Lists Home |
Date Index |
> May I suggest that that problem is solvable?
Lots of problems are solvable in theory but politics has a nasty habit
> As these dialects have traditionally made very extensive use of large
> entity collections,
If you think docbook and xhtml use large entity collections, take a look
> 4) approach developers and maintainers of parsers and validators with the
> feature request to support the new mechanism
That's where I disagree. I don't want to use non-xml systems 9which is
what you are suggesting here).
there have been several proposals in thsi area eg Simon's ents
or even for that matter my generation of the Mathml and draft ISO entity
sets from unicode.xml
I think alternative schema languages are only partly the issue, there is
an official xsd schema for mathml and unoffical relax ng ones, but the
main issue to me is fragments: I want to be able to tell somene that
they can type
to get an alpha, and I want that fragment to be a legal piece of
"something" that they can cut and paste into an xml document.
Currently it is just a non-wellformed error condition according to the
xml spec, although you can of course cut-and-paste it as text, which is
what people are forced to do, but I'd rather people use xml tools to
handle xml (even though I just use emacs for everything:-)
Howver much I'd like to be able to have some non-dtd mechanism, and in
particular a mechanism that was not tied to the head of a document,
I don't want to lead mathml users into a gheto of non-standard parsers
handling non-xml, so it seems to me that the political issues of making
this legal xml are more important and need to be solved first.
Probably there has to be some syntax, presumably in xml form, to declare
the things, but the exact syntax is not really important at this stage.
This e-mail has been scanned for all viruses by Star Internet. The
service is powered by MessageLabs. For more information on a proactive
anti-virus service working around the clock, around the globe, visit: