Computers have infiltrated almost every aspect of modern life. Not only do they serve as powerful tools for computation and social networking, but now allegedly contain insight into a very human characteristic: musical taste.
According to a recent article in The New York Times Magazine, the rise of digital algorithms, or computer-run formulas, led to the creation of programs that supposedly can calculate the type of music one prefers.
Popular programs include iTunes Genius and Pandora.
Advertisements for Genius claim that the program can "make a playlist of songs from your library that go great together," according to the Apple website. The Genius Sidebar also suggests songs not yet purchased that match one's general taste in music.
According to Boston University computer science professor Wayne Snyder, the process of matching similar songs isn't quite as formulaic as one might think.
"Humans respond to music in complex ways, psychologically, emotionally and culturally," Snyder said.
"Computers are currently not able to analyze anything but the simplest surface characteristics, such as the tempo and the tonal characteristics of the sound file," he said. "You'd do about as well by simply predicting that, say, John likes songs that last between 180 and 200 seconds."
MY TASTE IS YOUR TASTE
How then do these programs cope with the complexities of music? Simply put, they rely on other people's taste.
The Genius feature on iTunes is basically a database for everyone who allows Genius to look at their music libraries, Snyder said.
Genius compiles vast amounts of data on every person that turns on the feature, he said. The program can then determine statistical patterns from the data, he said.
The computer program essentially pulls information from people who have similar tastes, giving Genius users song choices from those people's libraries, he said.
In this light, Snyder said, the Genius feature becomes more of a social sharing between like-minded people.
People with similar tastes tend to like each other's songs, especially when large statistical samples are taken, he said.
While Genius' recommendations are largely accurate, there are inevitable limitations to the program, he said.
"It's like if you walk into a CD store and randomly pick up a ska CD, and you find out that you really like it," he said. "That's just something you lose if you use this program."
MAY I RECOMMEND...
Some BU students said they use Genius' recommendations as a reliable tool to increase their music libraries.
College of Engineering freshman Paul Perry said he uses Genius while searching for interesting new music.
"I use it as a starting point when I'm looking for new music because if I have no idea about that type of music, it lets me go through and see what it's like," he said.
Other students say they think Genius works fairly well at picking out songs that they like, with a few hiccups now and again.
"It's usually pretty accurate," said Erik Lindgren, a College of Art and Sciences freshman. "It will match certain elements of a song but not all, and sometimes I get some that I have no idea about," he said.
CAS freshman Allison Lorenc said she uses Genius to find similar-sounding songs.
"I usually will find a song that I like, or that a friend recommends to me, and then I'll use Genius to see if there's anything like [it]," she said.
CAS freshman Daniel Nayebi said he feels Genius opens up new possibilities in terms of song preferences.
"It kind of acts as a way to get me to see music that I wouldn't have thought of on my own," he said.
This is an account occasionally used by the Daily Free Press editors to post archived posts from previous iterations of the site or otherwise for special circumstance publications. See authorship info on the byline at the top of the page.
