W3C home > Mailing lists > Public > w3c-dist-auth@w3.org > April to June 2001


From: Clemm, Geoff <gclemm@rational.com>
Date: Wed, 25 Apr 2001 14:36:39 -0400
Message-ID: <3906C56A7BD1F54593344C05BD1374B1018E2386@SUS-MA1IT01>
To: WebDAV WG <w3c-dist-auth@w3.org>
I believe it is simpler and more desireable to deprecate the use of allprop
in all situations, not just Depth:infinity.


-----Original Message-----
From: Jim Whitehead [mailto:ejw@cse.ucsc.edu]
Sent: Wednesday, April 25, 2001 2:20 PM

This issue is currently stated as:

If a server has some computed properties which are expensive to compute,
then PROPFIND allprop, especially PROPFIND allprop with Depth infinity can
be an expensive operation.  There should be an implementation note added to
the document noting this problem.  Suggestion that servers might want to be
conservative in their implementation of such compute expensive properties.
Clients should only perform PROPFIND allprop when necessary, not by default.

It was raised by Ken Coar at the Advanced Collections breakout at the
Orlando IETF meeting.

But, since that time there has been a substantive discussion of this issue,
begun by Lisa Dusseault, starting at:


My understanding of the rough sentiment of the working group is:

1) PROPFIND allprop requests with Depth infinity should be deprecated. That
is, PROPFIND allprop with Depth infinity will remain in the WebDAV
Distributed Authoring Protocol, but there will be a requirements added that
new clients MUST NOT use this capability.  When existing clients are
revised, they SHOULD be rewritten to avoid using PROPFIND allprop Depth
infinity requests. This will only apply to PROPFIND allprop with Depth
infinity -- PROPFIND allprop for Depth 0 and 1 will still be OK.

2) The default behavior of PROPFIND will be changed so it is Depth 0, not
Depth infinity.

- Jim
Received on Wednesday, 25 April 2001 14:34:36 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 20:01:22 UTC