Some time ago I created Template:Hardy and Wright to save myself and others time. It would be nice to be able to add the edition as a parameter, and select the appropriate bibliodata, but unfortunately I'm not fluent in template-speak. Is anyone willing to expand it? The data I have for other editions is
I posted a suggestion at Talk:Glossary of category theory but realised that it would apply more generally. Many but by no means all of the glossary articles in Category:Glossaries of mathematics list entries in alphabetical order. It seems to me that for those ordered by subtopic, the reader who doesn't know exactly what a concept is has to scan the article: the editor who can't fit an entry into an existing subtopic has difficulty placing it. Would uniformly alphabetical order be preferable? Deltahedron (talk) 19:13, 5 June 2014 (UTC)[reply]
Is anyone getting odd LaTeX error messages of the form Failed to parse (PNG conversion failed; check for correct installation of latex and dvipng (or dvips + gs + convert))? I'm seeing them on Bott–Samelson resolution, for example, where X_{\overline{w}} seems to work but
Ah, I see now. I thought it was a problem with the implementation, rather than an actual error! It should have been :Z_{\overline{w}} \to X_{\overline{w}} giving of course. Deltahedron (talk) 11:56, 8 June 2014 (UTC)[reply]
The Encyclopedia of Mathematics is now a wiki supported by Springer-Verlag. The text from Springer remains copyrighted to Springer but any new articles added and any changes made to existing articles will come under the Creative Commons Attribution Share-Alike Licence. The articles need work because the formulae in the Springer version now exist only as PNG and the TeX source needs to be recreated. However, every article renewed in this way comes under the new licence, and so is presumably compatible with Wikipedia.
Just for a record, I'm planning to import some of articles from eom to here (cf. my TODO.) Right now, I'm preoccupied with other stuff. -- Taku (talk) 22:07, 15 May 2014 (UTC)[reply]
Hi. I was contacted as a volunteer (since I do a lot of work with copyright) and asked to review this. I just wanted to make sure that their license statement isn't misinterpreted. :) New changes come under Creative Commons Attribution Share-Alike but do not change the copyright of the underlying material. So, articles renewed are not autoamtically relicensed - only the new content is. See, for example, http://www.law.cornell.edu/uscode/text/17/103:
The copyright in a compilation or derivative work extends only to the material contributed by the author of such work, as distinguished from the preexisting material employed in the work, and does not imply any exclusive right in the preexisting material. The copyright in such work is independent of, and does not affect or enlarge the scope, duration, ownership, or subsistence of, any copyright protection in the preexisting material.}}
So the original copyright is untouched and the content from the original would not be compatibly licensed, but new material added would be. See also Derivative work. This is consistent with the source's statement that "The original articles from the Encyclopaedia of Mathematics remain copyrighted to Springer but any new articles added and any changes made to existing articles within encyclopediaofmath.org will come under the Creative Commons Attribution Share-Alike License." It doesn't say that the existing articles will come under the Creative Commons license - only that the "changes" will. :) --Moonriddengirl(talk)13:00, 20 May 2014 (UTC)[reply]
Strictly speaking it's the licence, not the copyright we need to know about. I agree it's not clear whether modified articles will come entirely under the new licence: perhaps someone will clarify that. Deltahedron (talk) 19:19, 30 May 2014 (UTC)[reply]
If I remember correctly, any change to an existing article renders the article to be under cc-by-sa. The logic is like this: licensing doesn't work word-by-word. Say, you added a single word to the original eom article; it's not like that that single word is licensed under cc-by-sa and the rest is not under cc-by-sa. In a case of math, sometimes, contributing a single word can be a substantial contribution (e.g., you added "complete" to metric space.) In other words, by editing an article, you created a "new" article and any new article will come under cc-by-sa. If you like an analogy, it's like editing a picture. You don't license the picture by pixel-by-pixel; that's just absurd. -- Taku (talk) 23:09, 6 June 2014 (UTC)[reply]
Is this understanding of the license somehow supportable? It seems rather overly permissive. My understanding would err on the side of caution: as a rule new articles will be covered under the new license, but old articles are not. There might be some exemptions to the latter. For instance, a wholly new paragraph in an older article might be cc-by-sa. But I think it is almost certainly not true that changing a single word makes the entire article fall under a compatible license. Sławomir Biały (talk) 12:25, 8 June 2014 (UTC)[reply]
List of Italian Mathematicians Question, :Those already on Italian Wikipedia listed here.
I was looking at List of Italian mathematicians to see which ones were red linked and needed articles. I tried Ugo Barbuti by chance, and noticed that, even though he is red in English, he is already listed in the Italian Wikipedia. I decided to check the list to see which ones were available in English and which in Italian. I did the list in two parts, one for the red links and one for the blue, just in case. All of the red links seem to be available in the Italian Wikipedia. Could these all be automatically redirected?
I don't know about other people, but even though my parent language is English, I use Chrome for my browser and right click to translate when I run into any other language. It gives OK English, mostly readable. As a service to visitor could we give them the Italian link for now?
Is there a translation project for these pages? Could a simple page be created for each in English that points to the Italian page and mark it as a stub? What about automatic translation, as poor as it might be, and put a banner at the top to say it was auto translated and needs a human editor? That way if someone finds it through Google or other search, they might be prompted to edit someone they know, or were looking for.
I moved the list to Algebraic equation, keeping a few linear equation links, and adding a link to Algebraic equation. I think it looks better now. I also reordered the links in order of usage, to my best guess. Hope that works for you. I will let you tackle the templates. But I think it is fixed for now.RC711 (talk) 16:45, 14 June 2014 (UTC)[reply]
In this article there is a link for a random math article. It points to http://tools.wmflabs.org/jitse-bot/rpim but seems to be broken. I would like to look at random articles in a general category like mathematics. I presume it is looking at the list of articles and selecting from that. Can that be done for any subject?
Apparently this special page selects a random article in the specified category, not in the subcategories. Called on "mathematics" the random choice is done among 18 categories and 13 articles, that is 31 items. Absolutely not useful! D.Lazard (talk) 20:28, 13 June 2014 (UTC)[reply]
Oh. Thanks for the details. I was being hopeful that it would pick among the some 30,000 math articles in the list of math articles, or among the categories and drill down. I would like a search on Wikipedia to find pages in any broad topic. Like a random page about chemistry, or one about birds, or one about snakes, or one about European cities - something like that. If we can define lists of pages with common themes, then it is simply a matter of picking a random number and then looking for that index to the particular list. Mathematics is the only subject I found with a master list of pages like this. I have been trying to drill down in category trees from a top page like Category:Mathematics or Category:Science to find all the child pages, working through all the subcategories. Work in progress.RC711 (talk) 20:50, 13 June 2014 (UTC)[reply]
User:Jimbo Wales posted the following in response to a question about WMF plans for mathematics software development:
It might be helpful to give more specific details. "png is still standard" is a tantalizing clue but... what do you recommend instead? What does the dream solution look like? What is currently state of the art on the web in terms of math editing and rendering software? The last time I looked into this (admittedly quite some time ago) what most math editors wanted was LaTex support, and rendering to png was a reasonable way to render. So, that's what we have now. What would math editors prefer today? I'm happy to help but it would be delicious if I had an NPOV summary of the current state of the art, how it compares with what we support, and some basic first step explanations of what the steps are to get from where we are to where we want to be, what help we might be able to engage from the broader math community, and what engineering costs we might expect to shoulder on our end. We have a new CEO now, specifically chosen for tech/product focus, and so a lot of things will be up for discussion over the next year or two.--Jimbo Wales (talk) 20:08, 6 May 2014 (UTC)[reply]
What I want is mathjax + fast rendering. It isn't problematic on other sites (MathOverflow, MathSciNet, arXiv, etc) and has become a de facto standard among professional mathematics sites. So why does it have to be so difficult here? —David Eppstein (talk) 21:11, 6 May 2014 (UTC)[reply]
What David said. It's technologically possible if there is will. (Everything is possible with will?) -- Taku (talk) 21:38, 6 May 2014 (UTC)[reply]
That would be a significant improvement, but unfortunately it doesn't solve the serif/sans problem. I'm not sure there actually is a solution to that, aside from a very strong recommendation that if you can in any reasonable way avoid it, even if it requires significant variation from your ideal word order, just don't put mathematics into running text.
I don't want the perfect to be the enemy of the good here, but I also don't want people to think, "oh, as soon as MathJax comes along, we'll be just fine". --Trovatore (talk) 21:17, 6 May 2014 (UTC)[reply]
Yes. In the default parameters for mathjax they are distinguished by using \( ... \) for inline, and \[ ... \] or $$ ... $$ for display (they don't automatically enable $ ... $ delimiters because dollar signs have other uses, but the delimiters are easy to change). See e.g. http://docs.mathjax.org/en/latest/start.html. On web pages outside Wikipedia it's extremely easy to use mathjax — just add a couple of lines to the header of your web page, copied from the docs that I linked, and go. —David Eppstein (talk) 21:49, 6 May 2014 (UTC)[reply]
It may also be worth pointing out that the community does not want or need a WYSIWYG formula editor. Most new editors of mathematics already know LaTeX, but may not know wiki syntax. I realize that a more user-friendly interface with the LaTeX source is now what visual editor seems to offer, but it may be worth emphasizing at what point feature development on that part of the project should stop. Sławomir Biały (talk) 21:40, 6 May 2014 (UTC)[reply]
Yes, I think this is very important. WYSIWIG editors are less flexible and more cumbersome than TeX. Honestly I don't think it's unreasonable to ask that anyone who wants to enter equations learn some basic TeX – it's standard and not hard. Ozob (talk) 01:59, 7 May 2014 (UTC)[reply]
I am not up to speed enough yet to know what the community may or may not want regarding a WYSIWYG formula editor, but in principle I agree with you that such a thing is not nearly as high a priority as making sure that proficient math editors are appropriately supported with the latest and greatest, subject of course to the path forward being cost effective and acknowledging that nothing is likely to be perfect anytime soon. But if there are things that can be done to make significant progress for you all, relatively painlessly, then I think that's an obvious thing to do.--Jimbo Wales (talk) 02:00, 7 May 2014 (UTC)[reply]
I wouldn't rule out WYSIWYG editors. There's one I have that works pretty well for simple equations, the one built into Apple's Grapher (in the Utilities folder on any modern Mac). You can do a suprising amount with it. It's keyboard driven rather than mouse, so you're typing and using shortcuts to place things rather than manually placing them but that works well as it means the app does all the layout, LaTeX like. There's also a toolbar for inserting things you don't know how to type, much like our edittools.
I would question the assertion that most new math editors know LaTeX. I didn't pick it up while doing maths at school or as an undergraduate, only later, and I imagine this is fairly common at undergraduate level, where at least in maths your main tasks are to learn things to pass exams. An equation editor though would help non-mathematicans more; right now you need LaTeX for things like matrices, all but the simplest of formulae, even when they appear in non-maths articles.--JohnBlackburnewordsdeeds02:33, 7 May 2014 (UTC)[reply]
Hello, Jimbo --- interesting to see you posting here. To see examples of how MathJax works, look at math.stackexchange.com . It works really well. It matches the size of the surrounding text and it doesn't get misaligned the way inline math notation on Wikipedia often does. (However, on that site, one sees some amazingly unskilled uses of it. I marvel at the fact that it is humanly possible to acquire such extraordinary unskillfulness.) Michael Hardy (talk) 03:20, 7 May 2014 (UTC)[reply]
I agree with you that most new math editors probably do not know any TeX. Despite that, I think it is reasonable to ask them to learn a little. Learning basic TeX is like learning basic wiki markup. It's nice to have an editor where you can push a button to turn on italics, but learning to insert two single quotes is not difficult. Similarly, learning to enter \frac{1}{2} to generate the fraction is not hard. Nor is learning to enter \times if you want a times symbol . And that's as much as many people will need and as far as many of them will get with TeX. In a perfect world, it would be nice if there were a WYSIWIG editor because it would lower the bar to editing. But the bar is already pretty low, the effort involved in creating a WYSIWIG equation editor is pretty high, and ultimately the mathematics is likely to be translated into TeX for rendering anyway. Ozob (talk) 03:44, 7 May 2014 (UTC)[reply]
The counter-argument to that is the visual editor exists, at least in part, so editors don't have to learn wikimarkup. Even without the visual editor you can get a lot done with no experience using the toolbar above the edit window, to insert bold, italic and other formatting so you don't need to learn their markup, and making it easy to learn as you go along because it shows you how to do it.
But hitting the '√n' button just gives you <math>Insert formula here</math>, with no guidance, no example(s) to learn from, not even a link to a help page. There's a limit to what could be done as 'Insert formula here' could be replaced with a wide variety of things, while bold, italic etc. only work one way. But this is also a motivation for doing more as even if an editor learns some equation editing, such as how to make a fraction, it doesn't mean they know other LaTeX such as how to make an integral.--JohnBlackburnewordsdeeds10:36, 7 May 2014 (UTC)[reply]
I think we have two essentially unrelated problems, and that these should be handled independently: rendering from the markup (presently TeX), and capture. Rendering currently is a major headache/obstacle (very slow MathJax, sizing and alignment of PNG, with some font complaints), and this I find really puzzling. Capture relates to interaction with the editor, and I think that here we should stick to an editor-visible and -modifiable markup. A WYSIWYG editor and/or toolbar can be treated as separate exercise. If we separate these two points (which only involves a decision on what markup to use), and even if we deal with only the rendering problem initially, we will have a major improvement. Treating them as intertwined may contribute to continued stagnation. —Quondum11:10, 7 May 2014 (UTC)[reply]
A problem, which is related to both rendering and editing: In most article the mathematical variables looks differently when displayed (serif italic) and inline (sanserif italic). The use of {{math}} allows inline serif, but few editors know it, and those who know it are often too lazy to use it (14 characters to insert a single variable, as {{math}} does not imply italic). A simple solution could be to introduce $a$ as a shorthand for {{math|a}}, and similarly for more complicated formulas that may be rendered in html. D.Lazard (talk) 11:43, 7 May 2014 (UTC)[reply]
For myself I see the slowness as something that will be fixed eventually by the speed of processors and in the not too distant future, so if there is a problem it really is what should we do about low specced smartphones used in the middle of beyond by poor people? That problem will be around for a while at least and it is one of the aims of Wikipedia to disseminate knowledge widely. Should we just keep pngs for that for the moment? As to having a visual editor for entering maths I would like to see one implemented and the main thing it should do for starters is not ruin a formula if they display it and a small bit is edited - i.e. that it be robust and not generate a huge pile of rubbish just to get the same display. After that it should be easy to use, and after that it should be possible to put in anything (rather than just cope with letting through anything). There's lots of simple maths around where people can contribute who've never heard of TeX. And I agree it should be easy to mark variables in the text as maths symbols. I would prefer them to be italic serif even in non serif text, but if we can have a user CSS style set which dictates that it would solve such problems for like them to match the surrounding text. It may sound trivial to some but there are people who find the mismatch quite annoying. Dmcq (talk) 12:18, 7 May 2014 (UTC)[reply]
Huh? Serif/sans is rendering and not editing, and as per Dmcq can be treated via individual preferences; treat it is a third independent thing if you wish. Adding a shorthand alternative writeup language is a peripheral problem, not part of the fundamental one; it could be added at any time. Writing off MathJax's slowness (and frequent complete failure to render!) as "it's fine - we'll just wait until more processor power is on tap" doesn't fly for a number of reasons. Let's focus on fixing what is really holding us back. —Quondum13:42, 7 May 2014 (UTC)[reply]
My two cents is that whatever is chosen, please keep it to one choice, and make that choice work, and work well. This includes ease of use, visual appearance and performance. I don't belong to the category of people that believes performance isn't an issue "since future CPU's and internet connections will be faster". If you manage to use up CPU power that is enough to numerically solve Einsteins field equations for 17 colliding black holes when displaying f(x) = x, then something is fundamentally wrong. Such software is simply not sustainable/maintainable in the long run. More importantly, it is very difficult to further develop such software because it is deemed to be a total mess (spaghetti code).
POV: WYSIWYG editors are nice up to a point. But think about it. What would you really prefer in the long run, MS Word or a edge-of-the-wedge LaTeX editor? Not saying that building wrappers to simplify for new users is wrong/bad, but please let advanced users have access to the "core" of whatever becomes chosen. This is not the same as having multiple edit options with competing editors. YohanN7 (talk) 16:25, 7 May 2014 (UTC)[reply]
Besides obvious bug fixes and performance improvement, what I want are (1) the support for commutative diagrams (I've heard the topology coverage here is less satisfactory than, say, algebra one and this might explain this.) (2) the support for simple figure drawing like circles and triangles. In terms of latex, both can be achieved by installing appropriate packages and thus there shouldn't be a technological hurdle. One more thing: the ability to use mathcal fonts that became standard among topologists such as J. Lurie. (I'm thinking of mathcal O or L that are not default ones; I can elaborate if I'm not making sense.) -- Taku (talk) 00:06, 9 May 2014 (UTC)[reply]
Topologists' mathcal I don't think so. This is what I had in mind. (I needed a PC to upload the pic.) -- Taku (talk) 10:49, 12 May 2014 (UTC)[reply]
That's overwhelmingly similar to the standard \mathcal (for reference: , ). A basic typographic principle is to not use too many fonts in the same work. According to that principle, either all uses of \mathcal should use the font you've suggested (whatever it is), or that font should not be used anywhere on Wikipedia. That's a vastly more invasive change than you seem to have in mind. I'm not convinced it's worth it, either; what about this font do you find so enticing? Ozob (talk) 02:30, 13 May 2014 (UTC)[reply]
Similar but different, which is not a contradiction :) I just wish it were an option (not changing the standard one.) -- Taku (talk) 21:19, 13 May 2014 (UTC)[reply]
It seems to me obvious, but perhaps worth spelling out, that the only viable form of internal representation of mathematics today is some form of LaTeX markup. A graphical interface or formula editor that can work with underlying markup is important, especially if VE takes off, but for some applications direct access to the markup will continue to be essential. Deltahedron (talk) 21:25, 10 May 2014 (UTC)[reply]
Yes, agreed. What is less obvious to most is the resultant separability of the various aspects of maths on WP. I would suggest that WP's greatest benefit will come from fixing the rendering problem, and treating everything else as secondary: get MathJax working, or find a replacement that works. Even the exact choice of underlying markup is not an immediate issue. If there is appetite for more other areas such as graphical editing interface, these can happen later or separately in parallel. Whatever happens, sort out the rendering as a priority. —Quondum22:19, 10 May 2014 (UTC)[reply]
As I already noted in the first reply in this thread, MathSciNet and arXiv also use MathJax. In the case of arXiv, it's for the display of the abstracts of preprints, after previously only allowing ascii text; the papers themselves are available in postscript or pdf, but generally not html. I don't remember what MathSciNet was using before MathJax, but LaTeX+PNG seems likely. I just checked the All-Russian Mathematical Portal, and they're also using MathJax. On the other hand, Dagstuhl still seems to be limited to ASCII text for the abstracts of their publications. —David Eppstein (talk) 22:54, 10 May 2014 (UTC)[reply]
It does not seem as though there is any technical difficulty (i.e., presumably other sites have MathJax working). If we can set down a clearcut objective of, say, getting MathJax rendering of LaTeX working within stated constraints (performance, platforms, development timescales), the rest of Jimbo's ask should not be too difficult to quantify (for those in touch with it). I think that the only thing really lacking is a the setting of a clearcut, clearly attainable objective, and I'm hoping that this would not be difficult if it is judiciously limited. —Quondum23:33, 10 May 2014 (UTC)[reply]
ZMATH used to have TeX and MathML rendering options but seems to have to reverted to showing scans of the older pages and very badly converted HTML for more recent, the latter in particular being almost useless. Deltahedron (talk) 16:20, 11 May 2014 (UTC)[reply]
Thanks for the link on fr:Projet:Mathématiques/Le Thé. MathJax is a really useful tool, I can't deny that (and I'm glad to use it). But
it can be slow on webpages with lots of formula (e.g. on Wikipedia)
it requires JavaScript (as far as I know)
just try to copy-paste a paragraph with math formula inside...
Compare with such a page (sorry, it's French). I cannot pretend to get a perfect rendering on every browser, but you can read it without JS (the real tool is in CSS), and copy-paste gives you a valid expression on most simple cases. The gap is a converter from latex (possibly inside wikicode) to HTML. I've not written it yet, but anybody can do it, since all these files are under CC-BY-SA license. Of course, there is still work to do (management of graphics), but in my humble opinion this path seems more fitting to mathematics on the web than MathJax machinery. Ambigraphe19:01, 12 May 2014 (UTC)[reply]
And on my very old-fashioned laptop which struggles with the current MathJax on WP and on EoM. Congratulations to the authors. I just had a somewhat disappointing conversation with User:Mdennis_(WMF) in which she she did mention that she could help with the WMF "grants programs, if you want to seek funding for development of mathematics software" and that might be of interest in developing this further. Deltahedron (talk) 06:38, 13 May 2014 (UTC)[reply]
I share your disappointment at the response in that conversation. I would go further, and suggest that I am appalled at the apparently purposeful disengagement and evident lack of understanding of where a little involved coordination and direction could make all the difference to existing efforts that have been stalling. We all understand that with Brownian motion a particle usually gets somewhere (unplanned), but as the WP community gets larger, this is like assuming that undirected local crowd behaviour is a useful governing principle. —Quondum14:04, 13 May 2014 (UTC)[reply]
Yay, that French thingie! Now I don't know whether its my crappy internet connection or the rendering that takes 0.5 seconds. That page is easily several seconds of work for (our) MathJax, at least on my system. (After playing around some more, rendering appears to take 0 seconds within human approximation, just as it is supposed to.) YohanN7 (talk) 15:35, 13 May 2014 (UTC)[reply]
The notice below saying "This discussion has been closed" refer to the discussion below, seen when you click on "show", not to the discussion above.Michael Hardy (talk) 15:35, 17 May 2014 (UTC)[reply]
This discussion has been closed. Please do not modify it.
The following discussion has been closed. Please do not modify it.
WMF plans for mathematics
A couple of weeks ago some comment were made here by editors concerned about the development of mathematics rendering and editing. The point was made that currently WMF allocates essentially no resources to this and it continues entirely on volunteer effort, which is made less effective by the way it is not integrated into WMF development. At that time I asked [1] what plans WMF had for developing mathematics-based text. Unfortunately neither you nor anyone else was able to answer before the question was aged off [2].
This is very disturbing. Mathematics support is a key component of writing a serious encyclopaedia and it is quite unacceptable that WMF should devote no resources to its effective development and have no plans to do so. Please would you ask the WMF to reconsider its policy on this matter, and allocate a suitable proportion of its resources to the maintenance, sustainability and development of mathematics rendering and editing? Deltahedron (talk) 08:33, 4 May 2014 (UTC)[reply]
I'm going to chime in here to support everything that was said above. Mathematics on wikipedia is already difficult enough to read and write as it is, any efforts to simplify this process should be encouraged. I understand that WMF is busy with everything on their plate, but at least have someone poke around and see what options are available to improve how math can be better communicated on wikipedia. Tazerdadog (talk) 06:20, 5 May 2014 (UTC)[reply]
Quite. If any money is going to be spent on mathematics on Wikipedia, it should go towards hiring people who actually know how to write mathematical articles for a general audience. — Scott•talk16:27, 5 May 2014 (UTC)[reply]
An excellent plan, and I would be happy if WMF were to spend money on that too. But support for mathematics rendering and editing would still be required. Currently it seems we have none. Deltahedron (talk) 16:56, 5 May 2014 (UTC)[reply]
I'm an excellent mathematician (even if I say so myself) and I rarely understand a word of them too. That's the nature of the mathematical game these days I'm afraid. Still it's true that many articles, even on elementary topics, could do with some Sqrt(1 + Tan^2(x))ing up (this is an excellent mathematician BTW who believes we can get on just fine with the ten digits and twenty six letters the good lord gave us like we used to on Usenet - nevertheless support Deltahedron because we must move on with the nooths I suppose, for better or worse ). Still there are some excellent articles out there, Logarithm for example. Coat of Many Colours (talk) 17:37, 5 May 2014 (UTC)[reply]
Grants are available to anyone who wants to do technical work, but, as I understand it, the WMF never funds content creation, no matter how sorely needed it is.
It's possible for dedicated people to form their own non-profit and seek funding for content creation, and if you're serious enough about it, then you might want to look into that. I doubt that it would work in practice, though: making some articles completely impenetrable to the lay person (and making sure that even the most trivial facts in it are all sourced to equally impenetrable sources) appears to be a goal held by some long-established editors, so efforts to write good, intelligible articles is likely to produce significant resistance. It would be unfortunate if you went to all that trouble and expense only to have some WP:OWNish editor revert it all to the impenetrable versions. On the other hand, mathematics may be the one area where this is least likely to be a problem. I've personally encountered several editors who really are trying to make these articles more accessible (with variable success).
As a mid-point between these two extremes, it might be possible for you to convince the WMF to fund a structured training program for making mathematics articles accessible, if there were enough editors interested in working on this. meta:Grants:IEG is probably the place to start that inquiry process. WhatamIdoing (talk) 18:23, 5 May 2014 (UTC)[reply]
Thanks for that, although perhaps I might re-iterate that my original request to Jimbo was entirely about WMF planning and resourcing technical elements for mathematics rendering and editing. Deltahedron (talk) 18:28, 5 May 2014 (UTC)[reply]
Many of the math articles we have are very useful to people who actually need to know something about the topic. Some years ago Sean Carroll wrote on his blog that he was doing a computation away from his usual workplace and he needed to know the explicit form of some spherical harmonics, and he found them on Wikipedia.
The problem with math really is that the general audience is math illiterate and generally not really interested to learn about the topic. It's therefore pointless to aim too much at the general audience, as we cannot make up for a deficient educational system here. What we can do is present the material in such a way to make it as useful as possible. This means that we relax the Not Textbook rule a bit and write up articles such as Methods of contour integration or Rational reconstruction (mathematics) that are very useful to people who are already into these topics who need to learn more. Count Iblis (talk) 19:15, 5 May 2014 (UTC)[reply]
(Notice WMF's usual non-answer answer.) I think we've been sidetracked here. The issue is squarely about the math rendering. Deltahedron has been too polite, so I will be more blunt. Basically, the math support here "sucks" in terms of performance and appearance (png is still standard), compared to other notable sites like math.stackexchange. This is more than a practical problem:
It gives an impression that Wikipedia is less hip (at least used to be). This decreases our ability to attract new editors.
It gives an impression that WMF doesn't care about the editors, especially those working on serious encyclopedic subjects like math.
(If I'm allowed to say a bad joke, unfortunately, non-math editors and admins are not smart enough to understand the problems that we math editors are having.)
It might be helpful to give more specific details. "png is still standard" is a tantalizing clue but... what do you recommend instead? What does the dream solution look like? What is currently state of the art on the web in terms of math editing and rendering software? The last time I looked into this (admittedly quite some time ago) what most math editors wanted was LaTex support, and rendering to png was a reasonable way to render. So, that's what we have now. What would math editors prefer today? I'm happy to help but it would be delicious if I had an NPOV summary of the current state of the art, how it compares with what we support, and some basic first step explanations of what the steps are to get from where we are to where we want to be, what help we might be able to engage from the broader math community, and what engineering costs we might expect to shoulder on our end. We have a new CEO now, specifically chosen for tech/product focus, and so a lot of things will be up for discussion over the next year or two.--Jimbo Wales (talk) 20:08, 6 May 2014 (UTC)[reply]
There are a number of problems with the typography of the PNG rendering. A few examples:
texvc / png
MathJax
The most obvious problem with inline formula is the baseline of the formula which is significantly lower than the surronding text. A secondary problem is the font size which is larger the text.
With MathJax the baseline is better, but still 1px too low. The font size matches correctly.
With display formula texvc performs better. There are problems with aliasing giving the brackets and all characters a jagged appearance.
There is the same font size problem.
There are a few subtile differences: the superscript on the is a little higher in MathJax.
PNG display gets worse if you scale the webpage on the client side. As these are static images you get an upscaled image which looks blurred.
Good stuff. Based on that, it seems like MathJax support is a no-brainer. But are there downsides to MathJax that I should be aware of?--Jimbo Wales (talk) 01:54, 7 May 2014 (UTC)[reply]
Addendum: This would lead one to believe that MathJax support already exists and perhaps just needs to be made the default? Or... what is the current status? Apologies that these are basic questions but I haven't looked into this in a long time, and I assume the same will be true for many readers of this page.--Jimbo Wales (talk) 01:56, 7 May 2014 (UTC)[reply]
Yes there are some problems with MathJax at the moment. It requires javascript, which not everyone has so some form of fallback is needed. It can be a bit slow to render a page Help:Formula (a page with very heavy use of mathematics) takes 20s to render because of this @Eloquence: Erik Moeller has put a WONTFIX on T38496 the main bug to make MathJax the default. There is a plan to do server side cacheing which should speed up the rendering with the Mathoid package and a major update to the mw:Extension:Math package @Physikerwelt and Gwicke: know more about this. The sticking point now seem to be getting this update production ready and getting it though code review. There seems to be very few developers who have the interest/expertise in mathematics rendering to move this forward.
To answer some specific points. "It requires javascript, which not everyone has so some form of fallback is needed" - I think we can basically ignore this issue right now. The number of people without access to Javascript is extremely small, and while they do need a fallback, making things worse for 99.9% of all people in order to help the .1% is probably not the best choice. Let me know if there's something wrong with that thinking. So what you'd like is some help with resolving the issues of T38496 and a commitment from the Foundation (Erik, really) that if we get someone to fix that bug, they're open to implementing it.
To make my role in this process clear: I totally trust Erik's judgment on the allocation of the limited resources available to him, and it would in any case be foolhardy for me, with no real knowledge of his production timetables, to put pressure on him to elevate this issue just because I've taken an interest in it. But, two things - I can try to help you campaign to find a community developer interested in this issue. Not sure how effective that will be but I'm willing to try to shine a light on it. And as I said above, we have a new executive director now, one brought on board to ramp up investment in product/engineering capacity. It seems likely that some of the constraints that Erik and the engineering team have faced in the past will be expanded significantly in the next two years, and so now is a good time to make a reasoned case and proposals for improvements that we in the community find important. I've got a few wishlist items of my own. :-)--Jimbo Wales (talk) 14:17, 7 May 2014 (UTC)[reply]
There is also a paper that describes the developments in the new version of the Math extension in detail at http://arxiv.org/pdf/1404.6179v1.pdf. In general I think the code-review process could be improved. Even though me and a few other voulunteers have the permission to push code changes to the repositories, we have doubts that changes we tested locally might influence the performance and stability of Wikipedia in a whole. Therefore we need code review from people with knowelege about the technical details of the MediaWiki installation at WMF.--Physikerwelt (talk) 17:35, 8 May 2014 (UTC)[reply]
OK, here goes. I suggest comments go in the next section, and that any edits to the draft text, unless quite trivial, be accompanied by a comment. Deltahedron (talk) 20:13, 6 June 2014 (UTC)[reply]
About 1% of Wikipedia's 4.5 million articles are assessed as being in "Mathematics and Logic". Probably a similar number are in theoretical physics and in computer science. So in about a hundred thousand articles, the ability to render mathematics is indispensible to the reader: the ability to write and edit mathematics is indispensible to the author and editor.
Currently the predominant mathematics markup system in all forms of document preparation is some flavour of LaTeX. It may be presumed that any serious mathematics content contributor will be thoroughly familiar with LaTeX. LaTeX is rendered on web pages in a variety of ways: currently Wikipedia uses two of the more popular methods, rendering formulae as PNG images and rendering dynamically using MathJax. There are deficiencies in the current implementation of each of these methods.
The stability and usefulness of current mathematics rendering is reduced by the following
Incremental development of reader and editor interfaces is apt to degrade the reader or editor experience without warning.
Major changes in editor interfaces, such as the introduction of Wikipedia:VisualEditor and Wikipedia:Flow, may be radically incompatible with existing LaTeX markup practices.
Effort to support mathematics editing and rendering comes entirely from the volunteer community. Currently one volunteer is working on mathematics rendering, and support for mathematics editing in VE consisted of one GSoC summer volunteer.
We are reliably informed that WMF has no plans for development of mathematics rendering and editing. That is, there is no plan to coordinate volunteer effort; no plan to integrate volunteer effort into existing products; no plan to ensure the sustainability of mathematics rendering and editing through major changes to the software and user interface.
As a consequence of the lack of plans, there is no allocation of WMF developer effort to the maintenance, sustainability or enhancement of mathematics rendering and editing. It is assumed that volunteer developers will undertake any tasks that are necessary, even though there is no plan to coordinate those efforts.
It is reasonable to say that there is considerable expertise and experience in mathematics rendering and editing in the existing editor communities. There is no explicit mechanism to capture that experience and make use of it in planning, development or review. Such efforts as have been made to do so are limited in extent and driven by the user community rather than WMF. The role of Community Advocates in linking the editor community and WMF planning and developers in this context has not been effective.
WMF designate a fixed contact person at WMF that cares about math related questions and a brief to maintain regular and frequent contact with volunteer community.
Short-term
Fix MathSource mode is currently disabled: see [5] which resolves this issue.
Fix issues with experimental mathoid (MathML + SVG) support on the Beta Cluster.
Excellent summary. I have one small criticism that point number two under "Specific" is not likely to make much sense to someone reading the summary who is not already familiar with the details of past discussions. I think this should be stated more plainly. Sławomir Biały (talk) 10:23, 7 June 2014 (UTC)[reply]
workflow
From my perspective the most important thing is to establish a workflow for further development and deployment of the math extension. There are two volunteers Frédéric Wang and myself actively working on the Math extension. From WMF side the Visual Editor team takes care of the Visual Editor related aspects. For Frédéric and me the main problem is that we have no fixed contact person at WMF that cares about math related questions. I think a very brief weekly Skype meeting with a fixed contact person would eliminate most of the problems. This would eliminate most of the randomness involved in the current development process.
roadmap
I tried heavily to promote the https://www.mediawiki.org/wiki/Extension:Math/Roadmap page to have a more coordinated development process. Looking at the history indicates that I have changed my username. If anybody here is willing to contribute to the roadmap pleas give me a signal and I'll update the page. Otherwise I see no need to update this page for my own use only.
source rendering mode
I apologize that the MathSource mode is currently disabled. There is a fix at https://gerrit.wikimedia.org/r/#/c/139439/ which resolves this issue. As soon as this is backported by some WMF employee the source rending mode will be back.
MathML support
We manged to have experimental mathoid (MathML + SVG) support to the master branch. It works well on private wiki installation, but on the Beta Cluster only the SVG is displayed correctly. The MathML elements like <mo><mi>... are removed for some reason. If anyone has an idea why this happens I'd be very happy about comments either here, at my user talk page or at the bug report at https://bugzilla.wikimedia.org/show_bug.cgi?id=66495. (I'll try to visit this page more often)
It is certainly a thing that exists, I.e., there are websites about it easily findable by Google. I don't think that mathematicians are the best-possible evaluators of the content: it's mathematical but not mathematics. I don't know what sort of sources would be most likely to establish notability -- management journals? --JBL (talk) 13:23, 7 June 2014 (UTC)[reply]
A cause for concern is that the main contributors to the article are Qhayez (talk·contribs), Bmaresc (talk·contribs) and Hugosbento (talk·contribs), all of whom have contributed exclusively or almost exclusively to this topic: and the linked web site shows that Prof B. Mareschal, Quantin Hayez and Hugo S. Bento Pereira are associated with this topic and the related company D-SIGHT. Deltahedron (talk) 15:45, 7 June 2014 (UTC)[reply]
It seems like a thing that exists, yes. But one has to wade through two thirds of the article to find out what kind of thing it even is. Perhaps a few sentences should be added to the lead about what the subject of the article actually is? Sławomir Biały (talk) 16:27, 7 June 2014 (UTC)[reply]
Please note: This is an updated version of a previous post that I made.
Hi all,
My name is Adi Khajuria and I am helping out with Wikimania 2014 in London.
One of our initiatives is to create leaflets to increase the discoverability of various wikimedia projects, and showcase the breadth of activity within wikimedia. Any kind of project can have a physical paper leaflet designed - for free - as a tool to help recruit new contributors. These leaflets will be printed at Wikimania 2014, and the designs can be re-used in the future at other events and locations.
This is particularly aimed at highlighting less discoverable but successful projects, e.g:
• Active Wikiprojects: Wikiproject Medicine, WikiProject Video Games, Wikiproject Film
• Tech projects/Tools, which may be looking for either users or developers.
• Less known major projects: Wikinews, Wikidata, Wikivoyage, etc.
• Wiki Loves Parliaments, Wiki Loves Monuments, Wiki Loves ____
• Wikimedia thematic organisations, Wikiwomen’s Collaborative, The Signpost
The deadline for submissions is 1st July 2014
For more information or to sign up for one for your project, go to:
One of the sources, ETC, is reliable for this sort of thing, but (like OEIS for integer sequences) not very selective. So I don't think OR is quite the right description, but whether it's notable is a different question. —David Eppstein (talk) 06:19, 22 June 2014 (UTC)[reply]
If the ETC and cut-the-knot have picked it up, it is has seen some review and the ETC is considered reliable, so from that perspective no OR. It seems to be just a very recent result in a niche area of mathematics. There might be a notability issue though, since the electronic version of the ETC (contrary to the printed book) contains a few thousand newer results in elementary geometry and not all of them might be considered notable on their own.--Kmhkmh (talk) 06:52, 22 June 2014 (UTC)[reply]
Surely a nuanced version of the notability requirement is needed for cases like this, especially in mathematics? Where the result can be considered to be established and valid, and not covered by a more general result, regardless of its publication and review status, it seems like arguing for deletion would be wikilawyering. I realize that this places quite a burden of evaluation on the editors, but in a sense this burden exists in all cases. —Quondum14:48, 22 June 2014 (UTC)[reply]
Currently such theorems/result could still be incorporated as partial content in survey/summary articles like suggested List of triangle centres (note however the ETC contains a few thousand, the book version around 400 iirc). I have no objection against a more nuanced version, however there needs to an effective method for blocking unwanted content as their is potential danger that people will try to publish correct/established directly in WP (instead of elsewhere). Also one needs to keep in mind that there is an infinite number of correct but not-notable mathematical statements/theorems. While there are case where most editors would agree it is ok to have them in WP (despite officially failing the current notability), there will be probably even more where editors disagree. In particular for the latter group a more nuanced policy should provide an somewhat objective criteria for the decision about the inclusion in WP.--Kmhkmh (talk) 16:31, 22 June 2014 (UTC)[reply]
The problem here is not WP:V, nor WP:RS, but notability. I agree with Kmhkmh: that we need better criteria for deciding notability for results that have been reliably published and whose correctness is established. I would suggest something. One could considered as non-notable a result that has not been used in secondary sources. Also, I consider that an article is WP:ORPHAN in nature if it may be naturally linked to only in lists or in "See also" sections (this is the case for this article, and also for the draft linked to in next section). Therefore, my suggestion is, in the case where notability is dubious, to consider as non-notable a notion or a result that is a quasi-orphan as it cannot be linked to from the text of other WP articles, without ad hoc expansion of these articles. D.Lazard (talk) 17:11, 22 June 2014 (UTC)[reply]
A possible way to nuance to notability (especially for content rather than articles) is: "Does it have reference value?" I think this ties in with D.Lazard's perspective: his suggestion gives ways of answering the question that I've given here. —Quondum17:18, 22 June 2014 (UTC)[reply]
Indeed: if a reader is likely to want information on the subject then the question of links from other articles is not so important. Deltahedron (talk) 17:21, 22 June 2014 (UTC)[reply]
As far as notability is concerned, the German Wikipedia for instance considers any (math) subject notable, that has been published in established journal or a properly published textbook. However that approach may not quite comprise the nuance Quondum is asking for above. Nevertheless i think it is good rule of thumb, which seems to be implicitly followed in en.wp as well anyhow.--Kmhkmh (talk) 18:07, 22 June 2014 (UTC)[reply]
