ffwd - Long tail recommendation and discovery for video
There's
an ever growing number of videos on the web. YouTube seems to
have an endless supply of everything from professional documentaries to
phonecam captures of the latest campus tasing. Despite this video
glut, finding interesting videos is not so easy. My current
approach is to go to YouTube or Digg and just watch the most popular
videos of the day. It's not too satisfying. A video of a 30
second car crash has a much better chance of making these charts
than a thoughtful, well-crafted film. Clearly tools for helping us
explore this long tail of video will be increasingly important.
ffwd is new service (still in private beta), that is hoping to help you find discover web video. I've taken it for a quick spin. They have a pretty nifty enrollment process - where you click on shows that you like so they can get an idea of your taste. Once you've selected your shows they assign you a video personality based upon your selections (I'm a 'comedy writer', apparently I like comedies). Once you are enrolled you can start to discover videos.
The video discovery is labeled as 'alpha' and they aren't kidding. The recommendations don't seem to be related to what I like at all. They offered reality shows (The Apprentice, Extreme Makeover Home Edition), weight loss success stories from San Antonio, a 'sexy webcam girl', nothing that I was interested in. The videos are coming from video sharing sites like iFilm and YouTube. Since they are in alpha, we can forgive them their recommendations, especially since they do seem to be looking for the right type of people to build their recommender. They have a job posting for an AI expert with skills like:
- computational linguistics
- collaborative filtering
- behavioral profiling
- relationship mapping
- semantic clustering
- symbolic systems
- Bayesian statistics
- feedback analysis
- personalized search
That's a pretty good sign. I think ffwd is targeting a real void - video discovery on the web - it will be interesting to see if they can do better than what we see now on Digg or YouTube. Oh yeah, The ffwd blog is pretty interesting too.
BTW, this service is founded by Patrick Koppula who was one of the founders of Garageband/iLike.
Posted by Toby on February 07, 2008 at 11:22 AM EST #
Sounds really interesting. MovieLens, LibraryThing, etc. have been around for a while, but even though in a way they are 'Web 3.0' services (peer recommendations), in another way they are still stuck in the past, when it comes to integration: sure, you get access to the IMDb or Amazon page for the item, but that's the extent of integration offered.
ffwd promises that -- the video is available in the same interface, so you can quickly act on a recommendation. There's still a problem, in that the service still relies on network effect in what is nearly a zero-sum game: more users here means less users for other peer recommendation services. Some people would be using multiple services, which means their data likely won't be up to date in all, which means the services will offer suboptimal recommendations (so it's bad for all concerned).
Take the new Data Portability initiative into account, though, and a market in recommendation services could soon spring up. Use various services, and give them authorization to pull your rating data from each other. The user gets to experiment with which site has the algorithms and user base to provide good ratings in which situations, and the services get to compete on how good they do their jobs, not merely how well they can gobble up and lock their market down (think Wintel).
--
Michel
Posted by Michel S. on February 11, 2008 at 11:58 AM EST #