Re: Application-specific media types

[Rick Jelliffe <ricko@xxxxxxxxxxxxxxxxxx>]

On Fri, 7 May 1999, Shane P. McCarron wrote:

> So, my question for you all is, would anyone here object to the creation
> of an application specific media type text/xhtml, registered by the W3C,
> that would be used to announce resources that are within the XHTML
> family (as that is defined by the W3C)?

Well, without arguing on merits or needs, like I said, I think the form 
needs to be manageble: so application/xml.html (using an XML registration 
tree or an xml subtyping mechanism) would a much preferable form. 

We should create a media-type name framework in which it is easy for any 
existing format to have an XML version (e.g. application/xml.vrml), and 
for any ordinary user to be able to read the media type and figure out 
what is going on.  The trouble with application/xhtml is that is 
disguises what is going on: users may think it means experimental HTML, 
or they may miss that it is HTML att all.

When a browser does not have a handler registered for a media type, it 
asks the users. This gives moderately sophisiticated users the chance to 
try some other application: they need to be able to look at the media 
type name and make some sense of it.  So  application/xml.html is better than
application/xhtml, and application/xml.vrml is better than 
application/vrxml or application/xvrml (and I think model/xml.vrml is 
better than both those).

This is why I think and "xml." prefix (which would probably require an 
XML registration tree) and some simple automated registration process (so 
that people can have instant access to a name) is the most workable 
solution: it seems to fit into MIME specs, works with existing 
technology, and provides a nice convention for conveying its meaning to 
humans.  (People can still register XML media types under IETF, 
experimetal or vendor trees, so a registration tree just adds convenience.)

Rick Jelliffe