Next message: Clemm, Geoff: "RE: Localized baselines"
Message-ID: <65B141FB11CCD211825700A0C9D609BC01D4D72E@chef.lex.rational.com>
From: "Clemm, Geoff" <gclemm@Rational.Com>
To: "'Tim_Ellison@oti.com'" <Tim_Ellison@oti.com>, ietf-dav-versioning@w3.org
Date: Fri, 25 Feb 2000 17:54:06 -0500
Subject: RE: Protected properties
Actually, my intent was that a protected property cannot
be updated by a client on any resource. So a client could
not set a DAV:checkin-date on the working resource. I think
this is both simpler to implement by a server, and better
behavior for a client. Do you agree?
Cheers,
Geoff
-----Original Message-----
From: Tim_Ellison@oti.com [mailto:Tim_Ellison@oti.com]
Sent: Friday, February 25, 2000 5:16 PM
To: ietf-dav-versioning@w3.org
Subject: Protected properties
My interpretation of a protected property is that it is owned by, and
defined by the server. If a conflict occurs, the server wins.
For example, a revision has a protected property called DAV:checkin-date,
but a working resource doesn't. If a client chooses to set a property
called DAV:checkin-date on a working resource they are free to do so (since
it is not protected), however, when the working resource is checked in, the
server adopts that property as its own, and overwrites the user's value
(thus loosing their data).
Seems fair to me.
Tim