Hi, (Forgive me if this has been discussed at length already and my googling did not pick it up).
I'm interested in how Couch could be used to store temporal data. I guess this question boils down to whether you can turn it into a time machine. As I see it, you could approach the problem by: 1. Making regular snapshots (e.g. daily) and just compare 2 points in time; 2. Implement some kind of version control as part of the document metadata; Option 1 seems to be related this remark about piggybacking off ZFS snapshots (http://stackoverflow.com/questions/121599/couchdb-backups-and-cloneing-the-database/395392#395392). I was wondering wether you get Couch to navigate to an arbitrary snapshot that you could load from the filesystem. Option 2 is more fine grained and allows different semantics, but I don't know how scalable this would be - I don't know if you could just naively store historical deltas within a document. Has anybody had any experience with this all? Thx, Ben
