BARBELITH underground
 

Subcultural engagement for the 21st Century...
Barbelith is a new kind of community (find out more)...
You can login or register.


Underground > Radio & Music >

Do you like whateveryone likes?

  
 
Our Lady Has Left the Building
13:07 / 12.02.06
From New Scientist, a study that suggests that people will like tracks more if other people like them, though this does not have any bearing on whether the song will be a hit or not.

The problem is it seems too difficult to judge what is meant by 'liking' a tune, and then translating that over to whether one likes it enough to part with cash for it.

Anyone else have an opinion?
 
 
grant
14:05 / 12.02.06
OK, here's the methodology:
Sociologists Matthew Salganik and colleagues at Columbia University in New York, US, recruited more than 14,000 people to visit a website with 48 songs by relatively unknown bands. People could listen to songs, rate them, and then decide whether to download them.

One group of participants saw only the names of songs and musical groups. Other participants also saw how many times a particular song had been downloaded by others. Both groups broadly agreed about which songs were good and which were bad.

But participants who could see how often a song had been downloaded tended to give higher ratings to songs that had been downloaded often, and were more likely to download those songs themselves. That created a snowball effect, catapulting a few songs to the top of the charts and leaving others languishing.

What they found, after dividing the "sees # of downloads" group into 8 sub-groups, was that the same songs didn't always wind up being most-downloaded, even though they were "rated" roughly the same early on. They're assuming this means that even though one song can be just as "good" (widely palatable) as another song, the whims of the marketplace -- the random nature of "buzz" and "hype" -- will take over at some point, substituting popularity for quality.

Cue sociologists making Britney/Tatu/Gwen Stefani jokes.
 
 
grant
13:32 / 14.02.06
There's an interesting footnote to this phenomenon in this New Yorker story on blogging.

About halfway down that page, the discussion turns to "power-law distributions."

Power laws are not limited to the Web; in fact, they’re common to many social systems. If you chart the world’s wealth, it forms a power-law curve: A tiny number of rich people possess most of the world’s capital, while almost everyone else has little or none. The employment of movie actors follows the curve, too, because a small group appears in dozens of films while the rest are chronically underemployed. The pattern even emerges in studies of sexual activity in urban areas: A small minority bed-hop, while the rest of us are mostly monogamous.

The power law is dominant because of a quirk of human behavior: When we are asked to decide among a dizzying array of options, we do not act like dispassionate decision-makers, weighing each option on its own merits. Movie producers pick stars who have already been employed by other producers. Investors give money to entrepreneurs who are already loaded with cash. Popularity breeds popularity.

“It’s not about moral failings or any sort of psychological thing. People aren’t lazy—they just base their decisions on what other people are doing,” Shirky says. “It’s just social physics. It’s like gravity, one of those forces.”

I think this is what the sociologists found in pop music.
 
 
the permuted man
13:51 / 16.02.06
I read that article too and I really wanted more information. What interests me is how many people think they don't listen to popular music.

The whole death of the hit, internet music sharing era, has everyone setting out to find undiscovered, under-represented artists. So often I hear people say "There's more music than just Britney and Gwen" or similar. It reminds me of Zizek's imagined other. It's like we all have this generalized idea of a popular music listener and we're sure we're not them, but if this is what everyone's thinking then this popular music listener doesn't exist.

I just heard about last.fm and signed up there -- loving it -- and it's really interesting to see all the musical data and think about how it came about. Like when I first got neighbors, I was amazed how even though they only had four or five bands on my list, their lists were remarkably similar. Of course, the first thing you want to do when you get neighbors is look for and listen to the bands they have that you haven't heard of, but I realized this would only make my own profile closer to theirs. Is this how they got to be so similar in the first place? Ok, now I'm rambling, but I do think there's a huge feedback effect on last.fm due to listening to other people's profiles building one's own profile and so on.

Anyway, back to the imagined O, I surf through people's musical journals and probably every eigth post is So-and-so sucks. Of course, the contents are basically: "Everyone likes them" or (the real critic) "They have no talent, Why does everyone like them?" Usually these journals are accompanied by profiles with 10 to 15 similar bands or the most popular bands of a different genre.

Now, I hate to jump to any conclusions, but I think it's popular to hate popular. And while this seems to run opposite to the study, I find in actuality it turns out the same. People run from this imagined popular to the same "undiscovered" gems as everyone else. But I'm not sure how that would come across in this type of study. If they were allowed to post comments or reviews it would be obvious, because you'd see stuff like "There are more bands than just X and Y, listen to Z!". If it was just numbers I wonder how it went. Something rockets up and then something else suddenly grows, then back to the first. Who knows. Like I started, too little information.
 
 
bergkamp clec
18:13 / 20.02.06
I think the reason the 'uncool to be popular' factor isn't working in this study is because while it stays as just the list of downloaders they're shown it is a bit like finding an undiscovered gem on last fm. Noone you know is talking about it and it isn't in the mainstream consciousness.

I think if you extended the study into the outside world over a period of time where these lucky bands had started to generate hype and became recognised as such you would start to see the figures drop off.
 
 
the permuted man
19:15 / 20.02.06
You make a really good point. If I find something online via databases, blogs, or even message boards, I still consider it as me discovering it. Whereas if I'm talking music with someone and they tell me about a group -- not that much different from a message board -- I feel like I'm late to the party. If a few people tell me about something, or I see it on TV, print magazine, on display in a store, I get to thinking it can't be that great: everyone's talking about it.
 
  
Add Your Reply