Date: Fri, 2 Apr 1999 12:51:11 -0500 Message-Id: <9904021751.AA26351@tantalum> From: "Geoffrey M. Clemm" <gclemm@tantalum.atria.com> To: sv@crystaliz.com Cc: ietf-dav-versioning@w3.org In-Reply-To: <012b01be7cb0$75514020$e6ea7392@honeybee> (sv@crystaliz.com) Subject: Re: Version issues From: "Sankar Virdhagriswaran" <sv@crystaliz.com> At one point during this thread Chris for the record stated his desire to manage the 'workspace' like information on the client to support scalability. I think this is an interesting question and requires further thought. My initial reaction was that the key trade-off here is between scalability (of the ilk that Chris is concerned about) and consistency. I would love to hear what others think about this issue. I believe that we are committed to a protocol that is compatible with both "workspace on the client" and "workspace on the server" (and it ain't easy to do so! :-). This was at the heart of the "what is a workspace" thread. I believe that the protocol now does work effectively in both of those scenarios, where a workspace with a read-only empty RSR provides the "lightweight" workspace needed for the "workspace on the client" scenario. Cheers, Geoff