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

the DAV:comparison report

From: Geoffrey M. Clemm <geoffrey.clemm@rational.com>
Date: Fri, 29 Dec 2000 14:28:22 -0500 (EST)
Message-Id: <200012291928.OAA19226@tantalum.atria.com>
To: ietf-dav-versioning@w3.org

During a proof-read pass over 10.13, it occurred to me that the
DAV:comparison report is too vague to provide a significant degree of
interoperability.  The example suggests it is useful for comparing two
collections, but how often does one really have an "old" and a "new"
copy of a collection on the server?  A smart client will keep track of
the changes made on the client, and only update the server with the
changes, not a new copy of the entire collection.  And even if one did
have both collections on the server, one usually wants a "smart"
compare that notices when a collection is "moved", instead of
reporting a a delete and an add for each member of the moved
collection.

As suggested in the text of this report, it is really for
comparing two baselines, which is of significant value since the
DAV:version-set of a baseline can be huge (and thus not something
one would want to download to a client), but the difference between
two baselines will often be manageable (and useful) information.

So I propose that we convert the DAV:comparison report into a
DAV:baseline-comparison report, and move it from the "general option"
section into the baseline option section, where it will be of
significant value for interoperability.

Comments?

Cheers,
Geoff
Received on Friday, 29 December 2000 14:29:10 GMT

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