The only data I need—my bookmarks and tags—are already backed up in a JSON file. You can get this via the API. It would be good to have the crawled archives as well, but I don't really need it. A large number of my bookmarks are probably dead anyway, and could use pruning, so an archive of these is not that valuable.