I think I understand now the disjoint thoughts on how the jcr-server should be designed. The existing jcr-server is a webdav view of a JCR repository, whereas it seems Brian is looking for a general webdav server that happens to use JCR as its storage interface. The reason for the confusion is because JCR and WebDAV do not share the same hierarchical models -- either one can emulate the other, but they are not the same.
What I suggest is that we start a separate contrib project for a general webdav server implemented using JCR and see how much of the code can be shared between the two, gradually merging them together until we reach a point where the two are (hopefully) distinguishable only by configuration files. What do you think of that as a plan? ....Roy
