W3C home > Mailing lists > Public > www-html-editor@w3.org > January to March 1999

RE: Content-Document-Type: was (Re: MIME types vs. DOCTYPE)

From: Jonathan Borden <jborden@mediaone.net>
Date: Fri, 26 Feb 1999 09:23:09 -0500
To: "MURATA Makoto" <murata@apsdc.ksp.fujixerox.co.jp>, <xml-dev@ic.ac.uk>
Cc: <www-html-editor@w3.org>
Message-ID: <000001be6193$89002f10$d3228018@jabr.ne.mediaone.net>
MURATA Makoto wrote:
>
>
> I am a co-author of RFC 2376 (XML media types).   I am attaching two
> of my e-mails about text/xml and application/xml.
>
> I am quite sympathetic to Jonathan, but I do not think that the URI
> of the DTD is always appropiate.  Tim's suggestion (a namespace URI plus
> the root element type) sounds very interesting.

	I was not suggesting that the URI be the URI of the DTD, rather that this
be a 'standalone' URI in the same fashion of the namespace URI.

	This provides the same type of unique document type identification as does
the namespace URI without predicating the existence of a DTD. In some cases
the DTD URI might be appropriate, in other cases a schema URI might be
appropriate, in other cases a uuid.

	In your attached e-mails there are several excellent solutions to this
problem in a similar vein. Use of a Content-Type parameter for text/xml and
application/xml has the same expressiveness as a distinct header, and since
we are in reality subtyping the content type this is perhaps more
appropriate, and would easily fit into RFC 2376.

	I assume that this parameter could be used with other content-types such as
text/sgml?

Jonathan Borden
http://jabr.ne.mediaone.net
Received on Friday, 26 February 1999 09:29:17 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:16:44 GMT