Jump to content

Wikipedia talk:WikiProject Mathematics/Archive/2015/Aug

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

"Thus"

[edit]

The usage and primary topic of Thus is under discussion, see talk:Thus (company) -- 67.70.32.190 (talk) 06:56, 1 August 2015 (UTC)[reply]

good article nominations

[edit]

There is at the moment (july-october 2015) a GA cup running for reviewing good article nominations Wikipedia:WikiProject Good articles/GA Cup but there is a distinct lack of mathematics article nominations Wikipedia:Good article nominations#Mathematics and mathematicians there are only 3 (one nominated by me, another I think a straight fail for another ). Are there no other mathematics articles to nominate ? WillemienH (talk) 07:57, 1 August 2015 (UTC)[reply]

Section "Selected publications" in biographical articles

[edit]

Hello everybody. What do the Wikipedia's policies have to say about the articles about mathematicians having a section with selected publications? It's common practice to add one (I personally like it, because in a quick look at just the titles of the articles and journals one can already get some ideas about the research of the academic), but it seems that there is no specific policy about it. 189.6.202.87 (talk) 23:24, 29 July 2015 (UTC)[reply]

See Wikipedia:Manual of Style/Layout#Further reading. JRSpriggs (talk) 23:45, 29 July 2015 (UTC)[reply]
  • As I discussed with the IP elsewhere, in a prettier place, I believe we should not be a resume service. Published books, maybe--published articles, definitely not. Or, where's the limit? Do we include conference presentations--and at what kinds of conferences? No, this is not a good idea. Drmies (talk) 00:21, 30 July 2015 (UTC)[reply]

I think this is a good idea, and have added such sections to many articles. The part that's a bad idea is having a "Publications" section that is not selective and lists everything (as a cv would). My tendency is to aim for a selecton of 4-6 publications (fewer if there are not enough important ones, more in rare cases when someone has many very significant works) and only include ones that have very large numbers of citations, are published in top journals, have specific mention in secondary sources as being significant results (especially if these results are also mentioned in the rest of the article), or have been given noteworthy awards. For some mathematical subjects such as theoretical computer science where conference publication may be more important than journal publication, yes, I'll definitely include conference papers too, but replacing some of the 4-6 journal papers rather than adding to them. —David Eppstein (talk) 01:37, 30 July 2015 (UTC)[reply]

  • Thanks David. Of course, all the editors in chief will be arguing that theirs is the top article... But awards, for instance, or "widely cited" (shouldn't be too hard to prove), I think that's a good reason for inclusion. Somewhere on our beautiful project it says something about "has exerted great influence", and we should honor that of course, and we can, in this case, too. For example, even if Tolkien's Beowulf speech ("The Monster and the Critics") had never been separately published, it would have been one of the most notable conference key note addresses ever and worthy of being listed. Drmies (talk) 01:44, 30 July 2015 (UTC)[reply]

I have one small gripe, that could be fixed by some industrious WikiGnomes no doubt. We often do not provide useful links to old works whose copyrights have expired. (Google books, Archive.org and Project Gutenberg all maintain free collections, and there are other projects with more specific holdings, like the Euler archive.) For example, very few of Leibniz' works are linked. The significant Nova Methodus pro Maximis et Minimis has its own article, which also lacks a link. The Latin works of Euler are mostly missing as well. It seems like these would be useful links for scholars in the field. Sławomir
Biały
12:41, 30 July 2015 (UTC)[reply]

  • See, for the dead ones there isn't much of a problem as far as I'm concerned--there's not much resume padding for dead people. A link can be added, in your example, to the article on the book--there is nothing wrong with that; we're really talking about a different problem. I suppose WP:EL can allow for such links, though the risk is always that articles get turned into linkfarms. But here we're talking (I assume) about living people and their lists of publications, where we should be very wary, esp. in the medical field, for instance, or in physics, where we need editorial discretion to prevent people from listing a million things. Or think of those books about God and prayer you can pick up at the pharmacy (in the US, anyway)--those authors churn those out by the dozens, and we shouldn't be listing all of them. Drmies (talk) 15:25, 30 July 2015 (UTC)[reply]
  • Certainly works which themselves meet the notability criteria should be listed. I would think that works for which the subject is known should be included as well. As an example, I just happened to have the article on Giusto Bellavitis open; there is a Works section with four entries and that seems a bit light. But even for copyright expired works there should be a line; we don't need to list someone's comment on somebody's solution to a problem posed by whosit. For actors there's no problem with listing every bit part and every guest appearance in a TV episode; but I guess, for some reason, most people are more interested in actors than mathematicians. --RDBury (talk) 19:47, 30 July 2015 (UTC)[reply]

Thanks to everyone who participated! I've already memorized David Eppstein's criteria. This edit should be fine I guess. :D 189.6.202.186 (talk) 04:06, 1 August 2015 (UTC)[reply]

Maybe this one is listing too many. Anyway, I won't change it, I've just pointed it for the case someone feels inclined to edit there :D (that article is really needing a lot of work). 189.6.202.186 (talk) 04:37, 1 August 2015 (UTC)[reply]

I generally agree with "David Eppstein's criteria", but the two examples in the last posts show that more deserves to be said: Most article about mathematicians have a section describing their main contributions. Sometimes, these contributions are the object of a specific article, which is normally linked. But, when it is not the case, (and also in this case), it is useful to read the original article. Even when a short selected list of publication exists, such as in James Harris Simons, it may be difficult to know which article corresponds to which main contribution.

Therefore, I would suggest to organize the selected list of publications as a list of references, linked from the section on main contributions. The criterion for inclusion, will then to have one, or at most two, linked publications by subject described in main contribution section. As usual, this section requires also references to secondary sources, but, IMO, this is a case where some primary sources are useful. An example of the bad result of systematically avoiding primary sources in this case is Andrew Wiles article: he is known for one paper; this is amazing that one cannot find the reference of this paper in the article about him! On the other hand, there are mathematicians that are known for many publications, each of them being relatively minor. In this case, a short list of selected publications is difficult to establish (and would be original research), and it is better to omit such a list. Paul Erdős is a clear example of such a case. D.Lazard (talk) 09:09, 1 August 2015 (UTC)[reply]

Even for Erdős some of his papers are more famous than others. The one with the elementary proof of the prime number theorem; the one introducing the Erdős–Rényi model of random graphs; the one that first uses the probabilistic method, all are particularly noteworthy. —David Eppstein (talk) 16:18, 1 August 2015 (UTC)[reply]

AfC submission

[edit]

Does anyone have any objections to my decline reason on Draft:Integrative Propositional Analysis? Best, FoCuSandLeArN (talk) 16:28, 3 August 2015 (UTC)[reply]

Pretty sure this should be under Linguistics or Philosophy or some other project. In any case, the decline reason wasn't clear to me; are you claiming it's from a single source? Self-promotion? --RDBury (talk) 20:14, 3 August 2015 (UTC)[reply]

Project template

[edit]

Please reconsider the use of {{WikiProject Mathematics}}; it's a disservice to our fellow editors to expect them to use a different protocol for one project, than the one they use for most others. Also, the claim that "The list of mathematics articles already has a list of all math articles" is false; it does not include new articles of relevence to this project. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 11:20, 27 July 2015 (UTC)[reply]

What's going on with Jitse's bot? Any update? Should we try to get something else going? Sławomir Biały (talk) 12:41, 27 July 2015 (UTC)[reply]
That template has been obsolete for six years (see the history); what's the issue? Ozob (talk) 13:51, 27 July 2015 (UTC)[reply]
The issue is that when an editor uses the template, expecting it to work like most every other project's equivalent, it does not. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 14:57, 27 July 2015 (UTC)[reply]
Oh, I see, you want us to use {{WikiProject Mathematics}} instead of {{maths rating}}. Sorry, I thought you meant something different. It shouldn't be hard to overwrite the current {{WikiProject Mathematics}} and replace all current uses of {{maths rating}}. I wouldn't object to that, though perhaps my opinion shouldn't count as I've never been involved in that kind of work. Ozob (talk) 01:02, 28 July 2015 (UTC)[reply]

List of mathematics articles no longer links to the list of mathematics articles but instead redirects to something else. Michael Hardy (talk) 22:30, 27 July 2015 (UTC)[reply]

And now I've recreated something similar to the former page that gave the alphabetical list of pages. List of mathematics articles now redirects to that. Michael Hardy (talk) 22:56, 27 July 2015 (UTC)[reply]
AIUI, article-space redirects should not go to Wikipedia: space pages. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 09:55, 28 July 2015 (UTC)[reply]

I have just done this edit. The statement I deleted has been grossly false for several years. I edited the page that redirected list of mathematics articles to make it a disambiguation page including a link to the actual list. But a user called The Banner insists on reverting without addressing the inappropriateness of the target of the redirect or of the fact that we had that statement on a template directing many thousands of articles to that inappropriate redirect page. If Jitse's bot does not get reactivated, we need to decide what to do about that. Michael Hardy (talk) 20:22, 3 August 2015 (UTC)[reply]

It links to Wikipedia:WikiProject Mathematics/List of mathematics articles. In my opinion, that is linking to a wikiproject. So I was bold enough to revert your edit. My only concern was to free Template:Index of logic articles from a link to a disambiguation page. And that was exactly what I have done. Who screwed up and how to solve that is not my concern. The Banner talk 20:25, 3 August 2015 (UTC)[reply]

And now I've done this edit, linking to the actual list of mathematics articles. Michael Hardy (talk) 20:57, 3 August 2015 (UTC)[reply]

Future of MathJax on wiki

[edit]

A volunteer has proposed removing client-side MathJax from Special:Preferences#mw-prefsection-rendering in phab:T99369, in favor of modern (up-to-date and probably faster) browser plugins. This will probably affect very few readers and editors, but it's likely to affect more people in this group than anywhere else, so I wanted to make sure that you heard about it. It will also be announced in the next m:Tech/News. Whatamidoing (WMF) (talk) 01:10, 17 July 2015 (UTC)[reply]

How can a "browser plugin" possibly work if there is no preference setting to produce output in a form that the browser plugin can parse? How does this relate to the "no special browser setup required" core goal of the MathJax project? How can this possibly avoid causing problems rendering Wikipedia pages that also include dollar signs as text characters? Where are the user tests showing the availability and working condition of these supposed browser plugins? And how does this interact with recent network-wide security alerts advising us to disable all auto-running plugins? To me this sounds less like an improvement and more like gratuitously breaking something that works because the person proposing the breaking doesn't care whether it works. Additionally, is there any hope of fixing the current problem that it is not possible to set the user preference to render math as MathJax without at the same time breaking the display of math formulas in the mobile app (which does not support MathJax)? If the developers' attitude to the decade-long disaster of bad math formatting on Wikipedia (and the only-very-recent and grudging support of decent MathJax formatting) is to step backwards and punt to someone else's browser plugin then I have very little hope for the future here.
And the framing of the proposal completely misses the point. It doesn't matter what we individually as mathematics editors are using to view Wikipedia; what matters is the default view, and how easy it is to format articles for that view. As editors of mathematics, what we have now are three incompatible and partially broken systems for getting the equations viewable by users: inline wiki-formatting (very limited and tricky to format but always works and produces a rendering compatible with the inline text), a system of templates (somewhat limited and even trickier but with better appearance), and <math> (can format essentially all equations using much better markup but underused because the default appearance is so ugly). Incorporating MathJax into the wiki at least gave us the option of full math with good appearance, and the hope that MathJax might eventually become the default view giving us a single good editing option. Now you're pulling the rug from under us, saying we'll be stuck with these editing incompatibilities forever, and that only power users who install special plugins will ever see well-formatted <math>. This seems like a huge step backwards. —David Eppstein (talk) 01:37, 17 July 2015 (UTC)[reply]
It says in that messy link,
Client side MathJax rendering is outdated. There are browser plugins that support MathJax rendering for chrome. The advantage with those plugins is that they use the most recent actively maintained version of MathJax and do not require that users log in.
(my emphasis) As far as I can see, this means that the default rendering mode will be MathJax (now PNG). This is not a small step as it will affect the vast majority of readers, at least those who use Chrome. (I just suppose the server can figure out that it is Chrome that requests a page, and that the appropriate plug-in is present.)
It will probably affect, as David notes, editors in this group the least, since many use MathML by choice, and others PNG by choice because it has, by far, the fewest bugs of the current options. YohanN7 (talk) 03:08, 17 July 2015 (UTC)[reply]
I think you are mistaken in claiming that this change will mean "the default rendering mode will be MathJax". Browser-plugin MathJax requires that the server sends the browser the LaTeX source code. So if this were to become the default rendering mode, users without the plugin would just see LaTeX source instead of rendered equations. This is such a bad choice that I think even the Wikimedia developers would eschew it. My interpretation of your quote about not requiring logins is that the developer did not think about this and did not realize how bad that would be as a default before proposing to kill any alternatives. —David Eppstein (talk) 03:22, 17 July 2015 (UTC)[reply]
It is not a claim of mine. It is an assumption based on what I quote. Client and server do actually have a little chat before the bulk data is sent, so, it is , at least in principle, possible to send plain LaTeX to clients having Chrome and appropriate plug-in. YohanN7 (talk) 11:02, 17 July 2015 (UTC)[reply]
While in principle it's great that the WMF is dropping by to see what the users want, it is less encouraging that the users have been saying the same thing for years now: give us working mathjax. There was a detailed proposal made last year that summarized the user's needs, back when WMF was apparently threatening to remove mathematics support altogether. Does this mean that WMF actually now cares enough about the users to take what they ask for seriously and budget for the addition of fully-fledged mathjax support for mathematics rendering? Or is this post just a token gesture, so the WMF can pat themselves on the back and say that the users were consulted, only to ignore them again? Sławomir Biały (talk) 12:12, 17 July 2015 (UTC)[reply]
Sławomir, "the WMF" is not doing this. As I clearly stated in the very first words of my message, "a volunteer" has proposed this.
Also, the WMF has never considered removing support for mathematics. I remember the hyperbolic claims that editor and his sock made – leaping straight from a designer saying that he wasn't certain how many different maths systems Flow would ultimately support straight to the destruction of all maths content in all namespaces – but those claims never had any basis in truth. In fact, at the time that he was spreading that drama, Flow already supported maths in the formats that he preferred. Whatamidoing (WMF) (talk) 17:45, 17 July 2015 (UTC)[reply]
But the WMF should be doing this: maintainance + development of math rendering architecture. Shouldn't it? Why do we need to rely on volunteers? If the WMF can devote engineering resources to stuff like Visual Editor, then surely it can devote some small fraction of the resources to the math support. The message we're getting is that math is not a priority; I understand it is less of priority than flow or VE. But I'm not quite happy that it is not a priority of any kind (and thus delegated to outsiders). -- Taku (talk) 18:35, 17 July 2015 (UTC)[reply]

It was Physikerwelt who suggested this change. I'd like to invite him to provide an explanation. @Physikerwelt: As I said on the task page, as best as I can tell, the plugin you're talking about is a third-party creation not supported by the MathJax team, and there are no corresponding plugins for other browsers. Why do you think that switching to this plugin would be an improvement? Ozob (talk) 14:56, 17 July 2015 (UTC)[reply]

Quouting myself from the Phabricator ticket:

"The currently enabled MathJax availible to the users is a old version that branched of from an old version. It's also unmaintained i.e. not supported by the MathJax team. The customization of the MathJax codebase is quite heavy. Updates made to MathJax (for example security updates) are not reflected in the customized wmf clone.

From a conceptual perspective this heavy customization of MathJax seems to suboptimal and the past has proven that is is unmaintained.

I would be interesting, if there are Firefox users that prefer MathJax over MathML.

For the future, we are going to do exactly that was proposed on the discussion page and also envisioned here T78046."

A more detailed overview about the new MathML rendering mode is availible from here http://arxiv.org/abs/1404.6179 I promise to review every code that is contributed to the math extension. So if someone is willing to update the current MathJax plugin this is defenity an alternative. However, my current focus is on the MathML rendering mode that still has some issues that need to be resolved. --Physikerwelt (talk) 15:30, 17 July 2015 (UTC)[reply]

@Physikerwelt: you are wasting your own time by chasing a chimera and in the process taking Wikipedia farther away from the path to working math markup. That arXiv preprint on mathoid, in its user requirements, omits the main one that has caused all of the problems for all of these years: the math generated by it must not look ugly. After all, <math> has existed for the last decade and has produced valid mathematical formulas for most of that time (modulo a few bugs here and there). What has not existed is math markup that is so well integrated into the text (in terms of font matching, baseline matching, size matching, scalability, etc) and so well spaced (up to the standards produced by Knuth in TeX), even in the default not-logged-in-vanilla-browser view, that nobody would consider formatting mathematics any other way. Instead, we have this horrible profusion of different editing styles, precisely because the default not-logged-in formatting of <math> is ugly. MathML is not good enough. It is not formatted as well as MathJax and in any case is a non-solution for the default view because Chrome does not natively support it. On the other hand, MathJax is well formatted and (in theory) usable by almost all. MathJax for all would solve our problems, and I don't think anything less would be good enough to get math formatting back on track here. If the problem is that Wikimedia's MathJax code is old and crufty and overcustomized, why not put some effort into updating it? The bottom line is, if you're not working towards the goal of making the default not-logged-in view at least as high-quality as MathJax, then you're not helping, because that not-logged-in view is the beginning and end of all our math markup editing difficulties. —David Eppstein (talk) 16:23, 17 July 2015 (UTC)[reply]

Here a side by side comparison of the different maths rendering modes.

Comparison of different maths rendering modes of a section of https://en.wikipedia.org/wiki/Help:Displaying_a_formula

If it were not for the baseline, aliasing and font size problems the PNG rendering looks good, MathJax is good with a couple of glitches, but to me the firefox MathML looks like a project still in need of a lot of work to get it looking really good.--Salix alba (talk): 01:14, 20 July 2015 (UTC)[reply]

Thanks for this comparison. I would expect that, in principle, SVG could be better at the aliasing issues than PNG, because it can be rendered into pixels after already knowing what kind of display it will be shown on, allowing subpixel-resolution color-fringing tricks. In practice, the view I get from Wikipedia's SVG (pretty much what you show in the Chrome view of your screenshot) looks overly-thickened and overly fuzzy. Not to mention that it still has bad font sizing and baseline issues. —David Eppstein (talk) 03:23, 20 July 2015 (UTC)[reply]

The MathJax option has now gone. If you want client side MathJax in Chrome you can use the Wikipedia with mathjax extension. (github page.)

MathML rendering in firefox is improved quite a bit with the mathml-font extension.--Salix alba (talk): 12:29, 24 July 2015 (UTC)[reply]

This news is very disappointing. Unfortunately, WMF has already made clear its antipathy to the needs of the users. Over a year ago, I told WMF representatives that working MathJax should be made an immediate priority. Later, this project put together a detailed plan for the future of MathJax. Apparently, these recommendations were not passed along to the engineering team. I think it's time to bring Jimmy Wales into this again. I think Salix's picture illustrates nicely why the current status quo is not acceptable. Sławomir Biały (talk) 12:53, 24 July 2015 (UTC)[reply]
(I'm probably stating the obvious, but anyway). The volunteer developer (developers?) seems to have the attitude that he knows what is the best and not particularly interested in what math editors or readers of Wikipedia would like to have. The WMF has, essentially, no plan ("no plan" is still a plan?). It would be "nice" if we can have some kind of long-terms plan. Of course anyone or any community can come up with a plan. For it to be implemented, a commitment is needed from someone and some organization. If that's not WMF, I don't know what else. @Whatamidoing (WMF):. I get improving (actually just fixing) math rendering would not increase, say, female editors. But a small amount of commitment should be reasonable in my humble opinion. -- Taku (talk) 19:48, 24 July 2015 (UTC)[reply]
Pilot studies have shown that improvement of MathJax rendering will increase the participation of female, disabled, minority and socioeconomically disadvantaged mathematics editors and readers by over 5000%. Studies have also shown that the old style PNG rendering is preferred primarily by white Anglo-Saxon males, which already constitute the bulk of our editing community. In diversity studies, MathML features only marginally better, largely with the small but growing "hipster" crowd, consisting mostly of white Americans (both male and female) in their late 20s, who enjoy the "retro" look of the MathML font. But our Wikipedia mission is to improve access to all the peoples of the world, regardless of their race, creed, color, gender, socioeconomic status, or language. And reinstating and improving the MathJax extension is vital for the WMF in its continued long term mission of improving access. This is especially important in the mobile market, already the primary way readers interact with the project, and one that is projected to have long term growth potential. In fact, because of this long term growth potential, most of the web has already opted against server-side rendering in favor of light-weight client side apps, relying heavily on AngularJS, AmberJS, and Backbone extensions. MathJax rendering support is vital for the long-term viability of the Wiki foundation projects, one that will bring back in multicultural, ethnic, and multi-gendered support to our mathematics editing and reading community. Cloud computing and Microsoft Azure technologies can help engender this, as we quantum leap forward into the new millennium. Sławomir Biały (talk) 23:29, 24 July 2015 (UTC)[reply]
Do you actually have links for any of this? If so I'd be very interested in seeing them. —David Eppstein (talk) 00:09, 25 July 2015 (UTC)[reply]
http://www.seven.co.nz/assets/image/full/iiuqGEd.jpg I mean, really... "quantum leap forward"? Sławomir Biały (talk) 00:21, 25 July 2015 (UTC)[reply]
Well yes, hence my use of "actually". But there should be usability studies of MathML vs MathJax in different populations, and I'd be interested in seeing what such studies might have to say. —David Eppstein (talk) 00:28, 25 July 2015 (UTC)[reply]

It turns out to be quite simple to use MathJax with a user script. All you need is to set the raw tex preference and add

window.MathJax = {
    tex2jax: {
      inlineMath: [ ['$','$'] ]
    }
  };

mw.loader.load( 
	'https://cdn.mathjax.org/mathjax/latest/MathJax.js?config=TeX-AMS_HTML');

to your Special:MyPage/skin.js. This might not be the fastest way of doing things but its developer independent.--Salix alba (talk): 07:59, 25 July 2015 (UTC)[reply]


@TakuyaMurata: You mentioned "no plan" I have been trying to get input on a plan for several years now ... jude by yourself https://www.mediawiki.org/wiki/Extension:Math/Roadmap ... for me capturing math semantics is much more important than the technical details of the rendering. I think mathoid will improve the rendering for most users and especially for those using browsers that fully support HTML5. However, much more important than the rendering itself is a cleanup of the markup to capture more semantics and allow for additional services such as math search or import export function to computer algebra systems. But therefore we need to get rid of all the problems (and there are a lot more than you might imagine) that the Math extension at this very moment. With the current code basis it is very hard to make progress at all. Therefore I'm trying to recruite a second volunteer how can help with code review or with parts of the implementation. I even wrote a guid to demonstrate how simple it is to review the math extension code at [1] but until now... I'm still searching.

Everyone is free to decide what's the right way. Either blame me and WMF that we are not making sufficent progress or open an IDE and help coding.--Physikerwelt (talk) 19:52, 28 July 2015 (UTC)[reply]

I think your priorities are wrong and that those wrong priorities in you and others have played a large role in why mathematics markup and rendering here has been so badly screwed up for so long. The top priority should be: is mathematics presented well to all readers. Second should be: is it easy for editors to generate good presentation of mathematics. Maybe third should be performance issues. A far distant fourth, useful mainly only to the extent it can help in the other three priorities, is: does the server-browser communication channel have a clean semantics. Because frankly who but a developer would or should care about that? —David Eppstein (talk) 20:40, 28 July 2015 (UTC)[reply]
To be clear, I said the WMF has no plan, and I am assuming that a volunteer such as yourself @Physikerwelt: is not a part of the organization (yourz're not in the payroll). This seems to be a main issue: since it creates a response that if there is a problem, fix it yourself. Personally I don't have any technical problem (I use MathML with SVG on iOS devices and are perfectly happy). As David Eppstein said (and in fact, I'm mostly simply repeating him here), this is about the presentation of math to the readers. The WMF's perception that some math "editors" may get inconvinicened by the removal of MathJax is thus missing the point.
I think the work on the semantic aspect is important and interesting; it may even make sense to have some computer algebra system support to generate some figures (I have heard nowadays one uses the computer to do concrete calculations with, say, polytopes in the representation theory.) It seem that you are more interested in that type of work and there is nothing wrong with that. I'm merely saying (as David Eppstein does) that this aspect is somehow irrelevant to the issue in hand. -- Taku (talk) 22:33, 28 July 2015 (UTC)[reply]

@David Eppstein: I think we agree on the priorities for math rendering some more details:

1) "is mathematics presented well to all readers"
all readers includes readers with limited vision, limited bandwidth, old hardware, working in secure environments
2)"is it easy for editors to generate good presentation of mathematics"
if LaTeX is referred as easy I agree. I personally think that clean semantics should imply good presentation i.e. i prefere $\sin{x}$ over $\mathrm{sin} x$. I'm a little bit sceptical about visual editing tools since some make it much harder to edit more complex formulae i.e. if there is no way back to LaTeX
3) performance
as long as the performance is reasonably good that does not matter at all... otherwise it's a problem of type 1
4) server browser communication
this is not a priority at all. However, since different users use a huge variety of browsers some form of standard is required for browser client communication. Otherwise the maintenance effort explodes.

@TakuyaMurata: cas integration is one thing I'm planning to work on in my regular job. I'm looking for a master student and we are going to start with the DRMF and Wolfram eCF project. --Physikerwelt (talk) 08:46, 29 July 2015 (UTC)[reply]

I just posted a long and somewhat ranty blog post about these issues. If you care, you can find it at http://11011110.livejournal.com/314841.html — anonymous responses should be possible there (but will initially be invisible until I review them as a spam-prevention measure). —David Eppstein (talk) 06:24, 5 August 2015 (UTC)[reply]
Quite interesting. A small remark: Wikimedia is indeed the nonprofit organization... but the software on which Wikipedia runs is rather Mediawiki. Boris Tsirelson (talk) 07:03, 5 August 2015 (UTC)[reply]
Thanks, corrected. —David Eppstein (talk) 07:15, 5 August 2015 (UTC)[reply]
Nice post, David. A few remarks:
  • I use Firefox with MathML for my daily editing on my desktop and Chrome with SVG fallback on my smartphone (desktop, not mobile version). Barring a few minor issues, which nevertheless should be fixed, I am quite happy with either combination. The foremost reason that MathML is not a universal solution is that most browsers don't natively support it. It seems one reason why the MediaWiki developers try to push MathML is to convince browser developers to natively support it, but I don't think that this will work out. The SVG/PNG fallback looks good, even better than MathML.
  • Having clean semantics would be nice, but that's not a problem of MathML but of LaTeX. Actually it's not a problem of LaTeX either but of the fact that formulas are entered without semantics in the first place. If we would agree on a set of macros and replace for example \cdot by \multiplication, \innerproduct, \groupoperation or whatever the meaning is (and the editor of an article is the best source for this information), we could also provide semantics from inside LaTeX. But changing all formulas accordingly would be a terrible amount of work. Also, agreeing on a standard set of macros and convincing the MediaWiki software engineers to include them could be problematic.
  • The current SVG/PNG fallback does not have the font size and alignment issues of the current PNG support. Actually, I don't understand why the current PNG support is not fixed in this respect since there obviously is a MediaWiki solution for these problems. Other limitations of inline images still apply of course.
  • Using HTML markup for mathematical formulas is an incredibly terrible idea due to inconsistent display and also for semantic reasons. In the German Wikipedia we try to avoid these, despite the current ugly PNG support. There are still some older articles with such markup, though. We also don't use the math templates.
  • The main problems with MathJax are rendering speed and text reflow. In my opinion these are not minor problems but they significantly reduce the fun of browsing articles and editing them. This especially applies to mobile devices. But this is no reason to discontinue the MathJax support in MediaWiki since MathJax might improve in this respect in the future. By the way, MathJax can also parse MathML, but this seems to be slower than directly parsing LaTeX.
Best wishes, --Quartl (talk) 11:10, 5 August 2015 (UTC)[reply]
Maybe for you on Firefox the SVG fallback (and whatever rendering preferences you've set) doesn't have sizing/baseline issues but for me on Chrome with my settings it clearly does. —David Eppstein (talk) 14:00, 5 August 2015 (UTC)[reply]
I have selected "MathML with SVG or PNG fallback" in Preferences/Appearance. On Firefox, I get the MathML display. On Chrome, which doesn't have MathML support, I get SVGs. Compared to selecting "PNG images" in the preferences, the font is cleaner, significantly smaller and better adapted to the rest of the text. The baseline is not perfect, though, and in places off by few pixels. Best wishes, --Quartl (talk) 14:33, 5 August 2015 (UTC)[reply]
Here's a simple test: does look normally formatted to you, with no font changes, misalignment, or other special emphasis on any of its words? Is that still true if you use your mouse or other pointer device to select the text of the sentence? If so, congratulations, your math formatting setup works better than mine. —David Eppstein (talk) 14:58, 5 August 2015 (UTC)[reply]
I get the following:
  • Firefox: 'this sentence' has a different font with slight serifs, slightly smaller than the rest of the text, but perfectly aligned. Selction, cut and paste works.
  • Chrome: 'this sentence' has a different font with pronounced serifs, slightly larger than the rest of the text, about two pixels lower. Can not be selected (SVG).
  • Explorer: 'this sentence' has a different font with pronounced serifs, significantly larger than the rest of the text, several pixels lower. Can not be selected (SVG).
The results in Firefox and Chrome would be tolerable to me if 'this sentence' was a formula, especially in comparison to the current PNG rendering. The result in Explorer is only marginally better than the current PNGs. It probably all depends on the font setup. In Firefox I do have an add-on called 'MathML-fonts 2.1.1-signed', but turning it on or off seems to have no effect in this case. Best wishes, --Quartl (talk) 15:37, 5 August 2015 (UTC)[reply]
A further test: How looks , compared with this sentence? For me, with Firefox and "MathML with SVG or PNG fallback", the first one with <math>, and the second one with {{math}} have rather similar fonts (slight serif), well aligned, but the first one is significantly smaller than the current text, and the second one slightly larger than the current text. Copy and paste gives: "How look this sentence {\displaystyle {\text{this sentence}}} , compared with this sentence?" Note the space before the comma introduced by <math>, and the \displaystyle introduced by the copy and paste (this probably explains why so many articles have this \displaystyle in every <math> formula, producing tiny and almost unreadable fonts).
Because of this this behavior, my preferred edit option, is, for the moment, to use {{math}} when possible, and <math> when impossible or for complicated formulas. The small font of <math> is somehow an advantage for complicated formulas, because of more compact formulas. D.Lazard (talk) 16:25, 5 August 2015 (UTC)[reply]
In Firefox, when I just select and copy 'this sentence' I just get 'this sentence', but if I select the the whole sentence I also get the displaystyle stuff. Probably, the MathML text annotation generated by MediaWiki is not good here or this is a feature of the MathML browser support. The extra space before the comma is a MediaWiki bug that was reported long ago. \displaystyle should have no effect on normal text, if anything it should produce larger symbols like compared to . Best wishes, --Quartl (talk) 16:52, 5 August 2015 (UTC)[reply]
I have no idea how to tackle the additional whitespace problem. I can not reproduce the problem on any test system. See for example http://math.beta.wmflabs.org/wiki/User:Admin. It might be connected with HTML tidy or custom sylesheets. If anyone has additional (in which processing step the additional whitespace is introduced) please comment on https://phabricator.wikimedia.org/T106855. A similar problem is https://phabricator.wikimedia.org/T103269... I have no idea where the information displayed in the section heading come from.--Physikerwelt (talk) 17:27, 5 August 2015 (UTC)[reply]
Using HTML markup for mathematical formulas is an incredibly terrible idea due to inconsistent display...

It is also the only workaround in existence. Most users get PNG rendering and what they see is not "inconsistent display". They see INCONSISTENTDIS PLAY with inline Latex, and this will not change the forthcoming years as it looks. I find articles with much inline Latex unreadable. (I mostly use PNG because the articles I edit (not to mention my sandbox) tend to be big (making MathJax impossible due to poor performance) and contain plenty of formulae (making MathML impossible due to bugs)). HTML has, after all, acceptable appearance - even tough it doesn't match 100% with what's in the displayed equations (in whatever mode). YohanN7 (talk) 17:50, 5 August 2015 (UTC)[reply]

Yes it's a (bad) workaround and its existence has exonerated the software developers from providing a viable solution for the math rendering problem. I completely agree with David on that point. Should MathML, MathJax or anything else become the standard rendering system at some point in the future (hopefully soon), all formulas typeset in HTML will still look as inconsistent as they do now. Best wishes, --Quartl (talk) 18:11, 5 August 2015 (UTC)[reply]
Yes, formulas typeset in HTML will look as inconsistent as now, but no worse. That is my point. The inconsistency is aesthetically bearable, while inline PNG Latex is not. I don't see anything becoming standard anytime soon. When something comes around, it should take a skilled volunteer (which is what we rely on) no more than a week to develop a working HTML->Latex robot. To be specific, I am referring almost exclusively to the template "math", which isn't used for complicated formulae, but rater writing x = y inline. Cheers! YohanN7 (talk) 18:59, 5 August 2015 (UTC)[reply]
You are a true optimist :-). A few months ago, I manually converted the German version of Combinatory logic from HTML markup to LaTeX. Even using advanced search and replace commands this was a horrible job. Best wishes, --Quartl (talk) 19:48, 5 August 2015 (UTC)[reply]
(ec) And that's exactly what I wrote it for. My hopes are that someone will implement MathJax server-side, so that it can spew out HTML/CSS to the browser, simple as that (webfont still needed though). -- [[User:Edokter]] {{talk}} 19:51, 5 August 2015 (UTC)[reply]
If I understood correctly, MathJax is already running server-side to perform the LaTeX → SVG and LaTeX → MathML conversions. Best wishes, --Quartl (talk) 20:03, 5 August 2015 (UTC)[reply]
I keep forgetting that. Then all that is needed is to send the HTML/CSS to the browser instead of an image renderer. -- [[User:Edokter]] {{talk}} 08:30, 8 August 2015 (UTC)[reply]

Removed fullstops at end of equations

[edit]

An anonymous user removes fullstops at end of equations in "Expected value". Is this a good idea to do? Boris Tsirelson (talk) 12:14, 10 August 2015 (UTC)[reply]

Looking closely I see that the first of these fullstops was indeed spurious. But the others were not. Boris Tsirelson (talk) 12:17, 10 August 2015 (UTC)[reply]

In my own writing I feel that every sentence should end with appropriate punctuation, even if it ends in an equation. --JBL (talk) 12:22, 10 August 2015 (UTC)[reply]
Per MOS:MATH#PUNC, "Just as in mathematics publications, a sentence which ends with a formula must have a period at the end of the formula." A footnote explains "This style, adopted by Wikipedia, is shared by Higham (1998), Halmos (1970), the Chicago Manual of Style, and many mathematics journals." Sławomir
Biały
12:39, 10 August 2015 (UTC)[reply]
Well, if so, I'll revert it. Boris Tsirelson (talk) 12:49, 10 August 2015 (UTC)[reply]
There is also the annoying habit of prefacing every equation with a colon (:). I am not talking about indentation here, but it is in spread use at WP (and unfortunately in the literature). YohanN7 (talk) 12:53, 10 August 2015 (UTC)[reply]
Sorry, I do not understand you. In wikitext we start with the colon for indentation, yes; but did you really see the colon in the rendered articles? If so, could you please give us an example? Boris Tsirelson (talk) 13:01, 10 August 2015 (UTC)[reply]
Sorry about being unclear. It also does not apply to this article. But you can find articles that say
Newton's second law is:
instead of
Newton's second law is
.
Ah, yes, I see. Somehow I've imagined this
:
that I newer saw. Boris Tsirelson (talk) 14:40, 10 August 2015 (UTC)[reply]
The colon is highly disruptive in the flow of reading. Sometimes, it is okay, but not as a standard as some use it. It is then just grammatically wrong. YohanN7 (talk) 13:18, 10 August 2015 (UTC)[reply]
I agree that this is an abuse of punctuation. See, for example, Fourier transform, where every displayed equation is prefaced with a colon, most of which are unnecessary. Even more egregious is Poisson summation formula, which features not only colons before every displayed equation, but colons in bold. I think the MSM should offer some guidance about when and when not to use colons. Sometimes, colons before displayed equations are helpful (e.g., if the resulting sentence would be grammatical with a colon if it were read aloud). For example, the two displayed equations in Fourier transform#Definition. But prefacing every displayed equation with a colon totally defeats the purpose of punctuation. Sławomir
Biały
13:30, 10 August 2015 (UTC)[reply]
I guess that non-mathematicians are inclined to treat displayed equations rather similarly to images and tables, as something that accompanies text and cannot sit within a phrase. Boris Tsirelson (talk) 18:51, 10 August 2015 (UTC)[reply]
In fact this usage is quite common in writing by bona fide mathematicians, as well. --JBL (talk) 20:13, 10 August 2015 (UTC)[reply]
And that would make it right? No way! YohanN7 (talk) 20:34, 10 August 2015 (UTC)[reply]
My post includes no content on its appropriateness or correctness. --JBL (talk) 20:41, 10 August 2015 (UTC)[reply]

Regardless of appropriateness, it would be nice if we had some guidance from other style manuals. Halmos recommends natural punctuation in mathematics writing, which would argue against a practice of universally colonating displayed mathematics. But if there are other recommendations in the literature, they might give a usefully different perspective. Sławomir
Biały
21:02, 10 August 2015 (UTC)[reply]

Krantz, A Primer of Mathematical Writing (pp. 23–24) also recommends natural punctuation and against colonating displayed equations. —David Eppstein (talk) 21:26, 10 August 2015 (UTC)[reply]

Is an GA rating higher of lower than a B+ rating ?

[edit]

On Wikipedia:WikiProject Mathematics/Wikipedia 1.0 the Summary table puts B+ left of GA (higher grade position ) of GA while Quality grading scheme puts B+ below GA (lower grade position ) the summary table seems to be automatically generated so i don't know how to change this, am I correct in this? (also see Mathematics B+ rating at Wikipedia talk:Version 1.0 Editorial Team/Index#Mathematics B+ rating ) WillemienH (talk) 08:47, 1 August 2015 (UTC)[reply]

The table is from an included page. One idea would be to include User:WP 1.0 bot/Tables/Project/Mathematics instead. I don't know if the Math project's criteria are consistent with this version though. --RDBury (talk) 02:47, 3 August 2015 (UTC)[reply]
Why do you have a B+ rating (which apparently means "probably a GA")? Why not just nominate such articles for GA status like other projects do? Kaldari (talk) 00:59, 4 August 2015 (UTC)[reply]
Many years ago, there was the perception that mathematics articles would never be accepted for GA status. GA reviewers need to understand what they read in order accurately assess the article, and mathematics articles, even well-written ones, often require mathematical background that GA reviewers will often not have.
B+ isn't used much. I think while the idea is good in principle, it hasn't really worked out in practice. I wouldn't object if someone were to remark all of those articles as B. Ozob (talk) 02:31, 4 August 2015 (UTC)[reply]
Incidentally, there is a bit of a backlog at Wikipedia:Good article nominations#Mathematics and mathematicians, but not one that is out of line with the other categories there. I think we should be pushing more of our articles for good article status. For instance, despite a lingering issue in its subsection on the Lee–Yang theorem, and despite being quite technical, I think Riemann hypothesis is in much better shape than the typical B-class article (I can say this because most of the effort to get it into shape was not mine). We already have a few mathematical good articles (see Wikipedia:WikiProject Mathematics/Recognized content) but they're kind of a random sample rather than really being the best of our articles. So one way that we could clean up this B+ issue would be to look through the list of B+ articles and see whether some or all of them should become good articles instead. (Like Ozob, I wouldn't object to getting rid of B+ as a class.) —David Eppstein (talk) 21:37, 10 August 2015 (UTC)[reply]

Some math articles created after Wikipedia:WikiProject Mathematics/Current activity stopped being updated

[edit]

A list I got by browsing the history of User:AlexNewArtBot/MathSearchResult:

I think that here are listed approximately 30% of the math articles created after 15 June. Dertemivivahirry (talk) 07:06, 28 July 2015 (UTC)[reply]

Thanks! You might get a more comprehensive listing from User:Mathbot/Changes to mathlists — Mathbot still seems to be running even though Jitse's bot isn't. —David Eppstein (talk) 07:20, 28 July 2015 (UTC)[reply]

Looking at the more comprehensive listing I observe that (at least) "X+Y" and "Exaggeration" should not belong... Boris Tsirelson (talk) 18:45, 28 July 2015 (UTC)[reply]
I think X+Y does belong, at least as much as the contents of Category:Mathematics fiction books (it's a movie about the International Mathematical Olympiad — I enjoyed watching it). But there are also some articles on ceramic tiles that definitely don't. —David Eppstein (talk) 19:17, 28 July 2015 (UTC)[reply]
By the way, does "Mathematical elimination" belong? Boris Tsirelson (talk) 07:23, 12 August 2015 (UTC)[reply]

Jacobian conjecture

[edit]

I have suggested a modification to the article describing the birational and Galois cases. See Talk:Jacobian conjecture#The birational and Galois cases. As usual, will need help on this.

L.Andrew Campbell (talk) 18:27, 13 August 2015 (UTC)[reply]

On the talk page, I have provided a more elementary, though admittedly less elegant, proof of the theorem, that is much more easily understood by elementary vector calculus students.

Generally, though, I think this article could use a lot of work (as many of you probably do). It should discuss some physical applications of it, such as the equivalence of the differential and integral forms of the 3rd and 4th of Maxwell's equations.--Jasper Deng (talk) 10:01, 14 August 2015 (UTC)[reply]

Troublesome editor

[edit]

[This here] concerns the math project as well, since Cuzkatzimhut (and also I) edit both math and physics articles. YohanN7 (talk) 13:24, 15 August 2015 (UTC)[reply]

I'm not generally a fan of these refimprove templates. Often they are clearly wrong, and occasionally they can be disruptive, especially when there is disagreement about how many references are enough. An easy solution that avoids most of the drama is to add a footnote or two, and then remove the template. Sławomir
Biały
14:13, 15 August 2015 (UTC)[reply]

It might be a vast improvement if the templates were reconfigured to go to the Talk pages, instead. They are frequently used for political purposes and by frustrated readers who decide to deface the entire page with ref templates, in the vague hope something they may lack the training to understand will lead them to the mother load reference that makes everything clear. Instead, 8 times out of 10, it is the WP article which is clearest. Cuzkatzimhut (talk) 18:57, 15 August 2015 (UTC)[reply]

Policy on red-linked personal names in Math articles

[edit]

In a recent edit, an IP user added yet another red linked item to the list of doctoral students in the infobox at Grigory Margulis. Do we have any policy in place regulating such matters? My own feeling is that infobox should not serve as a replacement for Math Genealogy or person's scientific biography (and bibliography), and in this case the list is already excessive (and also inaccurate). Personally, I would even be hesitant to add red linked names which are not likely to get their articles per our interpretation of WP:Notability to the main text. Arcfrk (talk) 04:57, 14 August 2015 (UTC)[reply]

Re the students to list in infoboxes: my preference would be to include only bluelinks. As for whether to redlink names elsewhere in the article: Wikipedia:Red link says to do it when you believe the subject of the link to be notable even though no article exists, and I think that's a pretty good rule. —David Eppstein (talk) 05:07, 14 August 2015 (UTC)[reply]
Thank you, David. Obviously, you and I are in agreement on this, but it would be nice to have a policy or a guideline addressing this issue (for infobox). In this particular instance, I have cleaned up the infobox, removing redlinked former students and two redlinked prizes. Arcfrk (talk) 04:14, 16 August 2015 (UTC)[reply]

Input needed

[edit]

A recent dispute over content in List of female mathematicians escalated to the point where the page was protected; see Talk:List of female mathematicians#Joshi edit-warring and more general issues of line length and image inclusion and following sections. Input from more editors would be welcome. RockMagnetist(talk) 03:25, 19 August 2015 (UTC)[reply]

Related to this, there is a new discussion starting over who should be included in the "mathematician" categories. Please see Talk:List of female mathematicians#Who is a mathematician? and contribute your opinions. —David Eppstein (talk) 17:59, 20 August 2015 (UTC)[reply]

I was thinking of renaming Algebra over a field to Algebra (mathematical object), and also merge Algebra (ring theory) in there as well; then we could make it a WP:CONCEPTDAB. The concept of an algebra is fairly standard, apart from whether or not we assume unital / associative / commutative / finite-dimensional / over a field / over a commutative ring.. Basically an algebra is something that has an binary addition, binary multiplication, and a scalar multiplication of some kind. Since many readers probably don't quite know what they are looking for, and the precise definitions depend on the sources, I think we should have a broad concept article that covers all "algebras".

We also have Non-associative algebra and Associative algebra. I suppose a Nonassociative ring is trivially an algebra over Z, so we could naturally merge that article into Non-associative algebra.

Any objections to any of the above? Mark MacD (talk) 11:41, 18 August 2015 (UTC)[reply]

I'm skeptical of these types of merges. The problem is that objects which share formal similarities may appear in very different contexts, and merging can create an incoherent mess. Algebra over a field is already an impossibly broad topic, and besides, it would plainly be inadvisable to try to create an article on everything in mathematics called an "algebra". Would this article also subsume the article vertex operator algebra, whose subject is neither of those things? --Sammy1339 (talk) 03:06, 19 August 2015 (UTC)[reply]
Thanks for the reply; fair point about vertex algebras, but as I understand it, they are not considered to be "algebras". I still don't think Algebra (ring theory) should be a separate article. Someone else has tagged Algebra (ring theory) with a merge tag, and I would like to merge its content elsewhere. The most common mathematical concept of "an algebra" is not impossibly broad; I have just expanded on the brief explanation in the even-broader concept article called Algebra. Mark MacD (talk) 09:12, 19 August 2015 (UTC)[reply]
We have vector space as a separate article even though it is a special case of module (mathematics). This is how we do things in Wikipedia, even if that is against Bourbaki's style. This makes sense from the pedagogical point of view since students learn elementary topics before advanced topics (I know "topos" is more elementary than topological space but that's another issue).
It is true that algebra (ring theory) is underdeveloped; I think the article title is not helpful for fostering healthy development. What matters with these objects are really whether they are associative or not. Thus, it's better to have two articles associative algebra and nonassociative algebra and spread materials accordingly . -- Taku (talk) 23:39, 19 August 2015 (UTC)[reply]
Taku, you were the one who added the merge tag to algebra (ring theory) in the first place! :-) I agree we should keep associative algebra and non-associative algebra separate from each other. Mark MacD (talk) 12:21, 20 August 2015 (UTC)[reply]
Yes, I think I wasn't clear about my position :) My proposal would be to have the three articles, and only those three, that are devoted to the topic of "algebra": (1) algebra over a field (2) associative algebra and (3) non-associative algebra. (1) covers both associative (e.g., a ring that is a vector space) and non-associative (e.g., Lie algebra) algebras, while (2), (3) are over an arbitrary (commutative) ring. There are going to be considerable overlaps between (1) and (2), (3). But I think that's ok for the reason mentioned above (mainly the pedagogical reason and here in Wikipedia we don't necessarily strive for the the strict implementation of the Bourbaki style). -- Taku (talk) 12:42, 20 August 2015 (UTC)[reply]
I see; we're pretty much on the same page then. I've just merged Algebra (ring theory) into Algebra over a field. Somebody might like to check it. Cheers, Mark MacD (talk) 09:41, 21 August 2015 (UTC)[reply]

RfC on definition of a mathematician

[edit]

I started an RfC here on the issue mentioned by User:David Eppstein above. --Sammy1339 (talk) 00:26, 22 August 2015 (UTC)[reply]

In other words, the discussion at Talk:List of female mathematicians has turned against Sammy, so he's going forum-shopping in hope of getting a more favorable audience. —David Eppstein (talk) 00:39, 22 August 2015 (UTC)[reply]
I'm not forum-shopping. The issue is obviously broader than that topic so I posted it at what I thought was the appropriate place. I linked to that discussion from the original one - I'm not hiding anything. --Sammy1339 (talk) 00:42, 22 August 2015 (UTC)[reply]

Order of a polynomial nominated for deletion

[edit]

I have nominated for deletion Order of a polynomial. See Wikipedia:Articles for deletion/Order of a polynomial. D.Lazard (talk) 10:56, 24 August 2015 (UTC)[reply]

AfC submission

[edit]

What do you think of Draft:Biweight midcorrelation? Best, FoCuS contribs; talk to me! 18:59, 26 August 2015 (UTC)[reply]

Bilingual glossaries: For translating content about elementary through high school-level mathematics

[edit]

Dear Wikipedians,

I found some glossaries that can be useful in translating material about mathematics at the elementary and high school levels: http://www.p12.nysed.gov/biling/bilinged/bilingual_glossaries.htm WhisperToMe (talk) 16:38, 31 August 2015 (UTC)[reply]

Exponential type

[edit]

Can someone answer my question at Talk:Exponential type? Eric Kvaalen (talk) 09:13, 31 August 2015 (UTC)[reply]

Yes we did. Boris Tsirelson (talk) 10:10, 31 August 2015 (UTC)[reply]
Oops, no, sorry; we did not! Experts in complex variable are needed... Boris Tsirelson (talk) 15:21, 31 August 2015 (UTC)[reply]
Probably, the question is already solved by Eric Kvaalen himself! Boris Tsirelson (talk) 17:27, 31 August 2015 (UTC)[reply]