- From: Geoffrey M. Clemm <geoffrey.clemm@rational.com>
- Date: Fri, 9 Feb 2001 23:07:04 -0500 (EST)
- To: ietf-dav-versioning@w3.org
From: "Lisa Dusseault" <lisa@xythos.com> > Just for interests sake, what would be the locking behavior of these > "shared" (a term I'd slightly prefer over "global" or "mutable") > properties? If you lock a version, does that prevent change to that > property on all versions? Are shared properties unaffected by write > locks? It depends how it's defined. I'd prefer for a "shared"/"global" property to exist on the VCR only, not on any of the versions. That's the easiest way for it to stay in synch across all versions. Then, clearly, a LOCK on the VCR would affect the property as well. A LOCK on a version would not. Sorry, my typo. I meant "If you lock a version-controlled resource, does that prevent change to that property on all other version-controlled resources for that same version history (you have stated that you want the property to appear on all version-controlled resources for that version history). Is the property on the version history itself? If not, does it disappear if you delete the last version-controlled resource for that version history (i.e. it's not there if you create a new version-controlled resource for that version history)? (What is the meaning of a LOCK on a version, anyway? If versions aren't mutable???) Versions have live properties which can be changed, and if the property is defined as being affected by a lock, a lock on that version would prevent changes to that live property. But as you say above, we should be talking about version-controlled resources, not versions. Important note: a "global" property is NOT the same as a "mutable" property. A mutable property may exist on versions, may have different values on different versions, and most importantly, may be changed on a version without causing a new version to be created. Fair enough. I shouldn't have confused the two. Note that the protocol allows the server to have both global/shared and mutable properties by just making them be live properties. But I agree that there is currently no interoperable way for a client to define that they want a property to be live, and no way to define the semantics of their live properties (e.g. affected by a lock, mutable, shared/global, etc.). But I believe that protocol for this should be done in a general WebDAV forum, since defining characteristics of live properties (kept on copy, single valued, whatever) is a general WebDAV issue, not a versioning-specific issue. > Note that although I'm always interested in exploring worthwhile > extensions to the WebDAV protocol, I'll point out that the versioning > protocol has received some criticism for being too complex because of > too many options, so I'm very reluctant to add even more options > before the IESG has a chance to review the ones that have already been > designed and reviewed. I strongly feel that BOTH mutable and global properties are required functionality for many source code and document versioning scenarios. So are ACL's and Searching, but that's happening in a general WebDAV context, because it involves general functionality that shouldn't be tailored just to versioning needs. Cheers, Geoff
Received on Friday, 9 February 2001 23:08:36 UTC