Wikipedia talk:WikiProject Mathematics/Archive/2019/Aug

Draft:Artin-Tits groups

Someone might want to rescue Draft:Artin-Tits groups from languishing in draft state. It looks well-sourced and reasonably close to publishable to me (although the title should be Artin–Tits group with an en-dash and no plural, and it needs more wikilinks), but a request to publish it was declined by Theroadislong for the bogus reason of being "largely incomprehensible" (it is a technical subject that would not reasonably be expected to be comprehensible to non-mathematicians). —David Eppstein (talk) 21:13, 27 July 2019 (UTC)

Thank you for your input, I declined it because I didn't think there were sufficient sources, I merely commented that it was largely incomprehensible, I have added some Wiki linking and moved it to article space. It does need better sourcing and I am sure it could be written in a more accessible way as is Tits group. Theroadislong (talk) 21:44, 27 July 2019 (UTC)

How is this distinct from Artin group? XOR'easter (talk) 22:17, 27 July 2019 (UTC)

It's not. At first I thought the definition at Artin group only allowed for relations between consecutive generators, but I think that's just a quirk of how it was written and not actually what's intended. Otherwise, I guess we should just try to see if there's anything different enough in the new article worth merging in, and redirect to Artin group (with no opinion on what the actual ultimate name should be). –Deacon Vorbis (carbon • videos) 01:16, 28 July 2019 (UTC)

I moved it from Artin-Tits groups to Artin–Tits groups and then from there to Artin–Tits group. I am uncertain whether that last move is right.

So far no other articles link to this new article. Michael Hardy (talk) 13:49, 28 July 2019 (UTC)

Definitely needs to be merged with Artin group. This is just an alternate name. — MarkH21 (talk) 16:11, 28 July 2019 (UTC)

I am (enthusiastically) willing to work on merging "Artin groups" and "Artin-Tits groups", which by coincidence have been created almost simultaneously. I just need a few days, and I'll post a proposal. OK ? I am a specialist of this area of mathematics, and should be able to make something coherent. Patrick Dehornoy 15:11, 1 August 2019 (UTC)

PS. I enterely agree with the critics about the first draft of "Artin-Tits groups", of which I am the author. I am not yet very familiar with the editorial features of Wikipedia, but I should learn fast... Patrick Dehornoy 15:15, 1 August 2019 (UTC)

Yes, please do, and thank you beforehand! Boris Tsirelson (talk) 15:27, 1 August 2019 (UTC)

I now posted a tentative version of the new, completed page merging "Artin group" and "Artin-Tits group". Please check and criticize. Patrick Dehornoy 19:47, 3 August 2019 (UTC)

P.Aczel reference to Recursive definition

Please see Talk:Recursive definition#(Aczel 1978:740ff). --CiaPan (talk) 16:58, 4 August 2019 (UTC)

Opinions of an edit

Any opinions of this edit? I wonder if there's a third way of phrasing the thing that's better. Michael Hardy (talk) 03:18, 8 August 2019 (UTC)

I don't think the edit is an improvement; I suspect that that is the sort of sentence which can be tweaked all sorts of ways without making it much more or much less clear. XOR'easter (talk) 03:28, 8 August 2019 (UTC)

I took a red pen to the opening paragraph. I tried to make it somewhat better, but there's still plenty that could be done I'm sure. As I said in the edit summary, please feel free to revert or tweak further as you like. –Deacon Vorbis (carbon • videos) 03:42, 8 August 2019 (UTC)

I like your version better than what was there before, and further tweaked it. --JBL (talk) 13:20, 8 August 2019 (UTC)

More on Artin–Tits group

The original creator of the article titled Artin–Tits group has told me that he merged it into Artin group. After that I moved it to Artin–Tits group. So two questions arise:

• Is that the best title for the article?
• Was the merger done well, and what further work may be needed?

Michael Hardy (talk) 02:49, 6 August 2019 (UTC)

I did some minor cleanup, but I am not sure which is the more prevalent name. The article mostly consists of a list properties and results that does not seem entirely appropriate here too. Thoughts? — MarkH21 (talk) 03:24, 6 August 2019 (UTC)

Is there the right place to answer these questions? (I am an old mathematician, but a newcomer on Wikipedia...).

About the title, there is no obvious answer. Artin-Tits groups are a vast subject, and many people work on them. More importantly, there are several classes of such groups, and several specialized subcommunities. "Artin-Tits groups" is more common in recent references. But an important subcommity, those working on "Right Angled Artin-Tits groups" ("RAAGs") uniformly use "right-angled Artin groups", and not "right-angled Artin-Tits groups". Ideally, the two names should be possible. What is the better solution for this situation?

About the content, your comment "The article mostly consists of a list of properties and results that does not seem entirely appropriate here too" is very interesting. This is precisely what I aimed at doing: it seems to me that the paper, as it stands, would be all I expect to learn about these groups if I were a newcomer in the subject, namely the state-of-the-art results which are considered important (top journals references). But you are a hugely qualified Wiki author, and I would like to know your opinion: what would be typically missing, or useless? As I intend to invest myself in writing on several subject of my expertise area, I am much willing to understand what is the exact aim and philosophy of Wiki (as far as mathematical subjects are concerned). Thank you very much for any comment....

A minor point: your talk page is wonderfully structured, with a summarized "abstract" presenting you. How could I do the same for me? Patrick Dehornoy 17:30, 8 August 2019 (GMT)

Thanks for your contributions! Wikipedia has many idiosyncrasies, and I'm not sure anyone is completely familiar or comfortable with all of them, no matter how long they've been here. Discussing the article here is fine; another good venue is its Talk page, but it's possible that more people will notice your comments here than there. (Wikipedia has the handy feature known as the Watchlist, which is a list of pages that you select as important to you so that the software will automatically notify you when they are changed. Many more people "watch" this page than Artin–Tits group, so feedback may happen faster here.) Regarding the "list of properties and results" comment, I think the concern is a matter of prose style more than anything else. We like Wikipedia articles to be prose paragraphs, with a logical flow from one to the next, rather than bulleted lists, which too often read in a choppy way (like PowerPoint slides filled up with text). Perhaps the mathematics articles that have received the "Good article" stamp of approval may be of inspiration. XOR'easter (talk) 16:56, 8 August 2019 (UTC)

You can put the text "{{Talk header}}" at the top of your own Talk page to make a header. This is an example of a template, like the {{citation}} and {{cite journal}} templates you have already encountered. XOR'easter (talk) 17:01, 8 August 2019 (UTC)

Indeed, that is what my comment meant. Wikipedia articles should be well-written prose that introduces and describes domains of knowledge, rather than an academic reference or a large collection of bullet-point facts.

I am far from well-qualified WP editor, and I appreciate the compliments but your contributions here may be equally or more important than my own. As XOR'easter mentioned, the top of my talk page has the aforementioned talk header template. If you mean my user page itself, feel free to copy bits that you may want to include on your own. The procedure to write on it is the same as writing on any other Wikipedia page! — MarkH21 (talk) 18:00, 8 August 2019 (UTC)

Australian IP edit-wars in Quaternion

Insists that knows the only correct notation and the rest of the world the preceding version is “wrong”. Look at https://en.wikipedia.org/w/index.php?title=Quaternion&action=history&offset=2019081216&limit=9 and talk:Quaternion‎‎ #Exponential, logarithm, and power functions, please. Incnis Mrsi (talk) 15:42, 12 August 2019 (UTC)

Strange editor

Please look at these edits: Special:Contributions/Hanumantw; not vandalism, but... something strange? Boris Tsirelson (talk) 16:07, 11 August 2019 (UTC)

It may not be deliberate destruction motivated by hatred of the good, but damaging things as a result of carelessness or ignorance can be just as harmful. JRSpriggs (talk) 01:21, 12 August 2019 (UTC)

Now blocked indefinitely. Boris Tsirelson (talk) 08:15, 17 August 2019 (UTC)

Factorial and double exponential

See recent history of factorial. An editor there insists that the constant function f(x)=1 is an example of a double exponential function and on using that example to change the statement that the factorial grows "slower than double exponential functions" to the overly-pedantic "slower than many double exponential functions e.g. (example)". I don't think this is an improvement, but additional opinions might be more helpful than my repeated reversion of these edits. —David Eppstein (talk) 22:46, 17 August 2019 (UTC)

Very depressing bug in the editor

I was editing Talk:Axiom of union#Independence of the axiom of union. I had planned out what I wanted to say in my head and was rushing to get it committed to writing before I forgot it. As usual, there were times when I needed to look at another article to get the correct spelling or latex symbol. This time, I went to the Axiom of powerset to get the symbol for the powerset. I had to enter the editor to see the source of that article to get "\mathcal (P)". When I tried to back out of that edit and return to my original edit, it would not allow me to do so. Ultimately, I had to "resend" to escape and as a result I lost most of what I had already written. I cannot say in words just how discouraging this is. For several minutes I sat stunned, enervated, unable to do anything. Eventually, I forced myself to re-enter an approximation to what I had written before.

This is not the first time this has happened. It has happened several times before. But it is infrequent enough that I forget to take precautions against it. Is there some way to get this bug fixed? JRSpriggs (talk) 01:20, 13 August 2019 (UTC)

Which editor(s) are you using? It may make a difference. I haven't had problems except when my browser exceeded available memory on my machine and locked up.... 01:36, 13 August 2019 (UTC)

(edit conflict × how ironic) Not quite sure I'm following your steps exactly. Do you mean that from the edit window, you went directly to another article (maybe via the search bar), and then hit the back button on your browser only to find that it hadn't saved the state of what you were doing? If so, then that's not really a bug; there's no guarantee that a browser will keep the current state of what you're doing cached if you move forward to another page. Ideally it will try to, especially for simple stuff, but that's not something you can rely on. On the other hand, if you go browsing in another tab or window, then all should be fine, but if not, it's likely browser-related, not a Mediawiki issue. Or have I misunderstood what the problem is? –Deacon Vorbis (carbon • videos) 01:42, 13 August 2019 (UTC)

To Arthur Rubin: I was just using the normal editor provided by Wikipedia, the old one, not the visual editor.

To Deacon Vorbis: Yes, I think you got it right. My browser is Firefox. I was thinking that there are three things I might do to work around this: (1) copy the source of my edit into a file on my disk before going on such expeditions, just in case this happens; (2) do the search and cutting (before pasting) in another copy of Firefox; or (3) save my edit before I am really done and then edit again after I have done the search. What do you recommend? JRSpriggs (talk) 06:13, 13 August 2019 (UTC)

This depends on the browser, and even on the browser version. With Safari, non-saved edits are sometime lost, sometime not lost when one come back to a page with the back button. If one has left the page after clicking "show preview", edits are sometimes kept, but with the most recent versions they are kept only if one does almost nothing (no search in the page, no diff, no look to the source, ...) with the visited page(s). Therefore, I have now the habit to use another tab for navigation during an editing process, and, when searching or following a link from the edit window, to to it with the right button and "open the link in another tab". D.Lazard (talk) 07:51, 13 August 2019 (UTC)

That's more or less my habit as well. XOR'easter (talk) 15:22, 13 August 2019 (UTC)

To Lazard: Thanks for the suggestion. I will try the right-click on a link to get a separate tab method. JRSpriggs (talk) 01:29, 14 August 2019 (UTC)

I’m late but I can confirm I had the same type of (very disheartening) experiences before on multiple occasions; they may have been in Wikipedia or some other website. Since I couldn’t figure out the behavior (how unsubmitted text is handled), my habit has been to copy large text into the clipboard before hitting submit (if edits are small, I don’t bother). —- Taku (talk) 22:54, 17 August 2019 (UTC)

Puzzled by watchlist

This WPM talk page is on my watchlist, in the sense that I do see the corresponding blue star between "View history" and "More". And nevertheless it does not appear on my watchlist. This is a new phenomenon (the last week or two). Mostly, my watchlist looks as before; but some items are missing, I do not know why. If I click the blue star ("remove this page from your watchlist") and then click the (no more blue) start again (making it blue again), it helps; the page returns to my watchlist. But afterward it disappears again. Why so? Boris Tsirelson (talk) 06:51, 19 August 2019 (UTC)

Or maybe not quite so. I just tried to click twice the star on Talk:Normal distribution (edited by me yesterday, and by a bot today); it did not help. Boris Tsirelson (talk) 06:57, 19 August 2019 (UTC)

Empty square root

The empty square root ${\sqrt {\;}}$ generated as <math>\sqrt{\;}</math> is very badly aligned (I have found this in nth root). The reason seems that the alignment is done on the center of the argument and that the space character is viewed as a zero-height character placed at the bottom of the line. For having a normal alignment such as ${\sqrt {{~^{~}}^{~}\!\!}},$ I have used <math>\sqrt{{~^~}^~\!\!}</math>. Do someone know a less weird method for a similar result? D.Lazard (talk) 07:42, 19 August 2019 (UTC)

P.S. A similar result (${\sqrt {\color {white}{x}}}$) is obtained with <math>\sqrt{\color{white}{x}}</math>, but it is also semantically doubtful. D.Lazard (talk) 07:54, 19 August 2019 (UTC)

Try <math>\surd</math>, giving $\surd$. No horizontal bar, but it does better align it with the text. If WP handled real TeX, \sqrt{\phantom{x}} or \sqrt{\hspace{1em}} would be reasonable ways to create an empty square root.--{{u|Mark viking}} {Talk} 08:14, 19 August 2019 (UTC)

Convex hull lower bound

Please see Talk:Convex_hull_algorithms#Lower_bound_on_computational_complexity --GunterS (talk) 08:43, 21 August 2019 (UTC)

Already answered there. D.Lazard (talk) 09:21, 21 August 2019 (UTC)

Cocycle of a group action

It is a pity that cocycle of a group action is not treated; neither in "Cocycle", nor in "Group action (mathematics)". See Talk:Cocycle#Cocycle of a group action. Boris Tsirelson (talk) 08:22, 24 August 2019 (UTC)

Pollard algorithms

Are Pollard's rho algorithm for logarithms and Pollard's kangaroo algorithm about the same thing? Bubba73 ^{You talkin' to me?} 15:35, 24 August 2019 (UTC)

I am not acquainted with these; but in "Pollard's kangaroo algorithm" I read: "Pollard's kangaroo algorithm ... introduced ... in the same paper [1] as his better-known Pollard's rho algorithm"; if so, then they are two different algorithms "for solving the same problem". Boris Tsirelson (talk) 17:21, 24 August 2019 (UTC)

The way it seems to me is that Pollard's rho algorithm is for factoring numbers and the other two use a similar method for finding discrete logarithms. Bubba73 ^{You talkin' to me?} 18:09, 24 August 2019 (UTC)

Draft:Genus of a group

Is this a topic that deserves a separate article? Genus (mathematics)#Graph theory already covers the topic but a quick Google search shows the topic is of independent interest. (I admit I’m not a specialist on this area so the others might know better.) — Taku (talk) 22:52, 26 August 2019 (UTC)

No separate IMHO, but definitely as little as redirect to Cayley graph, especially because the article already mentions the genus of it (albeit without a wiki link). Incnis Mrsi (talk) 16:21, 27 August 2019 (UTC)

Where in Cayley graph is the genus of a Cayley graph discussed? --JBL (talk) 17:09, 27 August 2019 (UTC)

In fact, nowhere the genus of the graph (I suffered a glitch). Anyway the concept is defined via Cayley graphs and may deserve a section there. Incnis Mrsi (talk) 17:24, 27 August 2019 (UTC)

I'd start by writing about the concept at Cayley graph and then breaking it out into its own article if that material grows too big. XOR'easter (talk) 17:45, 27 August 2019 (UTC)

I agree that Cayley graph is a better initial home for such material than Genus (mathematics)#Graph theory. --JBL (talk) 18:46, 27 August 2019 (UTC)

Started! XOR'easter (talk) 19:18, 27 August 2019 (UTC)

Thank you for the works. I will now make the draft page a redirect. —- Taku (talk) 23:07, 27 August 2019 (UTC)