W3C home > Mailing lists > Public > www-html@w3.org > November 2004

RE: Identifying (X)HTML without MIME

From: Ian Hickson <ian@hixie.ch>
Date: Wed, 10 Nov 2004 23:43:15 +0000 (UTC)
To: David Lieberman AWDSF <david@awdsf.com>
Cc: 'James Cerra' <jfcst24_public@yahoo.com>, www-html@w3.org, 'Lachlan Hunt' <lachlan.hunt@iinet.net.au>, trejkaz@xaoza.net
Message-ID: <Pine.LNX.4.61.0411102341260.8631@dhalsim.dreamhost.com>

On Wed, 10 Nov 2004, David Lieberman AWDSF wrote:
> I was under the impression that 'type-mixing' in MIME types is a not a 
> good idea because it mixes up two separate concepts: basic data types, 
> and processing guidelines that you use with the type.
> Is my thinking outdated here? I haven't taught this sort of thing in a 
> while.

In general, you're supposed to use the appropriate MIME type (where 
"appropriate" mostly means "intended to be used with the namespace that 
the root element has), so for an XHTML document (even if it contains 
fragments of SVG and MathML) you should use application/xhtml+xml, and for 
an SVG image (even if it contains fragments of XHTML or XForms) you would 
use image/svg+xml. But at the UA level, they should all be handled the 
same way, with dispatching being done via namespaces.

Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'
Received on Wednesday, 10 November 2004 23:43:17 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:06:09 UTC