Re:MathML and CSS

Ian Hickson wrote:
> You know what I think -- I don't see any reason, as an implementor, to
> implement CSS extensions rather than just implement MathML itself.

One can add that it is easier to implement XHTML then XHTML + CSS.
But what is the point? 
IMHO it is better to have one metamarkup language and powerfull style 
language then plenty of markup languages XHTML/MathML/DocBook/TEI
will be easily merged into single rendering scheme. No need to waste
resources on individual markup languages. 
Isn't it purpose of style language?

> CSS extensions are significantly more complicated to implement and test
> than the equivalent MathML. MathML is self-contained, you only need to
> test how it interacts with the MathML around it. CSS extensions have to
> apply to all namespaces, and have to interact with the rest of CSS.

Agree. But it is worth to do especially when extensions are not 'ad hoc'
extensions targeted on mathematics but can be used in other markup languages
apart of MathML. And one is talking about few not too sophisticated 
CSS extensions. No one requires to rebuild whole CSS visual rendering model.

> So IMHO it's easier to convince someone to implement a subset of MathML
> than it is to convince them to implement extensions to CSS that allow them
> to then implement a subset of MathML using a stylesheet.

Then we will need to convince someone to implement CML 
(Chemical Markup Language), DocBook, TEI and who knows what.
But is not it better to have one capable rendering scheme,
rather to convince someone each time we need something new.
If LateX can, if DSSSL can, if XSL can, what is wrong with CSS?


-- 
_____________________________________________________________
Web-based SMS services available at http://www.operamail.com.
From your mailbox to local or overseas cell phones.

Powered by Outblaze

Received on Wednesday, 19 May 2004 04:52:44 UTC