Hi, At some point we need to discuss where to start with Jackrabbit 3. I guess the most important deliverable is an embedded (not server; that is, no remoting) JCR 2.0 repository that uses a database as the storage backend. This would be 'one project' as I wrote in the subject (and one jar file).
The question is how we want to re-use existing Jackrabbit 2 modules (or generally "code") within Jackrabbit 3. Personally, at the beginning I would favor "copy code when needed", because only this approach allows to really refactor the code without risking to break Jackrabbit 2. I think we should continue to develop and change Jackrabbit 2 as if Jackrabbit 3 doesn't exist, at least at the beginning (things may be different later on). Regards, Thomas
