3 ms·

This kind of thing is just laughably stupid. Psychology is basically still a guessing game and claiming your iPod most played list gives an insight into a perso

This kind of thing is just laughably stupid. Psychology is basically still a guessing game and claiming your iPod most played list gives an insight into a person is likely valid and mostly useless. For example, when someone's asked to recall a memory they generally look up and to the right. When they lie they look down and to the left. However, if the person's left handed then it's wildcard because their cognitive functions are splayed between the two hemispheres. Even then this isn't even 80% accurate with right handed people. Just looking at someones iPod doesn't tell you what they actually listen to. My iPod has songs in my most played that I've never played before because my wife uses my iPod when hers runs out of battery. Just because a woman's playing soft jazz on an iPod where the top 25 songs are death metal is likely to mean she doesn't try to have sex with someone everyday of the week. I mean if she plays soft jazz when she invites you back to her place, it not being in the top 25 means she doesn't invite people back very often more likely meaning you won't get herpes. I'm a big fan of California Punk Rock, but it's one of the lowest scoring in number of plays. My father-in-law loves the Rolling Stones, yet I've only heard him play 1 song by them. I'd hedge a bet that what's on someones iPod that they don't play says more about them than what they do play.

derefr 18y ago
Well, it has sensible elements. The Netflix contest algorithms, for example, aren't helped at all by throwing in demographic data about the users or their tastes, because that information can apparently be "derived" from their movie ratings. Assuming you only own music you like (and throwing in your star ratings for good measure), I bet Apple's iTunes Genius department could tell you just as much about yourself. A single playlist probably won't tell you much, though, unless the library it was built from contains remarkably little variety, such that it can reliably be expressed with 25 samplings.

derefr 18y ago

Well, it has sensible elements. The Netflix contest algorithms, for example, aren't helped at all by throwing in demographic data about the users or their tastes, because that information can apparently be "derived" from their movie ratings. Assuming you only own music you like (and throwing in your star ratings for good measure), I bet Apple's iTunes Genius department could tell you just as much about yourself. A single playlist probably won't tell you much, though, unless the library it was built from contains remarkably little variety, such that it can reliably be expressed with 25 samplings.