Wikipedia talk:WikiProject Mathematics/Archive/2015/May

a flawed and clumsy definition

How, if at all, should one rephrase the definition quoted in this comment? Michael Hardy (talk) 14:57, 30 April 2015 (UTC)

As for me, this is a common illusion of non-mathematicians: to think that there are such mathematical notions as "variable", "discrete variable", "continuous variable". And therefore, an article with such title should not start with "in mathematics". Maybe "in elementary teaching of math" or something like that? Boris Tsirelson (talk) 17:45, 30 April 2015 (UTC)

(A bit off-topic, but not completely:) I was astonished to see that most non-mathematicians believe that such notions as "journal", "article", "professor" etc. exist among mathematical notions! Boris Tsirelson (talk) 17:48, 30 April 2015 (UTC)

On the other hand, we have an article "curve" with such phrase: "Various disciplines within mathematics have given the term different meanings depending on the area of study, so the precise meaning depends on context." Maybe something like that? Boris Tsirelson (talk) 19:10, 30 April 2015 (UTC)

Yes, but curves, even without full agreement on the precise definition, are an actual object of study. Does anyone study discrete and continuous variables, per se? I generally dislike articles that exist mostly to document jargon.

It's occasionally OK to have a short article that's mostly definition plus a couple of interesting facts about the objects meeting the definition; these won't ever be very good articles but they might sometimes be the best feasible solution. That's different from just observing some aspect of mathematical usage in practice and deciding to turn it into an article.

So, bottom line, I probably support redirecting to variable, unless someone can explain why not. --Trovatore (talk) 00:26, 1 May 2015 (UTC)

Presumably you mean variable (mathematics). In any case the topic of discrete (usually meaning integer-valued) vs continuous (real-valued) variables is an important one in optimization — see e.g. discrete optimization — in part because continuous variables allow methods like linear programming to be used but discrete variables are harder to handle. Searching Google scholar for "discrete variable" also finds very heavily cited works with this phrase in their title involving differential equations, and less-heavily but still well-cited works using this distinction in biostatistics. So I think it's a notable topic, but it needs to be rewritten to make clear that the distinction applies only to numeric variables. As for "documentation-of-jargon articles" that are in even stronger need of some sort of action: does anyone have a good idea what to do with of the form? —David Eppstein (talk) 01:01, 1 May 2015 (UTC)

Draft:Ricci soliton

Hi, this draft seems to be about a mathematical topic and I would greatly appreciate if anyone could provide feedback about whether the draft is suitable for Wikipedia. Thanks! Darylgolden^(talk) Ping when replying 13:51, 29 April 2015 (UTC)

Yes, it's a notable topic. Push it out to mainspace please. Sławomir Biały (talk) 17:28, 29 April 2015 (UTC)

 Done and it is now stubbily awaiting improvement at Ricci soliton. I've put it in Category:Riemannian geometry to start with. Arthur goes shopping (talk) 10:29, 2 May 2015 (UTC)

Draft:Convenient Analysis in infinite dimensions

Hello, mathematicians. Here's another of those old drafts about to be deleted. Anything worth keeping here? —Anne Delong (talk) 19:44, 21 April 2015 (UTC)

"Convenient vector spaces" gets about 200 hits on GScholar, including some secondary references, including books by Frolicher and Kock. I'd say the topic is notable. The article itself is fairly technical and likely written by one of the experts in the field. Unfortunately, that expert seems to have gotten discouraged by the rejection and left WP. The article could use more wikification and more non-Kreigl/Michor refs, but this is a matter of editing that is best done in mainspace. It's not a slam-dunk, but I think it would be good to bring it into mainspace. --Mark viking (talk) 20:04, 21 April 2015 (UTC)

And definitely not OR. Boris Tsirelson (talk) 20:14, 21 April 2015 (UTC)

Yes, I've heard of these. These have been studied extensively by Peter Michor and Andreas Kriegl (there is an AMS monograph here). I object slightly to the title. I think it should be titled convenient vector space. But otherwise this is good content that should be kept. Sławomir Biały (talk) 20:35, 21 April 2015 (UTC)

Definitely the title of the draft can't stay. What's an infinite dimension? --Trovatore (talk) 20:47, 21 April 2015 (UTC)

I changed the initial "A" in "Analysis" to a lower-case "a". Michael Hardy (talk) 04:42, 22 April 2015 (UTC)

Really? That bothered you worse than "infinite dimensions"? --Trovatore (talk) 04:51, 22 April 2015 (UTC)

It's not always a question of which bother me more. It might be a question of which one can be changed easily. Michael Hardy (talk) 23:17, 2 May 2015 (UTC)

The difficulty of changing it seems about the same to me. BTW I posted a related query on WP:RD/Language, which no one ventured to answer, which surprised me somewhat. See here — it's off-topic on this page, but I'd be interested in your thoughts on the subject, as I know you are also interested in language. --Trovatore (talk) 23:25, 2 May 2015 (UTC)

Please visit Draft talk:Convenient analysis in infinitely many dimensions. Boris Tsirelson (talk) 05:07, 22 April 2015 (UTC)

• Mark viking, Boris Tsirelson, Trovatore, Michael Hardy, it seem that everyone agrees that this is a notable topic. Before it goes live, I think it needs a less essay-like and more understandable lead section. I have written a suggested wording on the talk page, HERE. Let's continue the discussion there as Boris Tsirelson has suggested. —Anne Delong (talk) 11:05, 22 April 2015 (UTC)

Indeed, we should be thankful to Anne Delong; she is our only "mechanism" (sorry Anne) for saving valuable new articles in such cases. Boris Tsirelson (talk) 13:33, 22 April 2015 (UTC)

Well, you could cut out the middlewoman here, and use the search box below. To find drafts, type math-related keywords such as "mathematics", "integer" "vector"... well, you can think of some. All I ask is that if you find one that is of no use, and it has the pink notice that it's eligible for deletion under db-g13, just nominate it for deletion or let me know, but don't edit it - that removes the eligibility. I'll be glad to help with non-math-expert aspects of any you find. —Anne Delong (talk) 23:06, 22 April 2015 (UTC)

Tapering (mathematics)

Deletion of Tapering (mathematics) has been proposed. Is the article worth keeping? Michael Hardy (talk) 14:24, 4 May 2015 (UTC)

The article is indeed rather unclear in its current form, but tapering is certainly an important concept. However, it is covered in Window function. FireflySixtySeven (talk) 22:33, 4 May 2015 (UTC)

Tapering is well-known operation in solid modeling in computer graphics. It is a shape deformation operation, like bending or twisting [1], that alters the mesh representing solid, often deforming part of the shape down into a conical or prism-like section.. It is more akin to the tapering one does in a machine shop than tapers used in, for instance, signal processing. Definitely has nothing to do with exponential decay. I'm going to deprod, but a move to Tapering (computer graphics) may be warranted. --Mark viking (talk) 23:37, 4 May 2015 (UTC)

Watch out for crank user 108.242.169.13

108.242.169.13 (talk · contribs) came to my attention because of these edits to Talk:Van_der_Waerden_number; the author claims to have solved the (open) problem of the minimal n such that any k-coloring of the integers 1…n must contain a monochromatic k-term arithmetic progression. Since Wikipedia policy restricts talk page discussions to discussions about the article, this revelation was removed, once by mfb (talk · contribs) and once by myself. I also posted a relevant admonishment on 108.242.169.13's talk page. He is irate, and perhaps a response is in order. I tried to write one, but it came out unacceptably rude so I am not going to post it. The best response may be no response at all, just remove his material without engaging him.

The user claims to be Bill Bouris, a high school and community-college teacher of mathematics; his web site http://www.oddperfectnumbers.com/ is filled with similar crankery, boasting many unintelligible solutions of longstanding open problems. For example, he claims to have proved that there are no odd perfect numbers. He has also begun discussions on similar crankery at at least four other pages, which you can see in his user contributions page. mfb (talk · contribs) has reverted most of it, except at Talk:Langford_pairing, where David Eppstein (talk · contribs) opted to bring it back. —Mark Dominus (talk) 23:03, 6 May 2015 (UTC)

He has also appeared as 99.135.160.136 (talk · contribs). —Mark Dominus (talk) 23:06, 6 May 2015 (UTC)

We can remove again from the Langford pairing talk if you prefer. I brought it back only because it was a removal of talk page content without an adequate edit-summary explanation for what was being removed and why, if I remember correctly. I agree that it's crankery and is unlikely to be usable to improve the encyclopedia. —David Eppstein (talk) 23:21, 6 May 2015 (UTC)

I have no preference either way, and I did not mean to criticize your action, only to report on the current status. —Mark Dominus (talk) 23:30, 6 May 2015 (UTC)

Platonic solid - Classification

Please offer a view at Platonic solid#Classification. The issue concerns whether the existence of the five Platonic solids can be answered easily by an explicit construction, or cannot. Johnuniq (talk) 06:53, 5 May 2015 (UTC)

As for me, yes, it can (easily or not - this may be controversial). Boris Tsirelson (talk) 10:29, 5 May 2015 (UTC)

Agreed, "cannot" makes no sense here. --JBL (talk) 15:31, 5 May 2015 (UTC)

Likewise. I have corrected the statement. —Mark Dominus (talk) 15:00, 7 May 2015 (UTC)

PigTex

Is this just my lack of understanding of Tex?

Below, the "A" and "B" are supposed to be absent.

${\begin{aligned}A[G_{m}^{i},G_{n}^{j}]&={C^{ij}}_{k}G_{m+n}^{k}+\delta _{m}^{ij}\delta _{m+n,0}C\\B[C,G_{m}^{i}]&=0.\end{aligned}}$

Remove "A" to obtain

${\begin{aligned}[G_{m}^{i},G_{n}^{j}]&={C^{ij}}_{k}G_{m+n}^{k}+\delta _{m}^{ij}\delta _{m+n,0}C\\B[C,G_{m}^{i}]&=0.\end{aligned}}$

Remove "B" (but keep "A"), then (Removed crashing Tex to save eyes)

?
(Using PNG) YohanN7 (talk) 13:52, 7 May 2015 (UTC) Interestingly, MathML gives a more informative error message:

Failed to parse (Conversion error. Server ("http://mathoid.svc.eqiad.wmnet:10042") reported: "Error:["TeX parse error: Bracket argument to \\\\ must be a dimension"]"): {\begin{aligned}A[G_{m}^{i},G_{n}^{j}]&={C^{{ij}}}_{k}G_{{m+n}}^{k}+\delta _{m}^{{ij}}\delta _{{m+n,0}}C\\[C,G_{m}^{i}]&=0.\end{aligned}}

This is different from what PNG reports. YohanN7 (talk) 13:59, 7 May 2015 (UTC)

Solution?

Insert some air in the form of a pair of braces {}, "A" -> "{}", "B" -> "{}":

${\begin{aligned}{}[G_{m}^{i},G_{n}^{j}]&={C^{ij}}_{k}G_{m+n}^{k}+\delta _{m}^{ij}\delta _{m+n,0}C\\{}[C,G_{m}^{i}]&=0.\end{aligned}}$

Both pairs of braces are necessary. I guess this is due to my lack of understanding of Tex. The square brackets are usually used to pass additional arguments to an "environment", right? YohanN7 (talk) 14:08, 7 May 2015 (UTC)

Presumably the problem is with how the parser treats the \begin{align}. It looks to me like it is expecting an optional argument (which would be enclosed in square brackets). I don't think align accepts any arguments in LaTeX, so this must be something specific to our implementation. Sławomir Biały (talk) 14:41, 7 May 2015 (UTC)

A bracket that follows immediately a Latex command is generally considered as an optional parameter to this command, even if this command does not accept any optional parameter. In the case of "\begin", which does not has any optional argument, this optional argument is simply ignored (it is possible that some parsers consider the [ as a part of the text, but this is clearly not the case, here). In the case of "\\", the optional argument, if present, indicates the size of the vertical space. Therefore, the error signal is correct, as well as the suggested correction. D.Lazard (talk) 15:24, 7 May 2015 (UTC)

Thanks. YohanN7 (talk) 15:44, 7 May 2015 (UTC)

But if the command has no optional parameters, latex interprets brackets as ordinary brackets. The above commands all work as intended in ordinary latex. This is not what appears to be happening here. Sławomir Biały (talk) 16:30, 7 May 2015 (UTC)

Articles by Gisling

I would like to solicit opinions on how to handle the numerous articles by Gisling which consist largely of Maple 16 calculations and graphs. On his talk page several editors have raised concerns about these articles. In some cases, such as at Eckhaus equation, an editor went over the article, corrected it, and produced something respectable. In other cases, such as at Fujita-Storm equation, the concerns were not addressed. I started a discussion at Wikipedia:Articles for deletion/Bogoyavlenski-Konoplechenko equation asking to delete an article (and probably some related articles) on the flimsy grounds of WP:TNT in cases where I can't determine if even the subject of the article is accurately described (as it was not at Eckhaus equation - even the definition of this equation was erroneous.) I gutted several articles last night, but stopped short of going over all of them as I expected some resistance (which did occur this morning.) Note that Gisling has also contributed a large amount of quality material on the history of Chinese mathematics and other topics. --Sammy1339 (talk) 16:01, 9 May 2015 (UTC)

• • Majority of nonlinear differential equation and special function articles are sourced from US Government source:

United States Government: National Institude of Standards and Technolgy, Handbook of Mathematical Functions, Cambridge University Press 2010 printed edtion, 950 pages.( I bought this printed book)

there is also a web edition

US Government NIST Handbook of Mathematical Functions, full of color diagrams of various mathematical functions

It is a common practice in mathematics to plot graphs using either Matlab,Mathematica or Maple, for instance the graphs in Mathworld is generated wityMathematica, if you don't know how to make graphs using one of these, you have poor qualification , you are not qualified to edit any mathematic articles on wikipedia --Gisling (talk) 18:09, 9 May 2015 (UTC)

I don't think anyone objects to a few carefully selected plots that illustrate things well, but we shouldn't add plots just for their own sake. This is especially true of animations, which should be selected very carefully (see WP:IUP#ANIM) because of their generally larger file size, the inability to print or display on different media, as well as the fact that most readers find animations distracting. None of the animations in question seem to be good illustrations suited for the articles that they inhabit. Some samples are the galleries at: Discrete_q-Hermite_polynomials, q-Charlier_polynomials, Little_q-Laguerre_polynomials, q-Hahn_polynomials, Discrete q-Hermite polynomials, Continuous q-Jacobi polynomials, Continuous q-Laguerre polynomials, Affine q-Krawtchouk polynomials, Little q-Jacobi polynomials, dual q-Hahn polynomials, Al-Salam–Chihara polynomials, q-Racah polynomials. Even the stills at Coshc_function show poor judgement in the scale of the coordinate plane. (Who the hell plots a hyperbolic function over a domain like [-10,10]??) Clearly, your own advice is relevant: "if you don't know how to make graphs using one of these, you have poor qualification , you are not qualified to edit any mathematic articles on wikipedia". So, in short, I agree with the original poster that these galleries are a problem and many of them (including all of the animations) should be removed entirely per our image use guidelines.

Also, there is 唐戈 (talk · contribs · deleted contribs · logs · filter log · block user · block log), who seems to be stalking this issue and reverting any attempt to remove the galleries, otherwise defending Gisling's contributions at AfD, and inserting similar lists of unreferenced, poorly-formatted formulas into articles. I strongly get the impression that this is a sockpuppet or meatpuppet. Sławomir Biały (talk) 19:30, 9 May 2015 (UTC)

This talk page interaction is pretty discouraging, though I guess the final outcome was right. I agree that the animations I've seen on the pages linked in this discussion are useless. --JBL (talk) 21:45, 9 May 2015 (UTC)

• • Comment" I don't think this gentleman [2] is qualified to judge animations, because he never did one. Don't listen to what he talk, see what he did: he contributed only very simple 2D graphs in wikicommons, I don't think he has the ability to make 3D graphs in complex space, let alone animation.

Why animation ? When you have a function with more then 3 variables, the simplest way to visualize is make one parameter changes, then make animation plots. Apparently, this gentleman never makes a single graph with more than three variable

Any comparision ?? [3]

--Gisling (talk) 00:00, 10 May 2015 (UTC).

There are several fallacies here that need pointing out. First of all, the ability to create animations is not necessary to judge their suitability for an article. Secondly, as I have said, Wikipedia articles generally use animations very sparingly. So as a rule, I generally do not upload animations. But, thirdly, your entire methodology is flawed. I have uploaded animations, of a substantially higher quality than any of yours by the way. (I think the link you kindly provided above to your long list of bad, jerky, animations, with poorly selected aspect ratio, coloring, and mesh underscores this point. Any mathematical content of conceivable encyclopedic value in these animations is lost just from the sheer ugliness of this animatory effluent.) Sławomir Biały (talk) 01:07, 10 May 2015 (UTC)

Incidentally: the user whose username consists of two characters has been blocked indefinitely while Gisling has been blocked for three days for sock puppetry. --JBL (talk) 01:34, 10 May 2015 (UTC)

The quality of the TeX code at Fujita–Storm equation is still deficient, but far better than it was originally. If Gisling would improve his or her TeX skills, that would be a step in the right direction. For example:

${\sqrt {(}}x^{2}+5*x+15)$

should be changed to

${\sqrt {x^{2}+5x+15}}$

etc. Michael Hardy (talk) 21:30, 12 May 2015 (UTC)

Pearcey Integral graphs

US Government National Institude of Standard and Technology:NIST Handbook of Mathematica Functions has nice graphs of Pearcey Integral

http://dlmf.nist.gov/36.3

Can some one who claimed to be "Mathematica expert" provides similar graphs ?, Othewise, remove "Mathematica expert" claim please--Gisling (talk) 22:59, 9 May 2015 (UTC).

Is there some point to this question, or is it just trolling? --JBL (talk) 23:24, 9 May 2015 (UTC)

Aren't these images public domain anyway, if they are published by the US Government? Why reinvent the wheel? Sławomir Biały (talk) 23:48, 9 May 2015 (UTC)

If the federal government is the author of the work, as would be the case if the work were made for hire by its employees while on the job, then it is not subject to copyright. But the fact that the federal government publishes it, i.e. distributes it to the public, is not exactly the same thing. The federal government can buy a copyright or acquire it by bequest or donation, etc., and publish the work, and it's subject to copyright. Michael Hardy (talk) 21:07, 12 May 2015 (UTC)

Yes, I see http://dlmf.nist.gov/about/notices does not allow commercial re-use. Sławomir Biały (talk) 21:48, 12 May 2015 (UTC)

• Not that simple, you need to explain how that graph is obtained, I cannot just copy it and paste it to wikicommons, without knowing why and how. Honestly, I don't know how these plots were obtained(I never claimed myself as Maple expert), may be you can, Mathematica expert ? How about give it a try please, it is not that simple--Gisling (talk) 00:09, 10 May 2015 (UTC).

I am willing to make suggestions if you actually show a genuine interest, but I'm not getting that feeling from the interactions with you thus far. Instead, you have described me in the above thread as a "hooligan", in the AfD as "bloody ignorant", and this very thread seems like trolling in an overt attempt to bait me. (Why you have chosen to single me out, I do not know.) If you are genuinely interested in working it out, please post a comment over at the reference desk (this is not the right forum for such questions), including any details you think might be relevant to the computation, and I'll gladly see what I can do. Sławomir Biały (talk) 00:50, 10 May 2015 (UTC)

Inline Latex again

${\mathcal {}}_{P}l$e${}^{a}\mathbf {s}$e s${}_{e}{}^{e}$ Talk:Spectral theorem. YohanN7 (talk) 23:09, 14 May 2015 (UTC)

I would have expected an objective answer and/or discussion on the article's talk page, as you were the one who told me to put that issue there…--*thing goes (talk) 23:26, 14 May 2015 (UTC)

There is the place to talk, not here or on my talk page. This here is an invitation to others to join in there because it isn't a dispute between you and me. YohanN7 (talk) 23:37, 14 May 2015 (UTC)

How can I join?

Please join me here. :) — Preceding unsigned comment added by Sophie Concepcion (talk • contribs) 11:36, 16 May 2015 (UTC)

Stella (software)

Can anyone more familiar with our guidelines on software determine if Stella (software) is notable? I can't escape the feeling that Wikipedia is being used as a marketing platform here. Sławomir Biały (talk) 13:10, 8 May 2015 (UTC)

.

I personally not familiar with Stella, but I know that scientists use Stelle to model natural enviroments

[http://www.uvm.edu/~jphoffma/GSA/Generic.pdf Dr. E. Alan Cassell,Short Course System Dynamics Modeling of Natural Environments: An Introduction to STELLA Sunday 11 March 2001 Geological Society of America Northeastern Section 36th Annual Meeting, So. Burlington, VT.]

• Pharmacokinetic modelling using Stella on the Apple Macintosh

• [ https://books.google.ca/books?id=e6PaBwAAQBAJ&pg=PA41&dq=Stella+modelling&hl=en&sa=X&ei=i1xOVYmEJY6fyATKuoDQCQ&ved=0CFMQ6AEwCA#v=onepage&q=Stella%20modelling&f=false simulation of economic]

• Simulation in social sciences

• I don't go by "feeling" to judge an article. It is extremely irresponsible hooliganism--Gisling (talk) 19:22, 9 May 2015 (UTC).

Note to Gisling: the STELLA simulation software that you mention is not the same as the Stella polyhedron modeling software discussed in Stella (software). I agree that the STELLA simulation software is likely notable--at least I have seen a number of reviews, and as you note, books. But other than passing mentions at Geometry Junkyard and in a book, I've been unable to find secondary reviews of the Stella polyhedron software that are reliable. Probably not notable. --Mark viking (talk) 19:34, 9 May 2015 (UTC)

Well, given that the best you could come up with is a completely different software, I take it you agree that this is non-notable WP:ADVERTISING. And, fwiw, compliance with WP:NOT is not "hooliganism". That's just a baseless personal attack, which looks particularly silly given that you apparently failed to understand what the article was about. Sławomir Biały (talk) 21:06, 9 May 2015 (UTC)

I have Stella software, and I've used it for making many of the polyhedron images on Wikipedia. But on notability standards, I'm not in a position to offer an opinion. Tom Ruen (talk) 02:55, 13 May 2015 (UTC)

It is now at AfD: Wikipedia:Articles for deletion/Stella (software). Sławomir Biały (talk) 13:07, 16 May 2015 (UTC)

Actuary FAR

I have nominated Actuary for a featured article review here. Please join the discussion on whether this article meets featured article criteria. Articles are typically reviewed for two weeks. If substantial concerns are not addressed during the review period, the article will be moved to the Featured Article Removal Candidates list for a further period, where editors may declare "Keep" or "Delist" the article's featured status. The instructions for the review process are here. SandyGeorgia (Talk) 02:09, 17 May 2015 (UTC)

Floyd–Warshall algorithm

I could use some more eyes at Floyd–Warshall algorithm, please (see recent article history and talk page comments). —David Eppstein (talk) 23:08, 16 May 2015 (UTC)

Sigh. Still needs attention. --JBL (talk) 02:35, 17 May 2015 (UTC)

KasparBot

A bot is running amok and adding a template called 'Authority control' to the bottom of pages. It generates links that make little sense. See e.g. This version of Topological group (at the bottom). Is this legitimate? (If it is I'd say its legitimate bs, therefore bs, hence should not be here or anywhere.) YohanN7 (talk) 19:32, 16 May 2015 (UTC)

I mean, what is a link to National Diet Library doing there? whatever this is isn't much better. YohanN7 (talk) 19:52, 16 May 2015 (UTC)

Think of it as being like an interwiki link, except that instead of going to the German-language Wikipedia entry on the same topic it goes to the German National Library's entry on that topic. We've had these on biography for a while now, but this is the first I've seen of them being added to abstract topics. —David Eppstein (talk) 20:03, 16 May 2015 (UTC)

They sometimes link to pages in Japanese. Legitimate nonsense or not, it is nonsense. YohanN7 (talk) 20:12, 16 May 2015 (UTC)

What language would you expect a catalog entry from a Japanese national library to be in? Reading authority control might or might not help clarify what's going on here. —David Eppstein (talk) 20:35, 16 May 2015 (UTC)

Doesn't matter if there is an explanation buried somewhere. What is the purpose? How does this improve the articles? Has this diet library (as they call themselves) donated $1 000 000 to English Wikipedida? YohanN7 (talk) 20:51, 16 May 2015 (UTC)

You don't understand the purpose of a link from an article to library information about the same topic? This seems obvious, at least in the abstract. --JBL (talk) 21:01, 16 May 2015 (UTC)

How does these links help me? Do they offer a free course in Japanese? You are right I don't understand. If there are good places to link I link them (manually). YohanN7 (talk) 21:05, 16 May 2015 (UTC)

Possibly you should consider the possibility that some people speak English and Japanese, even if you are not among them. -JBL (talk) 21:12, 16 May 2015 (UTC)

That's no answer. How does it help me and the (few) others that don't speak Japanese and English (or Bemba)? YohanN7 (talk) 21:22, 16 May 2015 (UTC)

In exactly the same way that interwiki links to Japanese versions of articles don't help you. They're useful information for a subset of our readers that doesn't include you. —David Eppstein (talk) 21:38, 16 May 2015 (UTC)

Just to provide some background, naming authority files provide a standard naming ontology for library catalogs and are good for semantic web stuff, too. An example is the Integrated Authority File from the German National library. Working toward a standard ontology is a good thing for topic cross referencing and is one of Wikidata's objectives, IIRC. The question I have is: why are we linking to the National Diet Library of Japan naming authority, when the Library of Congress naming authority, called LCNAF, seems just as good and would be more immediately useful to English speakers? --Mark viking (talk) 22:12, 16 May 2015 (UTC)

I have no inside knowledge, but the {{authority control}} template is fairly agnostic about such choices, merely reporting what is available on the wikidata entry for the article. So my guess is either (1) someone took the effort to enter Japanese naming authority data on wikidata and nobody has done the same thing for LCNAF, or (2) someone took the effort to program the automatic transcription of Japanese authority control wikidata into our authority control template, and nobody has done the same thing for LCNAF. —David Eppstein (talk) 22:30, 16 May 2015 (UTC)

That makes sense, thank you. --Mark viking (talk) 00:23, 17 May 2015 (UTC)

Since this occupies a box of height at least three lines, it could at least spell out what it is about and where the links go. E.g. NDL → National Diet Library. It is also inconsistent. Sometimes it links to, not a library wiki-entry but to Integrated Authority File. Instead of saying "Authority control" it should spell out what the hell it is supposed to be. "Library catalog entry" or whatever. A parenthetical (in Japanese) might be appropriate when applicable.

When I first encountered this, I clicked the links and immediately took it for vandalism/some sort of unauthorized promotion. YohanN7 (talk) 09:03, 17 May 2015 (UTC)

As it is now it is just amateurish littering (yes, I am now aware that there are people around speaking Japanese—even German—thanks all for patiently explaining this to me). While we are at it, there is also the LIBRIS authority file. That would serve Swedish-speakers well. YohanN7 (talk) 09:03, 17 May 2015 (UTC)

"This metadata template links Wikipedia articles to various library catalogue systems. At the moment, it is used almost exclusively in biographical articles and on user pages." (Quoted from Template:Authority control.) Really? Boris Tsirelson (talk) 11:08, 17 May 2015 (UTC)

This is not true: It has been added to Graph theory‎, Real number‎, Integer‎, Number theory‎, Diophantine approximation, Division (mathematics)‎, Numeral system‎, Binomial‎, Number, Angle, Equation, Arithmetic, Logarithm (this is the list of the articles of my watchlist to which the article has been added yesterday or today). D.Lazard (talk) 12:32, 17 May 2015 (UTC)

I raised the issue at Template talk:Authority control. Sławomir Biały (talk) 13:53, 17 May 2015 (UTC)

The essay Wikipedia:Authority control has more information about this. It seems like authority control was first implemented on the German Wikipedia, and the template is now being propagated via the interwiki links. I don't know enough to say if this is a good idea. I think a legitimate concern is that, as far as I know, interwiki links are not very well validated. But that would seem to defeat the purpose of authority control. A more immediate concern though is that the template itself is very confusing to readers (as evidenced by the existence of this very thread). It is quite possible that a reader can see the template, click the link to the article Authority control, read that article, and still have no idea what the damn thing is about. This at least should be fixed, perhaps by replacing the link in the template to Wikipedia:Authority control instead, and improving that essay. Sławomir Biały (talk) 13:02, 17 May 2015 (UTC)

AfC submission

Care to offer insight into Draft:Topological Functioning Model? Thanks! FoCuSandLeArN (talk) 20:01, 19 May 2015 (UTC)

Full rank

My colleagues and I agree that the property of being "full rank" makes perfect sense and has a conventional definition for rectangular matrices. However, none of the books I have on hand give a definition. Can anyone produce a RS? (Refer: [4].) --JBL (talk) 00:50, 13 May 2015 (UTC)

The book by Gentle on Matrix Algebra, section 3.36, discusses the notion of full rank for non-rectangular matrices. --Mark viking (talk) 02:58, 13 May 2015 (UTC)

That book refers to a "full [hyphen omitted] rank matrix" rather than to a full-rank matrix. Punctuation was taught in elementary school when I was there; the reason why one should write about a "full-rank matrix", with a hyphen, and also say that a matrix "is of full rank", without a hyphen, is quite simple, and the presence or absence of the hyphen can effectively convey a lot of information in some cases (e.g. the difference between a "man-eating shark", which scares people away from beaches, and a "man eating shark", who is a customer in an exotic restaurant). I think people are still accustomed to seeing the traditional standard use of hyphens in magazines, newspapers, and books on subjects in which the non-technically-trained copy-editor is not afraid to say too much, although writers of advertising copy and package labeling do not use it. I think maybe it could still be saved, if an effort were made. Michael Hardy (talk) 17:52, 14 May 2015 (UTC)

@Mark viking :

• What do you mean by a "non-rectangular matrix"?
• Does that book anywhere give an explicit definition of "full rank" for matrices that are not square?

BTW, there's a pretty bad typo in equation (3.122) in that book. It says

$(xA)^{T}C(Ax)\,$

where it should say this:

$(Ax)^{T}C(Ax)\,$

Michael Hardy (talk) 18:11, 14 May 2015 (UTC)

• Ah sorry, that was a typo (or perhaps a thinko). I meant to say non-square.
• On page 77, last paragraph, the books says if the rank of a matrix is the same as its smaller dimension, then the matrix is of full rank. Then it goes on to note that "full row rank" and "full column rank" are typically used when discussing non-square matrices. No hyphens in any of these definitions in the book as far as I can tell. --Mark viking (talk) 20:02, 14 May 2015 (UTC)

Our article Rank (linear algebra) seems a little confused on whether a matrix having full rank is (a) both of full column rank and full row rank (and hence a square matrix), or (b) either of full column rank or full row rank (and hence can be non-square). I find the second repugnant – it seems like what would be used when someone is too lazy to use the term full column rank or full row rank as appropriate. What is the dominant use? (I already put in a note, but no-one of knowledge has chipped in.) —Quondum 00:41, 15 May 2015 (UTC)

Over a field, a square matrix is invertible if and only if it is full-rank (right?) So, I don't think "full-rank" is particularly useful for a square marrix. For a non-square matrix, a "full-rank", I think, has the usual meaning, meaning the rank (row rank) is the maximal possible; i.e., the matrix defines a surjection when it is viewed as a linear transformation. For example, to check the submersion theorem applies one checks if the Jacobian matrix has full-rank, meaning it is surjective; see for instance [5]. At least, this (full-rank = surjective) is how I use the term in my day life. -- Taku (talk) 19:41, 15 May 2015 (UTC)

I don't feel that just because in some instances another definition happens to be equivalent that one should consider one of them "not particularly useful". Your argument of being equivalent to being surjective is far more persuasive. (Your first argument would argue against using a new term for 'surjective', though!) In the context of matrices, one cannot call a matrix 'surjective' though: it is left-multiplication by that matrix which would be surjective, or right-multiplication my that matrix. So again, one is looking at calling it 'full row rank' or 'full column rank', with 'full rank' being only sensible where the two are equivalent (e.g. square matrices over a field). —Quondum 20:24, 15 May 2015 (UTC)

Full-rank ⇔ surjective is just not right. But I have never seen 'full row rank' or 'full column rank'. 'Full rank' is unambiguous, but it may be more common to call it 'maximal rank', at least in differential geometry (it is still referring to matrices). YohanN7 (talk) 23:00, 15 May 2015 (UTC)

The term 'maximal rank' makes much more sense than 'full rank' when this meaning is intended, and if it is more common, the article could be updated accordingly, subject to sourcing/dominance. —Quondum 01:04, 16 May 2015 (UTC)

Right or not, my point was using "full-rank" for "surjectivity" seems fairly common at lease in differential geometry. The reason I think is that it doesn't make sense to say whether a matrix is surjective or not; thus, "full-rank" becomes a shorthand for the linear transformation given as the left multiplication by the matrix being surjective (doesn't roll well on the tongue, does it?). For a square matrix, there is no need for the term "full-rank" (except in the pedagogical context) since "invertible matrix" is simpler and more precise. -- Taku (talk) 02:32, 16 May 2015 (UTC)

You need to be careful about the precise meaning: words used in mathematics are often used imprecisely, with a lot implied by context. For example, 'Jacobian matrix' may be used to refer to the matrix, but implied may be the mapping between tangent spaces that it represents, which implies 'left multiplication by'. This does not apply to matrices in general, where properties often are not referenced to the properties of the operators they might represent, but typically more directly in terms of the components of the matrix. So in matrices, my perception is that the rank seems to be mostly defined in terms of the dimension of the space spanned by the rows or columns, respectively. This is not the same thing as the dimension of the image of its left and right multiplication, or whether it is surjective, because these are determined by the dimension of the domain and the dimension of the codomain, which can be less or more respectively than the number of columns and rows. So the fit is really quite poor. —Quondum 03:19, 16 May 2015 (UTC)

There seems some confusion. By definition, I agree, the row/column rank is the dimension of the space spanned by rows/columns (they turn out to be the same number). But the rank of the matrix can be equally characterized by the dimension of the image of the linear transformation determined by a matrix. Via the use of a transpose, we only need to consider the case by the left multiplication. Then the rank of the matrix is the dimension of the image of the matrix viewed as a linear transformation. In other words, there might be some "a priori" distinction that can be made from matrix point of view and operator point of view, but the distinction is not too important to be concerned in practice. A case in point: one speaks of finite-rank operator (by the way, as Michael Hardy noted, the universe collapses if you forget hyphen here) even though it is not really a matrix. Taku (talk) 14:10, 17 May 2015 (UTC)

You seem to be missing what I'm saying. The rank of a vector space can be less than the dimension of its representation (it is equal to the size of the basis, and hence the rank of a map can be less than the rank of the matrix chosen to represent it). When defining the rank of a matrix, Bourbaki explicitly uses vector spaces of dimension equal to the dimensions of the matrix. Without this, your second statement in the above paragraph does not hold.

Bourbaki covers the rank of both a linear map and of a matrix in detail, but does not mention the concept of maximal rank in any form. I'd posit that the concept of a maximal-rank matrix has little utility – little enough that I would trim the definition to an observation that when the terms maximal rank or full rank are used, these terms typically mean "[description here]". We have sufficient sources to make this diminished statement. —Quondum 00:25, 18 May 2015 (UTC)

It seems to me like you're extrapolating too heavily from your own experience. "Full rank" was immediately recognized and understood by my office mates; I wouldn't bat an eyelash seeing (or using) it in a research paper. And we have at least two RSs in this thread with definitions (thanks very much to those who found them!), while I doubt very strongly that anyone will produce a RS that deprecates the term in the way you suggest. --JBL (talk) 01:10, 18 May 2015 (UTC)

I think a suitable definition of maximal rank has become clear (and I suppose we can assume that 'full rank' is a synonym). But what do you make of four out of six RSs mentioned in this thread that cover matrix rank simply failing to mention it? Should we present it as though it is mainstream and significant as if every RS had mentioned it? I was hoping some consensus would emerge, but it seems to be slow in coming; the proposals I make are merely strawmen for consideration. —Quondum 03:32, 18 May 2015 (UTC)

Okay, never mind. I've tweaked the article in the direction of maximal or full rank being defined for nonsquare matrices. At least that gets rid of the internal inconsistency, and does not change the article much. I'm not going to belabour this any further. —Quondum 04:26, 18 May 2015 (UTC)

Thank you, I am happy with the final result. --JBL (talk) 20:46, 18 May 2015 (UTC)

In case you're still looking, I found a reliable source: David C. Lay, Linear Algebra and Its Applications, 1994, Addison-Wesley, p. 242, exercise 26: "In statistical theory, a common requirement is that a matrix be of full rank. That is, the rank should be as large as possible." [Italics are in the original.] Mgnbar (talk) 00:01, 16 May 2015 (UTC)

And by the way books by Kolman, Hoffman and Kunze, and Halmos don't seem to mention full rank. Mgnbar (talk) 00:04, 16 May 2015 (UTC)

I suggest we define either of 'full rank' or 'maximal rank' and provide the other one in an or-clause. While 'full column rank' and 'full row rank' both make sense, they don't seem to appear in the literature. Both imply 'full rank' and 'full rank' implies one or the other (or both) of them (if they were defined) supported by the theorem that says row-rank = column-rank. Lee defines 'maximal rank' for an m × n or n × m-matrix, n > m as one having a m × m minor of rank m. YohanN7 (talk) 17:05, 16 May 2015 (UTC)

I would hope we make no definition without it being a fairly standard term. That was why I initially requested some input: some sources clearly define full rank as the maximal rank for the matrix size, but if that is a minority definition, I would prefer to see to noted as such. If most sources use a different term, or define it some other way, we should document it accordingly. So far, no-one seems to have done more than consider an isolated source or so. I we have a few highly notable secondary sources defining it (how widely accepted is Lee?), we could do as you say. —Quondum 23:12, 16 May 2015 (UTC)

Well, Lee is no Bourbaki, he defines things on the fly (and is therefore readable as opposed to Bourbaki). My point is that it doesn't matter very much, just take one "reputable" reference at random, use it for the def and provide the alternative term. It can't get very much wrong. Unfortunately, my own supply of books in linear algebra is limited. YohanN7 (talk) 14:58, 17 May 2015 (UTC)

With a risk of complicating the discussion further, I would like to mention yet another point of view: "full-rank" = "maximal rank" = "generic-rank". Here, I'm using "generic" in the following way. Let X be the (vector) space of all matrices of some fixed size (possibly non-square). We view X as an (affine) algebraic variety (X is simply a vector space.) Then the matrices of maximal possible rank form an open subset with respect to Zariski topology (it is the complement of the vanishing locus of minors.) So, a matrix in a general position has maximal possible rank and that's the generic rank of a matrix. (Do I make sense?) By definition, a matrix is full-rank if it has generic rank or equivalently maximal possible rank. -- Taku (talk) 14:38, 17 May 2015 (UTC)

This vaguely resembles an example in Lee's book. The set of m × n matrices with full rank is open in M(m, n) in the subspace topology, hence a submanifold. YohanN7 (talk) 14:58, 17 May 2015 (UTC)

It is correct that "full rank" = "maximal rank", when considering the set of all m × n matrices. But, in the case of the a differentiable mapping, it may occur that the Jacobian matrix is never full rank, that is "maximal rank" < "full rank". On the other hand, when "generic rank" is defined (that is in the context of algebraic geometry), it is true that "maximal rank" = "generic rank". As an example of a situation where the term "full rank" is useful, and probably widely used, is the following result: Given a set of polynomials that generate a prime ideal, the algebraic variety of their common zeros is a complete intersection if the Jacobian matrix is "full-rank" at, at least, one point of the variety. In this case, the singular points are exactly the points where the Jacobian matrix is not full-rank. D.Lazard (talk) 09:56, 18 May 2015 (UTC)

In essence then, "maximal rank" is generally used for the supremum of ranks of a set (e.g. Jacobian matrices of a map) of matrices, while "full rank" applies to one matrix? YohanN7 (talk) 11:53, 18 May 2015 (UTC)

There is another case to consider: infinite-dimensional matrices. Because we can zero a column of a full-rank matrix without changing the rank, the rank remains maximal, but because the column span is no longer that of the resulting vector space (the map is no longer surjective), it is no longer of full rank. But this argument depends on whether the rank of a matrix is defined for infinite-dimensional matrices. —Quondum 14:19, 18 May 2015 (UTC)

I don't think "rank" is helpful for infinite-dimensional matrices unless finite; but my "natural" inclination is to consider "injective" rather than "surjective" as the definition of "full rank" for infinite-dimensional matricies. — Arthur Rubin (talk) 20:34, 18 May 2015 (UTC)

Isn't the matter settled? Don't we have two reliable sources for full rank (Gentle and Lay), ignoring bad spelling in the former? More sources don't explicitly define the term because they don't need it or its meaning is obvious? (I don't want to dictate conclusion of discussion. I'm just trying to figure out whether I need to keep paying attention.) Mgnbar (talk) 13:45, 18 May 2015 (UTC)

I'm treating the original question as settled (despite my personal reservations), though any further discussion may lead to tweaks, e.g. distinctions between "maximal" and "full". —Quondum 14:19, 18 May 2015 (UTC)

I agree. --JBL (talk) 20:46, 18 May 2015 (UTC)

I realize I'm late to the party, but thought I would chime in. The term "maximal rank" can be ambiguous. Consider the following statement:

Suppose that $f:\mathbb {R} ^{n}\to \mathbb {R} ^{m}$ is a continuously differentiable function, and let U be the set of points where the derivative of f has maximal rank.

A rather trivial example of the ambiguity is the constant function $f(x)=0$ for all x. The rank of the derivative is zero everywhere, so the maximum value of the rank of f is equal to zero! Thus (under this interpretation) $U=\mathbb {R} ^{n}$. Now, clearly for "most" applications, this is not the interpretation that would be intended by the statement. Rather, we would mean

Suppose that $f:\mathbb {R} ^{n}\to \mathbb {R} ^{m}$ is a continuously differentiable function, and let U be the set of points where the derivative of f has full rank.

Here full rank means that the rank of Df is as large as it can possibly be for an $m\times n$ matrix. Thus, the restriction of f to U is a submersion, if $n\geq m$, and an immersion, if $n\leq m$. Sławomir Biały (talk) 15:47, 18 May 2015 (UTC)

Considering the reservations expressed here (by D.Lazard and Sławomir Biały), I have removed my unsourced addition of 'maximal rank 'to Rank (linear algebra), leaving the definition of 'full rank'. —Quondum 03:57, 21 May 2015 (UTC)

Category ICM 2014 Plenary and Invited Speakers

I've just created the category for the ICM 2014 Plenary or Invited Speakers. A list with the names of the speakers can be found on http://www.mathunion.org/db/ICM/Speakers/SortedByCongress.php . If someone wants to help to add the category to more articles, be more than welcome! Lolaszvodikech (talk) 00:44, 22 May 2015 (UTC)

Off-topic: A mathematician named László Erdős is listed there! I wonder if he is a relative of Paul Erdős! Anyway, with such a name he must attract a lot of attention :). Lolaszvodikech (talk) 01:28, 22 May 2015 (UTC)

Face configuration

The article title Face configuration appears to be a neologism: neither Cundy & Rollett nor Williams, both cited, use the term. Rather, they use the symbol to identify the related polyhedron (typically a Catalan solid). I do not have Grünbaum and Shephard to hand, but I have never heard the term in this connection. The article makes much of Cundy & Rollett's (non-existent) usage. Do we accept this kind of apparently fabricated usage, or should this kind of article be nominated for deletion? — Cheers, Steelpillow (Talk) 11:47, 19 May 2015 (UTC)

Vertex configuration is another article in the same genre it would seem. Although possibly slightly less WP:ORish, the term "vertex configuration" (or the other synonyms listed there) do not appear to be in wide use. I think this article should be redirected to vertex figure. The notation can be mentioned there, without creating a neologism. I don't know if there is a suitable merge target for face configuration, though. Sławomir Biały (talk) 13:34, 19 May 2015 (UTC)

Face configuration is a fairly well known concept in a facial recognition system, possibly a notable topic in the geometric/parametric approach to facial recognition. I could not find sources describing the polyhedral version. --Mark viking (talk) 16:39, 19 May 2015 (UTC)

As stated, the two sources are Cundy-Rollett for Archimedean dual polyhedra and Grünbaum and Shephard for Monohedral/Lave tilings. Williams repeats the Cundy-Rollett symbol usage. Mathworld calls it a Cundy-Rollett symbol. I'm open to merging Vertex configuration and Face configuration as one article and calling both Cundy-Rollett symbol, like at [6]. Vertex figure is functionally different than vertex configuration, not a symbol but a n-dimensional polytope existing at a vertex. Tom Ruen (talk) 16:56, 19 May 2015 (UTC)

I cannot find reference to "Cundy-Rollett symbol" on Mathworld, only a passing reference to the "Cundy and Rollett symbol" for the Archimedean solids. Either way, use for the odd table column heading does not establish notability. The "functional difference" alluded to above is trivial: a "vertex configuration" is just a symbol denoting a certain kind of vertex figure, so I agree with Sławomir Biały that merging is a good way to go there. But Face configuration is not a recognised term and the thing it denotes is not notable either - witness the fact that the purported sources use it in a very minor way and don't even bother to give it a name. If anything in the article worth keeping can be found, it can be merged across into Face (geometry), but AFAICT there is nothing even to justify a redirect and the article itself should be deleted. — Cheers, Steelpillow (Talk) 19:22, 19 May 2015 (UTC)

It is a symbol, not a figure. And I'd rather keep it separate from Face (geometry) and collect it with vertex configuration, whatever might be called. Here's another newer reference for Archimedeans Cundy-Rollett symbol. Tom Ruen (talk) 19:44, 19 May 2015 (UTC)

"Cundy-Rollett symbol" gets just ten hits on Google, most of which are scrapings from here - and Popko's book, which mentions it only briefly. That is not enough to establish notability. — Cheers, Steelpillow (Talk) 21:13, 19 May 2015 (UTC)

Whatever you'd like to call it, it's used EVERYWHERE! Would you prefer something like A universal symbol that represents the sequence of vertex valances around a face of a polyhedron or tiling?! Tom Ruen (talk) 21:40, 19 May 2015 (UTC)

When you say "it's used everywhere", you really mean that you, personally, are responsible for adding it to a large number of our articles, right? Did you have sources when you did so? —David Eppstein (talk) 22:53, 19 May 2015 (UTC)

No, by "everywhere", I means every book or paper that talks about about regular, semiregular, uniform polyhedra and their duals. Tom Ruen (talk) 23:57, 19 May 2015 (UTC)

Surely such a prominent concept has an established name. Ozob (talk) 02:06, 20 May 2015 (UTC)

For the more popular vertex configurations, I've found 9 descriptive names for the symbol: vertex configuration, vertex figure, vertex type, vertex symbol, vertex arrangement, vertex pattern, face-vector, vertex description, and Cundy-Rollett symbol. Tom Ruen (talk) 05:58, 20 May 2015 (UTC)

That is because the symbol is part of the theory of uniform vertex figures and has no significance outside of that topic. Moreover its mathematical significance even within that topic is so trivial that nobody has ever bothered to agree an accepted name. All those other terms are mere ad hoc descriptions, because that is all it deserves. The "face configuration" is even less justifiable. — Cheers, Steelpillow (Talk) 07:24, 20 May 2015 (UTC)

Sourcing?

Now I find myself being reverted. The article at Vertex configuration gives "Cundy-Rollett symbol" as a synonym and cites mathworld. As I pointed out above, mathworld does not in fact use this term and the cite is therefore incorrect. As I also pointed out, one lone author does not establish notability. However when I removed the reference Tom Ruen chose to revert my edit without comment. Is this behaviour acceptable to this Project? — Cheers, Steelpillow (Talk) 07:04, 20 May 2015 (UTC)

I restored the removal, and added a second source, and commented on the second source in the revert. Tom Ruen (talk) 08:48, 20 May 2015 (UTC)

No you have not restored my removal. You have moved it to the article lead and wrapped a whole load more trivial sources around it. I have to ask again, is this behaviour acceptable to this Project? — Cheers, Steelpillow (Talk) 19:05, 20 May 2015 (UTC)

I said I restored what you removed and I added a second reference along with my revert, and I named the second reference in the comment. I did not revert "without comment". After that I continued to improve the content as a good editor should. You can see "trivia" as you like, but my intention was to demonstate varied ways the symbol was named. If someone reads the article, they can check the usages and decide for themselves which name has the best legitimacy or whatever. Tom Ruen (talk) 03:40, 21 May 2015 (UTC)

Nevertheless, you reverted me. The tenor of this discussion is that thse sources are inadequate and you should not have done so. Yet you persist in embellishing them while the discussion continues. — Cheers, Steelpillow (Talk) 08:08, 21 May 2015 (UTC)

I'll let smarter people than me validate your confusing charges against me. I've done NOTHING but try to please you, and all you see is insult. Tom Ruen (talk) 09:59, 21 May 2015 (UTC)

So, reverting me is trying to please me. I'll remember that next time I want to try and please you. — Cheers, Steelpillow (Talk) 12:24, 24 May 2015 (UTC)

Indeed. removing-without-improving and reverting-with-improving might weigh as equal acts of kindness. Tom Ruen (talk) 19:28, 24 May 2015 (UTC)

Another sourcing

Tom Ruen (talk) 09:50, 20 May 2015 (UTC)

Physical Metallurgy: 3-Volume Set, Volume 1, edited by David E. Laughlin,

Page 16 The Cundy and Rollett symbol of a vertex configuration nm means that m n-gons meet at a vertex. The vertex configuration can also be written in the form of the Schlafli symbol {n,m} or (n,m). The eight semiregular Archimedean tilings are uniform. This means they have only one type of vertex configuration, i.e. they are vertex transitive; they consist of two or more regular polygons as unit tiles. In the case of layer structures, where one layer type corresponds to one of the Archimedean tilings, the layer next to it will preferentially be the respective dual tiling (Catalan or Laves tiling). The dual to a tiling can be obtained by putting vertices into the center of the unit tiles and connecting them by lines. If the tiling is regular, then the dual tiling will be regular as well. The dual of the regular square tiling is a regular square tiling again, so this tiling is self-dual. The dual to the hexagon tiling is the triangle tiling. While the uniform semiregular tilings are described by their vertex configuration, their duals consistent of just one type of polygon (are isohedral), but have more than one vertex configuration. Therefore, they are described by their face configuration, i.e. the sequential numbers of polygons meeting at each vertex of a face. For instance, the dual to the Archimedean snub square tiling 32.4.3.4 is the Cairo pentagon tiling, V32.4.3.4. Its face configuration V32.4.3.4 means pentagonal unit tile with corners, where 3,3,4,3,4 squashed pentagons meet.

page 20 The Archimedean solids can all be inscribed in a sphere and in one of the Platonic solids. Their duals are the Catalan solids, with faces that are congruent but not regular (face-transitive); instead of circumspheres like the Archimedean solids, they have inspheres. The midsphere, touching the edges are common to both of them. The face configuration is used for the description of these face-transitive polyhedra. It is given by the sequential listing of the number of faces that meet around each vertex around a face. For instance, V(3.4)2 describes the rhombic dodecahedron, where at the vertices around one rhombic face 3,4,3,4 rhombs, respectively, meet.

I'm not sure this is the best source for establishing standard usage in geometry. Apart from the obvious issue that this is a metallurgy textbook (and the terminology does not seem to be standard in metallurgy either), the neologism appears only in the fifth 2014 edition, long after we had an encyclopedia article on the subject. Also, the description that appears on page 20 is a paraphrase of our article on the subject, including the same choice of example. So it seems that one source has merely picked up our neologism. An encyclopedia should merely reflect what is already standard; it should not be in the business of inventing new standards. Sławomir Biały (talk) 12:04, 20 May 2015 (UTC)

It looks like a large fraction of the content above was referenced and taken from Crystallography of Quasicrystals: Concepts, Methods and Structures by Walter Steurer, Sofia Deloudi, (2009) p.18-20, p.51-53. Tom Ruen (talk) 10:23, 21 May 2015 (UTC)

Yes, this seems like a better source. Sławomir Biały (talk) 13:15, 22 May 2015 (UTC)

The oldest source I can find using the phrase vertex configuration is this 1993 paper on the uniform star polyhedra: [7]. Pretty much ALL the online constructions of the uniform stars trace back to this paper, of course the original source is the Coxeter 1954 paper on uniform polyhedra which draws the diagrams as polygons rather than listing the n-gons. Tom Ruen (talk) 10:33, 21 May 2015 (UTC)

• Uniform Solution for Uniform Polyhedra*
• Zvi Har’El
• Department of Mathematics, Technion − Israel Institute of Technology, Haifa 32000, Israel, E-Mail: rl@math.technion.ac.il
• ABSTRACT: An arbitrary precision solution of uniform polyhedra and their duals is presented. The solution is uniform for all polyhedra given by their kaleidoscopic construction, with no need to ‘examine’ each polyhedron separately.

I'd summarize and conclude Cundy and Rollett symbol is a good name for this notation for giving credit for their 1952 book Mathematical models, first published book expressing the convenient notation for the (convex) regular, semiregular, and semiregular dual polyhedra, as integers a.b.c. ... Then Zvi Har’El's 1993 paper extend that usage as vertex configuration and may get first credit for the first consistent notation for the uniform stars and duals, which include full orientation information (prograde/retrograde) and wrap information, which makes the stars explicitly constructable: like (a/b . c/(d-c) . ... x/y)/z. Roman E. Maeder then ported it into Mathematica around 1995, using the same terminology, while Robert Webb used it in Stella around 2001, but instead calling it a vertex description. Tom Ruen (talk) 11:22, 21 May 2015 (UTC)

In parallel, and a bit earlier Tiling and Patterns (1987) uses a similar more general notation for Euclidean tilings, calling the notation tile symbols or of type. It also has hollow tilings with star polygons and retrograde orientation, using negatives for retrograde, so somewhat similar to Zvi Har’El's star polyhedra usage. Tom Ruen (talk) 11:44, 21 May 2015 (UTC)

First article about Indonesian mathematician on WP-En

Hello. I've just created the first page about any mathematician from Indonesia at the English Wikipedia. The name of the page is Moedomo Soedigdomarto. My English is not very good, and the sources are all in Indonesian (my 3rd language, which I don't know very well too...) Anyway, if someone wants to help. Please feel free to join the effort. Chinese-Indonesian (talk) 08:46, 26 May 2015 (UTC)

Looks like a nice start, especially for someone who is not a native English-speaker, good job! But links (like the first two references 1 and 2) in Latin-alphabetized Indonesian are not too helpful for English speakers in English Wikipedia. Also, strong claims like "He was one of the first Indonesians" need to be cited. If possible please add more English sources. M∧Ŝc²ħεИ_τlk 10:34, 26 May 2015 (UTC)

Lecture notes as a reference?

See recent edits at Brouwer fixed-point theorem, and also my talk page. I think lecture notes are sometimes okay and sometimes not depending on what is available. In this case, a simple Google search will give millions of hits, and I don't see the need to have lecture notes linked. YohanN7 (talk) 11:38, 24 May 2015 (UTC)

Any lecture notes, particularly the terabytes of pdf notes floating around, should just be external links not references, they are not published mainstream secondary sources. M∧Ŝc²ħεИ_τlk 10:37, 26 May 2015 (UTC)

These are external links (didn't start out that way though), so I guess it is fine. Not a huge fan of the practice though. Why should we "promote" or "endorse" them? YohanN7 (talk) 10:45, 26 May 2015 (UTC)

It isn't "promotion" or "endorsement", I thought the purpose of external links is to point to other sources of info which are not considered reliable and are not secondary but have at least some usefulness. M∧Ŝc²ħεИ_τlk 10:49, 26 May 2015 (UTC)

There is a difference between neutral links to MathWorld and the like and lecture notes. But I have myself committed the crime (actual references, not external links) on occasion when nothing else has been available to me. See this more like me asking a question than pushing a POV (thought my posts look and sound like POV pushing) YohanN7 (talk) 11:18, 26 May 2015 (UTC)

I would say it depends on the circumstances. Even lecture notes by a well-known authority can be used as references (with care, and assuming there are no better sources available), per WP:SPS. I would say that this source is a good one. The chapter discusses the degree of a mapping from the perspective of multivariable calculus. The Brouwer theorem is a consequence of the homotopy invariance of the degree of a function. This is something that could, in principle, be added to the article, since the connection with degree is not really made clear. Sławomir Biały (talk) 12:19, 26 May 2015 (UTC)

I can buy that in full. As a side note, both Steven Willard (General Topology) and John M. Lee (Introduction to Topological Manifolds) handle it using homotopy theory (unless my memory fails). Especially Willard's proof was kind of neat, I recall, (and super-short). Will check this out. YohanN7 (talk) 13:38, 26 May 2015 (UTC)