv has an air of respectability that you don't get from your home page on geocities etc, but it's not deserved, and we shouldn't be misleading people into thinking that the arXiv is a reliable resource. Dmharvey 02:16, 17 March 2006 (UTC)

I agree. One should only use references to books and peer-reviewed journals. Oleg Alexandrov (talk) 02:54, 17 March 2006 (UTC)

That's not how it works in mathematics research, and I see no reason why Wikipedia should adopt stricter rules for citations in its mathematics articles than most of the mathematics community itself. Wikipedia would only be shooting itself in the foot. --C S (Talk) 05:08, 25 March 2006 (UTC)

Well, we should prefer refereed references. Of course for journal references that are also on the arxiv, we should provide an arxiv link (not everyone has access to an academic library). Furthermore, there are worthwhile things on the arxiv which don't get published in journals. A lot of times, Witten, for example, publishes a lot of his papers only through the arxiv, he doesn't feel that journal referees are qualified to vet his papers. And there are précis on the arxiv which are very good resources but not original work, and therefore not appropriate for journals. But of course, there is also crackpottism on the arxiv, so care is certainly required. -lethe ^{talk +} 04:07, 17 March 2006 (UTC)

You definitely need a lot of care when citing papers by a guy who "doesn't feel that journal referees are qualified to vet his papers". :) Oleg Alexandrov (talk) 15:49, 17 March 2006 (UTC)

That is a dangerous attitude; but in the case of Witten I suspect many of the referees would agree, and are probably relieved that they do not have to try to keep up! --KSmrq^T 16:17, 17 March 2006 (UTC)

Off-topic: but Alexander Grothendieck stopped publishing in journals as well. linas 23:32, 17 March 2006 (UTC)

The arXiv is mostly reliable, except for the general mathematics (GM) section which is where the crank articles seem to get listed. I removed all the links to Diego Saá's papers that I could find; they were added by User:Diegueins, who claims to be his son. R.e.b. 05:57, 17 March 2006 (UTC)

In the last six months, I have found there a paper proving P=NP and another proving P$\neq$NP. No comments... pom 16:18, 18 March 2006 (UTC)

Please sign up on the participants list!

If you have this talk page on your watchlist, then you should add your name, field(s) of expertise and interests to the Wikipedia:WikiProject Mathematics/Participants page! I know there are some newcomers who haven't yet signed up, and I suspect there are some old-timers as well. linas 22:15, 17 March 2006 (UTC)

I meant to sign up at some point, but I glanced over the list and, frankly, many of you guys seem to be so good that it's kind of scary (I'm only an undergrad student) :-) - only half joking. But now, if you say so... AdamSmithee 00:20, 18 March 2006 (UTC) And after signing up, I see that my nick and the alphabetical ordering puts me on top of the list :-D AdamSmithee 00:28, 18 March 2006 (UTC)

I would join but you see, I'm on vacation. Good luck to you all. -- 127.*.*.1 01:17, 18 March 2006 (UTC)

Let me also add, feel free not to add yourself to that list or any others, for any reason. I myself don't see what purpose the list serves, and don't like adding myself to lists like that, though I did so eventually. -lethe ^{talk +} 03:41, 18 March 2006 (UTC)

Obviously the list doesn't have any kind of official status, but it does create a kind of community, as well as crystallizing one's own role in the Mathematics project in one's own mind. Mostly it seems sort of like the ritual of everyone gathering in a circle and placing hands one above another to seal a pact. And I'd encourage AdamSmithee to put his name on the list simply because he feels out of place; doing so will put him correctly in place :) Ryan Reich 06:18, 18 March 2006 (UTC)

Actually, I got into a discussion recently about how many particle physicists there are working in WP; looking at the participants list help put a lower bound on the number. This is a lot like any department directory or phonebook or census: rarely looked at, but terribly useful when its really needed. That, and indeed, the community feeling of the historical "I was here" thing. In 20 years, the list may be interesting to review: "I remember old so-n-so." linas 02:58, 19 March 2006 (UTC)

Statistics on User:WAREL

I submit the following statistics as an argument to block WAREL for, I suppose, a few days.

User:WAREL was born 17th Feb 2006. He/she has a total of 242 edits since then. The following survey includes 99 of those edits (41%), plus a few of User:DYLAN LENNON's edits (WAREL is a reincarnation of DYLAN LENNON).

• Perfect number: 23 edits. At least 13 immediately reverted. Many prior edits by DYLAN LENNON going back to July 2005 -- generally not reverted
• Twin prime conjecture: 17 edits, all reverted
• Real number: 12 edits, 11 reverted (As DYLAN LENNON: 14 edits, 10 reverted)
• Decimal representation: 11 edits. 10 reverted.
• Zeta constants: 6 edits, 3 reverted
• Finite field: 6 edits, 3 reverted, the other 3 self-reverted
• Proof that 0.999... equals 1: 5 edits, all reverted
• Decidability (logic): 5 edits, all reverted
• Riemann hypothesis: 5 edits, 1 reverted
• Chen prime: 4 edits, 2 reverted
• Decimal: 3 edits, all reverted
• Halting problem: 2 edits, both reverted.
• Fermat's last theorem: 1 edit, reverted
• Soliton: 2 edits, both reverted
• Wiener's tauberian theorem: 1 edit, reverted
• Cousin prime: 1 edit, reverted
• List of real analysis topics: 1 edit, reverted

Of these 113 edits, there are at least 88 reversions, which is 78% of the edits listed above, or 36% of all edits logged.

He/she was even reverted twice on his/her own talk page.

WAREL has been reverted by at least 17 distinct editors: User:Jitse Niesen, User:JoshuaZ, User:Dmharvey, User:EJ, User:Schildt.a, User:Arthur Rubin, User:ANTI-WAREL, User:Oleg Alexandrov, User:Elroch, User:Mfc, User:Trovatore, User:Zundark, User:Fropuff, User:Fredrik, User:Paul August, User:KSmrq, User:Melchoir, many of whom you will recognise as being respected contributors to mathematics articles.

On the other hand, I note that WAREL has also made several nontrivial, non-reverted contributions to several mathematics articles: Riemann hypothesis, Perfect number, Hilbert's fifth problem, Perfect power, Proof that the sum of the reciprocals of the primes diverges. He/she also makes plenty of edits to articles in which I am not competent, especially relating to Japanese mathematicians and musicians. Therefore, in my opinion, a permanent block is not (yet) warranted, even given the fact that he/she was permanently blocked on the Japanese wikipedia.

Dmharvey 01:23, 20 March 2006 (UTC)

I wrote a note on his talk page a few days ago about his revertions at decimal representation, and Jitse wrote one today about perfect number (see User talk:WAREL).

I have a silly suggestion. How about writing a petition on his user talk page, telling him that if he engages in any disruptive activity again, at any article, he will be blocked for 12 hours? Then we could all sign it, and then, should he disrupt again, any of us administrators would be able to block him with a clear heart. Wonder what you think. Oleg Alexandrov (talk) 06:53, 20 March 2006 (UTC)

Your suggestion is not silly. I think it would be important to emphasise in this petition that although some of his/her contributions have been appreciated, his/her almost complete disregard for other editors' opinions is not. I've spent enough time on this now; if someone else writes it, I will sign it. Dmharvey 13:11, 20 March 2006 (UTC)

<math> rendering bug

Just noticed at perfect number (at the bottom of the section on odd perfect numbers), this math tag:

 <math>2^{4^{n}}</math>

is getting rendered as this html:

 <span class="texhtml">2<sup>4</sup><i>n</i></span>

to appear as:

 24n

.. which is clearly wrong.

I wasted some time tracking down the paper to check the clearly wrong result before realising that it was the rendering rather than the text that was at fault. I don't know if this is a well known bug, but a brief search on Mediazilla didn't throw up any candidates. I have reported it to the Wikitech-l mailing list mailing list. Hv 16:12, 20 March 2006 (UTC)

I've noticed this before. It's actually not a bug in the LaTeX => HTML converter. It has to do with HTML tidy, which is a program that processes the HTML after the converter is done with it. The correct translation would be something like 2^4ⁿ. I think what happens is that HTML tidy sees the second ^{and assumes that the author forgot the slash. So it inserts an extra slash producing 2⁴n}</sup>. Then it sees the next </sup> and can't find a matching ^{so it kills that one too. Finally the last} dies. This is just a theory, but I'm pretty sure that texvc gets the conversion right in the first place. Dmharvey 18:30, 20 March 2006 (UTC)

See for example http://bugzilla.wikimedia.org/show_bug.cgi?id=108. Dmharvey 18:35, 20 March 2006 (UTC)

Thanks for the pointer. Is this HTML Tidy we're talking about? Because if so I'm surprised there's no mention there that it is being used on WP. (I also had a quick browse of the HTML Tidy bugs database, and saw no related item there.) If not, can you point me at some details of the HTML tidy you mean? I'd like to track this problem further ... Hv 19:52, 20 March 2006 (UTC)

Yes, that's the Tidy I mean. There is a flag $wgUseTidy in the mediawiki source which enables use of HTML Tidy. I'm pretty sure they use it on WP itself. You could try asking User:Jitse Niesen, I know he's at least one person who's been thinking about Tidy recently :-) Dmharvey 20:16, 20 March 2006 (UTC)

Indeed, I do know about it. This is fixed in the current version of HTML Tidy, but that is not yet installed on the MediaWiki servers. Details are in mediazilla:599. I haven't yet seen your post to the mailing list (perhaps it's help up in a queue), but the solution is to upgrade HTML Tidy. -- Jitse Niesen (talk) 23:16, 20 March 2006 (UTC)

Cool, I even managed to find the changelog that fixed it ([2]) but I guess that's redundant now. (I also followed up with a "never mind" to my wikitech mail, so it may never get through to the list.) I look forward to the new version. Hv 23:23, 20 March 2006 (UTC)

Decimal representation or decimal expansion?

There is a discussion on which name is more appropriate at talk:decimal representation. Comments welcome. Oleg Alexandrov (talk) 03:39, 21 March 2006 (UTC)

Problem at transfinite number

There is an editor, User:Jagged 85, whom you may recognize as being interested in the contribution of Indian mathematicians. At transfinite number he has been making edits that attribute the concept to certain ancient Jaina mathematicians/philosophers. The evidence presented is, in my estimation, of the sort that would be accepted only by someone who either has an agenda, or who does not really understand the contemporary concept. I'd appreciate it if some interested folks would drop by and take a look. --Trovatore 21:46, 22 March 2006 (UTC)

I fully agree with your assessment. In fact, I'll go further: this is obvious crackpotism. Various ancient philosophers have made dubious or meaningless claims about infinity (I had found a quote by Aristotle stating that the number of grains of sands on a beach was "infinite"), but none of them corresponds to what we now view as transfinite numbers; and Indian mathematicians were so proud of their invention of the decimal system that they had fun writing very large numbers as cosmic cycles, and sometimes they confused them with infinity, but obviously this has nothing to do with the modern concept. I support any move toward removing the incriminated section. --Gro-Tsen 22:04, 22 March 2006 (UTC)

It is no more (and no less) nonsense than Galileo's work on infinite numbers, in which he found that the natural numbers were equinumerous with a subset (the set of squares) and recoiled in horror. It is not the transfinites. Septentrionalis 22:41, 22 March 2006 (UTC)

Well, if it could be documented that the Jaina had the notion of equinumerosity (as witnessed by one-one matching), that would already be a step in the right direction, though I still don't think it would be enough to use the word "transfinite". As I understand it the historical context is that Cantor didn't want to use the word "infinite" because he was talking about things that were not absolutely infinite. They were trans-finite, beyond a limit, but not in-finite, without limit. That last sentence may be a bit of retrospective etymology on my part, but I think it really is the basic idea, whether or not Cantor had that specific etymological reasoning in mind. --Trovatore 22:46, 22 March 2006 (UTC)

A section reviewing the general history of eastern and western ideas about infinity, including Aristotle's ideas, as well as Gaileo's shock, would not be out of place somwhere on WP. We do, after all, have Category:History of mathematics and the topic of infinity, just like the question "what is four dimensions", was a legit intellectual excercise over the millenia. No doubt Immanuel Kant had some pronouncemnts as well. linas 00:54, 25 March 2006 (UTC)

Never mind. That article exists, its called infinity, and the Indian stuff should be moved there. linas 01:02, 25 March 2006 (UTC)

Another tedious orthography question

Vladimir Arnold or Arnol'd? Vladimir Drinfel'd or Drinfeld? We should be consistent: and preferably across all references to them in WP. (In both cases we currently use the apostrophe sometimes, but far from consistently.) —Blotwell 06:46, 24 March 2006 (UTC)

For Арнольд, we may as well defer to the way it appears on his books and web page, "Arnold". --KSmrq^T 02:02, 25 March 2006 (UTC)

Transliteration of the soft sign ("ь")—which does not so much represent a sound as a modification—is problematic, and conventions vary. But for names, it appears that in a context like this, appearing before a consonant, it would typically be omitted. Wikipedia allows us to choose that one as primary, for the article name, and use redirects for the variants. --KSmrq^T 18:35, 25 March 2006 (UTC)

Springer Encyclopaedia of Mathematics

I just stumbled across the Springer Online Encyclopaedia of Mathematics it claims to be

the most up-to-date and comprehensive English-language graduate-level reference work in the field of mathematics today. This online edition comprises more than 8,000 entries and illuminates nearly 50,000 notions in mathematics

and seems to live up to its description. It seems like this could be a useful resouces for many articles. --Salix alba (talk) 00:14, 25 March 2006 (UTC)

Yes, and its pretty good too, at least for the 3-4 articles I looked at. I created a template fr this, which may be usd as the following (for example:) {{springer|id=f/f041440|title=Fredholm kernel|author=B.V. Khvedelidze, G.L. Litvinov}} which results in

B.V. Khvedelidze, G.L. Litvinov (2001) [1994], "Fredholm kernel", Encyclopedia of Mathematics, EMS Press

linas 00:47, 25 March 2006 (UTC)

Would be a good idea to add those entries to Wikipedia:Missing science topics. I will try to look into that these days. Oleg Alexandrov (talk) 06:45, 25 March 2006 (UTC)

First article I hit was the normal distribution [3] I was quite disappointed in that it doesn't have a single graph of it. That said, it'd be worth copying the index into a new article or added to the missing science topics. Cburnett 06:56, 25 March 2006 (UTC)

No, you can't do that; this came up before with MathWorld. It's a copyright violation.

The Springer encyclopedia seems pretty weak in set theory. --Trovatore 07:29, 25 March 2006 (UTC)

Also compare the article on Self-adjoint operator in WP to the one in Springer. Tell me which one is better.--CSTAR 14:45, 25 March 2006 (UTC)

Ours is definitely more self-adjoint:

C^*=C.

Oleg Alexandrov (talk) 19:30, 25 March 2006 (UTC)

Is it worth an article SpringerLink Online Encyclopaedia of Mathematics? --Salix alba (talk) 20:21, 25 March 2006 (UTC)

I would think it's probably worth an article (I never heard of it before this discussion, but we're not talking about something put up by some random hobbyist; this is Springer). The issue is how to write a neutral review that's not original research. That's a problem to which I have not thought of any good answer (it's why I slapped my own article on Kunen's book, Set Theory: An Introduction to Independence Proofs, with an OR tag). --Trovatore 20:53, 25 March 2006 (UTC)

See what reviews it has in the scholarly press. Scholar.google.com should have something (this should solve the Set Theory problem, anyway.) If that fails, it can be put in WP space, as a resource. Septentrionalis 21:26, 25 March 2006 (UTC)

They have a lot of great articles. They're beating us in a lot of areas, and already kick the crap out of mathworld (soon it'll be time to put mathworld out of its misery). However, have you seen their diagrams? Complete garbage! -lethe ^{talk +} 17:23, 26 March 2006 (UTC)

I have merged their lists of entries into the Wikipedia:Missing science topics. I highly doubt that this is a copyright violation in any way, as while their lists may be copyrighted (the order of entries I guess :), individual items in the list are not, and after merging together the mathworld links and the springer links and removing the bluelinks, little if any resemblance is left to their orginal lists.

By the way, I brought some order in that Wikipedia:Missing science topics by completing incomplete entries (mathworld had those), putting things in lowercase, regularly removing the bluelinks, and providing links to google search and google books for each entry. Those lists can be rather good at suggesting new redirects, new articles, or judging where we are lacking. Oleg Alexandrov (talk) 21:28, 26 March 2006 (UTC)

Actually, I will send Springer an email asking if they mind using their list as a resource for our redlinks list. Just to be safe. :) Oleg Alexandrov (talk) 00:19, 27 March 2006 (UTC)

I've looked things up in the library's copy one or two times; good to see I don't have to go all the way there now... :-) Anyone know if the online edition differs significantly from the one in print? Fredrik Johansson 00:32, 27 March 2006 (UTC)

none of the springer links seems to work. how does one get to it from the springer website? thanks. Mct mht 07:15, 5 April 2006 (UTC)

The Springer server is down every now and then. Will come back eventually. Oleg Alexandrov (talk) 02:36, 6 April 2006 (UTC)

blahtex 0.4.4 released

Major changes since 0.4.3 are:

• support for Japanese and Cyrillic in PNGs
• much faster PNG output, because we're using dvipng rather than dvips/imagemagick

Useful links:

• test wiki at http://wiki.blahtex.org (recently got attacked by spammers :-))
• page illustrating blahtex's features
• blahtex home page
• updated error list (currently 337 errors)

Dmharvey 14:10, 26 March 2006 (UTC)

The dx in \int f(x) dx doesn't look right in the MathML output (it's rendered "d x"). Fredrik Johansson 14:45, 26 March 2006 (UTC)

Which browser+version are you using? This was a known problem with earlier versions of Firefox. Dmharvey 14:48, 26 March 2006 (UTC)

Firefox 1.5.0.1. Fredrik Johansson 14:58, 26 March 2006 (UTC)

Hmmm... does the same thing happen at all font sizes? Dmharvey 20:54, 28 March 2006 (UTC)

Normal, no style, enlarged.

Essentially. Increasing the text size a few times doesn't change the absolute width (it stays at 3 pixels); it looks normal if I use an obscenely large font. By the way, the space gets one pixel narrower if I disable the page CSS style (but still looks too wide, though this could be in my imagination). See image. Fredrik Johansson 21:28, 28 March 2006 (UTC)

That's a bummer. Thanks for pointing this out. I looks like Firefox is interpreting the "d" and "x" as belonging in separate "frames" and doesn't want to overlap them; therefore because the "d" is italicised and tall, it pushes the "x" to the right. I'm not totally sure about this, especially since there's a one pixel overlap in your second example, but that could just be some rendering thing that happens after the frames have been positioned. I will put it on my list of bugs to pursue; it's probably something that the Firefox folks will need to deal with. Dmharvey 03:39, 2 April 2006 (UTC)

gradient issues

There is some disagreement on what to include in the gradient article. It is argued by some parties that it should be a disambig. Comments welcome at talk:gradient#Should gradient be a disambigutation page? Oleg Alexandrov (talk) 17:20, 26 March 2006 (UTC)

Programs for linear algebra illustrations

What programs would people around here recommend for making images to illustrate geometry and linear algebra concepts (and the like)? I'd like to manually input coordinates for vector arrows, line segments, points, etc., choose colors and line styles, and output the result to SVG. Eukleides looks good, but it doesn't do 3D and I need that. Fredrik Johansson 23:45, 26 March 2006 (UTC)

Matlab gives you complete control, 3D, and output to color EPS. Here is a (free) program which it seems outputs to svg [4]. May be more. Of course, Matlab costs money, but should be available at any university, if you are in academia. Here are some pictures I made with it. Oleg Alexandrov (talk) 00:16, 27 March 2006 (UTC)

Yeah, I have access to Matlab, but not at home (not conveniently, anyway). Fredrik Johansson 00:22, 27 March 2006 (UTC)

You could learn a scripting language and roll your own tool. It shouldn't be that difficult. Dysprosia 02:41, 27 March 2006 (UTC)

Next thing you build your own rocket in your backyard, and could as well write your own encyclopedia. :) Oleg Alexandrov (talk) 04:03, 27 March 2006 (UTC)

Been there done that SingSurf, good for algebraic surfaces. It relies on JavaView which is quite good for 3D maths and is free as in beer but not speach. Also see Interactive geometry software for others. --Salix alba (talk) 12:04, 28 March 2006 (UTC)

IE compatibility

I wonder what people think of a policy of changing unicode html tokens to tex tags in order to ensure compatibility with Internet explorer browsers which apparently have problems with some unicode symbols. I guess compatibility with IE takes precedence over our own MoS guidelines, right? What do you folks say? -lethe ^{talk +} 11:53, 28 March 2006 (UTC)

We shouldn't use Unicode gratuitously in articles anyway. Unicode is far from being a ubiquitous standard, and when someone tries to edit in something that isn't Unicode capable, it screws up the entire article. That's not good behaviour. Dysprosia 11:57, 28 March 2006 (UTC)

When I work on my Windows laptop I don't see some Unicode characters on Wikipedia, even though I use Firefox and not IE. I guess it is a problem of missing fonts more than browser.

Changing unicode to LaTeX may be a huge amount of work, and may yield expressions which are a mix of both html and TeX. It would be fine I think if people do it on a case by case basis, but I would not be sure about making that a policy.

To comment on Dysprosia's comment, Unicode is a fact of life on Wikipedia given interlanguage links and foreign names/words. Luckily not that many browsers screw Unicode anymore, maybe just Lynx or really old browsers. Oleg Alexandrov (talk) 18:52, 28 March 2006 (UTC)

Well, I happen to use Lynx some times when I don't have access to a graphical browser, or (less often for me), when I use other operating systems I may use a browser that may not support Unicode. I'm not saying that Unicode should be completely removed from articles, it just shouldn't be used when there are other more portable equivalents out there that won't be mangled if someone edits with something that's not Unicode compatible. For example, one shouldn't just use a Unicode alpha when an α will be just as suitable. Dysprosia 22:55, 28 March 2006 (UTC)

I don't understand your example. Isn't that a unicode alpha that you've displayed? We shouldn't use unicode when unicode will suffice? -lethe ^{talk +} 23:04, 28 March 2006 (UTC)

No, it's a HTML entity, edit the section and have a look: &alpha; renders as α. Dysprosia 23:08, 28 March 2006 (UTC)

Oh, I see. But uh, don't the web browsers render the HTML tokens with unicode? I thought they did, and so therefore HTML tokens and UTF-8 text are equivalent (for viewing purposes). Or am I mistaken? -lethe ^{talk +} 23:11, 28 March 2006 (UTC)

The difference is that the Unicode alpha is just another character in the text, like "t", or "q". The HTML entity is the string "&alpha;". All good computer systems should support ASCII, and the HTML entity consists of only ASCII characters, so no matter if you use a computer that supports Unicode or if you don't, the string will be unchanged. However, some browsers that don't support Unicode simply ignore the Unicode characters, so if someone edits with one of those browsers, it will look like all the Unicode characters in the article have suddenly disappeared. If the browser chooses to render "&alpha;" with a Unicode character, that's fine, but it doesn't mean that that Unicode character is somehow equivalent to the HTML entity -- they aren't. Hope that explains things a bit better... Dysprosia 23:16, 28 March 2006 (UTC)

Yes, I understand now. UTF-8 text will get lost in the edit box by some browsers, even though it renders the same. Thank you for explaining. -lethe ^{talk +} 06:58, 29 March 2006 (UTC)

Replacing Unicode would be bad policy. This question was already decided when the wiki software switched over to UTF-8 as a standard. The world has gone Unicode, and that includes even standards-flouting Microsoft. To the best of my knowledge, all contemporary browsers can display Unicode characters if configured with adequate fonts. Usually Code 2000 will suffice. --KSmrq^T 21:29, 28 March 2006 (UTC)

I brought this up because some user went on a crusade to replace all instances of ℵ with $\aleph$ inline and display mode alike. I didn't like it, but apparently IE doesn't display ℵ correctly even if you have a font for it (which we learned because it displays if he changes web browser). -lethe ^{talk +} 23:08, 28 March 2006 (UTC)

PNG shouldn't be used inline. Dysprosia 23:10, 28 March 2006 (UTC)

That is also my opinion, but do we not have an obligation to lower our standards to support IE? Some might say we do. -lethe ^{talk +} 23:16, 28 March 2006 (UTC)

The HTML entity ℵ looks like it works... Dysprosia 23:25, 28 March 2006 (UTC)

Are you saying that ℵ displays differently from ℵ in IE? Septentrionalis told me once that he couldn't see ℵ correctly (I don't know for sure what setup he was using). --Trovatore 23:33, 28 March 2006 (UTC)

Doesn't work for me, either. I certainly prefer ℵ, regardless, as it's difficult to distinguish ℵ from the Hebrew letter by inspection if they were in Unicode, and those may display differently on different browsers. — Arthur Rubin | (talk) 23:36, 28 March 2006 (UTC)

I'm saying that ℵ should work on IE, that is, it should actually display. It shouldn't matter that much that it "looks different". I don't have IE so I can't check this. Dysprosia 23:56, 28 March 2006 (UTC)

I do not see the point of distinguishing ℵ from the Hebrew letter. Next we will be wanting an α different from alpha. I'm using a computer in the same cluster; both ℵ and ℵ now display well (and almost identically) in this IE set-up, but the second is a little square box in the edit window. Septentrionalis 00:08, 29 March 2006 (UTC)

alefsym definitely looks better alongside roman text than a Hewbrew aleph. The Hebrew aleph is too big. Do you not also find it so? -lethe ^{talk +} 07:05, 29 March 2006 (UTC)

I don't understand what you're talking about. If you want an aleph, you have ℵ, which actually does work. Dysprosia 00:10, 29 March 2006 (UTC)

OK, when I reboot into Windows to look at this in IE, I just see a square for the ℵ character. This is in IE 6.0.2900.someothernumbers, SP2, WinXP Home Edition, Version 2002, SP2. I suppose to really figure out what's going on I should say what fonts I have installed, but there are too many to conveniently list. --Trovatore 00:43, 29 March 2006 (UTC)

It's probably not a font issue, since if you try another browser on the same system, it will display. It's an IE issue. Now the question is, do we want to replace inline HTML token/UTF-8 with tex to support IE? -lethe ^{talk +} 07:05, 29 March 2006 (UTC)

I am the "user [who] went on a crusade to replace all instances of ℵ with $\aleph$". I was just replacing characters which I could not read with IE in those articles which I was trying to clean up for other reasons. alefsym causes the same problem as "ℵ" in IE. Also there is an element symbol which does not display correctly; and a proves symbol. Although these are rare. Oddly, I think that the actual Hebrew letter aleph works (at least I see the Hebrew letters OK in Google when I switch languages). JRSpriggs 05:24, 29 March 2006 (UTC)

Why do novices constantly "fix" things that obviously are not broken for most people? If the Unicode characters are in the article, there is nothing wrong with the characters for the author, and presumably for most readers. Adjust your own browser, your fonts, your configuration. Common sense and common courtesy suggest you at least ask before launching an ill-conceived massive alteration campaign—especially if you haven't been editing long enough to create a User page!

Suggestion: Look at this page and adjust the things under your control so you see as few missing characters as possible. (Note: For me, none are missing. Again, I highly recommend Code 2000.) This is a page in my personal user space; do not edit it! --KSmrq^T 07:07, 29 March 2006 (UTC)

I have been editing here for about two months. I did not create a user page because I have no interest in talking about myself for the public. I have a User-talk page to communicate about our shared work here. You are wrong to say that these characters are "obviously are not broken for most people". Most people use Internet Explorer 6.02 or earlier. So most of our readers will not be able to read the characters in question. And remember, this is an encyclopedia for the general public, not a private domain for you and the other authors to glory in their own words. Do not worry, I will not edit your user pages. JRSpriggs 07:33, 29 March 2006 (UTC)

Don't get defensive. KSmrq has a good point. We have a community here with established conventions. You can do whatever you like, make whatever decisions you want, decide what's the best format to use in articles, but we have the same rights, and in order to keep from devolving into continual revert wars, we try to respect consensus and community guidelines. When you've been here a while, you get a stronger feeling for that. Now, obviously you feel that wikipedia has to conform to IE's capabilities. Maybe you should try to win people over to your view instead of fighting with them. At the moment, I'm on the fence, but about to fall on the other side. -lethe ^{talk +} 07:43, 29 March 2006 (UTC)

Several distinct issues are at play. One is the recurring integration of novices into the community, with the usual exuberant misstep and jaded correction. A second is the display of the rich panoply of Unicode characters, whether mathematical or otherwise, in articles as viewed with a diversity of browers and fonts. Almost always the problem is with the fonts and browser settings. The Unicode characters are here to stay, especially when BlahTeX generates MathML for Wikipedia. A third issue is what appears in edit windows. The wiki software could be conservative and convert non-ASCII characters to named or numeric entities, but a browser that can display a page with Unicode characters can probably edit them as well.

But my point is none of these. I'm genuinely puzzled by the hubris of editors who assume that the article is broken because their view of it shows missing characters, especially when the same character appears in many articles. Do they think everyone else is stupid or blind? I don't know the statistics for Wikipedia readers, but one browser watch site shows slightly over 50% IE6 users, so it would seem reasonable to assume that many people had viewed any given Wikipedia article in IE6 without complaint. Yet these editors inexplicably fail to draw that conclusion.

Which leads to a design question: Is there anything we can do to head off these edits before they occur? The insert menu already shows a large assortment of non-ASCII characters, but obviously that's not enough of a hint to some editors. Should every article page have a prominent link to help with missing characters? --KSmrq^T 09:10, 29 March 2006 (UTC)

I know what we need. Here pages that use indic fonts include a template which indicates that they're being used and that if you want to view the page, you have to make sure your system is ready. If we want to use stuff in a math article which doesn't have widespread support, we could have a template like that one. That would probably keep new editors from changing font stuff, right? -lethe ^{talk +} 12:26, 29 March 2006 (UTC)

This page contains Indic text. Without rendering support, you may see irregular vowel positioning and a lack of conjuncts. More...

What about the difference between ''x''&sup2; x² and ''x''² x²? I'd say the latter looks better on my screen. --Salix alba (talk) 23:39, 29 March 2006 (UTC)

Only a few superscript characters have Unicode points, so consistency weighs in favor of the <sup> tags. For example, look at x²x³ = x⁵ versus x²x³ = x⁵. Similarly, a few special fractions have Unicode points, while most do not. For example, compare ¾ (entity frac34) to ³⁄₇ (using entity frasl, and tags <sup> and <sub>). --KSmrq^T 01:21, 30 March 2006 (UTC)

Move of "Ruler-and-compass constructions" to "compass and straightedge"

John Reid moved the article "Ruler-and-compass constructions" to "Compass and straightedge". As the article currently stands, I think there are problems with the new name. I intended to move the article back to its original name, until we can reach a consensus, but I inadvertently left out the hyphens and moved it instead to Ruler and compass constructions. Please share your views on any of this at Talk:Ruler and compass constructions. I will volunteer to make any necessary changes after we arrive at a consensus about what to do. Thanks — Paul August ☎ 17:58, 28 March 2006 (UTC)

Poll on "ruler" vs "straightedge"

Some of us can't agree on how to properly call the article Ruler and compass constructions, with the other option being Compass and straightedge. "Votes" at Talk:Ruler and compass constructions are solicited. :) Oleg Alexandrov (talk) 21:49, 29 March 2006 (UTC)

Neusis

Please see Jim Loy's angle trisection page. He shows a few methods using forbidden tools; I call your attention to the so-called tomahawk and to the movable, marked carpenter's square. Is the use of these tools not equivalent to neusis? John Reid 18:33, 31 March 2006 (UTC)

Please! Neusis? Yes? No? John Reid 19:59, 7 April 2006 (UTC)