On Monday, 6 May 2013 14:12:48 UTC+1, Trevor Saunders wrote:
> On Mon, May 06, 2013 at 08:24:07AM -0400, Boris Zbarsky wrote:
>
> > >I am still waiting for the rebuttal of my arguments, in the original email
> > >in this thread, about how TeX is strictly better than MathML for the
> > >particular task of representing equations.
> >
> > How easy is it to build an accessibility application on top of TeX,
> > or even a restricted subset of it? Note that these exist for
> > MathML, but not so much for TeX.
>
> I actually think it would be easier to map tx math into the
> accessibility APIs we support than mathml.
There are several problems/issues here:
# Context
How do you differentiate/identify math powers (e.g. "a^2"), footnotes (e.g.
"some text^1") and code ("int c = a^b;")?
With MathML markup, you have clearly identified what the content of the
document/sub-tree is.
# Parsing
With a TeX-like format, a speech synthesiser/screen reader/web browser would
need to write a parser for that format.
With MathML, the parsing is already handled by the SGML/XML/HTML5 parser so the
application can process it via DOM/SAX/a reader API.
> currently we don't expose mathml at all other than as a an object that
> we say is an equation, and its not really clear how to fix that with
> mathml.
This is enough information for the screen reader/speech synthesiser to know
that it has MathML content, and thus walk the MathML DOM to read the math out
loud. It should also be enough to query associated CSS styles to handle any
Aural CSS or CSS Speech styles associated with the MathML.
Another important consideration is existing web content. If you are going to
start rendering text that has e.g. "a^2" as math, then all documents that use
that, e.g. "<p>You can use a^b in TeX to denote 'a raised to the b<sup>th</sup>
power'.</p>"
- Reece
_______________________________________________
dev-platform mailing list
dev-platform@lists.mozilla.org
https://lists.mozilla.org/listinfo/dev-platform
