On 12/7/07, Dave Pawson <davep@d...> wrote:
> Mark Baker wrote:
>
> >> Is it the case then that every time someone creates a new namespace
> >> they have to create a new MIME type to go with it?!
> >
> > Not necessarily a new namespace, but every new language should have a
> > new media type, yes.
> >
> Which definition of language Mark?

"vocabulary".  So by "new language" I mean a vocabulary which isn't
built upon - as an extension of - another.

> > I gave a presentation to the W3C CDF WG a while back that might help
> > shed some light on the subject;
> >
> > http://www.markbaker.ca/Talks/2004-media-types-and-compdocs/slide1-0.html
>
>  From which
> Many assume that applications will dispatch on the root namespace
> But implementations don't, in general, do this ... ok, newer ones do,
> mostly. Older ones don't though.
>
> and
> Such a document might be handled....
> and
> Obviously there's some wriggle room for interpretation....
>
> I like the arguments put forward, but you seem to accept it's all
> ifs and buts and maybes as of today?

Pretty much.

> Content-Type: application/compound+xml (??)
> <html:body cdf:type="image/svg+xml">
>   <svg:svg ... />
>
> Seems quite workable and extensible (given your security caveats).

Yes, thanks.  I wrote a section for the CDI (Compound Documents by
Inclusion) which speced out that solution, also my favourite.  I left
the WG a while ago though, and since their work isn't in public, I
don't know what's happened to it.

> Is it a TAG issue? They don't seem to have done much other than say
> register?
> http://www.w3.org/2001/tag/2002/0129-mime

Yep, it's an issue.  It started as one, but split into three;

http://www.w3.org/2001/tag/issues#mixedNamespaceMeaning-13

> Nothing about applications or dispatching.
> Suggests it's one of those that nicely falls in the cracks?

Yep.

Mark.
-- 
Mark Baker.  Ottawa, Ontario, CANADA.         http://www.markbaker.ca
Coactus; Web-inspired integration strategies  http://www.coactus.com