Is Justin Timberlake a Product of Cumulative Advantage?

[Guybrush]

A lot of sound points made, and I partly retract what I wrote about quality being redundant.

Fundamentally the music has to be strong, or at least interesting and in touch or expressive of something close or true to the listeners, even if it might appear superficial to others. The mass record buying public aren't dumb, at least not when it comes to the higher-order megastars (Prince, Michael Jackson, Madonna, Timberlake - who I think is as close as you can get to those guys these days - and so on).

I agree here, but what is interesting then is how strong and interesting a piece of music has to be to gain popularity under all circumstances, to be omnipotent, if you will. And conversely, how bad is ‘too bad’: how mediocre is successful music allowed to be? (That is, mediocre as judged by the common man in isolation, not knowing what others think.)

Because the long-run success of a song depends so sensitively on the decisions of a few early-arriving individuals, whose choices are subsequently amplified and eventually locked in by the cumulative-advantage process, and because the particular individuals who play this important role are chosen randomly and may make different decisions from one moment to the next, the resulting unpredictably is inherent to the nature of the market. It cannot be eliminated either by accumulating more information — about people or songs — or by developing fancier prediction algorithms, any more than you can repeatedly roll sixes no matter how carefully you try to throw the die.

This passage in the original text is a bit iffy. You could argue that on-line magazines and bloggers fill this pioneering role today—at least for some scenes—and that quality therefore is important, as these, ideally D), make their recommendations not knowing what others think of the music. (Let’s assume it’s an out-of-nowhere debut album.) (Bloggers are a terrible example, actually. Well, well.) Looking at it this way, you could argue that the Christina Aguilera album tanked because the early valuers thought it lacked quality, and spread the bad word; while the opposite happened to the Justin Timberlake album.

 

[borderpolice]

its a very well written and rounded article. its also good to hear the empirical evidence to back the ideas. but surely the arguments are not particularly groundbreaking?

This is not groundbreaking. modern sociology has understood the strong mimetic component in human behaviour for over half a century now. What's neat about this research is that it provides very clearcut empirical research, and a neat mathematical model.

 

[blunt]

I finally got around to reading this article yesterday, and whilst it's pretty interesting and feels 'true' in a common-sense-kinda-way, I can't help finding the approach they've taken to have some serious flaws:

- they use the popularity of downloads in the initial group as a measure of 'quality', and yet don't appear to have set up parallel listening groups to check for differences in the listening habits from one sample to the next (as they did in the second group)

- in the second group, how many people listened to stuff not because it was popular, but because it's at the top of the page? How many people were just too lazy to scroll down the list?

- what does it take to increment a tune's play count? Is it counted on the initial click? Or at the end of the tune? They don't say. Just because someone started to play a tune doesn't mean they liked it when they actually heard it.

I'm instinctively drawn to their final conclusion that predictions should be treated with scepitcism, and that if you played the history reel back twice you'd get different outcomes; but at the same time I'm suspicious of it as a piece of empirical research.

 

[mos dan]

At the risk of boringly belaboring the obvious, I think it's important not to think of this idea in relation only to generally ubiquitous (pop) culture phenomena such as JT, but also locally ubiquitous stuff as well.

definitely. a hypothetical on a very local basis:

Trim and Wretch's mixtapes are released. Lots of people on Dissensus like the former, and say so. No-one is that interested in the Wretch mixtape by comparison. The hype leads me to buy one and not the other (for I am not rich), I forget about Wretch and enjoy Trim's CD (remembering the good things that have been said about him as I do so), I write an enthusiastic blog post about Trim, dot-alt.com's readers become more likely to go out and buy Trim; soon it becomes increasingly 'known' in grime circles that Trim = good.

(all this might have happened if ukrecordshop had sent the damn CD to me)

But then this is all pretty self-evident. If anything isn't cumulative advantage MORE likely to carry weight in small circles, where 'taste-making' is at such a high premium, and you have to go on hearsay more often. Srsly at FWD>> the conversation is always "ooh have you heard Skream's new track?" "omg did you catch those new Coki dubs on N-Type's show the other day??". You invariably have a half-formed opinion of a song before you've heard it...

 

[turtles]

It just occurred to me that cumulative advantage is essentially also how google works. The more pages link to a given page, the higher up on google search results that page will appear, meaning it will be taken as the definitive page on the given topic, so more pages will link to it, Basically this makes google an accelerant for the whole cumulative advantage thing, i guess.

 

[Freakaholic]

It seems that technology and the present state of music seem to be pushing cumulative accumulation toward more and more influence, especially subconsciously. While a majority of people dont go out and try to discover a new genre, say freak-folk for example, they are constantly indundated with what their peers, local radio stations, and personal set of influencing websites are listening to, like it or not. But can it be said that the more inputs a person has, the more influential these inputs will be?

For example, when I go to buy a new batch of records, say dubstep, i first go with the artists and labels i know and trust: Skream, Tempa, Kode9, Dub Police, etc. As the market gets bigger and bigger, and as the pool of artists and labels get larger and larger, i will miss out on more and more simply because there is too much to keep tabs of. So, while the interweb, blogs, online retailers, itunes, and others can diminish my local radio's influence, it can create a deluge of new material that is too daunting to tackle. I used to be able to name off the top of my head every dubstep label out there. Now I doubt if, while writing a list, could name all the ones in my crate, let alone all those that are out there that I have not sampled.

This, to me, seems like a passive form of CA influence. Its not so much actively aiding my decisions and tastes, but passively filtering out what i focus on. There is so much music out there, and so many people talking about it, that my filters are overloaded. Perhaps it will come full circle, and they will just burn out, and i will cease to be affected by the public at all.