Today's podcast is a twenty-minute interview with Arthur van Hoff about the newly-released TiVo SDK. A founding father of Java, van Hoff became Marimba's CTO and chief scientist and then, in 2002, launched a small startup called Strangeberry. When TiVo acquired Strangeberry last year, it described the company thusly:
A small Palo Alto based technology company specializing in using home network and broadband technologies to create new entertainment experiences on television. [TiVo SEC filing]I'd been traveling when the SDK launched, and wasn't able try it until after this interview was over, but now that I've run through the samples I'm on the horns of a dilemma. It's tempting to jump in and start writing TiVo HME (home media engine) apps that turn the TV into an information kiosk. Projecting electronic calendars on the TV, for example, might be a nice way to augment the paper calendar hanging on our refrigerator. On the other hand, I'm inclined to wait until the SDK opens up more of TiVo's guts: scheduling, the media repository, video playback.
Some pullquotes from the interview:
On Rendezvous: The TiVo uses Rendezvous to discover applications on the local network.
On home control: I have an X10 home lighting system, and I've written a Java application that runs on a Linux server in my closet that the TiVo discovers. So I can now control all the lights in my house, and turn on the fountain in the back yard, and stuff like that.
On partner apps: In that case the app isn't going to run on your PC, it's going to run somewhere in the network. There are obvious applications in that space: shopping, movie ticket buying, things that make sense with your home entertainment world.
Towards the end of the interview I ask whether the SDK will enable me to write apps that people can use to integrate TV into the blogosphere. For example, yesterday's entry about John Abramson's book was originally motivated by his appearance on CSPAN's BookTV. I quoted from the book and would also have liked to quote video snippets from the show. But although we can quite handily deconstruct and reformulate print media, we can't do the same with audio or video. Of course there are tremendous technical and legal hurdles. On the technical side, it's hard to create a realm of shared discourse when content lives partly in the cloud and partly in local cache, with no unifying addressing scheme. On the legal side, we've yet to clear the way for fair-use quotation of rich media.
Given these limitations and the constraints currently imposed on the first TiVo SDK, I'm not yet seeing a huge opportunity to make TV into the kind of two-way experience that the Web has become. However, van Hoff does suggest that TiVo's tagging technology could be used to associate metadata with the shows we watch, and then export that metadata to the blogosphere. I'd definitely like to try doing that.
Former URL: http://weblog.infoworld.com/udell/2005/02/08.html#a1170