Re: Relation between markup and transport

On Thu, 5 Jul 2001, William F. Hammond wrote:
> Arjun Ray <aray@q2.net>, Mon, 2 Jul 2001 23:57:39 -0400, writes:

>>> 2. Namespace extensions.
>> 
>> A maguffin, frenetic W3C boosterism notwithstanding. 
> 
> The key point in the distinction between "text/html" and "text/xml"
> should be whether there is implied knowledge of markup vocabulary.

What the names (should) mean?  For text/html, yes; text/xml, no.

> Namespace extensions provide that.

Nope.  If they did, then the discussions on the xml-sig archive,
amounting to 1500-2000 messages, were conducted in bad faith by some
parties.
 
> Would you call it sniffing if a revised spec said that when the XML
> form of HTML is served through HTTP as "text/html" the first non-blank
> line of the body of the served object must be a line matching stated
> criteria.

Probably not, because then that would be part of the definition of
text/html itself, and therefore not only known to, but also expected
of, "text/html processors" (whatever these beasts might be.)

>> (Personally, I think the best answer may be a separate text/xhtml
> 
> Then how long would it be before content providers could assume that
> this reaches the same audience as "text/html"?  

They don't have to assume anything.  Either a client announces support
(typically in an HHTP Accept: header), or it doesn't.  

> Wouldn't the old type name remain as the default content type?

No.  It could be a policy implemented by a server.

> If this is done, would it also make sense to change the name of
> the root element from "html" to "xhtml"?    :-)

Only the namespace-besotted would worry about that.  Let 'em. :-)



Arjun

Received on Thursday, 5 July 2001 20:01:11 UTC