User talk:Ulf Rehmann

Another spammer

Please notify me about spammers by dropping a note here with a link to the spammed page like this:

YOLO loveING YOLO shouting this in a club cause Soul yolo‎ --Jjg 19:22, 11 May 2012 (CEST)

User:EddieLandau --Jjg 19:25, 11 May 2012 (CEST)

Antonio Miguel Ive been well thanks Russia How r u How was Brazil‎ --Jjg 21:04, 11 May 2012 (CEST)

---

Ulf - we have rather a lot of these low-lifes at the moment (my guess is that this site is now on an "easy to spam" list circulating the spammer websites), so I have taken to removing the links on their pages so as to deprive them of what they want, a chance google crawl of a page pointing to their pathetic porn or gambling site on a high-value website like this. So, if you see a remark (links cleared) or (page cleared) on a spam report, please check the history for the reason why. --Jjg 22:56, 9 May 2012 (CEST)

Wouldn't it be better to delete spam accounts (after checking their contributions) instead of only blocking them? Blocked-only spammers still "spam" the user list. --Peter Schmitt 13:36, 27 January 2012 (CET)

Deleting spam accounts would be good, but there is no tool to delete a user here, see this. Just "pages" can be deleted. There is this Extension, I can propose to the sysadmins to install this extension (I myself can't do that). However, today they announced to me they will install a "Double-opt-in" registration process in order to defeat spammers. --Ulf Rehmann 15:35, 27 January 2012 (CET)

Even if spammers are kept out in the future this extension would be useful to clean up the past. --Peter Schmitt 00:33, 29 January 2012 (CET)

Might be an idea to block their IP for a couple of years too. --Jjg 00:38, 9 May 2012 (CEST)

We do that. --Ulf Rehmann 00:47, 9 May 2012 (CEST)

Can't we mandate all who open accounts to pass some super-Turing test? E.g., asking to register at least one of the MathSciNet "area codes"? I would assume that spammers and spambots would not have an easy access to MSN, while the majority of sincere users will be able to prove their humanity one way or another... -- Sergei Yakovenko 08:13, 10 May 2012 (CEST)

I have notified the admins about the recent problems. --Ulf Rehmann 23:40, 10 May 2012 (CEST)

Could we also have such a meeting place?

Something similar to this would be useful. (Or like this. But please, not like this.) And by the way, did you see this topic? An EoM-relevant quote from there follows. --Boris Tsirelson 15:05, 13 December 2011 (CET)

"I am the product manager at Springer responsible for managing the technical implementation of the new EOM wiki. I have adapted the citation template documentation for the template:SpringerEOM to use the proper naming convention for the new site. I will be closely watching the site's incoming traffic from WP and updating the ids of WP citations that used the template so that their link-outs will resolve to not only the correct article in the new EOM wiki, but to the version of the article that is identical to the version previously hosted at eom.springer.de.--Nbrothers (talk) 16:13, 30 November 2011 (UTC)"

Categories

Something is wrong, I guess; we have now both Category:Dynamical Systems and Ordinary Differential Equation and Category:Dynamical Systems and Ordinary Differential Equations. :-)

Thanks. I guess I fixed this.--Ulf Rehmann 22:43, 9 January 2012 (CET)

And by the way: when all first chars are uppercase and when they are not? --Boris Tsirelson 22:23, 9 January 2012 (CET)

Aha, I was careless here. What is the Wikipedia convention? What happens if I rename a category? Will the renameing shift all the entries in its files?--Ulf Rehmann 22:43, 9 January 2012 (CET)

On WP they are generally lowercase; example. No, I am afraid changes in all pages need to be made by hand. And by the way, also replacement of "unofficial" templates to "official" ones (when ready) will be made by hand. Unless we'll create corresponding bots, which is feasible but not so easy (as any programming). --Boris Tsirelson 22:47, 9 January 2012 (CET)

Ok. I'l fix the cases issues. Concerning the templates we'll see later. I guess when programming bots one needs admin access to the machine where the wikimedia is installed? That I don't have (otherwise we'd already have the ParserFunctions ext.... --Ulf Rehmann 23:32, 9 January 2012 (CET)

You mean, the root access. No. Not even admin rights on the wiki. I did it once on Wikipedia, being just a plain user. --Boris Tsirelson 06:47, 10 January 2012 (CET)

But for organizational reasons bots are best run from separate "bot" accounts. This link may be helpful. --Peter Schmitt 15:16, 10 January 2012 (CET)

Thanks for your hints. I did find perl based bot which works.--Ulf Rehmann 12:25, 11 January 2012 (CET)

Keeping spammers away

It could help, to provide an account only after asking a math question and getting the right answer. The question should be replaced from time to time. Here is one, for example. :-)

Calculate the area of the set of points $(x,y)\in\R^2$ satisfying \[ \Big( \frac{x-5}{\sqrt{15}-2} \Big)^2 + \Big( \frac{y+3}{\sqrt{15}+2} \Big)^2 \le \frac3\pi \, . \] --Boris Tsirelson 11:12, 14 January 2012 (CET)

And we should not underestimate the problem. See e.g. [1], [2]. --Boris Tsirelson 13:27, 14 January 2012 (CET)

Something like the above has been implemented. Look at Special:Version, search for "ConfirmEdit". Questions are less sophisticated though. --Ulf Rehmann 13:35, 1 February 2012 (CET)

Well, I am glad it is considered seriously. But now (a) it asks me a question whenever I edit something, and (b) a bot can solve such questions easily. I'd prefer not to bother users that have accounts, but to ask harder questions when creating accounts. Like QuestyCaptcha. Of course, it is reasonable to reconsider accounts that exist when such system starts. --Boris Tsirelson 20:51, 3 February 2012 (CET)

I agree with your suggestions: it is inconvenient to have to answer a question every time one does a tiny edit. (In fact I was not aware of this fact until now.) I will mention this to the sysadmins and propose to modify this. However, all these things apparently need time. --Ulf Rehmann 21:24, 3 February 2012 (CET)

Unfortunately, my prediction seems to come true: "Typical solutions tend to be broken by newer spambots". --Boris Tsirelson 10:38, 28 March 2012 (CEST)

And again... --Boris Tsirelson 11:10, 10 May 2012 (CEST)

Obsolete pages

Please delete (include pages here for deletion):

• User:Boris Tsirelson/Anchor
• Jordan arc, simple arc, Arc simple arc Jordan arc

Suggested move

I suggest to move Surjection to surjective function (leaving a redirect). I think that the adjective is more often used than the noun. (Move is listed among the "user" rights, but there is no "move" button.) --Peter Schmitt 02:47, 14 February 2012 (CET)

I have a "move" button, thus I guess you have, too. But maybe it does not fit into the (not so wide) window. Maybe after "Read", "Edit" and "View history" you have a small black triangle; put the cursor on it, and hopefully you get "Move" and "Watch". --Boris Tsirelson 07:13, 14 February 2012 (CET)

(Blushing) I overlooked this. In fact, now I remember that I noticed it some time ago but did not think of it now ... --Peter Schmitt 17:36, 14 February 2012 (CET)

I propose not to move "surjection" to "surjective function", as there are the notions of "surjective mapping" or "surjective morphism" etc. as well, which are all covered by the notion "surjection". You may want to look into MathSciNet for occurences of "surjection" under "anywhere" or even under "title" to find that notion used quite commonly. --Ulf Rehmann 09:44, 14 February 2012 (CET)

On the contrary, I'd say. In my view, these related concepts are an argument for having pages for each of them and using "surjection" or/and "surjective" as disambiguation pages pointing to them. (By the way, I think that function and mapping need to be better "coordinated". What do you think?) --Peter Schmitt 17:36, 14 February 2012 (CET)

Oh, well, is there any need to split some page of almost empty content (about surjection) into several disambiguated pages? On the other hand, your suggestion to coordinate the pages function, mapping definitely makes sense. But there are much more pages involved, at least Multi-valued mapping, Correspondence (here the notion within algebraic geometry certainly has to be expanded into a complete section, if not into a page of its own), further: Permutation, Morphism, Isomorphism, Epimorphism, Monomorphism, Endomorphism and the like (at the end of Morphism there are more), and then you are in the endless realm of categories, which gives you even more. --Ulf Rehmann 18:36, 14 February 2012 (CET)

I have just operated the page Isomorphism. See there for Categorization and literature, which is may be of relevance for the item Bijection. Mutual links probably were nice. --Ulf Rehmann 14:54, 18 February 2012 (CET)

Mistake needs correction

By mistake I moved Helly number instead of Helly theorem to Helly's theorem. I could revert the move, but now the redirect has to be deleted. Sorry for that! --Peter Schmitt 23:53, 6 March 2012 (CET)

Done. --Ulf Rehmann 00:13, 7 March 2012 (CET)

Thanks. --Peter Schmitt 00:38, 7 March 2012 (CET)

Pretex request

Hi Ulf -- I find myself with a bit more time to contribute. Could you do the same "PreTeX"ing for me for Topological algebra, Banach space, Commutative Banach algebra, C*-algebra, Group algebra. I see that this is done by an emacs script, but unfortunately I've not been able to get "its just text" to work on my version of firefox. (if there is another way please let me know/point me to the relevant page). --Jjg 15:52, 16 April 2012 (CEST)

Hi Sergei, it seems you did use a wrong signature: The name of your user page is User:Yakovenko, not what is shown in your signature. But the way, in case you want to change that name, I could do that for you.

Concerning "It's all text": In case you use emacs: You have to give the emacs command "M-x server-start" before clicking on the little edit button in oder to get it to work. I personally don't use firefox, but rather chromium, for which a similar macro has to be applied. The emacs script is very helpful, because allows to use the emacs minibuffer to reedit code which had been keyed in a bit before (just by using the up/down arrows).

Concerning PreTex: Will do so. I'll put the stuff into the respective discussion page of either page, with a short preamble or so. In fact we could do this for every page, since this can be automated. (However, I think the method above allows to work much faster...) --Ulf Rehmann 17:31, 16 April 2012 (CEST)

Hi Ulf -- Actually the sig is right: I am "J.J. Green"!. Thanks for the emacs advice, but in fact the problem seems to be my version of firefox conflicting with addon (there is no "edit button"), I'll try chrome ... Cheers --Jjg 17:41, 16 April 2012 (CEST)

Ok, sorry, I realized too late I was talking to the wrong addressee... Seems I'm talking to too many at the same time. Maybe you put some id into your user page?

The toy for chrome is "Edit with Emacs": https://chrome.google.com/webstore/detail/ljobjlafonikaiipfkggjbhkghgicgoh

You should configure your emacs so that it honors utf-8, and also, that you don't close or leave chrome's edit window while you are editing in emacs, since this would mean you loose contact with the server, and if you close your emacs buffer than its content may be lost.

Another thing: I have, in my .emacs file, a line

(add-hook 'edit-server-start-hook 'turn-off-auto-fill)

because otherwise emacs tends to autofill the whole buffer, thereby destroying WP links. For convenient editing you can use M-x longlines-mode. --Ulf Rehmann 18:20, 16 April 2012 (CEST)

Hi Jim, I have prepared the pretex of the pages you mentioned in their respective talk pages. The Topological algebra had so little tex that I finished that rightaway (except that some MR/ZBL entries are too many, cold you please check and remove the nnosense?).

One more hint: The tex code entries in the above pretex files look like this "$ $". This is good for direct editing. The emacs macro won't work well for that. If you want to use that, you should make a global replacement '$ $' ==> '$_$' first. Then you can apply the macro, and it will traverse you trough all the tex places. --Ulf Rehmann 20:07, 16 April 2012 (CEST)

Help on sandbox

Do our help pages mention the idea of a (private) sandbox? This would be helpful, see for instance Talk:Normal form and User:Yakovenko/sandbox1. --Boris Tsirelson 07:43, 20 April 2012 (CEST)

I was absolutely thrilled by the simplicity and elegance of this solution! ;-) As a novice to Wiki, I also would suggest one-line help on how to create new pages and rename the old ones (with appropriate caveats about redirection). Sergei Yakovenko 12:44, 20 April 2012 (CEST)

Help on equation numbering

This also would be helpful; see Talk:EoM:This project#Hadamard theorem. --Boris Tsirelson 16:25, 21 April 2012 (CEST)

Pages that need some improvement

Hi Ulf. Is there a way to mark a page that needs to be updated or improved (or even created)? Or maybe to add it to a list somewhere? I've been doing some TeX conversion on two pages, and they both need a lot more than simple TeX conversion. They need to be updated and improved. Unfortunately, I have no time to do this right away, but maybe other people have. --Baptiste Calmes 22:41, 23 April 2012 (CEST)

We have to arrange something like a category "deficient". I'll do this within the next days, as there are several such pages which need modification, correction or urgent update. I think a category like this should work, assuming that if a page is deemed by someone to be 'deficient', reasons should be given on the associated talk page. --Ulf Rehmann 23:36, 23 April 2012 (CEST)

Keywords

Recently I realized that in many cases the terminology is manifold and, say, "fibre bundle" is almost completely synonymous with the "fibration"... Of course, redirects solve the problem, but may be it makes sense to create genuine "keywords" as, say, a template (rather than just a search term) which would be treated almost as the page name when searched for?

How to implement it? By changing the mediawiki software? Much too much for our busy Springer programmers. And yes, redirects solve the problem, and are treated as the page name when searched for. So, is there still a problem? --Boris Tsirelson 11:08, 10 May 2012 (CEST)

I don't know. The problem with redirects is as follows: the same keyword may appear on many pages, while the redirect can be only to one place. I now realize that instead of keywords one could use (sub)categories, but then the question would be again in the numbers, - how strongly branching trees of the form {Differential Geometry -> Bundles and fibrations -> connections -> characteristic classes} we could tolerate. -- Sergei Yakovenko 13:23, 10 May 2012 (CEST)

Then maybe a keyword should be the name of a disambiguation page?

But all that could be discussed rather on Talk:EoM:This project. --Boris Tsirelson 14:42, 10 May 2012 (CEST)

Another example: currently the pages dedicated to connections on vector bundles and connections on a manifold are a complete mess, with two pages on parallel displacement, parallel displacement(2) (which is itself a bad translation from Russian, the right terms should be "translation" or "parallel transport"). I would suggest keeping two closely interlinked pages on "Parallel transport" and "Connection on a bundle", but clearly indicating all other relevant keywords at the beginning. -- Sergei Yakovenko 08:45, 10 May 2012 (CEST)