It would not (entirely) surprise me to see a large portion of their business model coming from data-sets. Back at the end of 2008, an individual gathered every

It would not (entirely) surprise me to see a large portion of their business model coming from data-sets. Back at the end of 2008, an individual gathered every single Tweet that had been made to that point, simply by trying every possible Tweet-ID. He released the data-set, and promptly took it down, because the Twitter staff had contacted him. They apparently were in talks, and I never did see the end result. Here is the blog post on it: http://blog.infochimps.org/2008/12/29/massive-scrape-of-twitters-friend-graph/ http://blog.infochimps.org/2008/12/29/massive-scrape-of-twit... He collected 10 million Tweets, which was a side-effect when trying to gather a graph of user-following, and got permission from Twitter staff to release it, after which they changed their mind. Its rather disapointing, and would have been a great corpus to do work on. Either way, I think Twitter could develop a significant profit-stream by bringing in some stats staff, and running analysis on their end against the databases. If Twitter really does want to become "the heartbeat of the internet" or whatever they called it, that's the way to monetize it.