It occurs to me that some of you (anybody out there?) might be interested in this recent day-job article: I had a piece in the New York Times Magazine about Pandora, the Internet radio service. My interest was/is in their “Music Genome Project,” the engine that underlies what music you hear when you use the site.
On first listen, some things grab you for their off-kilter novelty. Like the story of a company that has hired a bunch of “musicologists,” who sit at computers and listen to songs, one at a time, rating them element by element, separating out what sometimes comes to hundreds of data points for a three-minute tune. The company, an Internet radio service called Pandora, is convinced that by pouring this information through a computer into an algorithm, it can guide you, the listener, to music that you like. The premise is that your favorite songs can be stripped to parts and reverse-engineered.
Anyway it’s a long piece, but if you’re interested, here it is.
A couple of years ago I did some research into the song "St. James Infirmary," wrote up what I found, emailed that essay to friends and posted it on my web site (as part of a series of "Letters From New Orleans," as I was living in that city at the time). Based on the feedback, I wrote a second version of the essay, and asked for more feedback. Based on that, I wrote a 
Just finished the article and loved it. I’ve always been intrigued by Pandora and its inner workings. And I knew the general gist of it, but until I read your article I hadn’t quite different their endeavor is from most other recommendation machines out there: They all use some flavor of data-mining algorithmic smarts to find likeness, things that are like what you’ve said you like. But from there, Pandora completely diverges from the rest. Pandora, is EXCLUSIVELY considering the content of the music with NO consideration of what other people think. (You like things that are like what you like.) The others are EXCLUSIVELY considering what others think with NO consideration of the content. (You like things that other people like you like.) Fascinating.
Hey there Slimbolala — Thanks for that, much obliged. Safe to see we’re interested in the same elements of the service. It’s interesting to explore the results. I wouldn’t want it to be my only option. But I like having it.