Re: [syndication] friendlier feeds

["S. Mike Dierken" <mdierken@hotmail.com>]

>
> OK.  I guess this should allow to map extended mime types into namespaces.
But
> isn't as nice as:
>
> Accept-xml-namespace: http://purl.org/rss/1.0/
>

XML namespaces don't identify the schema (structure of names/values) of an
XML Document instance.
Multiple namespaces will occur in one Document.

If a new 'accept' header wanted to use URIs, I'd be all for that. I'd make
it relate to 'content-model' and make it independent of 'content-type'. The
'model' is the concept of schema, independent of syntax. For example, Java
serialized objects still have a schema (after deserializing them) and it
would be nice to be able to talk about that without loading the darn object
and getting its class.

So I'd use something like:
Content-Type: application/xml
Content-Model: http://purl.org/rss/1.0/

(I know the content-type has a more specific rss mime type, but this is for
example. And I don't know what the 'model' value syntax should be to work
across content-types)