Beyond Napster, a Free Culture

Top-down cool. A huge glob of capital hovers on the 43rd floor of a skyscraper somewhere in America, telling you and me what we should think is hip this week. Welcome to the whole damn world. Fortunately or otherwise, our cool-chasing is built into our genes, and it can exist with or without money driving it. Maybe the grassroots can't destroy our money-driven musical culture, but I bet there's a way a natural culture can thrive alongside the one we're force-fed. (This is the last of three features on pop music; you may want to read part one and part two for background.)

The capital doesn't have any opinions, of course. It just perpetuates itself. It's potential energy; when it's spent on the mass media, it drags youth culture in its wake, and through some miracle perpetual motion machine, that energy is recaptured in the sale of T-shirts, CDs, movie tickets, Gap vests, makeup, and shoes.

There's an indefinable quality about this kind of "cool" I'm talking about. I can't really put my finger on what I mean, but the closest I can get is: looking up to someone because you perceive their opinions regarding dynamic cultural subjects you care about to have value.

If your peer group looks anything like mine, recognizing the importance of, say, RSS a couple of years ago would have set you apart as someone whose vision could be trusted by your peers. And local peer group is all that matters for cool.

On the other hand, if your local environment buys into the top-down idea of what cool is, bands-and-brands, you're kind of screwed if you invest your mental effort in evaluating networking hardware. Ninety percent of high school is realizing that catching up with the global cool is something you don't have the resources for, and instead finding a local cool that works for you. You heard it here first.

So how does someone become, in this narrow definition, cool? The key is that it involves dynamic culture. Trends come and go, and come back again. Being with the trend, or (better) just slightly ahead of it, scores big. I'm guessing the cool people who first wore flared pants a few years ago, two weeks before we suddenly all decided it was mainstream, are now watched by their peers to see what other trends they're going to predict.

On the other hand, in 1992, wearing bell bottoms (I'm sorry, it's the same damn thing) would have had the opposite effect on your social status. Fads rise and fall, and we need to buy low and sell high. I plead guilty to still liking Fatboy Slim, which I'm sure would have been very chic three years ago -- now it means I have no taste, apparently.

Corporations spend millions on getting in and out at the right times. They speculate on the meme market, buying a trend small and selling it back to us for real dollars when it's big. MTV is day-trading our culture.

Did you think you were the one who discovered (fill in the name of something you thought was cool)? Chances are, some executive paid for research three months prior, and upped the hip of some TV show or commercial by letting you see it. Sorry, but your discoveries have all filtered through money.

I'd really like to denigrate the cool-chasing impulse, and that's easy to do when it's a driving force in someone else's peer group or the characters on Daria (episode 505, "The Story of D"). But it's part of being human. We all like to play this game a little, some of us a little too much maybe.

It does affect us all our lives, starting when we're little kids "acting out" to draw focus and attention (few things are less cool than being ignored). It ends shortly after we arrange to have our ashes shot onto Mars, thereby becoming more cool after we're dead.

And it runs parallel to economic considerations. Money is a good way to motivate someone, and it has the advantage of working on complete strangers. But the peer-group drive was motivating human beings to spot trends tens of thousands of years before money was invented.

("People are saying the tribe over the hill is doing throwing spears this year. We should make some throwing spears too." Our vacuous impulse toward fashion and fitting-in is probably a spinoff from a survival instinct. I want to align with the one who can find roots and berries; that's pretty damn cool if my tribe is on the brink of starvation, and we'll all trust the root-finder a lot more when we're eating better. Berries, Spice Girls, same thing.)

The trust system works for music. I don't browse record stores anymore; they're 99% junk (to my ear) and I don't have time or energy to sift through it all. Instead I ask friends what they're into. Which friends? Cool friends.

How do I know they're cool? My brain keeps a ledger. One of the interesting ideas that sociobiology brings us, as it struggles to shed its ugly reputation from the 1970s, is that human beings are hardwired with the capability to keep track of about 150 other human beings. Perhaps that's the size of a typical village on the African savannah, 50,000 years ago.

The evidence is pretty anecdotal, but each person's internal map of pecking orders and trust networks seems to grow not much beyond that size. You and I can track coolness factors for about 150 of our closest friends, no more.

But a computer can track more.

What we need is a system that can store musical (and other cultural) recommendations for 150 million of our closest friends.

Napster doesn't address this at all; though some have found new music through it, I sure haven't. The few times I've tried to chat with people sending me something from an artist I already know I like, nothing's come of it. It's been great at finding junk I already knew about from hearing on the radio, though.

But: recommendation systems. Basically the idea is to accumulate preference vectors from a large number of contributors, and then provide a way to ask, "if I like A and B, and dislike C, what else might I like?"

There have been a handful of master's theses and dissertations written on the subject. There are academic projects like GroupLens, but, like GroupLens, they're all 3-4 years old and there's no source.

The best-known recommendation system was Firefly, which actually worked pretty well. My initial experience was telling it a few musical artists I liked, and seeing it start spitting back to me other artists I already knew I liked. "You enjoy Peter Gabriel and Tori Amos? Have you heard Kate Bush?" Firefly was swallowed by Microsoft and is now apparently part of Passport (which I don't use, so I don't know what it looks like now).

But what's needed to leverage cool-tracking into a free (speech and beer) culture is an open system that will integrate with existing communities on the internet.

I've gotten to recognize a number of Slashdot users' nicknames on sight, and it might be interesting to see what you-all are listening to. My latest example is an Apocalyptica CD I bought last week based on a reader's recommendation in a comment (it hasn't arrived yet, so I can't say if Barnes & Noble's 24kbps preview did it justice). I'd click through to user pages a lot more often if I knew you-all were sharing musical preferences.

Note that none of this is necessarily tied to music trading. It might be nice if my user page listing the fact that I like Sarge had the music itself, or a preview, a click away. But the important thing is the recommendation, not the trading, and to build a network uninfluenced by money, the distribution system has to be separable. If Barnes & Noble is the only supplier, and has any control over what recommendations I get, I won't be able to trust them.

So the software really should have a distributed database of people's recommendations, one that's not ownable by any one entity. Ideally, the users of Slashdot, Kuro5hin, and Ain't It Cool News should all be able to drop their cultural likes and dislikes into the same database (hierarchical namespace based on domain name?). Queries drawing on users with taste like yours should be able to come just from one community, or from the entire database, your choice.

And -- the important point, the key to it all -- the database needs to recognize who was on top of which cool when. If I drop in Jonatha Brooke in February, and then 10,000 other people discover her in June, I need credit for being on the rising curve of the trend.

I don't know if I need a little star by my name or anything (I'm far too modest for that, oh stop, no really) but my positive rating of that artist or album should count for more if I got in early. Time is crucial in measuring trends.

And following the directed graph in the other direction, if I tell the database that I like these six CDs, it needs to be able to tell me the top ten users who plugged in their recommendations for those same CDs first -- on the theory that, if they led the curve last year, their opinions for this year will be of interest to me (and 10,000 other people).

Corporate cool-chasers pay good money on research, every day, to find this stuff out. In a voluntary contributory system, everyone could have access to the same information, for free. It would undercut corporate cool the same way GPL'd software undercuts Microsoft: when it's free, nobody will pay for it.

Writing code that can be integrated into existing internet communities is key as well. Firefly failed because nobody went there. Everything from the database to the web interface was proprietary, and its information was only on one website. But allow Slashdot to tie into your system, and you instantly add hundreds of thousands of potential users.

There's no money to be made in this, or not much compared to the $150 billion that the big six make every year by selling us trends.

But if you're a programmer looking for a way to influence musical genre for years to come -- or rather, to remove the influence of the glob of capital and allow a natural culture to flourish on its own -- this is a great way to start.

Music distribution and economics have historically been the two major influences on the evolution of its style. The internet has reached a point where that doesn't need to be true. Culture can be abstracted from economics, style from money. There are 10,000 singers and musicians working day jobs right now who don't care if they ever make a million dollars. They just want to be heard, and while the internet lets them reach a billion people, there's no way for word to spread. They might as well not exist. They might as well go back to selling 3 CDs at a time at gigs.

If that can't change, it's a damn shame. The alternative is another century of small-time musicians giving up in disgust while a wad of cash stuffs our ears with the Abercrombie and Fitch song.

The entrenched money is here for good. We'll have the RIAA and its prepackaged bands until copyright law changes drastically. We'll have (quasi-)payola in major radio markets until the radio itself goes away. We'll never get rid of the top-down cool because there's just too much money in it.

But we can have a market of our own, not based on money. We can leverage the best part of the internet, its communities, to produce a grassroots cool.