Pandora

So I vaguely remember using Pandora in college, but I hadn't taken a look at it in awhile. It is awesome. And impressive/probably beyond my comprehension.

According to the creators, here's how it works:

"Together we set out to capture the essence of music at the most fundamental level. We ended up assembling literally hundreds of musical attributes or "genes" into a very large Music Genome. Taken together these genes capture the unique and magical musical identity of a song - everything from melody, harmony and rhythm, to instrumentation, orchestration, arrangement, lyrics, and of course the rich world of singing and vocal harmony. It's not about what a band looks like, or what genre they supposedly belong to, or about who buys their records - it's about what each individual song sounds like."

Has anyone done this with books? I've used Novelist and What Do I Read Next? (or whatever it's called), and I always assumed there were some book monkeys sitting in a room and manually putting together lists of read alikes, which seems much less effective than what Pandora does. (Maybe I'm wrong. Who knows. That happens from time to time.)

But if I'm right, why not do what Pandora does? Why not take the literary attributes that "capture the unique and magical [literary] identity" of a book: writing style, pacing, character development, plot structure, language, etc. then let a computer magic up a list of which titles/authors are similar to each other?

It would be a big but awesome undertaking. Or maybe some sites or databases out there already work this way, and they've just tricked me with the way they appear to work.

In any case, thanks for taking me back to Pandora, Web 2.0 committee. In the twenty minutes I've been playing and blogging, I've already found a couple new bands that are making me smile.