I’ve written before about what a revelation Pandora Radio has been for me. Following, in no particular order, some capsule reviews of new bands I’ve discovered and old bands I’ve rediscovered through this resource.
Back in February I experimented with Pandora Radio and loved it…enough that I bought a subscription within a few days. It’s my background music now; I might never own an analog radio again.
For a while I ran around telling all my friends about how Pandora was the greatest thing! since sliced bread! you should try it! But I’ve stopped doing that, because I’ve learned that it doesn’t work as well for other people…starting with my wife. I think I know why, and it reveals an interesting failure mode of all such systems.
I’ve been meaning to experiment with Pandora Radio for a while, and finally got to it today. It’s based on something called the “Music Genome Project” that categorizes music by how it expresses a large number of “genes” — traits that describe features like song structure, instrumentation, genre influences, and so forth. According to Wikipedia these are used to construct a vector, and similarity between tunes is measured by a simple distance function. You give Pandora a seed artist; it then apparently random-walks you through tunes and artists similar to that artist’s style.