How Songs Get Popular

An anonymous reader writes "Researchers created an artificial music market of 14,341 participants split into two groups to pick music from unknown musicians. In one group, the individuals had only song titles and band names to go on. The individuals in the other group saw how others had rated the songs. Turns out popularity bred popularity, which explains why there's so much crap on the radio."

Re:the variable that was changed

[mattmacf]

Not quite. From TFA:

The social-influence group was further divided into eight separate, non-interactive "worlds." Members of each world could not see the decisions of the other seven. The idea behind this was to observe multiple outcomes for the same songs and bands.

"If you look at Britney Spears, some people say she is really good. Others say she isn't good, she's just lucky," Salganik told LiveScience. "But by having just one argument, it's impossible to distinguish. However, if you have 10 worlds, and she's popular in all 10, then you can say she's actually good. But if she's only good in one, then you could say it was due to luck."

Although different songs were hits in each world, popularity was still the deciding factor, although the "best" songs never did very badly and the "worst" songs never did very well.

What you missed is the fact that "group B" was in fact subdivided into eight distinct, independent sub-groups. Rather than determining "WHETHER OR NOT the ratings were actually true" (Who is to decide whether a song is good? Critics? Fans? Other bands?) what the researchers did was take the same independent ratings (from group A) and give them to each subset of group B. It's not surprising that the "best" songs generally did well and the "worst" ones generally did poorly. What is notable is that different songs were hits in each "world," based (presumably) on the same set of independent data.

Re:Would the Beatles have made it today?

[Drishmung]

George Starostin describes his introduction to the Beatles [rinet.ru] somewhat differently. As in someone really not exposed to them who now has very definite views.

Summary is misleading - gotta RTFA!!

[porcupine8]

As I've had to point out to a couple people already, the summary is misleading. It makes it sound like highly-rated songs became popular. This was not the case. Songs that were downloaded more often kept getting downloaded more often and became popular - regardless of whether or not they were highly rated!

The whole point is that the ratings (ie, quality) of the songs had little or nothing to do with their popularity - low-rated songs became popular as often as highly-rated songs! And in different test groups (there were 10), different songs became popular, still independent of ratings.

Re:Just like /.

[zlogic]

When I get mod points, I usually try to mod up unmoderated posts. There is often really insightful/interesting stuff hidden under (Score:1) and at the bottom of the page, which most moderators seem to ignore. Modding a (Score:4, Insightful) with lots of replies further up doesn't make sense to me. It's already perfectly visible, and that's the whole point of moderation: to identify good stuff and make it more visible that uninteresting stuff.

Re:A key to music is the familiar.

[soliptic]

Not quite.

The key to good music the balance between the familiar and the surprising.

What is the soloist doing when he attempts to "build"? Actually the ideal process hardly ever takes place--that is, it is hardly ever the case that a conscientious soloist plays a thinking solo for a hard-listening hearer--but when this does happen, the key process is memory. The soloist has to establish for the listener what the important POINT, the motif if you like, is, and then show as much as he can of what it is that he sees in that motif, extending the relationships of it to the basic while never giving the feeling that he has forgotten it. In other words, I believe that it should be a basic principle to use repetition, rather than variety--but not too much. The listener is constatnly making predictions; actual infinitesimal predictions as to whether the next event will be a repetition of something, or something different. The player is constantly either confimring or denying these predictions in the listener's mind. As nearly as we can tell (Kraehenbuehl at Yale and I), the listener must come out right about 50% of the time--if he is too successful in predicting, he will be bored; if he is too unsuccessful, he will give up and call the music "disoganized."

Thus if the player starts a repetitive pattern, the listener's attention drops away as soon as he has successfully predicted that it is going to continue. Then, if the thing keeps going, the attention curve comes back up, and the listener becomes interested in just how long the pattern is going to continue. Similarly, if the player never repeats anything, no matter how tremendous an imagnation he has, the listener will decide that the game is not worth playing, that he is not going to be able to make any predections right, and also stops litening. Too much difference is sameness: boring. Too much sameness is boring--but also different once in a while.

-Richmond Browne