W3C home > Mailing lists > Public > ietf-dav-versioning@w3.org > October to December 2000

RE: Value of DAV:suported-method-set

From: Clemm, Geoff <gclemm@rational.com>
Date: Thu, 2 Nov 2000 13:20:49 -0500
Message-ID: <3906C56A7BD1F54593344C05BD1374B10D9F69@SUS-MA1IT01>
To: ietf-dav-versioning@w3.org
I'd like to be able to extend this at some point to include
things like "understood arguments to the method", and for this,
choice (a) is better, so I'd rather go with that.

Cheers,
Geoff 

-----Original Message-----
From: Jim Whitehead [mailto:ejw@cse.ucsc.edu]
Sent: Wednesday, November 01, 2000 6:06 PM
To: ietf-dav-versioning@w3.org
Subject: Value of DAV:suported-method-set


Currently, draft-ietf-deltav-versioning-10 specifies that the value of
DAV:supported-method-set is a series of empty XML elements, where each
element is in the DAV: namespace, and the name of the element exactly
corresponds to an HTTP/DAV/DeltaV method name.  I think this is less than
ideal, since it unnecessarily couples the DAV: namespace, and the space of
HTTP/DAV/DeltaV method names.  Prior to this property, these namespaces were
completely orthogonal. With the introduction of this property, with values
in the DAV: namespace, all new HTTP/DAV/DeltaV methods will need to be
chosen so they do not overlap an existing DAV: URI.  While there are
currently no collisions, I think it is valuable to keep these namespaces
separate, so future protocol extensions do not need to worry about this
namespace overlap.

The remedy is easy:

a) Place these names in a separate, dedicated namespace, such as
DAV://methods/.

b) Use some other marshalling.  A comma-separated list of method names might
allow reuse of server code that generates the value of the Allow HTTP
header.

My preference is (b), but either would address the coupling issue.

- Jim
Received on Thursday, 2 November 2000 13:28:25 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 13:57:39 GMT