Some thoughts on Internet TV

Today two interesting articles about Internet TV appeared online. Both of them are about the announcement of a joint collaboration between Netflix and LG to manufacture a Netflix STB. The articles can be found here (EETimes) and here (NYTimes).

This confirms what I've always thought about TV: we will gradually assist to the migration of satellite and cable TV to the Internet as a transport layer. Video On Demand using the Internet is not something new. Back in 1998 I remember working on a university project focusing on VOD. I had a chance to visit an Italtel lab where they set up a VOD network for about 20 users, using ADSL and a Sun workstation as a server. The system was running very smoothly and providing all its users a terrific experience. It was really cool, but, at the time the technology wasn't ready and the cost of the bandwidth and processing power were prohibitive. Today, 10 years later I think we are ready for that. We have enough bandwith in our homes and the cost of the necessary technology has fallen dramatically. YouTube brought us video to the web and right now, many startups are trying to diversify the offer in some way.

P2P or Client/Server ?

Initially, when I tried the first version of Joost I was impressed. The quality is just ok (it's not HD) but being aware of the complexity required to design a p2p real-time video delivery network I think they did an awesome job. Differently from Joost and other p2p video companies, startups like Hulu or Netflix are using a client/server model (as far as I know) to deliver high-quality videos. Who will win? I don't know, but I believe that the sooner they will be able to bring their technology to a STB platform, the easier will be for them to widespread the product. To be honest, watching video from a PC is an alternative, but having access to a whole video archive directly from your LCD is definetly a much better user experience.