W3C home > Mailing lists > Public > www-html@w3.org > November 2003

Re: Type Attribute

From: Ernest Cline <ernestcline@mindspring.com>
Date: Mon, 17 Nov 2003 02:50:44 -0500
Message-ID: <410-220031111775044625@mindspring.com>
To: "David Woolley" <david@djwhome.demon.co.uk>, "W3C HTML List" <www-html@w3.org>




> [Original Message]
> From: David Woolley <david@djwhome.demon.co.uk>
>
>
> Whilst I can't see many people using multi-valued types client side.
>
> > * q-values are specific to HTTP and thus not suitable for generic use.
>
> What is specific to HTTP is the ability to provide extra information to
> select a specific media type.  Even without HTTP, a browser could 
> combine local and remote q values (not many browsers support local
> q values) to choose a best match media type; it just doesn't have
> the ability to select the appropriate one.  One could, however, envisage
> a system where there were compound files from which a version could be
> chosen.

What protocol is there other than HTTP that supports
multiple versions of a resource available from the same URL?
Unless someone can present one that sees real use today,
adding to the burden that a user agent would face in
implementing XHTML by turning type into a multivalued
attribute with q-values is  a pointless bit of gee-whizery.

Even if such a protocol exists, unless it does not have
its own analogs of the Accept header and q-values
all the  comments about how the type attribute is redundant
with an http: URL would apply just as much.
Received on Monday, 17 November 2003 02:52:07 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 5 February 2014 23:40:10 UTC