table ids

Hi all, Is there any particular reason the ID columns for xwikidoc are generated (hashes of doc name) instead of just a sequence? Its causing us a little grief when we want to do things like rename a document - copying the doc seems like a hack, and breaks in subtle ways. I think things would be much cleaner if we could assume an inviolate id for each document, and would also have the benefit of making manual sql maintenance task a lot simpler. Let me know what you think, Matt