For the Win - Cory Doctorow (read novels website txt) 📗

Dibyendu's cafe. Instead, she took her walking stick and limped all around Dharavi, up and down the streets where the tiny factories would normally have been hives of activity. Many of them were, but many were not -- many had workers and crowds out front, and it was like a virus that was spreading through the streets and lanes and alleys, and now it was as if all the crying had lightened her so that her feet barely touched the ground, as though she might fly away at any instant.

She was just turning to go back to her army and maybe a few hours' sleep when they grabbed her, hit her very hard on the head, and dragged her into a tiny, stinking room.

Confidence is a funny thing. When lots of people believe something is valuable, it becomes valuable. So if you're selling game-gold and people think game-gold is valuable, they buy it.

But it's better than that. If there's a wide-spread belief that Svartalfaheim Warriors swords are valuable, then even people who don't think they're valuable will buy them, because they believe they can sell them to people who do believe that they're valuable.

And when people who buy to sell to others start to bid on Svartalfaheim swords, the price of the swords goes up. Of course it does: the more buyers there are for something, the higher the price goes. And the higher the price goes, the more buyers there are, because hey, if the price is high, there must be plenty of suckers who'll take the swords off your hands in a little while for an even higher price.

Confidence makes value. Value makes more value, which makes more confidence. Which makes more value.

But it's not infinite. Think of a cartoon character who runs off a cliff and keeps running madly in place, able to stay there until someone points out that he's dancing on air, at which point he plummets to the sharp rocks beneath him.

For so long as everyone believes in the value of a Svartalfaheim sword, the sword will be valuable, and get more valuable. As the pool of people who might buy a Svartalfaheim sword grows -- say, because they're getting calls from their brokers offering to sell them elaborate, complex sword futures (a contract to buy a sword at a later date), or because their smart-ass nieces and nephews are talking them up -- the likelihood that someone will say, "Are you kidding me? This is a sword in a video game!" goes up.

Indeed, this doubter might have other choice observations, like this: "If everyone has these swords, doesn't that mean that there's more swords than anyone could possibly use? Doesn't that mean that they're not valuable, but valueless?"

Or if the doubter is impossibly old fashioned, he might even say: "What if the people who run this Fartenstein game decide to change the number of swords available by just deleting a ton of them? Or by printing up a kazillion more? Or change the swords into toothpicks?"

Oh, the sword-speculators will reply, they'll never do that, it would ruin the game, they can't afford to do that. And here's the thing: they're half-right. So long as the game-runners believe that messing around with the swords will piss off all these people who own, speculate on, buy and sell swords, they can't afford to do it.

These cartoon characters run in place on air, shouting that the swords will always go up in value, shouting that the game-runners will never nerf or otherwise bork them, and they can stay there, up in the air, waving their swords, being joined by others who are convinced by their arguments and the incontrovertible fact that they are indeed not falling, until...

Until...

Until there's enough widespread confidence in the proposition that they will fall. Until the press starts to publish wide-eyed stories about the absurdity of ever believing in the value of these swords, pointing out that the fall is inevitable, that it was pre-ordained from the moment the first speculator bought his first sword.

Think of the belief in infallible swords as a solar system. In the center, there's the sun, gigantic and white-hot, exerting gravity on the planets and asteroids that spin around and around it. At the outer edge is the dandruff of planetesimals and asteroids, weakly caught in the gravity, only halfway committed to being part of the system. As the sun begins to cool off, begins to shrink with the force of disbelief, these outer hangers-on fly away. These are the tasters, the people who bought one or two little swords or sword-futures or "fully hedged complex sword derived securities" because everyone else was doing it. They hear that this thing is too good to be true and see the prices start to drop and so they sell off what they've got, take a small loss, and tell their friends.

Well, now there's a bunch of people saying that swords aren't really that valuable. Less confidence equals lower prices. And there's more swords on the market. More swords equals lower prices. The larger planets, closer in, the investors with a fair bit of money in imaginary cutlery, these folks see the prices dip and continue to fall. They hear the brokers and analysts scurrying around saying, "No, no, the sun will burn bright forever, the sun will never dim! Prices will come up again. This is temporary."

Here's the thing: if the brokers and analysts can convince these bigger investors that they're right, they will be right. If these bigger investors hold on to their swords, the market will stay healthy for a while longer.

But if they aren't convincing enough, if these bigger investors lose confidence and start selling, they'll never stop. That's because the first seller to get out of the sword-market will get the highest price for his goods. But once he gets out, his swords will be on the market (remember, more swords equals lower prices) and everyone else will get a lower price. And when they sell, the prices will go down further, panicking more investors, putting more swords on the market, forcing the prices down further.

Somewhere in there, the game-runners are apt to have a minor freak-out and then a major one. They'll start to mess with the sword-supply. They'll take swords out of the market, or put swords in, or nerf swords, or buff the hell out of them, anything to keep the fun from collapsing out of the game.

And that'll probably make things worse, because this isn't an exact science, it's a bunch of guesswork, and there are ten zillion ways to get this wrong and so few ways to get it right.

The sun gets smaller, and dimmer, and the close-in planets are feeling the tug of oblivion now, the call of deep space that says, "Spin away, spin away to forever, for the sun is dying!"

They don't want to spin away. They want to hang on. They have so many swords in the bank, they're practically made of swords. They've made a fortune buying and selling swords. Of course, they spent the fortune on more swords. Or different swords. Or axes. But whatever they've spent it on, it's basically the same thing, because every broker knows that you won't get in trouble for recommending that people buy things that have always been profitable.

If the sword market collapses, these planets -- these major, committed investors -- will die. They will be wiped out. They have pledged their lives and love and immortal souls to magic swords, and if the swords break their hearts, they will never recover. So as the market for swords gets crummier and crummier and crummier and crummier, they grow more and more insistent that everything is fine, just fine, it'll all be back to "normal" any day now. They can't afford to lose confidence, because they aren't going to fly off into space. They're going to fall into the dying sun and will be incinerated in its glowing heart.

But denial only works for so long. The sun is dying. No one wants your swords. Your swords are worthless. Even the people who need a sword to kill some elves or orcs or random wildlife critters are faintly embarrassed by the fact, because worthless swords are now the subject of numerous jokes about idiotic investment schemes and corrupt brokerages and loony investors who got swept up in the heat of the moment. These people go and kill monsters with bows and clubs for a while, because everyone knows how much swords suck.

How low can the value of a sword go? Subzero, as it turns out. Not only can a sword become worthless, it can actually cost you money to get rid of it. Oh, not the sword itself, of course, but the derivatives of the swords. The bets on swords. Where someone else has made a bet on whether your sword will go up or down in value, and then packaged it up with a bunch of other bets, just figuring out which bets are in which packages can cost so much money that you end up losing money, even on winning bets.

Confidence is great, but it isn't everything. Reality catches up with everyone, eventually. All suns extinguish themselves. All cartoon characters eventually plummet to the bottom of the canyon. And every sword is eventually worthless.

Command Central was bedlam. The game-runners snarled at each other like bad-tempered, huge-bellied dinosaurs, and ate like dinosaurs, too, sending out for burgers, pizza, buckets of chicken, huge thick shakes Anything they could scarf down one-handed while they labored over their screens and shouted insults at one another.

Connor hardly noticed. He was deep in his feeds. Bill's new security subroutines let him run every player's actions backwards and forwards like a video, branching off into other players' timelines every time they crossed paths in a party, a PvP combat session, a trade, or a conversation. It was an ocean of information, containing every secret of every player in every game that Coke ran.

It was too much information. He was looking for something very precise -- the identities of gold-farmers -- but what he had was every damned thing ever uttered or done in-game. It was a wondrous toy and an infinite distraction, and practically every spare moment Connor could muster was spent writing scripts and filters to help him make sense of it.

Just now he was watching a feed of every player who had PvP killed another player, where the dead player's toon had earned more than 1000 Mario coins in the previous hour. This was turning out to be a rich vein of potential gold-farmers and Webblies. He was just trying to figure out how to write a script that would also grab the player IDs of anyone who was nearby during one of these fights, when he realized that Command Central had gotten even noisier than usual, devolving into raw chaos.

He looked up. "What's wrong?" he said, even as his fingers moved to call up general feeds showing the overall health of the game and its systems. And even before anyone answered he saw what was wrong. Server load had spiked across every game-shard, redlining the server-clusters seated in air-conditioned freight containers all over the world. It seemed as though every single metric for server-load was at peak -- calculations per second, memory usage, disk churn. But on closer examination, he saw that this wasn't quite true: network load was down. Way down. Somehow, these vast arrays of computing power were all being made to work so hard they were in danger of collapsing, but it was all happening without anyone talking very much to the servers.

Indeed, network load was so