Next message: Boris Bokowski/OTT/OTI: "postcondition for PUT"
From: Jim Whitehead <ejw@ics.uci.edu>
To: WebDAV WG <w3c-dist-auth@w3.org>, http-wg@hplb.hpl.hp.com, ietf@ietf.org, ietf-dav-versioning@w3.org
Date: Tue, 15 Aug 2000 17:51:46 -0700
Message-ID: <NDBBIKLAGLCOPGKGADOJCENIDIAA.ejw@ics.uci.edu>
Subject: Improving HTTP/DAV status reporting
A common concern raised by developers of WebDAV clients and servers is the
deficiencies of HTTP/DAV status reporting. For example, two common problems
are overloading of HTTP status codes to mean several different conditions,
and the inability to precisely report many kinds of status using the
existing status codes. Even when an existing status code does cover a
condition well, there is no mechanism for passing supplemental status
information. WebDAV provides two operand methods such as COPY and MOVE where
an error could occur at the source or destination, a capability outside the
scope of HTTP error reporting. It is also possible that multiple error
conditions can simultaneously occur, yet only one can be reported in a
response.
Especially for 4xx series status codes, there is a limited number of status
codes, leading to conservative use of status codes within the protocol
development community, and leads to overloading of existing status codes. A
principled expansion of the HTTP status code space seems to be a cornerstone
of any effort to improve HTTP/DAV status reporting.
To evaluate and address these concerns, a new mailing list has been created
to discuss issues concerning improved HTTP/DAV status reporting.
To post to this list, send your email to:
report@webdav.org
Subscription instructions and archives are available at:
http://mailman.webdav.org/mailman/listinfo/report
This mailing list is welcome to all, and maintains a public archive.
While the impetus for this discussion is coming from the WebDAV community,
this problem affects HTTP broadly, and hence a new forum for discussion has
been created. It is expected that list members will determine where
specification work, if any, will take place.
- Jim Whitehead <ejw@ics.uci.edu>