- From: Geoffrey M. Clemm <geoffrey.clemm@rational.com>
- Date: Mon, 20 Nov 2000 13:46:19 -0500 (EST)
- To: ietf-dav-versioning@w3.org
From: Ross Wetmore <rwetmore@verticalsky.com> Actually, this was discussed earlier and it was agreed that version selectors could be shared amongst multiple workspaces but that this would be done through the binding protocol. We should distinguish two different concepts. The first concept is the standard 2518 concept of collection membership. Since a workspace is a collection, it has members, like any other collection. The second concept is that each resource can identify a single workspace in its DAV:workspace property. Let's call the first concept "member of a workspace" and the second concept "owned by a workspace". I think that Tim and I would be willing for a resource to be a member of more than one workspace (e.g. from nesting a workspace inside another workspace), but not for a resource to be owned by more than one workspace. The use for this was to insure consistent versions across several workig teams without requiring either each team to poll for changes, or each checkin to poll for and update all affected workspaces. For reasons like access control and locking, it is important to identify a single "owner", but I think shared membership is all that you need for the purposes you describe above. Putting such a limitation in the spec would invalidate this. The spec should either be extended to handle such cases at the current time, or it should be defined in such a way that this can be added without violating the current spec when bind functionality is defined. Defining the spec with a specific exclusion for this should be considered unacceptable. If we may it clear in the protocol that multiple membership is acceptable, but that multiple ownership is not, does that satisfy your requirements? Tim: Am I correct in assuming that this would be OK with you? Cheers, Geoff
Received on Monday, 20 November 2000 13:52:05 UTC