FWIW, this was written in July of 2010 (2+ years ago) -- CouchDB is in a very different place now than it was then. Reading the mailing lists of CouchDB, Redis

FWIW, this was written in July of 2010 (2+ years ago) -- CouchDB is in a very different place now than it was then. Reading the mailing lists of CouchDB, Redis, MongoDB and Cassandra are _very_ different experiences. CouchDB's list reads like 10 or so of the same people discussing very high level efforts like documentation and Windows builds, developing the DB at a glacial pace -- including merging in changes from all the spin-off CouchDB efforts that all seem to be defunct now (e.g. BigCouch and the sharding code). Tangentially, MongoDB/Redis/Cassandra mailing lists are NOTHING but "How do I..." questions, deployment questions, feature development questions, patch submissions, etc. (more-so Cassandra and MongoDB lists). CouchDB to me has found this life that feels very academic to me which I think is a good thing in the long-term for the project. The principles are in no rush to get to features and have the motto "slow and consistent wins the race". I would be surprised at all if a few years go by and then CouchDB gets rediscovered suddenly as the panacea to everything (something akin to how Jetty suddenly became hot business in the Java server world after being mostly ignored for 10 years) With the money behind Cassandra and Mongo it is probably not much of a surprise that there are much more new deployments going on and Redis has found a place somewhere between the two with what I would say is a Linus-like steward at the helm (props to Salvatorefor being everything that is right with open-source) I wouldn't build a commercial product on CouchDB tomorrow, but I am eagerly waiting to see where it goes in the next year. It is wonderfully designed, but I'd like to see some of the nagging "table stakes" issues like replication failures fixed before caring about Feature XYZ and release 2.0