Next message: jamsden@us.ibm.com: "Re: Questions on activities"
Date: Thu, 6 Apr 2000 18:13:48 -0400 (EDT)
Message-Id: <200004062213.SAA04086@tantalum.atria.com>
From: "Geoffrey M. Clemm" <geoffrey.clemm@rational.com>
To: ietf-dav-versioning@w3.org
Subject: Re: Stable URL's for working resources?
From: "Tim Ellison/OTT/OTI" <Tim_Ellison@oti.com>
I feel like I'm being awkward today... Working resources are first
class resources, they have an identity, and thereby a Stable URL.
I vote that we keep it.
First class resources: I agree.
They have identity: I agree.
They have a stable URL: Now there we diverge. Most first class web
resources with identity do *not* have a stable URL. The only way
we can get away with requiring them for versioning metadata is that
versioning metadata has to persist even after it is no longer
"current", and so inevitably there is some underlying versioning
store that we can require an implementor to expose in a standard way.
On the other hand, working resources do *not* persist after they are
no longer current (i.e. have been checked in or unchecked out).
Implementations very sensibly take advantage of that fact, and do
not associate the level of tracking and control applied to something
like an immutable revision.
Regards,
Tim
p.s. Although the resource is mutable, that means its contents/members
change -- the working resource itself (i.e., the 'thing' that is in a
checked-out state) is persistent and uniquely identifiable.
Having a unique id for a resource is easy. But a stable URL is not
just a mechanism for identifying a resource, it is also a mechanism
for *locating* the resource. And making a stable one of those is hard,
and not something normally done, even for first class resources.
Cheers,
Geoff