Joost Exec VP David Clark recently gave an interview to CBS TV showing off Joost. He started by describing the quality of the picture, and the universe of channels, and then headed off (at the end) into the hinterland of interactive TV by showing a bunch of features that would allow you to get more information about a television show, and interact with other users. Hmmm, said the skeptical interviewer, how is this different from the webTV efforts of the past. Clark's answer was, essentially, "I don't know why they didn't succeed but ours is really really cool".
There have been huge investments made in the past by various companies. At Microsoft we built a very elegant framework for adding interactive content to television shows, and pioneered a whole bunch of interactive shows with various studios like NBC. We built delivery mechanisms as well that allowed that content to be delivered by satellite, or even as a slow speed forward-error-corrected bit stream in the unseen portions of a broadcast signal. We also acquired WebTV for over $400 million to bring internet content directly to television. These efforts failed because they really didn't understand how users watched TV. By focusing on bringing PC like features to the big screen, they misunderstood that for most television viewers, TV is a passive experience — a "10 foot" versus "2 foot", or "lean back" vs "lean forward" experience. Most TV users wanted a better picture (bigger and clearer), and more shows to watch.
Since the late 90's, many more people watch television on PC's, internet based social networks have entered the mainstream, and streaming video is now mainstream. Interactive TV's time may be here finally, and Joost may be the company that finally succeeds with it. However, the road is littered with the corpses of those who've tried and failed along the way.