W3C home > Mailing lists > Public > www-math@w3.org > April 2008

Re: Supporting MathML and SVG in text/html, and related topics

From: David Carlisle <davidc@nag.co.uk>
Date: Tue, 15 Apr 2008 15:51:59 +0100
Message-Id: <200804151451.m3FEpxti011390@quetzal.nag.co.uk>
To: jirka@kosek.cz
CC: whatwg@whatwg.org, public-html@w3.org, www-math@w3.org, www-svg@w3.org


Jirka,

> Why not simply allow such features only in XML serialization where XML 
> namespaces could do the job without introducing any new features? 

That was of course the original model, circa 1998 that new content would
move to a more modular, clearer, XML world. However for various
reasons it hasn't happened yet (on the general web) one could argue
endlessly about the reasons for that.

However if HTML5 is standardised in its current form then there is no
chance of a general move to XML on the web ever happening. People,
authoring tools, web browsers, will all typically use text/html with the
more forgiving parsing that implies. This isn't necessarily a bad thing,
but it's definitely not the original vision at the time xml, xhtml,
mathml were conceived.

If the entire tool chain is html based, but svg and mathml are only
available in XML (which is the current situation, officially) then that
means that there is a massive barrier to putting mathemtics or vector
graphics on the web. Not only the mathematics/graphics, but the entire
page has to be authored in a different way, even if you just want to add
a fraction somewhere, and when you have authored your page you need to
make sure your web server is configured to send a different mime
type. Practically speaking this is a massive (fatal) barrier to
widespread adoption. Currently it's hard to put matheml on the web,
principally because it's hard to put xhtml on the web (the vast majority
of people who claim that they are serving xhtml are really serving
text/html). That works (some of the time) for (x)html but it doesn't work at
all for anything with mathml or svg.


So.... if the html community decides that it's not going to deprecate html
and move towards xhtml long term, then as a consequence I think that
there has to be a way to get mathematics and vector graphics markup into
html.  The current html5 draft solution with html and svg special cased
rather than a generic extension mechanism with svg and mathml being
specific examples is not what I'd have chosen, but it isn't my choice
and it is workable and will I think cover a vast majority of the use
cases.

David
Received on Tuesday, 15 April 2008 14:53:42 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Saturday, 20 February 2010 06:13:01 GMT