Next message: jamsden@us.ibm.com: "Re: DAV:revision-resourcetype"
Date: Wed, 26 Jan 2000 14:33:33 -0500
Message-Id: <10001261933.AA28886@tantalum>
From: "Geoffrey M. Clemm" <geoffrey.clemm@rational.com>
To: ietf-dav-versioning@w3.org
Subject: Re: "stable" href's
From: jamsden@us.ibm.com
This feels a little like exposing server implementation details to clients.
It's not a server implementation detail, but rather something the client
has to know in order to do its job. If it gets a stable URL, it can't
move it, and clients need to be designed accordingly. In particular,
a client can store these URL's in documents with a high degree of
confidence that they will continue to identify the same resource.
Some servers may need to move these URLs around for one reason or another.
I agree that they may be forced to do so, but with the understanding
that this will invalidate all cached copies of those URL's. What
is important is that *clients* cannot change those URL's, so that they
are only changed by some out-of-protocol administrative action.
So clients should only rely on the URL bindings they made and shouldn't be
messing with server URLs.
If we don't define which are stable, how will a client know which ones
they can and should mess around with?
I know we've said that there will be a stable URL
for each resource for versioning unaware clients,
These stable URL's are for versioning *aware* clients, so that they
can maintain reliable references to revisions, versioned resources, etc.
but I don't think those
URLs will be useful, especially to versioning unarare clients because they
won't resemble anything meaningful.
A client is just a computer program, so when it is storing and
retrieving a URL, it doesn't really care whether it is meaningful,
but only that it continue to reference the intended resource.
Cheers,
Geoff
"Geoffrey M. Clemm" <geoffrey.clemm@rational.com>@w3.org on 01/20/2000
11:02:22 PM
Sent by: ietf-dav-versioning-request@w3.org
To: ietf-dav-versioning@w3.org
cc:
Subject: "stable" href's
After thinking for a while about Neil's question about whether a
MOVE can be applied to a revision, etc., I now believe that it would
be worthwhile for us to define which properties contain "stable"
URL's, i.e. URL's allocated by the server that cannot be modified
by a client with a MOVE request.
Unless anyone objects, I propose to make a pass through the protocol
identifying those properties which I believe identify stable hrefs.
The value to a client is that it can cache these names with the
guarantee that another client cannot MOVE them somewhere else. A
server can of course chose (or be forced) to break these bindings, but
there's nothing we can do about that.
Comments?
Cheers,
Geoff