W3C home > Mailing lists > Public > w3c-dist-auth@w3.org > January to March 2003

RE: DAV properties / PROPFIND vs GET/HEAD

From: Clemm, Geoff <gclemm@rational.com>
Date: Thu, 20 Mar 2003 23:30:07 -0500
Message-ID: <E4F2D33B98DF7E4880884B9F0E6FDEE2023C97B1@SUS-MA1IT01>
To: w3c-dist-auth@w3.org

I agree that we should clarify what is meant by "accept headers".
But I'd be concerned about dropping this language, because it could
introduce incompatibility between new clients (that would then 
expect accept headers to be respected) and old servers (that would
ignore them).

Cheers,
Geoff

-----Original Message-----
From: Julian Reschke [mailto:julian.reschke@gmx.de]
Sent: Monday, March 17, 2003 1:45 PM
To: w3c-dist-auth@w3.org
Subject: DAV properties / PROPFIND vs GET/HEAD



Hi,

RFC2518 currently says that the values of get* properties must match those
HTTP headers returned upon a GET/HEAD request "without accept headers" (for
instance, see 13.3).

I'd like to see clarified

a) the rational for this restriction, and

b) a precise definition of what "accept" headers are. Looking at RFC2616,
the only "accept" header is actually the one called "accept", while
"accept-language" does not qualify.

My personal preference would be to get rid of this language, unless we can
agree on what it's supposed to accomplish, and rephrase it accordingly. I
personally think that PROPFIND should be just an extended syntax for HEAD,
supporting additional properties and allowing collection member retrieval,
but stay completely compatible with the GET/HEAD response headers where
possible.


Julian

--
<green/>bytes GmbH -- http://www.greenbytes.de -- tel:+492512807760
Received on Thursday, 20 March 2003 23:30:14 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 2 June 2009 18:44:03 GMT