Slashdot Mirror
MathML 2.0 Becomes W3C Proposed Recommendation
Nearly three years after the officialization of MathML's first generation, MSjogren writes: "W3C has announced the advancement of MathML 2.0 to Proposed Recommendation. Check out the W3C Math home page.
Now I just wish I could get it to work decently in Mozilla too :(" Part of the proposed recommendation is this explanation of some of the difficulties and aims of mathematical expression, especially when it comes to transmitting over the Web, which emphasizes the importance of a format which can be written to by various tools as appropriate, for reading by anyone.
I cannot but be pleased with this innovation; it enables the creation on the Web of a perfect logical notation to resemble my "Begriffsschrift", or "concept-script". This shall allow for the statement of propositions such that their grammatical form models their logical form, and as such is a development of the highest importance.
(p)(p v p)
I suppose that we could just have a tag that allows TeX formatted stuff to be used, but I think its best that web pages use something that is consistent with HTML's idioms.
"Homo sum: humani nil a me alienum puto"
(I am a man: nothing human is alien to me)
My only political goal is to see to it that no political party achieves its goals.
Mozilla (since M18 at least) supports MathML out of the box. I'm sure the support isn't 100% complete or bug-free, but looking at a couple of demonstration pages it can do some pretty impressive stuff.
:)
It supports only the "presentational" markup, not the "semantic" form, so unless you are amazingly patient you probably need a tool to generate it (there is a TeX to MathML converter available already) and you have to use XHTML in your web pages because XML-in-HTML is not supported by Mozilla. It also isn't included in all builds, and Netscape didn't choose to build it in NS6.
All that said, it works! I can view MathML pages today in my usual browser
Stuart.
Scientists, mathemeticians, doctors, and STUDENTS!!!
I'm only a little out of school, and I remember how math was one of the bottlenecks in my studies. For most of high-school, a good calculator, $50 - $100, was all I needed to breeze through Algebra. The simple interface and display was enough to do most problems, except for geometry. Computer tools really did help me understand concepts and check my work.
Calculus changed that. You feel like you need a quill pen just for the notation, and it feels like you are hacking your own tools to do simple problems. It doesn't get better as you go on, with more complex subjects adding more notation. And I thought only classical students needed to learn Greek.
I'm not saying this will change the way we teach the natural sciences and engineering, but it will facilitate a new generation of computer-based tools, the same way a graphing calculator was a leap over the basic scientific calculator.
Now we are closer to an age where the textbook, the notebook, the scratch paper, the homework assignment, the completed homework, the exam, and the calculator all have the same interface, and the student doesn't have to constantly translate between the media.
Exactly...I'm in charge of designing and building placement testing for my university and every damn year we look at MathML and find that it STILL isn't working as promised.
:-) I'm sick of outputting scores to GIF to demonstrate examples and stuff. Wasn't XML supposed to allow us to build this stuff without plugins and stuff????
I design a lot of adaptive testing (get one right get a harder question, get one wrong get an easier....but the branching algs are much more complicated) and it sucks to have to have an image file for each item in the testing bank. If I need to make a change, its into one of a number math softs and then photoshopping the results.
If I want to do truely adaptive and add some random elements, I can make the computer create a similar question to see if the student really understands (or doesn't understand) before giving them another level to look at. Its nearly impossible right now. I had to build a gif creator and a small scripting language which completely kills server performance with any ammount of students. With MathML, I could simply throw in the random bits and calculate the answers and let the client computer take care of the rest.
This is just my needs, but I can think of a dozen other uses that could directly benefit students. I've helped set up a few tutoring sites for folks and this would be great to build large libraries of questions without becoming too repetative. Most students learn by repeatedly doing something and if they are repeatedly doing the same questions, they are only learning to memorize the answers.
Blah...give me MusicalML and I'll be just as happy
clif
Manager of Development
Testing Center
Indiana University Purdue University Indianapolis
(and)
The other problem with doing equations as bitmaps is that it breaks the functionality of the web. Visually impaired people can make the font bigger, but the equations will stay small. You can't search through it. You can't do text-to-speech. You can't change your stylesheet and have all the equations change style as well.
As far as LaTeX,
- it's never going to be learned by more than 0.01% of the world's population,
- it represents a 1970's-style approach to making a user interface (ooh, you mean I get my own terminal instead of having to hand someone a stack of punched cards?), and
- its aggressive stance on separating form from content means that you have to jump through hoops to make a complicated layout turn out how you want it.
I can understand why math and physics journals encourage submissions in LaTeX, because they want to take away the authors' freedom to format their paper according to their own preferences. But it's just not appropriate for many other situations.The Assayer - free-information book reviews
Find free books.
...because of the many legacy documents in TeX, and because of the large authoring community versed in TeX, a priority in the design of MathML was the ability to convert TEX mathematics input into MathML format.
--
HTH
Brett
i think by keeping the HTML idioms you create alot of complexity/messyness in the code. i could see having a TEX tag like
<TEX>
insert raw tex input here
</TEX>
with standard wrapping/alignment/etc instructions in the opening tag. but adding more stuff is kind of silly if there is already a well documented way of doing the same thing.
use LaTeX? want an online reference manager that
-- john
Um, Netscape is up to at least 4.75...
But anyway, Mozilla hasn't yet hit optimizing stages yet, so it won't become fast for another few months. So be patience, and the browser you want will arrive. ^_^
For years, I've been wanting to put my class notes up on the web for my students, in a hyperlinked form. The problem has always been that I could get good looking equations by just putting up a PDF, or I could get hyperlinks by using one of the TeX to HTML converters, but I've never been able to get both.
Of course, even when MathML allows good TeX to HTML conversion, I'm still going to have a good time trying to get the converters to work with all the AMS-Latex extension macros (ever try to use HyperLaTeX and AMSLaTeX at the same time? Ugh!). It's going to be plain LaTeX for a long time, folks.
Maybe if you didn't care about it you wouldn't have written about it, but since you took the time to waste your time and ours with your lame comment you must care about this story in some form or another.
Every rule has an exception, and this is the only rule with no exceptions! Huh? -- Spatch
Every rule has an exception, and this is the only rule with no exceptions! Huh? -- Spatch
That's why I don't pay that much attention to the W3C. Great, it's a standard--in another year maybe software will start to implememnt it.
Bite the hand.
Vunerabilities such as specifing to do nasty things with pi, and e, and the number 1. Think of the chaos! ...<mo><mo><mo>... Buffer overflow problems are probably going to be the first things to break. ^_^
karma is for the weak >)
I think this is really cool, but vector graphics are more important. I hope someone codes a WYSIWYG editor for this, because my brain can handle only so man bad acronyms between . Now just convince MS to adopt this standard.
Have a gander at XHTML modularization (behind which MathML was a big driver). No more need to wait for specs to settle down for everything, just what you need.
Not Quite
It is getting close though, and with plugins and authoring support already coming from the major graphics application vendors, promises to have a chance of being used in the mainstream.
Combine SVG with a DOM and scripting support (the obvious being Java^H^H^H^HECMAScript) and you have the beginning of an open standard Flash killer.
Bleh!
You mean SVG? The non-proprietary W3C developed vector graphics format that I can view (admittedly only a limited subset of) in my browser today?
Stuart.
The M18 milestone of Mozilla has binaries with the MathML support compiled in, so you might like to give that a try. If you feel a little more adventurous, here are some pages with more recent binaries with MathML support - Linux and Win32. These also have SVG support compiled in as well for vector images and the Linux binaries have XSL as well.
For more general information, take a look at the Mozilla MathML page.
Cheers,
Toby Haynes
Anything I post is strictly my own thoughts and doesn't necessarily have anything to do with the opinions of IBM.
i would agree. after you get over the initial hump TeX (or rather LaTeX) is really easy. i would say that I think LaTeX's shortcoming is lack of advertisement. People simply just dont know about it. It's a shame really.
$\frac{1}{x}$ to write 1/x
is pretty easy, i'll have to check out the math ml and see what it's like. or i'll probably just write everything in LaTeX and use tex2html which has pretty nice output.
use LaTeX? want an online reference manager that
-- john
Phew! That makes me feel a LITTLE better. However, remember that Java is also at 2.0, and look at the problems that still linger (the most notable slowdowns are still due to the nature of Java itself).
"Ancillary does not mean you get to rule the world." --U.S. Circuit Judge Harry Edwards, speaking to the FCC's lawyer
Some of you might be interested in the MathML and SVG of Mozilla. That build requires NO PLUG-INS to do simple SVG and decent MathML.
The future of the web should be: built in SVG (and for math stuff: built in MathML). Most of the little graphics and font hacks that are currently necessary to build a web page would go away if we can get built in SVG in the major browsers.
When people think SVG and they hear that it is a replacement for Flash, they don't realize that the advanced features of Flash are not could be put to immediate use on most web pages. Scaleable fonts of a guaranteed size and simple graphics are must more valuable. Take the page that you are reading right now for example. There should be no reason that you have to download the title bar text "Slashdot News for Nerds..." as a pixmap graphic.
Plug-ins are not the best solution for MathML and SVG/Flash.
And 3 or 4 years down the road some butthole company will announce they have a patent on it and start trying to extract royalties from the community. Worked for Unisys...
I'm trying to teach myself to set people on fire with my mind... Is it hot in here?
"Wasn't XML supposed to allow us to build this stuff without plugins and stuff?"
no. xml is a standard. it's up to users to adopt it.
The W3C lives in a fantasy world where we all have good WYSIWYG tools for doing all of our HTML design. Complexity doesn't matter, as the tools handle it all for us. Unfortunately, at the current state of the art, the good tools suck and the bad tools (which everybody seems to use) produce HTML of Lovecraftian horror. We're gonna be doing it (or at least cleaning it up) by hand for a long time, folks.
MathML is another one of those things that tries to be all things to all people, all at the same time. Result is yet another markup language trying to be a page description language.
As a counter suggestion, how about something based on the old eqn preprocessor for troff? Simple and easy to understand, and far more in the "content, not format" spirit than MathML. Brian Kernighan was supposedly working on something like this. Any news?
--
Welcome to the Turing Tarpit, where everything is possible but nothing interesting is easy.
..because of the many legacy documents in TeX, and because of the large authoring community versed in TeX, a priority in the design of MathML was the ability to convert TEX mathematics input into MathML format.
i dont see why they cannot just take raw tex formatting and use that. why create a whole other language to do the same thing?
use LaTeX? want an online reference manager that
-- john
The article says: Modern mathematical notation is the product of centuries of refinement, and the notational conventions for high-quality typesetting are quite complicated.
That notations are complicated is clear, but it's not clear to me that there have been centuries of refinement. My impression is that it's more like months of refinement spread out across centuries of accepting the status quo.
When someone creates a new form of mathematics, she usually get to define the notation for it. If she has no ability in design or semiotics, you may end up with something very far from clear and optimal. In creating notation, you want to take the abstract and make it concrete. But many mathematicians thrive in the highly abstract, and find beauty in representing entire constructs with a single letter.
The strong sense of continuity and tradition in mathematics seem to ensure that the notation will likely be frozen in that state. If we were to take all we've learned about psychology, perception, and structured computer programming, we could probably come up with a unified notational system for math that streamlined everything considerably. Just a hypothesis.
We've got a package developed in-house that combines Perl, TeX, dvi2html and several equation formatting tools to be able to generate any function we need in an acceptable appearance. AND, it can do on-the-fly equations as well... so long as you count "on-the-fly" as "within 20 seconds or so when the server isn't busy". :)
-Chris
...More Powerful than Otto Preminger...
To see what I mean compare any hand written HTML document with a compliant XHTML document. What it comes down to is this: non-trivial XML documents basically have to be machine generated.
Whether this is a good or a bad thing is left as an excersize for the reader. I personanly think that no matter how many open-source/portable/perfect document generators there are, there will always be a market for typing this stuff in straight text, in a format designed to be hand-editable.
I really personally dont see what the big deal is with XML. Its a nice system for defining text document formats that follow a narrow set of rules and can be edited by hand (but doing that is error prone and unforgiving of mistakes). Its not going to change the world per-se, though it might help to unify formatted documents.
The Assayer - free-information book reviews
Find free books.
When I first heard of MathML I was very excited. I had been doing all of my math in LaTeX. But for writing I sometimes prefered HTML because of it was interpreted (faster), cleaner in some respects, and graphically stronger (nested tables, images, colors).
And then I laid my eyes on some sample MathML.
To write (a+b)^2 in MathML you would have to do this (paraphrased from the W3C's page).
<MATH>
<MROW>
<MSUP>
<MFENCED>
<MROW>
<MI>a</MI>
<MO>+</MO>
<MI>b</MI>
</MROW>
<MFENCED>
<MN>2</MN>
<MSUP>
<MROW>
</MATH>
Whereas in LaTeX, you would write $(a+b)^2$.
This is nothing new, MathML1.0 looked just as bad, pure shit infact. It was never intended to be written or read by hand, but to be a low level, generallity of math typesetting. And this begs the question, what the fuck is it for? It is ML and therefore it is highlevel, all those tags don't make it any more readable by a computer, like <MO>+</MO> wow plus is a math operator, no shit. TeX powerful enough to describe everything mathmatical, and simple enough to write by hand. Why?
Why, why, why? We don't write javascript with tags. I for one won't write anything unless I can do a <MATH LANG="TeX">.
i thought so too, and the alt tags hold the LaTeX code for that particular equation.
use LaTeX? want an online reference manager that
-- john
If Opera or iCab ever support it, that would probably finally motivate me to dump NS 4.72. But I'm sorry, I just refuse to downgrade to the latest NS/Mozilla/IE. Every browser is just bigger, slower, and more buggy than the last. What I really want is NS 4.73 -- you know, the bug-release version of NS 4.72 that would never crash, and would fix bugs like incorrect rendering of stylesheets.
I think we're looking at a really long delay -- maybe 10 years -- before anyone can really start writing MathML into their pages with any confidence that the typical user will have a browser that displays it correctly. Look at Java 1.1. It's been years since 1.1 came out, and I still have to have a warning at the top of my 1.1 applet's page explaining that it won't work with the NS+MacOS combination.
The Assayer - free-information book reviews
Find free books.
"Well, a browser plug-ins are no different than a software or hardware update."
When you're making a web site, you can't expect people to 1) visit the site that hosts the plug-in, 2) fill in a registration form, 3) download your software, 4) exit Netscape, 5) run "setup.exe" 6) restart Netscape, 7) Find your website again. They'll just go to some other website instead.
Furthermore, lots of KDE users just use kfm as their web browser, and lots of GNOME users use Galeon. Plug-in makers don't make plug-ins for anything but Netscape and IE. Sometimes you can't get a UNIX plug-in at all.
My grandmother gets free cable internet from their city government that runs through the TV cable and is accessed from an appliance that sits on top of the TV and also functions as the cable TV decoder. This device doesn't support plug-ins at all. So your argument breaks down completely in this instance.
Now in our case, we can probably install MathML plug-ins on our machines and administer the test, since we control the machines on the client-side. But most plug-ins are generally worthless, and forcing your website readers to use them is textbook example of a Really Bad Idea.
========
Stephen C. VanDahm
Yeah, here are some reasons why.
Knuth spent 10 years on TeX.
It was one man, not a commitee,
and he designed it with
much more than math equations in mind.
It is able to be hand written;
MathML is too bulky to be done by hand.
And as for vector graphics,
TeX is actually based on Metafont
which supports all kinds of graphics operations
tho' without colour...(yet)
Unfortunately, a TeX pluggin hasn't materialized...?
and probably won't: TeX is a HUGE system.
Also, TeX doesn't support semantic content,
but serious mathematicians have little use for that anyway.
Oh, Knuth, where are you now?
Save us from this pitiful XML nonsense!
!#
Finally, a way to teach terrorists about fission with web pages that make everything look correct! This is truly a great thing, praise the green bug god for this. Now to have my minions can study mathematics and chemistry as i force them to slave away scraping watch hands so that I can collect enough fissionable material!
"Science is about ego as much as it is about discovery and truth " - I said it, so sue me.
..until Microsoft releases an "Active Math" extention for this?
"Depression is merely anger without enthusiasm." - Anonymous
While I'm certain that this is a major boon for scientists, mathemeticians, and even doctors, I would much rather have seen a W3C Reccomendation for a non proprietary vector graphic format.
The next Slashdot story will be ready soon, but subscribers can beat the rush and slashdot the links early!
Should be shot... isn't there some sort of w3 recommendation against using fonts like that???
DrLunch.com The site that tells you what's for lunch!
...but please, PLEASE promise us that it'll run perfectly on the first version! I hate nothing more than a standard with bugs, loopholes, and vulnerabilities.
"Ancillary does not mean you get to rule the world." --U.S. Circuit Judge Harry Edwards, speaking to the FCC's lawyer
Every time a solution to my HTML problems comes out, it takes years to get supported, it's inconsistent (style sheets), and by the time it's actually reliable, it's darn near irrelevant. I, at least, need a faster method of implementing solutions. Waiting on the browser gang is like waiting on Death.
1Alpha7
Live to be Moderated
nah, they should just use tex
If you read the article, you can see that MathML 2.0 designing goals include
"Facilitate conversion to and from other mathematical formats, both presentational and semantic.
Output formats should include:
And I'll be able to throw away my PIII-G400-17" computer to look at these maths on my favorite VT320 (amber on black or green on black) !!!
#include "coucou.h"
Now I can write
Sigma n=0 -> Infinity ((x^n))/(n!) = e^x
without seeming utterly mad! Muahahaha!
... and people wonder why there are no decent browsers. How can browsers be compliant with the W3C standards if said standards change every other week? If the W3C would just quit twiddling around with stuff, perhaps we could get a few browsers written, debugged, and working properly before the spec changed yet again.
--LordEq
Tho' your promise count for nothing
Certainly, (La)TeX is better, easier, more widely known among mathematicians and scientists ... now. But when you can be sure that every desktop with a browser on it can render MathML, but less than 1% of them will have TeX, which will be taught? Which will be supported and used. Some of us will continue to use TeX until the alternatives are clearly superior. Unfortunately, many people will sacrifice some of the things that TeX does so well that aren't obvious for ubiquity. I can count on TeX to fill my paragraphs beautifully. Browsers don't always get it right.
The net will not be what we demand, but what we make it. Build it well.
on that note:
how are the articles about nanotechnology relevant to most of us? or what about D-VHS? synthesizers made from old commodore 64 parts?
this is news for nerds...
i think all of the above catagories including the incorporation of mathml specs into the html standard apply to nerds of one type or another.
use LaTeX? want an online reference manager that
-- john
The standard proposes to do lots, including:
Facilitate conversion to and from other mathematical formats, both presentational and semantic. Output formats should include
Anything which will allow input and output into Mathematica and TeX both (let alone the others) is going to not be something that you can not type directly by hand, so for this standard it would be unfair to expect that. Instead, it is important to make sure that the standard includes the important mathematical notions that will port from TeX and computer algebra systems. (to me, that means all of TeX and LaTeX except the page-layout specific features, and most of Mathematica, Maple, and Matlab...)
It may be that the standard is trying to do too much or that it would only be useful to the mathematical elite, but given the ambitious role it is clear that the standard will need to be complicated and presumably not suitable for unaided digestion or production.
See the standards page here, for the 12 line code for the expression for (a+b)^2.
It's psychosomatic. You need a lobotomy. I'll get a saw.
really thought:
e(i*pi)-1 = 0
is a beautiful equation. in incorporates all of the important numbers e, pi, i, 0 and 1. i've often thought that if i ever got a tatoo that would be it. yes i know i'm a nerd.
use LaTeX? want an online reference manager that
-- john
There is a TeX plugin from IBM, it's called TexExplorer or something like that. By the way, I don't see how you can do a search in a bunch of text files for a particular kind of formula (``search for all the formulas containing 'something' to the power three inside an integral'') ... I hope this MathML stuff will make that kind of things possible.
Again we run into this problem... trying to make HTML documents *look* the way we want.... people try and get around this sort of problem by doing silly things like putting invisible images in web pages to get things to line up right...It seems MathML is some kind of bastard stepchild between the original beginnings of HTML and the LaTeX philosophy.... I guess XML was supposed to solve all of that, but I think that needing special programs to generate a document because the tags are so complex is a little much... I know it's been said before, but maybe some LaTeXish thing would be better... relatively simple to code, yet robust enough to remain flexible... I mean $\frac{x^{2}}{y_{3}}$ just seems alot nicer than the junk you'd have to write to get the same result in the current specs for MathML. But then again, I guess it won't matter if I write latex2mathml.pl anyway. :)
"Needless to say, I'm not very comfortable with depending on browser plug-ins for anything."
:-)
Well, a browser plug-ins are no different than a software or hardware update.
If you disagree, than you should still be using a 1980s hardware and software.
Yes, it is a pain. But there is no gain with out a pain. Go figure.
Karma stuck at 50? Add 2-5 inches.. err.. 2-5x Karmas Count to your pen1es.. err.. Karma all naturally and private
Something like this would be schweet if it became widely accepted. HTML is OK for most things, but mathematical notation and things molecular diagrams in Chemistry have to be represented with image files which can, of course, be a real pain in the ass.
Our little college is looking into MathML as a possible way to give online placement tests to incoming first-year students, and I found that it required special browser plug-ins to work. So I'm guessing that a browsers like Opera, kfm, and my grandmother's TV internet appliance are just out of luck. Needless to say, I'm not very comfortable with depending on browser plug-ins for anything.
How soon is it going to be before browsers support this stuff out-of-the-box, without me having to download and install a stupid plug-in?
========
Stephen C. VanDahm
MathML markup is not primarily intended for direct use by authors. While MathML is human-readable, which helps a lot in debugging it, in all but the simplest cases it is too verbose and error-prone for hand generation. Instead, it is anticipated that authors will use equation editors, conversion programs, and other specialized software tools to generate MathML.
--
--
You are a fucking moron.
Of course, as per the linked document, MathML isn't supposed to replace TeX any more than PostScript is - it's meant to be machine-generated from easier-to-input forms such as TeX.
Well, I'd still prefer how TeX does it... much more compact. I am somewhat uncomfortable with using a huge number of tags for complex (extensive) formulas.
Moz.
see a Text Widget
Anyone else find it strange that Microsoft is on the working group, and Netscape is not? Where was Netscape on this one?
With the dominant browser being IE and MS' consumer focus I really don't see this standard ever being implemented for a meaningful percentage of people. And then MS will implement it just wrong enough to make it worthless.
I hate to say it, but if you're really doing this stuff you probably already know (La)TeX, so you can just write it up in that and generate a graphic, or distribute the (La)TeX. Using graphics to represent this stuff on the web isn't optimal, but at least you know it will work and display right. (Sorry to you freaks who browse text only ;)
ChrisWhat's the difference between an orange?