W3C home > Mailing lists > Public > ietf-http-wg@w3.org > July to September 2004

Re: MIME registration for application/gdiff

From: Lisa Dusseault <lisa@osafoundation.org>
Date: Tue, 28 Sep 2004 09:27:15 -0700
Message-Id: <421078F3-116B-11D9-8494-000A95B2BB72@osafoundation.org>
Cc: HTTP working group <ietf-http-wg@w3.org>, Linus Walleij <triad@df.lth.se>
To: Martin Duerst <duerst@w3.org>

>> How does <http://www.w3.org/International/iri-edit/draft-duerst-iri> 
>> work?
> There are three actual files on the server,
> http://www.w3.org/International/iri-edit/draft-duerst-iri.xml
> http://www.w3.org/International/iri-edit/draft-duerst-iri.txt
> http://www.w3.org/International/iri-edit/draft-duerst-iri.html
> and they get content negotiated, but of course can also be
> accessed directly. I'm not sure exactly what this is in your
> terminology.
Then I imagine the authoring model must be that the client can PUT an 
application/xml file to draft-duest-iri.xml, PUT a text/plain file to 
draft-duerst-iri.txt, and so on.  The resource draft-duerst-iri may not 
truly exist and it might be either forbidden or cause confusion if the 
client tries to PUT to that URL.    By that model, the PATCH method 
could work exactly the same with its existing syntax.  Only the 
GET/HEAD methods actually do content negotiation.  Probably PROPFIND 
and PROPPATCH and other WebDAV methods are undefined, or if they are 
defined they'd work independently on each resource, so that for example 
PROPFIND 'getcontentlength' to the text file would retrieve the text 
file's length.

It would certainly be interesting if some day this were 

Received on Tuesday, 28 September 2004 16:27:28 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 1 March 2016 11:10:38 UTC