The main-point is: To really do something useful with all the data we produce on the web daily, we need to have public APIs that allow to access old data.
In my opinion better than all Facebook of the world are things like HTML5, truly open standards that allow for simpler data-retrieval (even if its just web-scraping).
But if everyone (Twitter, FB, Tumblr, etc.) opened their APIs completely and released their data into the wild, who would profit then?
See sites like efreedom and other (basically) spammers: They make money with StackOverflow's contents.
Will Matthew's idea remain wishful thinking?
