yea, its more than 140 bytes! the average "tweet" object is about 3k, and includes a lot of meta data about the user, the tweet, location, etc... Its direct fro

yea, its more than 140 bytes! the average "tweet" object is about 3k, and includes a lot of meta data about the user, the tweet, location, etc... Its direct from the Twitter stream. think there is a bit of overhead of actually parsing the JSON file into a CLR object... since RavenDB actually knows about JSON it may be possible to just give it the data, and see what happens...