This isn't really a hacker/non-hacker issue, its just an issue of having vision, and having good people with good ideas. A company that relies solely on its fou

This isn't really a hacker/non-hacker issue, its just an issue of having vision, and having good people with good ideas. A company that relies solely on its founder for innovation is not a healthy company. I don't have much love for third party widgets. Their inclusion on Facebook has cluttered up the service. For people who are using websites as tools to accomplish tasks, I don't see 3rd party widgets as a compelling feature. Like Facebook, LinkedIn is also a social network, but it differs as it is fundamentally a tool rather than a toy. So far widgets have been toys, some of them excellent toys, but toys none the less. Its a hell of a genie's bottle to open up in exchange for apps like TopFriends, iLike, and the option to draw a mustache on your friend's profile picture. I realize that the creation of tool-widgets is possible, but I am still waiting to see if the effort of developing a public API is worth the return in tool functionality (LinkedIn will probably survive without the inclusion of TopConnections, and ConnectionsIdTotallySleepWith apps ;). To address a few other points, Friendster failed to get on the boat because it had immense technical problems leading to its stability/usability plummeting like Wall Street trader jumping out of a window on black Friday. They probably would have also missed the boat on developing an API, but this is what happens at poorly lead companies: they fail catastrophically. My last thought is that after a company becomes sufficiently large, having a hacker leader who wants to build the public API by himself may be less of an asset than a disaster. Facebook released their API quickly, and they released it incomplete and with shoddy support and documentation. They've happened to get a blankcheck for good publicity... but others may not be so fortunate.