Wikipedia talk:WikiProject Mathematics/Archive/2019/Apr

Math Portals

Which of these approx 30 Portals does the math project provide support to or are you interested in managing? From Category:Mathematics_portals.

1. Portal:Contents/Mathematics and logic
2. Portal:Algebra
3. Portal:Analysis
4. Portal:Angles
5. Portal:Applied mathematics
6. Portal:Areas of mathematics
7. Portal:Arithmetic
8. Portal:Category theory
9. Portal:Cryptography
10. Portal:Differential equations
11. Portal:Discrete geometry
12. Portal:Discrete mathematics
13. Portal:Equations
14. Portal:Functional analysis
15. Portal:Game theory
16. Portal:Geometry
17. Portal:Infinity
18. Portal:Information theory
19. Portal:Integrals
20. Portal:Irrational numbers
21. Portal:Knot theory
22. Portal:Mathematical optimization
23. Portal:Mathematics
24. Portal:Metallic means
25. Portal:Number theory
26. Portal:Probability distributions
27. Portal:Set theory
28. Portal:Tensors
29. Portal:Topology
30. Portal:Volume

Legacypac (talk) 22:45, 2 April 2019 (UTC)

Most of them seem reasonable, particularly those on specific areas of mathematics. I don't understand the purpose of 1 (same as Portal:Mathematics?), and the portals on 4, 13, 17, 19, 20, 24, 28 (which is a little broken), and 30 seem hyper-specific. They look like they are portals for essentially one article or about a topic that's not really cohesive (e.g. Portal:Equations is about any equation???). — MarkH21 (talk) 23:07, 2 April 2019 (UTC)

Name order for Chinese mathematicians

There is a move discussion at Talk:Tian Gang, with a disagreement about whether we should treat all mathematicians the same or all Chinese people the same in terms of their names. More input would be welcome. —Kusma (t·c) 07:33, 6 April 2019 (UTC)

Links to DAB pages

I have collected another batch of math(s)-related articles which contain links to DAB pages, where expert attention would be welcome. Search for 'disam' in read mode, or for '{{d' in edit mode; and if you solve one of these puzzles, post {{done}} here.

• Hybrid number  Done
• Metatheory  Done
• Separable algebra  Done
• Performance effects  Done
• Coulomb constant  Done

Thanks in advance, Narky Blert (talk) 01:19, 7 April 2019 (UTC)

Wow, performance effects is thick with management-speak. XOR'easter (talk) 15:34, 7 April 2019 (UTC)

Yes; but anyway, "sum of squares" there is just that... Boris Tsirelson (talk) 18:21, 7 April 2019 (UTC)

Another Portal

Wikipedia:Miscellany for deletion/Portal:Percentages Legacypac (talk) 06:39, 8 April 2019 (UTC)

Problems resulting from software that generates TeX code

A few years ago some software packages appeared on the web somewhere that generates TeX code for those who don't know how to do that by hand. I don't know what any of them are called, nor where or how they are found. But at first some of them generated code that looked as if it was written by someone with a major psychosis. That problem may have subsided. But we find things like this:

\dim\Sigma{\,>\,}4

I just fixed a whole bunch of things like that in the article titled Topological geometry.

Apparently one of these packages encloses binary operation symbols and binary relation symbols within {curly braces}, writing

a{+}b

instead of

a+b

and that affects what the reader sees, as follows:

${\begin{aligned}&a{+}b\qquad {\text{instead of}}\\&a+b\end{aligned}}$

(There are compelling reasons why TeX was designed to work that way, and I have found that some people don't know about those reasons.)

I suspect somebody sees that and "corrects" it by manually adding spaces, changing

a{+}b to a{\,+\,}b.

At any rate, multiple instances of the latter usage were in the Topological geometry article until I fixed them a moment ago.

Is there something that can be done to prevent this problem? Michael Hardy (talk) 18:23, 7 April 2019 (UTC)

@Michael Hardy: That notation has been present in the article from its very first version Special:PermaLink/812407777. However, its author Rboedi seems no longer active, so I doubt you'll manage to find out the reasons for it. --CiaPan (talk) 09:44, 8 April 2019 (UTC)

Here are more examples of backslash-comma-rightcurlybrace sequences (it uses a regular expression for the contents sought and an insource: magic word to search in the articles' definition instead of their rendered version) :

https://en.wikipedia.org/w/index.php?search=insource%3A%2F%5C%5C%2C%5C%7D%2F&title=Special%3ASearch&profile=advanced&fulltext=1&advancedSearch-current=%7B%7D&ns0=1

I have no idea of any solution to prevent such notation. --CiaPan (talk) 09:48, 8 April 2019 (UTC)

@CiaPan: I wonder if you missed the point. Not all instances of \,} are objectionable. The one I mentioned is objectionable for the reasons that I explained. Michael Hardy (talk) 19:46, 8 April 2019 (UTC)

@Michael Hardy: The above was kind of a proof of concept for my third approach to finding problematic math typesetting – and the first one going somehow towards useful results. I realize it was neither general enough nor specific enough – but it was just a try. If you don't like it, just forget it. I won't cry. And if you do, hopefully you'll be able to translate your objections' criteria to the language of regular expressions more precisely. --CiaPan (talk) 21:04, 8 April 2019 (UTC)

There is a chance that an Wikipedia:Edit filter could detect the syntax and give a warning. These are a little expensive so we really need to see how frequent it is to be worth the CPU time. Salix alba (talk): 21:55, 8 April 2019 (UTC)

Group theory: an oddly written sentence

See my remarks at Talk:Sims conjecture. Maybe someone here knows how to clean this up. Michael Hardy (talk) 17:40, 9 April 2019 (UTC)

Good news regarding the List of unsolved problems in mathematics

Every item in the List of unsolved problems in mathematics now has a bluelink, a reference or both. There's still an "other" section whose items could stand a proper classifying, if anyone feels like tackling that. XOR'easter (talk) 16:52, 7 April 2019 (UTC)

Invariant subspace problem is Functional analysis. Boris Tsirelson (talk) 18:24, 7 April 2019 (UTC)

Kaplansky's conjectures: the first (on group rings) is Algebra, the second (on Banach algebras) is Functional analysis. Boris Tsirelson (talk) 18:28, 7 April 2019 (UTC)

Erdős–Ulam problem and Scholz conjecture are Number theory.

Rendezvous problem is Game theory. Boris Tsirelson (talk) 07:38, 8 April 2019 (UTC)

Generalized star height problem is Formal language theory.

Kung–Traub conjecture is Numerical analysis. Boris Tsirelson (talk) 18:54, 8 April 2019 (UTC)

Thanks! I think all the sorting is done now. Work remains to be done on providing blurbs and copy-editing those that are currently there. XOR'easter (talk) 22:13, 8 April 2019 (UTC)

The current redlinks in the list: Taniyama's problems, Unsolved Problems on Mathematics for the 21st Century, DARPA's math challenges, Hartshorne conjectures, MNOP conjecture, Zariski multiplicity conjecture, Kung–Traub conjecture, Closed curve problem, Calluna's Pit, Birkhoff conjecture, Quantum unique ergodicity conjecture, Guralnick–Thompson conjecture, density hypothesis, Keating–Snaith conjecture, and Mazur's conjectures. In the "solved since 1995" list, we have Erdős sumset conjecture, Babai's problem, Anderson conjecture, Beck's 3-permutations conjecture, Kauffman–Harary conjecture, Normal scalar curvature conjecture, Böttcher–Wenzel Conjecture, Nirenberg–Treves conjecture, Alon–Friedgut conjecture, Kirillov's conjecture, Kouchnirenko’s conjecture, Deligne's conjecture on 1-motives, Erdős–Stewart conjecture and Harary's conjecture. Some of these could be articles, and some could conceivably be redirects to (sections of) existing pages, while perhaps some should be trimmed completely. XOR'easter (talk) 22:49, 9 April 2019 (UTC)

Replacing 2D renders with 3D models for solids

As some of you may know, a couple months back, Commons added the feature of uploading .stl 3D models. When opened, these files can be interacted with: zoom in, pan around, etc (see the 3D model on the right). I think they would be a great addition to articles about shapes and solids, such as Cube and Mobius strip. I've already replaced some of the lead images with 3D versions, and I hope to add more.

There are some issues to note though: The 3D models aren't interactive on mobile, and smooth surfaces aren't well replicated (Wikipedia's 3D file viewer doesn't smooth things). – XYZt (talk  |  contribs) – 05:58, 31 March 2019 (UTC)

I reverted those changes. They are black and white and inferior quality, and not interactive within an article. Tom Ruen (talk) 05:43, 10 April 2019 (UTC)

A new newsletter directory is out!

A new Newsletter directory has been created to replace the old, out-of-date one. If your WikiProject and its taskforces have newsletters (even inactive ones), or if you know of a missing newsletter (including from sister projects like WikiSpecies), please include it in the directory! The template can be a bit tricky, so if you need help, just post the newsletter on the template's talk page and someone will add it for you.

– Sent on behalf of Headbomb. 03:11, 11 April 2019 (UTC)

Draft:Quantile

Any use? If not just ignore it and it will go away. Thanks. Legacypac (talk) 06:53, 11 April 2019 (UTC)

• I doubt that this is even remotely useful. Reyk _YO! 09:05, 11 April 2019 (UTC)

New article assistance

Hi. Dimension of a scheme was just created. While I can check for structure, sourcing and copyvio stuff, would appreciate someone with more knowledge on the subject to take a look and make sure it's notable, and well, correct. Onel5969 ^{TT me} 14:51, 16 April 2019 (UTC)

To editor Onel5969: The stub is correct, and the subject notable. However, I am not sure whether it deserves having a stand alone article, as the definition is a trivial generalization of the dimension of an algebraic variety, and AFIK, anything that can be said on this topic results immediately from the second and third definition/properties listed in the linked article. Thus we must consider merging the new article as a section of the older one. D.Lazard (talk) 15:23, 16 April 2019 (UTC)

Thank you for your feedback, D.Lazard. Since I'm not up to snuff to discuss a merger, I'll leave that to you good folks over here at this project. Onel5969 ^{TT me} 15:44, 16 April 2019 (UTC)

I think it's conceptually better to have a separate article. Of course, the notion is a generalization of dimension of an algebraic variety. But already stuff like codimenion do not have a counterpart in the variety case. It's similar to the way we have scheme (mathematics) or morphism of schemes as separate articles from variety ones. Note the article is more than a definition. -- Taku (talk) 05:05, 17 April 2019 (UTC)

"Where triangle's area is triangle's area"

Seeing a recent note about "well-intended vandals" I recall an anecdotal case from my childhood. Our guest asked me a geometric problem; I solved it and wrote for him a solution. He read loudly: "...where triangle's area is triangle's area" and asked me menacingly: what's this? I asked: where? He showed me a line: "...where $S$ is triangle's area". After a moment of confusion, I realized that for him "$S$" means "triangle's area" and is read "triangle's area" by an eternal global indestructible convention. And, looking at him, I realized that he would aggressively defend this position. Being just a teenager (not a professor yet) I preferred to escape.

Back to Wiki. I think many (non-mathematicians) believe that all mathematical notations are an eternal global indestructible convention. Seeing a formula slightly different from their textbook they just "fix the error". I see no way to solve this problem. Do you?

By the way, I guess, many think (likewise) that words of different languages are related by the canonical bijective correspondence. But this is not our problem. :-) Boris Tsirelson (talk) 05:30, 18 April 2019 (UTC)

No, there is no way to solve this problem. (If you want to feel depressed, the article binary tree is a particularly hopeless case.) But I like your anecdote! --JBL (talk) 11:18, 18 April 2019 (UTC)

Template:Did you know nominations/Xiuxiong Chen

Please take a look at Template:Did you know nominations/Xiuxiong Chen. Thanks! Dennui (talk) 04:19, 13 April 2019 (UTC)

• @David Eppstein and Michael Hardy:

Sounds like something that could be appropriate for WP:ITN (although maybe not of wide enough interest), although they both look like they are eligible for WP:DYK. — MarkH21 (talk) 04:22, 13 April 2019 (UTC)

• Could be interesting for DYK, but I refuse to review DYKs that do not already have their full complement of QPQs. That's just laziness and I don't want to enable it. —David Eppstein (talk) 05:27, 16 April 2019 (UTC)

What are QPQs? Thanks. Dennui (talk) 15:16, 18 April 2019 (UTC)

See WP:QPQ and Quid pro quo.

I can do it. Dennui (talk) 15:58, 19 April 2019 (UTC)

Article product rule

Hello! In a book by Ivan Niven there is a proof which is probably the best proof of the product rule for beginners. It's not presented there. I would like to show it on the page, but English is not my native language, and I don't know how to paraphrase it very well. Would anyone be interested in adding it to the article? (for the benefit of the freshmen!). Dennui (talk) 19:26, 3 April 2019 (UTC)

Another think... In a book by Vladimir Arnold he shows a geometric interpretation of the product rule (with a rectangle... you know that image?...), which makes it "obvious". Dennui (talk) 19:27, 3 April 2019 (UTC)

Comments about a specific article belongs on that article's talk page. So please raise this issue instead at Talk:Product rule. Better yet, add the proof yourself, with citation, and let a more proficient English speaker clean up your text. Mgnbar (talk) 19:41, 3 April 2019 (UTC)

Thank you. I will do that. I just need to get some time (lots of exams already, on ring theory, calculus III, differential equations, etc... xD) Thanks again! Dennui (talk) 19:43, 3 April 2019 (UTC)

Which is the best proof for beginners made depend on which kind of beginner it is. I like this one: side of a rectangle moves while remaining parallel to where it was, thus causing the sides that it meets to grow or shrink. Now let two sides move. And then: The length of the side times the rate at which it moves equals the rate at which the area grows, but it's two sides. The lengths are ƒ; and g; the rate of growth of the side of length ƒ is g′ and the rate of growth of the side of length g is ƒ′. Michael Hardy (talk) 17:41, 4 April 2019 (UTC)

The length of the side times the rate at which it moves equals the rate at which the area grows should be "The length of the side times the rate at which it moves in the direction away from the other side and perpendicular to itself equals the rate at which the area grows." — MarkH21 (talk) 19:42, 4 April 2019 (UTC)

@MarkH21: It can move toward the other side, and then we would say that its rate of motion is negative. Michael Hardy (talk) 21:37, 5 April 2019 (UTC)

Yes, but my point was that movement parallel to the side won't change the area! Unimportant pedantry though. — MarkH21 (talk) 04:51, 6 April 2019 (UTC)

@Michael Hardy: That's essentially Arnol'd argument! Dennui (talk) 16:13, 19 April 2019 (UTC)

By the way, Niven's is more algebraic and is better than the ones presented. A guy named Howard Levi presents an algebraic one in his unusual calculus textbook. Dennui (talk) 16:14, 19 April 2019 (UTC)

Markov chain central limit theorem

I have created a new article titled Markov chain central limit theorem. It could use more work. In particular

• cited references
• So far just three articles link to it.
• Possibly a proof.
• Categories? I've included a bunch of these, but might I have missed some that will be obvious to others?

Michael Hardy (talk) 19:22, 20 April 2019 (UTC)

MathGenealogy link

I swapped the link in the {{mathgenealogy}} template (widely used in our biographies of mathematicians) from NDSU to AMS; please discuss this change at Template talk:MathGenealogy. —David Eppstein (talk) 20:08, 20 April 2019 (UTC)

Proposed restructuring on articles on plane curves

There is a discussion on the overlap and naming of the articles: algebraic curve, plane curve, curve, and differential geometry of curves (especially the first two). Much of the discussion is about the focus on plane curves in algebraic curve and curve. Some restructuring needs to be done because of the overlaps and redirects. — MarkH21 (talk) 05:06, 15 April 2019 (UTC)

(The discussion has been moved to the above link.) -- Taku (talk) 23:21, 20 April 2019 (UTC)

Conglomerate (set theory)

Here is an AfD that could use additional attention from those with expertise in mathematics and set theory. Eozhik (talk) 14:57, 21 April 2019 (UTC)

• Opinions of our experts in logic/set theory are very welcome there. Boris Tsirelson (talk) 05:34, 25 April 2019 (UTC)

Original research and examples in math articles

I'm pondering a rewrite of Discrete logarithm based on eight reliable sources that I've assembled at Talk:Discrete logarithm#Reliable sources. The rewrite raises some questions for me about what constitutes original research and plagiarism in math articles. Here are a few:

1. There is no mathematical reason to restrict the group to be finite, but all of the reliable sources (that I can find) do, so we must as well?
2. Therefore we cannot use the example G = {..., 1/100, 1/10, 1, 10, 100, ...}, which would otherwise be a very accessible example (because the discrete log there is just the common logarithm)?
3. Nor can we use the example of rational points on an elliptic curve? (Reliable sources do elliptic curves over finite fields.)
4. Reliable sources do examples in GF(2^4) and GF(3^2) but not GF(2^2), GF(2^3), etc. Can we do an example in GF(2^3), or would that be original research?
5. If we must pull our GF(p^k) example from GF(2^4) or GF(3^2), then must we use exactly the same instance of the problem? Or can we use a different base, for example?
6. If we must do exactly the same instance of the problem, then is there a danger of plagiarism?

I'm raising the issue here, because similar questions apply to many math articles, and I can't find specific policy. Mgnbar (talk) 14:20, 23 April 2019 (UTC)

Here you could apply WP:Common sense and WP:Be bold: if your new version is much better than the old one, you will not be reverted, or if you will, a compromise version could improve the article further. If your edit if controversial but can be modified for being an improvement, the discussion on the talk page, the precess WP:BRD can converge to a good result.

To answer to your specific questions and for getting an article that is not too technical, you must recall that for any element a of a group G denoted multiplicatively, the map $n\to a^{n}$ is a group homomorphism from $\mathbb {Z}$ to $G$ that induces an isomorphism from some $\mathbb {Z} /k\mathbb {Z}$ to some subgroup H of G. The discrete log is the inverse isomorphism. The two first examples of the article are the case where G is the group of positive real numbers. In this case, the discrete logarithm is simply a restriction of the usual logarithm of base a, and therefore does not deserve to be further studied. All what precedes is certainly not WP:OR, even if we are not able to source it exactly. So there is no reason for avoiding such examples. The example of $\mathbb {Z} /8\mathbb {Z}$ is also worth to be given, as the image H of the homomorphism has at most two elements, so the discrete logarithm is trivial. This explains why discrete logarithm is generally considered only when H is G or a large subgroup of it.

An other reason for which only finite groups are generally considered, is that the study of discrete logarithms is motivated by cryptography. So a large part of the study consists of searching for which finite groups the discrete logarithm is difficult to compute. So, it can be useful to give examples for which the discrete logarithm is not difficult to compute. D.Lazard (talk) 16:01, 23 April 2019 (UTC)

Another more trivial reason is that it should be a "discrete" logarithm, not just a logarithm in the usual sense from continuous mathematics. —David Eppstein (talk) 16:09, 23 April 2019 (UTC)

If the content you are adding is just a modest generalisation of well-documented material where the correctness of filling-in steps you make will be easily seen by mathematically well-informed editors, then I think this should be considered far from what Wikipedia intends original research to mean and furthermore not in need of additional sourcing. User:D.Lazard is right to reference WP:Common sense. I'm not clear just how far you want to go with your generalising, though

Wikipedia is meant to be a tertiary source; plagiarism is a non-issue.

A question in general: do we consider Wolfram MathWorld to be a reliable source? — Charles Stewart (talk) 19:06, 23 April 2019 (UTC)

It is widely used as a reference for at least certain kinds of articles. In my opinion this is not-so-good and it would be better to use better sources. The only time it's ever come up on WP:RSN seems to be this week, in a discussion of whether Wolfram Alpha is reliable. --JBL (talk) 19:18, 23 April 2019 (UTC)

I think there's a general sense that merely having an article on MathWorld isn't enough to merit a stand-alone article on Wikipedia (example). Nor is their choice of terminology always seen as reliable (example). XOR'easter (talk) 19:40, 23 April 2019 (UTC)

See WP:SCG, in particular the section Examples, derivations and restatements. It recommends that editors be given quite a lot of leeway. Jheald (talk) 21:40, 23 April 2019 (UTC)

Thanks, everyone. Special thanks to Jheald, whose link to WP:SCG is specifically what I wanted. Mgnbar (talk) 13:03, 25 April 2019 (UTC)

Be rash!

We are told to "Be bold" in editing. Not knowing algebraic geometry, I created a page called Kodaira map, redirecting it to Kodaira–Spencer map. If it needs further work, probably someone here knows what should be done. Michael Hardy (talk) 17:08, 18 April 2019 (UTC)

The redirect should be appropriate; I don't think there is a meaning of "Kodaira map" other than "Kodaira–Spencer" one. -- Taku (talk) 22:50, 20 April 2019 (UTC)

Sorry, I stand corrected. "Kodaira map" there means a map determined by a linear system. Authors in complex algebraic geometry use this terminology presumably motivated by Kodaira embedding while authors such as Hartshorne do not (I have therefore changed the redirect). -- Taku (talk) 23:38, 26 April 2019 (UTC)

Quasi-coherent sheaf on an algebraic stack

Another new article. Could use the eyes of you good folks here at this project. Thanks for taking the time.Onel5969 ^{TT me} 01:16, 27 April 2019 (UTC)

Versor

Versor (physics) has been proposed for deletion since April 18, and will probably be deleted (if nothing changes) today or tomorrow. Anyone want to save it? —David Eppstein (talk) 00:10, 25 April 2019 (UTC)

Surely "normalized vector" and "unit vector" are more common terms. I don't recall seeing "versor" being used to mean ${\hat {r}}={\vec {r}}/|{\vec {r}}|$ before. I'd suggest just making versor (physics) into a redirect to unit vector, but I doubt it's a common search query. XOR'easter (talk) 20:02, 25 April 2019 (UTC)

Someone removed the prod notice with the rationale that it survived AfD in 2016. I can't find this AfD. Rather than going through the whole rigmarole in order to delete an article that we don't need but which won't go away, I'm going to make it a redirect into unit vector. XOR'easter (talk) 16:29, 27 April 2019 (UTC)

Please do so. I don’t see how this is different from a unit vector. Also Scott Burley (talk · contribs) may have been looking at the talk page which seems to indicate that a user was planning on nominating the article for AfD without ever doing so. — MarkH21 (talk) 19:47, 27 April 2019 (UTC)

Talk:String theory#Semi-protected edit request on 29 March 2019

Hi! String theory is within this project's scope, so I was hoping somebody would have a look at the above edit request and implement/deny it as required. I understand pretty much nothing the abstract for the source is saying. Thanks in advance, NiciVampireHeart 22:36, 27 April 2019 (UTC)

I've weighed in (against), but I don't feel confident enough to close. — Arthur Rubin (talk) 23:36, 27 April 2019 (UTC)

(But string theory is't exactly math; it *uses* math but not math. -- Taku (talk) 01:10, 28 April 2019 (UTC))

That's why I didn't feel confident enough to close. The article says "perhaps" twice in the sentence in the body; essentially that the larger number may be the number of parameter sets, which may have associated vacuums (vacuua?). — Arthur Rubin (talk) 03:22, 28 April 2019 (UTC)