Wikipedia talk:WikiProject Mathematics/Archive/2019/Jan

Repairing recursions at Harmonic number

I posted details of my problem at the TP there, but additionally take the freedom to submit this here to a broader public:

Please, could someone more dignified than I am take care about adding either missing or more reasonable base cases in the recurrence relations in this article? Both my efforts to either generally have as the base case some $H_{0}=0,$ not particularly coined as one of the harmonic numbers, or to specifically introduce it at places in specific need, were promptly reverted, ignoring that $H_{0}=0$ is already in use at the end of this section. Thanks for taking in consideration. Purgy (talk) 10:31, 10 January 2019 (UTC)

Errors in Alhazen's problem article

Hello

Sorry, I don't speak english, so be kind when you read me.

There is an error in Alhazen's problem article and also in Ibn al-Haytham article: we can read "This (i.e Alhazen's problem) eventually led Alhazen to derive a formula for the sum of fourth powers". There is a lot of treatises written by Ibn al-Haytham. In the The book of Optics, we can find the Alhazen' problem. Its solution has nothing to do with sum of the fourth powers (See A.I. Sabra, Ibn al-Haytham' lemas for solving Alhazen's Problem). The sum of four power is in an another treatise ( fi misahat al-mujassam al-mukafi On the Measurement of the paraboloid). You can read the source (Victor J. Katz (1995), "Ideas of Calculus in Islam and India). The banned contributor Jagged 85 made this mistake in june 2007[1].

He added also a second error "Mathematicians were not able to find an algebraic solution to the problem until the end of the 20th century" It is non sens because before 20th, people knows already that intersection of two conics led to a quartic equation. (see Paul Bode (1892),«Die Alhazensche Spiegel-Aufgabe in ihrer historischen Entwicklung nebst einer analytischen Lösung des verallgemeinerten Problems to see all the solutions (algebraic, trigonometric, geometric...). p. 86 you can see an equation of the Huygens' hyperbole.

Can you fix these two errors ? Thank's. HB (talk) 21:10, 12 January 2019 (UTC)

Yet more links to DAB pages

I have collected yet another, but this time very small, batch of articles which include mathematics-related links to DAB pages. Expert attention in solving these puzzles would be welcome. If you solve any of them, remove the {{dn}} tag from the article and post {{done}} here.

• Uniformizable space  Done
• Termial  Done
• Hybrid number  Done

Thanks in advance, Narky Blert (talk) 04:05, 20 January 2019 (UTC)

Exponentiation

Totally unproductive sniping -- discussion should be at Talk:Exponentiation. Other editors are invited to participate there. --JBL (talk) 15:35, 20 January 2019 (UTC)
The following discussion has been closed. Please do not modify it.
User:Jasper Deng seems to think that explicitly mentioning the exceptions to a supposed "mathematical identity" is somehow inferior than just saying the equation holds "in general". I made some changes which he objected to for the use of elementary logical quantification, which I have now removed. He continues to revert for no apparent reason. He's also made some wildly wrong statements like "The convention in mathematics is to use intensional definitions" and has confused the logical negation of an equality with a quantified inequality. Stemdude (talk) 02:05, 20 January 2019 (UTC) @Stemdude: insists on an arcane interpretation of the ≠ symbol that is contrary to literally every textbook I've seen. I'm already at WP:3RR. I'd like someone with more experience in formal logic to chime in, however, I believe that his concerns are unfounded. Please leave any comments you may have on that article's talk page.--Jasper Deng (talk) 02:41, 20 January 2019 (UTC) Both of you get a trout for a ridiculously childish edit war. The discussion on the article talk page has only been open for a couple of hours, and the first comment by a third party (me) was just added to the conversation. Give the undo button a rest and wait for other editors to weigh in. Perhaps while you're waiting you could try to find some reliable sources on which to base the relevant parts of the article. --JBL (talk) 03:02, 20 January 2019 (UTC) All I'm trying to do is prevent Wikipedia containing obviously incorrect statements i.e. claiming that a mathematical identity can only hold "in general" rather than being true by definition. I'm disappointed that such an obvious fact seems to elude so many Wikipedians. But I'm not going to be bullied off this project because an admin wants to win a foolish edit war. Stemdude (talk) 09:13, 20 January 2019 (UTC) In case it wasn't clear, the crux of the issue is that you interpret $f(x)\neq g(x)$ as $\forall xf(x)\neq g(x)$. This is not at all the most common interpretation of the $\neq$ sign and is absurd when comparing the two functions as objects in their own right, since $f(x)\neq g(x)$ is perfectly valid as an inequality of sets. @Stemdude: I'd trust User:Joel B. Lewis who is a professional mathematician. Definitely not a "deranged undergraduate". By the way, I spoke with some graduate student friends of mine and they also do not support your view. Most importantly, if independent reliable sources have not corroborated your interpretation of that sign, then we cannot accept it. See WP:EXPERT.--Jasper Deng (talk) 09:45, 20 January 2019 (UTC) This has NOTHING to do with my edit to Exponentiation whatsoever. You seem to want to argue about irrelevant issues rather than talking about improving the article in question and have gone off on a wild tangent. And I never called JBL a "deranged undergraduate"; I was referring to you but I decided to revert that comment in the interests of keeping things calm. I'm amazed that an admin is behaving so childishly. Stemdude (talk) 09:56, 20 January 2019 (UTC) And if we need any more proof that Jasper Deng has no idea what he is talking about, check out this [this|https://en.wikipedia.org/w/index.php?title=Talk:Euler%27s_formula&diff=877389352&oldid=877387501] comment in which he confuses (complex) exponentiation with logarithm. ix isn't multivalued! Stemdude (talk) 10:09, 20 January 2019 (UTC) @Stemdude: okay, then it's time to ask for your qualifications because $i^{x}$ by its most conventional definition very much is, and I in particular am not the first one to state that there. What I said above most definitely does have everything to do with your edit, because your edit concerns a statement of that form. Don't be so self-righteous, especially when you have shown no evidence of consulting external resources to support your claim. You're wasting your breath without citing reliable sources in any case.--Jasper Deng (talk) 10:18, 20 January 2019 (UTC)

Polygons

Hi

1) It is César R. K. Stradiotto again.
2) To avoid any problems, now I just consulted the video tutorial to post complains about wikipedia pages. thanks.
3) The last post I did was about this page [List of two-dimensional geometric shapes][2], where the section

Polygons with specific numbers of sides -Quadrilaterals --Trapezus

were pointing to a Wikipedia "Pornhub"-about page.

4) I just saw the link were changed, but still not corrected: Now the link on Trapezus is pointing to a geographic place: Trabzon [3] (Just look for Wikipedia Trabzon, on Google).

That´s it.

Cordially

César R. K. Stradiotto — Preceding unsigned comment added by 189.85.185.93 (talk) 15:27, 21 January 2019 (UTC)

As far as I understand, there is no such polygon (and there is such geographic place). Thus I delete it from the list of two-dimensional geometric shapes. If anyone saw such polygon, correct me. Boris Tsirelson (talk) 15:50, 21 January 2019 (UTC)

"Trapezus" could be a misspelling of "trapezoid", "trapezius", "Trabzon", or "Trapezus (Arcadia)". JRSpriggs (talk) 09:20, 22 January 2019 (UTC)

Invalid variants of mathematical theorems

Look at the history of Lagrange's four-square theorem. I added a section revealing an invalid variant of the theorem. This means a variant of what the theorem says that would make it false. Three times, however, someone reverted me. Interestingly enough, there's an article, Beal's conjecture, which has a similar section that no one objected to. We need some kind of discussion on what the best rule for how articles on mathematical theorems should deal with sections like this. Georgia guy (talk) 15:50, 22 January 2019 (UTC)

I think you both are correct: the material can be included but not in the form you gave it. It would probabably be helpful if you framed this in the context of what is the algorithmic complexity of finding the four squares; then you could mention that greedy doesn't work, but something else (properly sourced) does. Another problem with your edit is that you did not provide a source for greedy not working, so it ends up looking like original research. You could try {{cite OEIS|A006892}} (Sloane, N. J. A. (ed.). "Sequence A006892". The On-Line Encyclopedia of Integer Sequences. OEIS Foundation.). That would also give you an alternative method of framing this in terms of the growth pattern of this sequence. I don't see a need for a general rule to handle this sort of case. —David Eppstein (talk) 16:08, 22 January 2019 (UTC)

Please show me a better form. Georgia guy (talk) 16:16, 22 January 2019 (UTC)

It involves doing some literature research. I don't have time for that this morning. —David Eppstein (talk) 16:32, 22 January 2019 (UTC)

As a general question, we certainly don't want to open the door to adding random text of the form "proposition P is true, but this stronger proposition Q that I just made up is false". There needs to be a reason to mention Q. The best sort of reason would be that multiple sources consider Q a natural thing to mention, but if we all agree that Q is an easy misunderstanding of the statement of P I might be willing to look the other way.

In this case, though, I just don't see it at all. There's nothing in the statement of four-squares that invites considering the greedy algorithm. And if you do wonder about it, it's trivial to convince yourself that the greedy algorithm gives unboundedly long sequences of squares, since you if you have a number that gives N squares, then just add a really big square number to that, and now you've got N+1. --Trovatore (talk) 17:14, 22 January 2019 (UTC)

I disagree. I think it is very natural to go from a statement that a number can be represented using few members of some set, to the question of how few members the greedy algorithm uses. I have a publication on exactly those issues (not suitable as a reference here because it doesn't mention sums of squares). There are many sequences in OEIS based on this principle of greedy representations as sums from some set, and published literature going back to Pillai in 1930. And as you point out sequences of natural density 0 (such as squares) never have a finite bound on the lengths of their greedy sums. The big problem with the proposed addition was not its interestingness or naturalness or triviality, but (as you point out) that it was totally unsourced and therefore appeared to be original research. You undercut that issue by pointing out how simple the calculation is, though. —David Eppstein (talk) 20:48, 22 January 2019 (UTC)

Well, it's not an unnatural direction to go, I suppose. But it's not directly suggested by the statement. Contrast with the section GG is referring to in Beal conjecture, where the "variants" section seems to be a (slightly clumsy) attempt to explain why some clauses in the theorem's hypothesis are necessary. That directly implicates Grice's maxim of relevance; this doesn't. --Trovatore (talk) 21:26, 22 January 2019 (UTC)

Maybe it's my background, but to me the question "how do we find it" is directly suggested by any existence theorem, and "does the most obvious method of trying to find it work" is not far behind. —David Eppstein (talk) 00:47, 23 January 2019 (UTC)

Category: Mathematical object

Jamgoodman adds systematically Category:Mathematical objects to many mathematical articles. As almost all mathematical articles could belong to this category, I have open a deletion discussion at Wikipedia:Categories for discussion/Log/2019 January 23#Category:Mathematical objects. D.Lazard (talk) 15:36, 23 January 2019 (UTC)

Please also see the discussion on my talk page. The category is a perfectly good category, but generally articles should belong to its subcategories rather than directly to it. It is a useful distinction e.g. from the category "Areas of mathematics" and should not be deleted. --JBL (talk) 16:35, 23 January 2019 (UTC)

Documenta Mathematica

Is this the same thing as Journal der Deutschen Mathematiker-Vereinigung? Or Jahresbericht der Deutschen Mathematiker-Vereinigung?

Mostly asking to see if redirecting to German Mathematical Society would be appropriate. Headbomb {t · c · p · b} 20:39, 24 January 2019 (UTC)

It's also published by the DMV but it has a different ISSN than JDMV (ISSN 1431-0635 for Doc. Math., ISSN 0012-0456 for JDMV) and is indexed separately from JDMV in MathSciNet, with both of them publishing at the same time. So I'm pretty sure it's not the same thing. Nevertheless, if we don't have an article on Doc. Math., and it's mentioned in the DMV article, a redirect might make sense. I don't think there is a "Journal of the..."; I've only seen it as Jber. —David Eppstein (talk) 21:42, 24 January 2019 (UTC)

Created the redirect. According to the [4], 'Journal der Deutschen Mathematiker-Vereinigung' is a subtitle. Headbomb {t · c · p · b} 03:02, 25 January 2019 (UTC)

List of constants

In case anyone's looking for something to do, I just stumbled upon List of mathematical constants, which is in fairly rough shape. The table is somewhat broken, and perhaps has some columns that could simply be removed. It's a bigger project than I have time for right now, so I thought I'd mention here. –Deacon Vorbis (carbon • videos) 15:07, 30 January 2019 (UTC)

Wow. --JBL (talk) 15:15, 30 January 2019 (UTC)

Similar issues at List of mathematical symbols and List of mathematical symbols by subject. Not only some columns could be removed, but also many entries should be removed: those that are far to be standard, and must be defined before being used (such as $\bullet ,$ ↯, ⨳, and many others).

Also Mathematical symbol redirects to List of mathematical symbols, while it deserves to have a regular article. D.Lazard (talk) 16:14, 30 January 2019 (UTC)