Wikipedia talk:WikiProject Mathematics/Archive/2013/Jan

Binary numeral system/Binary number system

In this edit, User:Apteva moved Binary numeral system to Binary number system. The edit summary mentions what I surmise may be Google hits. I think if an error is widespread, a Wikipedia article might help correct it. Students often write

${\frac {-b\pm {\sqrt {b^{2}-4}}\ ac}{2a}}$

when they ought to write

${\frac {-b\pm {\sqrt {b^{2}-4ac}}}{2a}}.$

If we found that error to be as widespread on the web (and I suspect it's out there in lots of places) as it is in turned-in homework assignments, would we follow the erroneous usage here?

(We had an article titled Number system that carried a hatnote saying that it is not about numeral systems. On March 17, 2012, User:Mlm42 redirected that to Number, and the redirect remains intact. We put a lot of work into finding links to number system that ought to link to numeral system, and there don't seem to be any of those now.) Michael Hardy (talk) 19:13, 19 December 2012 (UTC)

I strongly agree that "numeral" is correct. Is the right approach to find textbooks, articles, etc. that use "numeral", and then to redirect the "number" article to the "numeral" article? Mgnbar (talk) 19:55, 19 December 2012 (UTC)

Binary numbering system is used far more commonly in books than binary numeral system.[1] There is only one article. The discussion is on what it should be called. Binary number system is the most common term used.[2] Correct to one particular post-doc or mathematics professor is not important. What is important is what is most commonly used. Apteva (talk) 20:04, 19 December 2012 (UTC)

From the dictionary: wikt:numeral; noun 1. a word, letter, symbol, or figure, etc., expressing a number; number: the Roman numerals. Apteva (talk) 20:09, 19 December 2012 (UTC)

The policy on common names says that the title should agree with the most common use in reliable sources. I find that a Google Scholar search on "binary number system" turns up 4150 results as opposed to 391 for "binary numeral system". Of course, search results depend a lot on the choice of search string. Michael Hardy, what is your source for saying that "binary numeral system" is correct and "binary number system" isn't? RockMagnetist (talk) 20:10, 19 December 2012 (UTC)

Relatedly, what is the point of including "system" in the title? Why isn't it just binary number (or binary numeral, both currently redirects)?— Preceding unsigned comment added by David Eppstein (talk • contribs) 20:13, 19 December 2012‎ (UTC)

I can answer the question as to the source - it is 20 years of studying mathematics and having a PhD in Mathematics. A number is only a number. A numbering system is the mathematics used for calculation using that particular base. Apteva (talk) 20:16, 19 December 2012 (UTC)

Yes, you have explained the difference between a numeral and a numeral system, but you haven't answered my question. I didn't ask "what is a numeral system", I asked why the title should be about the more complex object (the system) rather than about the simpler objects (the numerals). —David Eppstein (talk) 22:21, 19 December 2012 (UTC)

Are you answering my question on sources for yourself or Michael Hardy? Personal qualifications are not relevant to Wikipedia policy. RockMagnetist (talk) 20:24, 19 December 2012 (UTC)

Subject experts are often needed and are always welcome. Apteva (talk) 20:28, 19 December 2012 (UTC)

Absolutely - but citations are still needed (and subject experts should be good at finding them). RockMagnetist (talk) 20:30, 19 December 2012 (UTC)

So I would choose the title "base 2" though,[3] except that it is not "encyclopedic". Apteva (talk) 20:39, 19 December 2012 (UTC)

You should choose something comparable like binary number compared to base 2 number compared to base two number as in [4] where you see binary number is a clear winner. Base 2 does not identify the topic well. Dmcq (talk) 03:05, 20 December 2012 (UTC)

See Talk:Binary numeral system#Requested move. Apteva (talk) 20:28, 19 December 2012 (UTC)

"Number system" is a misnomer, as shown by the dictionary definitions quoted by Apteva: a word, letter, symbol, or figure, etc., expressing a number. Michael Hardy (talk) 04:44, 20 December 2012 (UTC)

Lots of things are misnomers. So many that we even have a list of mathematical misnomers. But if they're in wide enough use, that's what we should call them anyway. —David Eppstein (talk) 05:11, 20 December 2012 (UTC)

That depends on how firmly entrenched it is. "Fundamental theorem of algebra" may be beyond hope, but in certain matters we can correct popular misunderstandings. For example, mathematical illiterates think "exponential growth" is synonymous with something like "surprisingly fast growth". Our article with that title sets the record straight. Michael Hardy (talk) 18:50, 21 December 2012 (UTC)

The article Number makes a clear distinction between "number" and "numeral". Moreover we have Numeral system. On the other hand, formally speaking, a "binary number" or "binary fraction" is a rational number with a power of two as a denominator. Thus, the correct title of the article would be "binary numeral system" (or "binary numeration" if one want to avoid "system"). Personally, I prefer "binary numeral system", but, if we want to name after the most common use, and if we decide the most common use after the number of Google hits, "binary number system" seems the best choice. D.Lazard (talk) 09:38, 20 December 2012 (UTC)

I think trying to use numeral in places where it is very uncommon in normal usage is just wrong in Wikipedia - we should use common language.. Decimal numeral for instance hardly ever occurs compared to decimal number, and it has been that way for the last two centuries at least according to [5] in Google ngrams. Numerals normally refers to the numbers as written using the symbols of different writing systems like Babylonian or Roman or the digits in our current system, not how they are put together. Dmcq (talk) 11:50, 20 December 2012 (UTC)

Actually "Number system" is not a misnomer. Michael Hardy has committed the fallacy of affirming a disjunct. In the OED, one of the definitions of number is "Something which graphically or symbolically represents a numerical quantity, as a word, figure, or group of these; a numeral; (also) a ticket or label bearing such signs." Thus, although "Number system" is less precise, it is a synonym for "Numeral system". RockMagnetist (talk) 01:16, 21 December 2012 (UTC)

What the word "number" means may vary with the context. In some contexts a "number" is a _song_. That doesn't mean it wouldn't be a misnomer if used that way in most math articles. Michael Hardy (talk) 18:47, 21 December 2012 (UTC)

The requested move has now been closed and the article moved to binary number. Probably it would be a good idea to edit the article (and especially its lead section) to discuss common usage vs correct usage here, and also to say what a single binary number is as well as what the whole system is. I suppose one advantage of the new title (not included in the discussion) is that it doesn't use the phrase "number system" so we can still use the more precise "numeral system" terminology for that while also using "binary numbers" to refer to numbers represented using this system. (Relatedly, a binary numeral is just a bit, not a multi-bit number, right? So why does it redirect here and not to bit?) —David Eppstein (talk) 04:42, 3 January 2013 (UTC)

"Bit" means "binary digit". A numeral is not necessarily just one digit, it can be multiple digits. So "binary numeral" is not limited to one bit. JRSpriggs (talk) 06:34, 3 January 2013 (UTC)

As an interested layman, I would expect the term "numeral" to refer to a typographic arrangement comprising one or more symbols, while a "number" is a mathematical idea which may be represented by the numeral. In practice, the numeral for a given number is also commonly referred to as the "number". For example in the decimal numbering system, I might write "6" where an ancient Roman would have written "VI". In binary I would write "110". Here are three different numerals representing the same number, and all commonly referred to (in their respective spoken languages) as "the number six". I'd suggest that provided we understand that the term "number..." is often used as shorthand for "the number represented by the numeral...", there should be no problem. For example WP:COMMONSENSE tells us that "No. 10 Acacia Avenue" is using the number/numeral as the name of a property. I'd also agree that we are well rid of the term "system" in the title. I don't know if that helps. — Cheers, Steelpillow (Talk) 17:17, 3 January 2013 (UTC)

Well, but the binary system is most frequently used in contexts which have very little to do with systems of writing — in computers. Are those numbers or numerals? —David Eppstein (talk) 17:27, 3 January 2013 (UTC)

I would suggest that since most patterns of information within an electrical circuit are not typographic arrangements then they cannot be numerals. But equally, they are only representations of the mathematical idea of the number, not the idea itself (I certainly find it hard to believe they might be sets of sets, for example). It would seem that, whatever the precise mathematical definition of a number, linguistically we use the term to cover a multitude of sins, without full regard to consistency. — Cheers, Steelpillow (Talk) 17:51, 3 January 2013 (UTC)

To Eppstein: Do you really mean that the common use of the binary system in computers is not a system of writing the numbers in the memory? What is the significant difference between writing on a paper sheet and writing in the memory of a computer? On the other hand one of the side advantages of computers is that the common use of several numeral systems (decimal, binary, floating point, ...) may help to clarify the important distinction between the abstract notion of a number and its representation as a numeral that has been perfectly summarized by Steelpillow. D.Lazard (talk) 18:20, 3 January 2013 (UTC)

It is a representation that is intended for calculation rather than for visual communication. In that sense it is more similar to a configuration of an abacus than to a sequence of written digits. —David Eppstein (talk) 01:53, 4 January 2013 (UTC)

QUOTE from David Eppstein: "Relatedly, a binary numeral is just a bit, not a multi-bit number, right?" END OF QUOTE.

No, a binary numeral can have more than one binary digit, just as a decimal numeral can have more than one decimal digit. Michael Hardy (talk) 19:02, 5 January 2013 (UTC)

Well, I have to admit, I still find the way you're making this distinction between numbers and numerals to be confusing. If it were "a number is an abstract mathematical object and a numeral is a written representation of a number" then fine, that would make sense to me, but then binary numbers would be numbers to me and not numerals. Because, binary numbers are abstract mathematical objects (sequences of bits) that are not primarily used as a system of writing, and that have an arithmetic that obeys the Peano axioms and other properties that one would expect integers to have. It is as if you defined a vector in a way that required it to only be an object in a coordinate-free vector space, but insisted on calling it a "vectroid" whenever it was given coordinates in a space with a basis. I don't see that as being a useful distinction. —David Eppstein (talk) 19:30, 5 January 2013 (UTC)

Multiplication algorithm

I would appreciate it if someone could give a third opinion on this addition.—Emil J. 16:28, 3 January 2013 (UTC)

Even if it were notable, and I don't believe it is, that article would be the wrong place. Something like the Mental calculation article would be more suitable. Dmcq (talk) 18:07, 3 January 2013 (UTC)

Perimeter: review wished

Hello,
I'm trying to translate French good article fr:Périmètre into English (see Perimeter, work in progress). As English is not my native tongue, a review may be useful, to correct my mistakes.

All the sources in fr:Périmètre are in French, so I didn't insert them in the English article, thinking we could find other ones in English. Did I do well? If you prefer, I can easily copy and paste French sources in Perimeter. --El Caro (talk) 16:52, 5 January 2013 (UTC)

Mathematician biography nominated for deletion

• Paul Jefferys (AfD discussion)

It won't get automatically listed by the 'bot. Uncle G (talk) 10:46, 6 January 2013 (UTC)

Trace inequalities

Trace inequalities is a new article and could probably benefit from more eyeballs. In particular: Which other articles should link to it? (I've added a few links including one from the list of inequalities, but I haven't attempted to figure out which others there should be.) Michael Hardy (talk) 18:24, 7 January 2013 (UTC)

Merging ring theory into ring (mathematics)

Hi all,

I would like to propose the merger above. ring theory has not developed well for years and most of stuff there is already in the "ring" article. There are some duplications like "history" section. I can see an argument for separate articles for "object" and "theory". But the current setup isn't this sort, and unless there are editors who want to write ring theory, there is little point having it. I'm also interested in hearing any open-ended suggestions (as to what we do about them.) -- Taku (talk) 21:28, 7 January 2013 (UTC)

As a general principle I would expect an article on Ring (mathematics) to concentrate on telling the reader about rings as mathematical objects: their structure, properties, morphisms, associated objects; whereas Ring Theory should describe the theory, in terms of historical development, applications, relationships to other theories, key results, comparisons with other theories. Deltahedron (talk) 22:28, 7 January 2013 (UTC)

My take is that this doesn't make sense. It wouldn't make sense to merge group (mathematics) and group theory or even number and number theory, (or music and music theory, for that matter).

The articles have very different purposes. The theory article should address questions like "How are rings studied and what has been discovered?" while the ring (object) article should answer questions like "what is a ring and what features and flavors do they have?".

One thing we can agree upon is cleaning up duplication between the two articles. Still though, I bet an almagam of the articles would produce a monster-sized article that is less useful than the separate articles. Rschwieb (talk) 14:28, 8 January 2013 (UTC)

I think a merger is the right thing to do, the current ring (mathematics) article is a bit long but I don't think ring theory is a reasonable subtopic article. I'd prefer to see the two merged and then have some discussion about splitting off a couple of major bits as subtopics. Dmcq (talk) 14:38, 8 January 2013 (UTC)

Agree with the first two comments -- probably some chunks of Ring (mathematics) should be moved over to ring theory (like, e.g., the sections on commutative and noncommutative algebra). --JBL (talk) 14:53, 8 January 2013 (UTC)

@Dmcq Seemingly, this logic would apply to all of these pairs, most of which are comparably sized with ring/ring theory: graph theory / graph (mathematics),knot (mathematics) / knot theory , group (mathematics) / group theory, category (mathematics) / category theory, field theory (mathematics) / field (mathematics) and several more I have not listed. If you do not agree that all these pairs should be merged, I am very interested in knowing why rings are different. Rschwieb (talk) 18:11, 8 January 2013 (UTC)

I can't say I'm keen on those. The split seems to be basic intro versus more advanced and I don't think that is a good way of going round the business. There are a number of articles like Introduction to special relativity and Introduction to quantum mechanics and they suffer from the same sorts of problems of not having a clear idea of the split and not knowing the level of knowledge needed at each stage in the articles and not being clear about how much should be included. They try to sidestep an editing problem by forking and instead created two worse problems. I think Wikipedia:Make technical articles understandable has a better approach in general. It says one can start up 'Introdyuction to ...' articles but does not recommend it in general. Dmcq (talk) 00:29, 9 January 2013 (UTC)

I think in most cases it makes sense to have separate "object" and "theory" articles. The split should not be based on more-or-less advanced, though, I agree with that. The "object" article should ordinarily be shorter than the "theory" article, but probably more "technical", not in the sense of "advanced", but in the sense that it's largely about the definition of the object and its basic properties. The "theory" article will be more narrative, and discuss what it is that interests foo theorists about foos.

On the other hand, the "object" article is the natural place to talk about applications of foos to things other than foo theory. For example, a lot of stuff that's done with groups is not group theory, and perhaps makes more sense at group (mathematics) than at group theory. --Trovatore (talk) 00:36, 9 January 2013 (UTC)

I still think there should be a hierarchy and in that case the theory should be under the object article. However if you look at the group theory article its says at the top 'This article covers advanced notions. For basic topics, see Group (mathematics).' and if you read the introduction it does not make clear that it is a subtopic or a split in the sense you say. Dmcq (talk) 00:44, 9 January 2013 (UTC)

I wouldn't argue that a ring is different. In fact, the same argument I made above applies to field theory. It hasn't developed much; "history" sections are duplicate. Why not merger it into field (mathematics)? The caregory theory is "much" more than category: think of functors, natural transformations. In contrast, the ring theory is all about rings. -- Taku (talk) 22:52, 8 January 2013 (UTC)

I am all with you on eliminating the duplication between these two articles (and any of the other pairs I mentioned, for that matter). But I still have no idea why doing that with a wholesale merger is warranted. You just delete duplicate places or abbreviate them with links to the body of the material.

Let's put the reasons for and against a wholesale merger on the table:

Against: The articles (with duplication removed) have plenty of content, and merging them would make an overlarge article. (I would also go so far as to say that putting them in the same article is shoehorning two distinct topics into one topic.)

Against: It would be inconsistent with the precedent set by a large number of pairs of articles that I mentioned above. If ring and ring theory should be merged, then so should all of those.

I haven't really seen that there is a strong reason for at all. If your reason is to remove duplicaton and overlap, then sure, by all means, that is a good thing to do... but not by a wholesale merger. The costs of the merger obviously outweigh the benefit of removed duplication.

If it helps make you feel the article theory is more useful, I would be interested in improving the ring theory article in the future, when I get some time. Rschwieb (talk) 14:55, 9 January 2013 (UTC)

You noted in the talk page for Ring (mathematics) that it might be worthwhile spelling out the purpose of the separate articles. I think this is the crux of the problem. It really needs to be spelt out or you get the current mess. I agree there is material for more than one article but we haven't done what should always be done at the start of the 'theory' articles which is say clearly what the topic of the article is. I think there would still be a need to move some stuff around in accordance with the topics. Dmcq (talk) 17:44, 9 January 2013 (UTC)

OK: that sounds like a good place to start. I'll be on the talkpage sometime soon to help with eliminating duplication and discussing content sorting. See everyone there! Rschwieb (talk) 18:10, 9 January 2013 (UTC)

Abelian algebra

User:r.e.b. has created an article in which in the first sentence we find the phrase "extensions of a commutative algebra". It's obviously a bad idea to write "extensions of a commutative algebra", since the linked article is about the study of commutative rings. But a hatnote at the top of that article says "For algebras that are commutative, see Abelian algebra." Currently Abelian algebra redirects to Algebra#Abstract_algebra. Is there a better target for that redirect? Or should a better target be created? Michael Hardy (talk) 16:50, 9 January 2013 (UTC)

I suggest to redirect it to commutative ring and to add a short mention of this additional structure to the lead of the latter article. Probably, with a corresponding section. Incnis Mrsi (talk) 16:59, 9 January 2013 (UTC)

List of things named after Johannes Kepler

List of things named after Johannes Kepler is a new article. It should get divided into appropriate sections. Michael Hardy (talk) 03:55, 1 January 2013 (UTC)

Done, by someone. Michael Hardy (talk) 17:53, 11 January 2013 (UTC)

Archimedean–Galileo spiral

Archimedean–Galileo spiral is a recent article whose deletion is proposed. Are there reasons to keep it, or other reasons to delete it than the one cited? Michael Hardy (talk) 18:57, 10 January 2013 (UTC)

As far as I can see, this is OR whose only source is [6], as reported in A202407, and as such the prod is appropriate. (Furthermore, the name appears to be only invented for Wikipedia, and I find it ungrammatical: if anything, it should be either Archimedean–Galilean spiral or Archimedes–Galileo spiral.)—Emil J. 19:58, 10 January 2013 (UTC)

I certainly agree about the grammar. Michael Hardy (talk) 17:57, 11 January 2013 (UTC)

Tsit-Yuen Lam

We currently have three links to Tsit-Yuen Lam, but the number of articles that mention him in references and otherwise is much largers, maybe dozens. Should we have an article about him? (If so, those references should have authorlinks.) Michael Hardy (talk) 17:55, 11 January 2013 (UTC)

Relevant articles exist in the German and French editions. Liege mooch (talk) 12:39, 12 January 2013 (UTC)

He very likely passes WP:PROF criteria #1 (many very highly cited publications, as counted e.g. in Google scholar), #2 (Steele Prize), and #3 (Fellow of AMS). Only one criterion would be sufficient. So I agree, an article is justified. It would probably be a better idea to stub out something for him first, rather than putting in all those authorlinks as redlinks, though. —David Eppstein (talk) 17:07, 12 January 2013 (UTC)

Is there any chance I can convince you two write "criterion #1"? Michael Hardy (talk) 00:28, 13 January 2013 (UTC)

Now it exists; here's a to-do list

I've created the article. Tsit-Yuen Lam now redirects to Tsit Yuen Lam. I translated the German article, which I think is pretty much identical to the French article (except that I moved one short paragraph to an earlier place in the article and made it part of a larger paragraph).

So the following should be done:

• Further expansion;
• Checking for accuracy by people who know things about him;
• Going to the many Wikipedia articles that mention his name and adding links to the new article.
• Probably other things as well.

Michael Hardy (talk) 02:29, 13 January 2013 (UTC)

Maybe there's aren't so many articles that mention him. I've linked from some of them to the new article. Here's a question: should one title the article Tsit-Yuen Lam or Tsit Yuen Lam? The latter is what he uses on his university web page. His publications seem to get cited using the hyphenated version, or else T. Y. Lam. Michael Hardy (talk) 02:43, 13 January 2013 (UTC)

Fundamental recurrence formulas

I suggest that Fundamental recurrence formulas should be merged either to Generalized continued fraction or to Convergent (continued fraction). There seem to be several articles covering the material already, and this one is unsourced and with a debatable name. I raised this at Talk:Fundamental_recurrence_formulas#Suggest_merger but no-one seems to be watching the article. Deltahedron (talk) 07:37, 14 January 2013 (UTC)

Seems fine to me. In fact I wonder why there is Convergent (continued fraction) either. People seem to stick in every term as if Wikipedia was a dictionary rather than an encyclopaedia. We've got redirects for that sort of thing. Dmcq (talk) 10:02, 14 January 2013 (UTC)

I would support both merging Fundamental recurrence formulas into Generalized continued fraction and Convergent (continued fraction) into Continued fraction. The first merge would benefit Generalized continued fraction as well: it includes the recurrence formula in the lead but does not discuss it in the body. RockMagnetist (talk) 17:14, 14 January 2013 (UTC)

Deltahedron, I think your experience is the reason that WP:MERGEPROP recommends discussing merges at the destination article's talk page. There are probably many more eyes watching Generalized continued fraction. RockMagnetist (talk) 17:23, 14 January 2013 (UTC)

While we're at it, Convergence problem seems too general a name for the criteria for convergence of continued fractions. RockMagnetist (talk) 17:32, 14 January 2013 (UTC)

Wikilinks in publication titles

See this edit. My feeling is that a wikilink only belongs on the title of a publication if it is a link to an article about that specific publication, because otherwise it looks too much like a link to the publication itself, and that wikilinks for topics discussed by a publication should instead be included in prose describing the publication. But I'd welcome further discussion or additional opinions. —David Eppstein (talk) 00:39, 15 January 2013 (UTC)

I think in this case, with the exception of Serre's Conjecture, the links aren't misleading because only part of the title is linked. But I agree - it's a poor substitute for discussing the work of this scientist in the body of the article. RockMagnetist (talk) 01:36, 15 January 2013 (UTC)

I agree with David. These edits seem to originate from an overzealous wikilink-er. These are titles and should be treated as wholes. If a subject that is relevent to the article appears in a title and is not already linked to in the article, then an appropriate place to put the link would be in the ==See also== section. If the subject is not relevent to the article (or to the understanding of the article) then it shouldn't be linked to in the first place (IMHO). Bill Cherowitzo (talk) 17:41, 15 January 2013 (UTC)

MathJax problem

I turned on MathJax rendering, and in general it looks good, but I noticed a problem in the articles Euclidean algorithm and continued fraction - the lines in the fractions are too long and overlap with other parts of the expression:

$a_{0}+{\cfrac {1}{a_{1}+{\cfrac {1}{a_{2}+{\cfrac {1}{\ddots +{\cfrac {1}{a_{n}}}}}}}}},$

${\frac {1}{2.25}}={\frac {4}{9}}=[0;2,4].$

If someone would notify the right parties, and tell me how to in the future, thanks in advance! - Virginia-American (talk) 19:14, 15 January 2013 (UTC)

Looks ok to me (in Chrome under OS X). Occasionally on other sites (but not here) I've had to reload a page to get mathjax to render properly without overlapping the surrounding text. —David Eppstein (talk) 19:18, 15 January 2013 (UTC)

I do not see the problem in any of three rendering modes. BTW, these "lines in the fractions" are called vincula. Incnis Mrsi (talk) 19:24, 15 January 2013 (UTC)

On a Mac, I see this problem frequently when I use Chrome, but not Safari or Firefox. RockMagnetist (talk) 20:36, 15 January 2013 (UTC)

I'm using Chrome Version 23.0.1271.97 m, under Windows 7 service pack 1 - Virginia-American (talk) 21:54, 15 January 2013 (UTC)

I'm using Chrome and I see the comma under the top line rather than after it which is a bit peculiar. Under Firefox it seems okay. Dmcq (talk) 00:44, 16 January 2013 (UTC)

Looks good to me. On stackexchange it sometimes comes out wrong when first saved, but looks OK after reloading. Michael Hardy (talk) 06:43, 16 January 2013 (UTC)

Now it's OK. - Virginia-American (talk) 12:31, 16 January 2013 (UTC)

I'm using Chrome 24.0.1312.52 m and it has a comma under the top line rather than at the end. I have the default experimental MathJax set in my preferences rather than using 'importScript('User:Nageh/mathJax.js')' in my common.js file, perhaps that's the difference. Dmcq (talk) 13:29, 16 January 2013 (UTC)

The difference, your said? It is easy to check that Virginia-American does not use any personal script. Incnis Mrsi (talk) 14:06, 16 January 2013 (UTC)

It seems that is a problem that is specific to Chrome: all users reporting it have this problem with Chrome and nobody report it on another browser. Myself, I use standard MathJax (set on the "preferences" page) with Firefox under Window 7, and above formulas look fine. D.Lazard (talk) 14:43, 16 January 2013 (UTC)

Affine function

Hello,
Affine function is a redirect to Affine transformation. Affine functions deserve their own article, don't they? Do you agree? --El Caro (talk) 14:47, 18 January 2013 (UTC)

What’s the difference between an affine function and an affine transformation? I thought they were synonyms.—Emil J. 15:20, 18 January 2013 (UTC)

In other words: the $\mathbb {R} \rightarrow \mathbb {R}$ functions defined by f(x) = ax+b deserve their own article on WP, in my opinion. What should be the title of this article? --El Caro (talk) 15:31, 18 January 2013 (UTC)

This family of functions are included as a simple case, being affine transformations on a one-dimensional space. Unfortunately the article places excessive emphasis on a geometric interpretation at the expense of a more general definition, particularly in the lead. I see no reason to have a separate article, though it may make sense to give this specific example under the Examples of affine transformations section. — Quondum 17:38, 18 January 2013 (UTC)

Possible plagiarism

Pointed out by anon at Talk:Riemannian manifold. No one responded there so, I figured I'd bring it to the attention of folks here. Mct mht (talk) 07:51, 19 January 2013 (UTC)

Proved or proven

Is "proved" or "proven" the preferred usage for mathematics articles? It seems that "proved" is more common in specifically mathematical contexts, but another editor has suggested this is a British/American difference. Is there an established convention? Deltahedron (talk) 19:15, 6 January 2013 (UTC)

There is no solid convention in mathematics, and both terms are commonly used. I would just treat them as synonyms and leave the original use unless it sounds particularly bad. — Carl (CBM · talk) 19:22, 6 January 2013 (UTC)

Thank you, but my question was specifically about whether there was a convention here on Wikipedia. Deltahedron (talk) 19:40, 6 January 2013 (UTC)

Not that I am aware of. — Carl (CBM · talk) 19:52, 6 January 2013 (UTC)

I think I tend to use "proven" only as an attributive adjective, and maybe not always then. As a past participle, I use "proved". Maybe this will go where "gotten" has gone in England, i.e. it's no longer used by respectable writers, and the ignorant masses there seem to think it's an Americanism. Michael Hardy (talk) 18:26, 7 January 2013 (UTC)

I find myself accused on my talk page of describing the English as ignorant masses. I did nothing of the sort. In every country there are ignorant masses; England is no exception. Michael Hardy (talk) 00:19, 9 January 2013 (UTC)

I wrote you appear to describe the English as "ignorant masses". This has nothing to do with mathematics and I am surprised to see the discussion continued here. Deltahedron (talk) 07:28, 9 January 2013 (UTC)

You complained to Michael that his comment here could be misconstrued as being offensive, hence Michael added here a clarification of his comment. This sounds like the most natural course of action to me.—Emil J. 13:00, 9 January 2013 (UTC)

I've encountered a couple instances where, when a point of correctness was concerned, two sides both pointed to each other and said "I think that's what they prefer." The one I remember the best is the use of "an historic," to which both Americans and English seem to say that it's the other side's preference. The other day I hit a webpage concerning the use of "matrixes" where the same finger-pointing happened between physicists (or it might have been chemists) and mathematicians. Rschwieb (talk) 14:35, 8 January 2013 (UTC)

If it’s still relevant, I would say “[name] proved [theorem]”, “[theorem] was proved”, “[theorem] has been proven”, etc. Standard English usage (I think). But if an article consistently uses one in the other’s context, go with that. —Frungi (talk) 06:38, 9 January 2013 (UTC)

I second that the verb forms are prove, proved, and proved, and that proven is an adjective. 192.35.44.24 (talk) 19:27, 22 January 2013 (UTC)

Participles are adjectives, so that's not really such a clear distinction. I would avoid making any firm rulings on this. For things that sound outright wrong, like "a proved fact", go ahead and change it, but for things that are reasonable, even if it's not the way you (generic you) would personally say it, just leave them alone. It might be partly an ENGVAR issue but it's clearly not a clear enough one to base the decision on whether the article as a whole is in American English or British English, so mostly just leave it alone.

As for Frungi's distinction, it does somehow sound better that way, but it's weird because there is no grammatical distinction. In both cases prove{d|n} is being used as a passive, so really it should be the same form in both cases — the verb form of to be is irrelevant to the verb form of prove. --Trovatore (talk) 21:13, 22 January 2013 (UTC)

König's theorem: two versions and implicit use of induced subgraph?

Hello,

I very much enjoyed the article König's theorem (graph theory) but I have some comments/questions:

It also mentions König's Line Coloring Theorem, which looks very different. I don't know if one of these theorems follows from the other, but now there is no comment on this in the article.(Perhaps the other deserves a separate article?)

I also thought the treatise of perfectness of graphs was interesting. The definition of perfect graphs relies on induced subgraphs, however, here only the full line graph (and its complement) of a bipartite graph are considered. Shouldn't it be explained why this does cover everything?

Many thanks, Evilbu (talk) 16:32, 21 January 2013 (UTC)

Willard Gibbs

Hi. I've been working for a while on the bio. of Josiah Willard Gibbs. First, I'd like to mention that I think the article should of rated as of High (rather than Mid) importance in math. As a mathematician we owe him the concept of the cross product, the formulation of vector calculus in terms of div, grad, curl, and all that, and many of the ideas of what later became convex analysis. And Gibbs is one of the people most responsible for rescuing the work of Hermann Grassmann from obscurity. His work on statistical mechanics also had a very deep and long-lasting impact on mathematical physics.

Second, I'd like to attract other editors who might wish to work on improving the article. Last Nov. I nominated it for FA. It got significant support, but some of the FA review regulars kept asking for independent copyediting and in the end the discussion just stalled and was archived. Gibbs's work as a theoretical scientist is extremely interesting and important, and I think his wikibio deserves more attention. - Eb.hoop (talk) 23:46, 17 January 2013 (UTC)

I agree high importance does seem appropriate for him. Dmcq (talk) 16:28, 24 January 2013 (UTC)

Need help at set theory

User:Multipundit is insisting on spamming the set theory article with an enormously disproportionate quantity of material on multisets and fuzzy sets, two topics that are extremely minor areas of concern to set theorists. Please take a look. I have not attempted to engage him because I don't think it would go well. --Trovatore (talk) 04:43, 18 January 2013 (UTC)

The article Named set could probably also do with some attention. Deltahedron (talk) 07:38, 18 January 2013 (UTC)

A few days later, this article now reads as a personal essay on the work on Burgin, making grand claims which do not appear to be supported by references to any independent assessments in mainstream scientific literature. Deltahedron (talk) 07:32, 24 January 2013 (UTC)

I just read the named set article and it strikes me as verbose and lacking in substance and I don't see why the whole lot of it can't just be treated as just one functor in category theory. The words 'vacuous waffle' come to mind. Dmcq (talk) 16:21, 24 January 2013 (UTC)

The editing history of User:Multipundit shows the account has some connection to Burgin, at least as an advocate of Burgin's work. Taking that for granted, I don't think the current edits are completely bad. I would support trimming down the named set article somewhat, but I think it is generally OK, and the detailed references are useful. The lede is particularly bad. For the main article set theory, the due weight of named sets is extremely small. — Carl (CBM · talk) 16:33, 24 January 2013 (UTC)

Hyperbolic Fibonacci and Lucas functions, aka a new maths article that's too "essay-like" for creation at the moment

Hullo mathematical peoples. A new editor has written Wikipedia talk:Articles for creation/Hyperbolic Fibonacci and Lucas functions, which looks very well referenced (I've pointed him to how to do inline citations properly). However, it's been rejected as "more like an essay than an encyclopedia article", even though I don't see distinctive signs of WP:OR in it. This is a rejection reason that's notoriously difficult to get past - can anyone help in assessing the suitability of the article in its current form, or in improving it to the point where it's acceptable?

(If you think, at any point, that in fact it's not too essay-like, but don't want to bother with the mechanics of the WP:AFC creation process, just let me know or say so here and I'll accept/create it, as I'm otherwise happy with it.) --Demiurge1000 (talk) 21:24, 18 January 2013 (UTC)

Keep this article. I question the whole process of article rejection because someone (who are they and what gives them this authority) decides that an article can be rejected on such grounds. It's terribly offputting for new authors. Has the deletionist regime got such a firm grip on this site that they can even get stuff deleted before it's even had a chance to get off the ground? --Matt Westwood 06:21, 19 January 2013 (UTC)

Comment. The article does read like an essay, and an essay based on the work of a small number of people. There is a disturbing similarity to this paper by two of that small group. The mathematics is not nearly as novel as is claimed and there is far more to the history of these functions. The wider question about Article Creation is interesting, but there is no requirement to go through that process. Anyone can simply write an article directly if they want. If they want other editors to comment first, they can ask, as happened here. Deltahedron (talk) 07:49, 19 January 2013 (UTC)

I find the idea of a link between the Fibonacci/Lucas numbers and the sinh/cosh functions to be interesting. Can we not keep it? JRSpriggs (talk) 08:56, 19 January 2013 (UTC)

I agree that it is interesting, and have thought so ever since first coming across it years before the work reported on in this article. It was certainly known to Édouard Lucas, and goes back to Lagrange and Legendre: see page 71 of Williams (1998). Any article on this subject should give a proper history of these ideas. Deltahedron (talk) 09:13, 19 January 2013 (UTC)

Deltahedron, I wanted to firstly thank you for thanking the time to review my article. I would like to first address your concern that there is a "disturbing similarity" to a paper by Alexey Stakhov and Boris Rozin. Their work is fundamental to the modern day developments of this subject therefore it played a major part in the creation of this article but it is definitely not the only source of information. The other concern is that this article is written in an essay style. This is my first article on wikipedia and I am still only a beginner at this, so still learning how to write in a more encyclopedic manner. I would be really grateful for any constructive suggestions on what to improve in the article, or if experienced editors could optimise what is already there. You also mention that this topic goes back to Édouard Lucas, Lagrange and Legendre, it would be great if you could possibly add some information regarding that as I am not that well informed about their works on this subject. All in all, I am glad you find this topic interesting and hope with the help of experienced editors like your good self and others we will be able to refine this article to a proper wiki standard. Matt Westwood & JRSpriggs thank you for the support. WIKIWIZDOM (talk) 22:38, 20 January 2013 (UTC)

Thanks to all for the feedback. I've accepted the article and created it at Hyperbolic Fibonacci and Lucas functions; I've also added templates indicating the current problems with it, and linked to Deltahedron's comments from its talk page. --Demiurge1000 (talk) 20:20, 19 January 2013 (UTC)

Demiurge1000 thank you very much for your help. WIKIWIZDOM (talk) 22:38, 20 January 2013 (UTC)

This article is about functions of the form $a\cosh(x\log b)$ and $a\sinh(x\log b)$. The integer sequences that appear in the article (λ-sequences) are special cases of Lucas sequences. Does this deserve a separate article? I have commented this in more details on the talk page D.Lazard (talk) 17:03, 22 January 2013 (UTC)

I have proposed the deletion of this article with the following concern:

The subject of this article on mathematics is not notable. The mathematical content of the article consists in results that are trivial consequences of known theories, typically that of linear recurrences. When these known theories are cited in the article, this results always of my edits. Instead of referring to knowns theories, the article cites only non-notable publications that, for most of them, are not reliably published. The part of the article devoted to phylotaxis is a blatent WP:fringe theory and I suspect that it is also pseudo science. See the talk page for more details.

D.Lazard (talk) 17:46, 24 January 2013 (UTC)

The prod tag being removed I have nominated this article for deletion, with the same concern D.Lazard (talk) 23:42, 24 January 2013 (UTC)

Moebius plane and Möbius plane

user:Ag2gaeh has created a new article titled Moebius plane when there was alrady an article titled Möbius plane. Both spellings are correct and in one sense equivalent (whereas Mobius is by some reasonable standards a third, and incorrect, spelling). The initial edit summary when the article was created suggests that the creator knew of the other article but simply preferred spelling it without the umlaut. The proper way to handle that situation is with the "move" button. I've put "merge" tags on the two articles. (I haven't yet looked to see what links to them.) Michael Hardy (talk) 18:35, 19 January 2013 (UTC)

I've gone ahead and changed Moebius plane into a redirect. Anyone who wants to merge content from the previous version into Möbius plane is welcome as far as I am concerned. --JBL (talk) 20:23, 19 January 2013 (UTC)

I undid that because it was not done properly. I recommend giving the author of Moebius plane time to respond and then (if there is still consensus) following the procedure in Help:Merge. RockMagnetist (talk) 20:44, 19 January 2013 (UTC)

That the articles need to be merged is obvious (well, unless by some chance there are two notions called "Möbius plane", which I doubt, but I haven't looked at the articles). So I think an RM is just unnecessary proceduralism here. Certainly, the major contributors might have views on how the merge should be done, and a courtesy notice on that point would not be out of line. --Trovatore (talk) 21:00, 19 January 2013 (UTC)

What is the point of pretending that the consensus will be anything other than for the author of the new stuff to edit the existing article? Moebius plane should be and will be a redirect sooner or later and anyone who wants to see earlier versions can go to the history; all you've done is to perpetuate a ridiculous situation. Also, please see Bill Cherowitzo's comments at Talk:Moebius plane: the material you've restored is almost certainly copyrighted. --JBL (talk) 21:07, 19 January 2013 (UTC)

(To clarify, JBL is responding to RockMagnetist here, not to me. --Trovatore (talk) 21:17, 19 January 2013 (UTC)

I agree that the consensus is almost certainly going to be for a merge, but what is the rush? Michael Hardy put the merge tags on, so you might as well give people a chance to respond. Having said that, I probably wouldn't have reversed the merge if it had been done properly. Just at a glance, Moebius plane has more material and more references than Möbius plane, and anything that would improve the latter should be moved over. RockMagnetist (talk) 22:39, 19 January 2013 (UTC)

Correction: anything that would improve the latter and isn't a copyright violation should be moved over. —David Eppstein (talk) 22:58, 19 January 2013 (UTC)

I wouldn't call a copyright violation an improvement. Is there copyright violation in Moebius plane? RockMagnetist (talk) 23:28, 19 January 2013 (UTC)

Never mind, I see it. I'll remove the material. RockMagnetist (talk) 23:38, 19 January 2013 (UTC)

Is the person who added the copyrighted material the owner of the copyright? That seems probable under the circumstances. If so, what is the proper notice he should give that it's being done with permission? Michael Hardy (talk) 20:06, 24 January 2013 (UTC)

Try Wikipedia:Donating copyrighted materials especially at #Granting us permission to copy material already online. Deltahedron (talk) 20:43, 24 January 2013 (UTC)

Since a redirect was created, I've merged the edit histories of the two articles. The article is now almost as user:Ag2gaeh left if. I suspect we need to consider whether some of the earlier material should be merged into the now-existing version. Michael Hardy (talk) 04:29, 20 January 2013 (UTC)

Wave maps equation

The new article titled Wave maps equation has a bunch of obvious problems. Among them:

• What should the link to connectivity link to rather than to that disambiguation page?
• Is a summation convention followed here, where the repetition of $\alpha$ means one is to sum over all values of the index $\alpha$?
• Some context is needed. Is this something from physics?? I don't know.
• The article is an orphan: no others link to it.

Michael Hardy (talk) 17:38, 23 January 2013 (UTC)

The ZMATH review of the article cited states in part

This is the simplest geometric nonlinear wave equation. [They] arise in the analysis of more difficult hyperbolic Yang-Mills equations either as special cases or as equations for certain families of gauge transformations. They arise as well in general relativity for spacetimes with two Killing vector fields.

Zbl 1065.35199 Deltahedron (talk) 17:49, 23 January 2013 (UTC)

Maths related DRN

Hello WP Maths,

I volunteer over at Dispute Resolution Noticeboard and currently we have an active dispute revolving around the Golden Ratio. It's quite a technical dispute so if some of you with a bit more mathematics knowledge than I would be able to swing by and offer any opinions we'd be greateful. Dispute here: Wikipedia:Dispute resolution noticeboard#Golden Ratio. Thanks, Cabe6403 ^{(Talk•Sign)} 08:49, 25 January 2013 (UTC)

deletion of Hyperbolic Fibonacci and Lucas functions?

Hyperbolic Fibonacci and Lucas functions is listed in Articles for Deletion. Express opinions on this page. Don't just say Keep or Delete; give your arguments. Michael Hardy (talk) 18:37, 26 January 2013 (UTC)

Destruction of {{math}} and similar templates

The activity of the self-styled "latex to html" user, especially that he exposes spaces to line-wrapping, is known for a long time. What is not well-known is that he destroys not <math> only, but also {{math}}, {{mvar}} and {{open-open}} [7]. As I can extrapolate his attitude, he will raze all what is not "standard HTML typesetting" in his views: {{frac}}, {{sfrac}}, {{sqrt}}. I know that distrust to templates is entrenched in minds of some Wikipedians, but the bare wiki code has some disadvantages:

• It is unexpressive because math symbols (such as variables) would have the same appearance as the usual italic text.
• In the wiki code, it does not provide easily noticeable marker and bounds of a math expression.
• It is susceptible to silly replacement patterns of AWB and user scripts, such as [8][9].
• The hard-coded stuff like (0,1) or √2 cannot be improved across multiple articles by few changes, unlike {{open-open}} and {{sqrt}}.

On the other hand, templates which use the "texhtml" class can be customized to use good fonts, for example fonts uploaded with MathJax (although is it some work yet to be performed). Incnis Mrsi (talk) 08:45, 13 January 2013 (UTC)

I think {{math}} should go away. It looks very ugly to mix serif and sans-serif in the same line. If the formula really can't be understood in sans-serif, then display it and use LaTeX. --Trovatore (talk) 08:52, 13 January 2013 (UTC)

This view is unconstructive here, as it in no way addresses the issue raised (and we have no evidence that it even reflects a majority view; my own preferences are contrary to these). At best, it is an argument to modify the template definition (e.g. to match/use MathJax fonts), not remove its use. Though I have not reviewed the edits of the mentioned user, I agree with the concern expressed by Incnis. A single user on a mission can destroy a lot of useful formatting detail over time in this way. — Quondum 10:04, 13 January 2013 (UTC)

Any wiki which is not actively working towards replacing all of its html math with the MathJax version of $\LaTeX$ is in danger of falling behind the technological curve. It's like the difference between programming in machine-code and/or assembler and using a high-level language: $\LaTeX$ renditions in MathJax (rapidly becoming the de facto presentational tool) can be universally ported to any other platform hosting it. That's just to start with. --Matt Westwood 10:25, 13 January 2013 (UTC)

For these guys lazy enough to make some efforts towards correct appearance of "texhtml" it, certainly, looks very ugly. See user:Incnis Mrsi/standard.css and then, look at MediaWiki talk:Common.css/Archive 14 #Definitions for class="texhtml". Unfortunately, I have not a sysop in any Wikimedia project to try something like integration of MathJax's fonts with class="texhtml". Incnis Mrsi (talk) 12:32, 13 January 2013 (UTC)

For displayed formulas, <math> with $\LaTeX$ is the gold standard. Unfortunately, inline formulas with <math> don't match nicely with the surrounding text. Currently I have the impression that the {{math}} template gives the best optical compromise (in particular concerning variable names and non-breakable spaces), together with supporting templates like {{open-open}} and {{sqrt}}. Of course, there is a serious demand to improve the current situation to make it more appealing to put mathematics into Wikipedia. Maybe other fonts for the <math> environment are a solution, but I leave this to the experts (is someone working on this?). For the time being, converting {{math}} to just bare italic text for variables is certainly a step in the wrong direction when the unavoidable <math> environment for displayed formulas is present. Schmock (talk) 11:26, 13 January 2013 (UTC)

The assumption here is that mathematics formulae need to match "nicely" with the surrounding text. My contention is to ask: why should it? As long as it's not glaringly jarring. In MathJax, mathematical symbols rendered thereby are obviously mathematical entities, and are rendered in a font and with a presentation that makes them obviously stand out from the surrounding text, but so that they fit in aesthetically neatly with that text. Contrast that with the MediaWiki presentation which does not do that so well, which is what many people are put off by.

When the mathematical symbols are in the same font (albeit italicised) as the surrounding text, it makes it very much more difficult to follow any mathematical argument being made. Also, if all mathematics were presented using MathJax and an attempt were made to make this completely consistent, then a colossal amount of tedious markup code (sub and sup, specific markings to make stuff italic, all that undesirable overhead) can be completely disposed of. --Matt Westwood 11:50, 13 January 2013 (UTC)

(Edit conflict with Westwoodmatt, that illustrates his first sentence) I agree with the majority that {{math}} and the relative templates is the best choice for inline formulas. To Trovatore: I strongly disagree with you: using the same font for text and inline math is confusing: it results in the fact that a variable looks not the same inline and in displayed formulas. Thus, if an article contains a, a or a (inline), a and a or a (displayed), the reader should understand that there are two variables (one scalar and one vector) and not one nor four (nor six). Moreover the distinction between "a" (article) and "a" (variable) may be unclear, especially for people which is not accustomed with mathematics typographical conventions, or inside emphasized text. D.Lazard (talk) 12:28, 13 January 2013 (UTC)

In published mathematics the norm is for the math to be in the same font family (although italic) as the usual text. For example, the default in TeX is Computer Modern text and Computer Modern math. Variables in math have exactly the same appearance as italic text, because they are taken from the same font files - the "math fonts" have symbols only, not text. There is some professionally published math that has a different font for math, but this is a small fraction of the overall literature. So arguments that say there will be confusion from having the same font are not compelling to me.

Personally, I am willing to accept the {{math}} template. The {{mvar}} template (and the use of var in general) never had any significant support, and is based on some misunderstandings about the HTML var tag. In the end, the goal is to convert everything to MathJax. Someone should poke the developers to see what the status of that is. — Carl (CBM · talk) 13:03, 13 January 2013 (UTC)

If you use MathJax for inline math then you're accepting the use of a special font for inline math which is different from what Trovatore wants and what the person being complained about is doing in changing math to html. Personally I prefer having the maths bits be obvious and I believe that the default MathJax fonts should be used by default, but we should also be trying to get MathJax to support other fonts so logged in users can make the maths look like the surrounding text. Some people have a real thing about finding different fonts jarring and we should try and cater for them. Maybe that nuisance that is changing the articles is one of those. Dmcq (talk) 14:03, 13 January 2013 (UTC)

Perhaps the best solution is (dare I say it) to make the default Wikipedia skin use a serif font, like any normal reference work. It will never happen, of course, but we can dream... anyway, I'm inclined to agree with Matt Westwood that for the long-term health of the project it's best to encourage the use of LaTeX markup. It does look awful, but then again so do all of our math typesetting options. Zueignung (talk) 17:38, 13 January 2013 (UTC)

@Dmcq: I can accept a different font for MathJax only because MathJax has so many other advantages. But I don't accept the argument that the math font is supposed to be different, or that it is better for it so be different, because professional publishing (e.g. journals and textbooks) almost always use the same font for math and text. If there was a way for MathJax to use a font that matched the body text, that would be even better. (For those who don't know, the main font issue is that TeX needs more detailed font alignment info, so it is a general problem to make fonts work with math in TeX, and MathJax must have inherited this issue, so it uses its own fonts which have the necessary typesetting info.) — Carl (CBM · talk) 17:43, 13 January 2013 (UTC)

One of the problems that I have always had with professional publications is precisely the fact that the same font is used for mathematical symbols as for the text surrounding. It makes it far easier to follow the mathematics if the fonts are different enough for it to be clear what's mathematics and what's text. And the exciting thing about the net is that it is not necessary to be bound by the conventions of the past, which are restricting - it costs no more to present the mathematical stuff in a completely different font from the text, and IMO this is as it should be. I don't see that "this is what has traditionally been used in books, so this is what we must continue to traditionally do" should be the slavish rule of mathematics on wiki sites. --Matt Westwood 19:02, 13 January 2013 (UTC)

It's an exciting thing about the net that we have the freedom to present stuff that looks ugly? That's nothing special about the net; professional publications on paper could do this font-switching if they wanted to. They don't, because it's ugly. If the default for Wikipedia text in general were a serif font, we'd never be having this discussion; everyone (or almost) would agree that we shouldn't switch inline.

So you, "personally", think what I propose "looks ugly"? Why should your complete lack of sense of style affect what is clearly and obviously an improvement over the way they did it in the mediaeval times? --Matt Westwood 20:07, 13 January 2013 (UTC)

It is true that sans is not ideal for mathematics, and it's true that serif is not ideal for web display, at least at current resolutions. That's the source of the conflict. I don't agree with resolving the conflict by presenting a mishmash that looks like a ransom note with letters cut out of magazines. --Trovatore (talk) 19:14, 13 January 2013 (UTC)

You'd obviously know all about that, as it's obviously your field of expertise. --Matt Westwood 20:07, 13 January 2013 (UTC)

I think you're getting upset, which is a pity; let's remember we're just talking about typography. I know I do it too. On the substance, let's imagine as a thought experiment that the default font family were serif. Would you really want to switch to sans inline for mathematics? --Trovatore (talk) 20:27, 13 January 2013 (UTC)

The example you cite is exactly what I don't want. If you'd said it the other way round: imagine as a thought experiment that the default font family were sans serif. Would you really want to switch to a serif font for mathematics?" The answer to that is: Yes, I seriously do. I'm not a fan of sans serif of any description, I think there's absolutely no excuse for its pointless existence. What's the point of a font where you can't tell the difference between I, l and 1? See what I mean?

Having the mathematics in a font where every character is uniquely identifiable is utterly essential in all mathematics. You can not guarantee that in one of those egregious sans serif fonts. --Matt Westwood 20:35, 13 January 2013 (UTC)

I would not oppose a global switch to serif. From the point of view of our project, it would be an improvement. But we'll never be able to convince the rest of Wikipedia; the general consensus is that sans is better for the Web. Maybe when pixel sizes shrink by a factor of two (in linear dimension) that might change.

In any case, we seem to have established that your preference has at least two sources, and while one may still be that you would like to distinguish mathematics from English, one of comparable importance is that you just generally prefer serif? --Trovatore (talk) 20:44, 13 January 2013 (UTC)

More accurately: a) Mathematics must be in serif in order to be completely unambiguous, b) it does not matter whether the text and the mathematics are in different fonts - the juxtaposition is not "ugly", it's just that it may take the eye a little getting used to. Putting that together: if you must insist on your text being sans serif, then you really ought to compromise and make your mathematics in a different (i.e. "serif") font. --Matt Westwood 21:38, 13 January 2013 (UTC)

On the first point, the de facto standard for slide presentations in mathematics is the LaTeX beamer class, which uses sans by default, so it can't really be "must". On the second point I'm going to flatly contradict you — yes, it is ugly. --Trovatore (talk) 21:58, 13 January 2013 (UTC)

Okay, you've pwned me. --Matt Westwood 22:06, 13 January 2013 (UTC)

It's not a competition, I hope. We're both trying to improve the presentation of mathematics articles. --Trovatore (talk) 22:27, 13 January 2013 (UTC)

@MattWestwood: the problem with simply asserting "this is how it should be" it that it isn't very convincing. As Trovatore said, since journals and books are electronically produced these days, they could also switch. Knuth spent a long time researching typography for TeX, and he could have used different fonts if he had wanted to (he designed all the Computer Modern fonts from scratch specifically for TeX, after all). It's not clear why Wikipedia would be the right place for experimental typography. If we could achieve typesetting here as good as TeX, that would be ideal. I think MathJax will move us in that direction, but there is a long way to go. Also, MathJax will mean all math needs to be in <math> tags; all HTML and template formatting is temporary until/unless MathJax is enabled. — Carl (CBM · talk) 19:26, 13 January 2013 (UTC)

Here is the compromise I would propose (which I think is also the current best practice): articles that primarily use <math> and HTML should remain that way, and articles that primarily use <math> and {{math}} should remain that way. Editors who start new articles can pick the style they want.

We should discourage the IP editor from wholesale changes between styles (although that editor also seems to do other useful things, like changing <math> to something else for single-letter formulas). — Carl (CBM · talk) 19:32, 13 January 2013 (UTC)

I don't actually think "changing <math> to something else for single-letter formulas" is useful, for those of us who view articles using mathjax. It makes the variables look different from the way they do in longer formulas. —David Eppstein (talk) 19:38, 13 January 2013 (UTC)

This is why I consider wikipedia is last-eon technology. Life moves on. Get with the beat, luddites. Cross the Atlantic to a more progressive continent. --Matt Westwood 20:07, 13 January 2013 (UTC)

Although I take it in good humor, comments like that are not very distinguishable from trolling... — Carl (CBM · talk) 20:37, 13 January 2013 (UTC)

@David Eppstein: yes, it is certainly worse for people who use MathJax, but I am personally more worried about the default appearance, because few people have MathJax enabled compared to the number who just read articles without logging in. For the latter group, an HTML letter such as x (in any font) will usually be better than $x$. If we ever get MathJax enabled by default, I will help convert these to use it. — Carl (CBM · talk) 20:37, 13 January 2013 (UTC)

I certainly much prefer to have the symbols look as far as possible the same wherever they are. They are symbols not words. The symbol x just looks like a different symbol from x to me, and a definitely looks different from a. I guess it isn't so important that they are distinguishable from normal text as usually you have a few variables in so the italics show it is a bit of math so it is fairly easy to spot. The MathJax site uses Georgia font as far as I can see which is a serif font so the body text is closer to the MathJax equations - Perhaps people could do that on Wikipedia who were annoyed by the change from non-serif to serif. I'm sure it must be fairly easy to change the body text to Georgia though of course the MathJax font is closer to Times so just saying serif might be okay. Dmcq (talk) 02:22, 14 January 2013 (UTC)

Taking a step back for a moment, shouldn't we separate this into two discussions? Namely: (a) whether the the IP should be discouraged from making wholesale style changes to whole articles (ref Incnis and Carl), (b) the age-old discussion around math fonts, which I predict will not reach consensus soon. — Quondum 07:18, 14 January 2013 (UTC)

Yes, this debate erupts periodically and never results in agreement. This case seems analogous to citation styles and spelling differences, where there can also be strong conflicting opinions. The recommendation in WP:CITEVAR to seek consensus and "if there is disagreement about which style is best, defer to the style used by the first major contributor," seems appropriate here. RockMagnetist (talk) 01:42, 15 January 2013 (UTC)

Of course we like the math to be laid out properly. Trouble is the typeset stuff on Wikipedia looks like it's been photocopied out of a text book. RF 09:08, 18 January 2013 (UTC) — Preceding unsigned comment added by 109.153.136.200 (talk)

You're probably looking at it with the default appearance preferences f 'default to PNG' which is like that. In the future those will be rendered using MathJax which can be got by logging in and setting the preference to 'MathJax (experimental; best for most browsers)'. There is an argument between those who think variables inline should match surrounding text and not math separate equations and those like me who like the symbols to look the same everywhere they are used. In print the problem doesn't arise so much because they use a serif font normally, but I think it is an overhead distinguishing between the various symbols in Help:Math#Alphabets and typefaces but not on the basis of serifs so I have to count a and a the same. I think people who want the fonts the same for running text and symbols should change their default font to Times or Garamond, which is quite easy I believe if you have a username and log in. Dmcq (talk) 15:17, 18 January 2013 (UTC)

I just tried changing my css. I edited the custom css from preferences appearance which for me was my vector.css file and put in

div#bodyContent { font-family: Georgia, serif !important; }

span.texhtml { font-size: 1em !important; }

And now my text has serifs and looks more like the math. One could easily put in other serif fonts like Times New Roman instead. There's some misalignment problems but they're not bad and I'm sure some expert could help if they worry. Dmcq (talk) 16:29, 18 January 2013 (UTC)

I tried your custom css and found it prettier but harder to read. RockMagnetist (talk) 16:51, 18 January 2013 (UTC)

I like it with Times New Roman, though. RockMagnetist (talk) 17:47, 18 January 2013 (UTC)

You might like Cambria then if you're using Windows just make the list 'Cambria, Time New Roman, serif'. Dmcq (talk) 19:08, 18 January 2013 (UTC)

I like it! Very clean. RockMagnetist (talk) 19:26, 18 January 2013 (UTC)

The response from an IP address is a timely reminder that the great majority of people visiting Wikipedia do not log in and do not set any preferences. Shouldn't their needs be our priority? When you're editing a page, it would be a good idea to view it in an incognito window so you can see how it will look to most people. RockMagnetist (talk) 16:56, 18 January 2013 (UTC)

WP:«math»

Please, contribute to classification and comparison of all types of math typesetting in Wikipedia which are known to me. Thanks for the discussion which helped to write it. Incnis Mrsi (talk) 15:31, 20 January 2013 (UTC)

Using <blockquote> to delineate theorems

The same editor seems also be going around and delineating theorems using the html <blockquote> construct everywhere, under the guise of a "formatting improvement". I don't think this is a standard practice. I've reverted a few such changes (e.g., [10]) where there was a substantial change to the intended way in which the theorems were to be displayed. That seems to be most in keeping with the standard philosophy of not making unilateral changes to article formatting. Sławomir Biały (talk) 12:50, 27 January 2013 (UTC)

TAFI

Hello, Please note that Remainder, which is within this project's scope, has been selected to become a Today's Article for Improvement. The article is currently in the TAFI Holding Area, where comments are welcome about ideas to improve it. After the article is moved from the holding area to the TAFI schedule, it will appear on Wikipedia's Main Page in the "Today's Article for Improvement" section for one week. Everyone is invited to participate in the discussion and encouraged to collaborate to improve the article. Thank you, TheOriginalSoni (talk) 07:36, 27 January 2013 (UTC) (From the TAFI team)

The remainder article is currently not rated by the project. Sławomir Biały (talk) 12:53, 27 January 2013 (UTC)

Cocycle and cochain (algebraic topology)

In 2005, Cocycle was a regular dab page, but during this 7 years it was tampered by two users. Meanwhile, I found Cochain (algebraic topology) (edit | talk | history | links | watch | logs) to be an ancient redirect to cohomology and almost without inbound links, which IMHO does not make much sense. A volunteer to fix these irregularities? Incnis Mrsi (talk) 19:24, 20 January 2013 (UTC)

I think these should just be redirected to singular cohomology. Oh, wait, the article doesn't exist :D -- Taku (talk) 01:10, 30 January 2013 (UTC)

n-simplex

There are a series of articles about simplexes (5-simplex 6-simplex etc.) which contain a sentence of the form "More simply, the vertices of the 9-simplex can be positioned in 10-space as permutations of (0,0,0,0,0,0,0,0,1). This construction is based on facets of the 10-orthoplex." The form is not completely regular, but some (all?) of these are wrong. The 10-simplex, for example, has 11 vertices, but there are only 10 listed, the 9-simplex is right (ten vertices in 10-space) except that there are only 9 coordinates are listed. I think it's intended that the (regular) n-simplex be constructed in (n+1)-space, like in the 8-simplex article. 173.227.48.5 (talk) 23:41, 28 January 2013 (UTC)