Thursday, September 6, 2007

Strange Goings-On In Online Worlds

Two related news stories caught my attention recently, and fired up my imagination. I thought I'd share them with you, and just muse a little.

They both involve Massively Multiplayer Online Role-Playing Games, or MMORPGs for short. (I kid you not, that is the commonly accepted acronym. I can't pronounce it, either.) Now in these games, you create a character, log in, and interact with the characters of perhaps millions of other people who have also logged in. Many of these games (World of Warcraft, Everquest) are set in fantasy settings, but not all; I understand that the designers/managers of the first MMORPGs noticed that many of the players who created characters for their fantasy worlds didn't want to go out and slay dragons; they prefered to do mundane things, like run the cobbler's shop or man the potions booth--things that allowed them to interact in non-threatening ways with other characters. So it is not surprising that some MMORPGs are simply set in everyday settings, where you make your character get up, brush teeth, go to work, etc. The game Second Life falls along these lines.

Now, some of these games have so many players, interacting in such sophisticated, complex ways, that it is unnecessary for the designers to work in "magical" solutions. Here's an abstract example. Suppose your fantasy character needs to buy a suit of armor. He goes to the smith to purchase one. So where did the armor come from? In most computer games (and most paper-and-dice role-playing games like Dungeons and Dragons) one doesn't worry about where the armor comes from; one simply assumes that there are smiths around that have what you need in stock. But in a truly massive online game, you don't have to assume this; the smith is being played by someone in Idaho, and he had his character craft your armor when he was online last Tuesday. When you buy the armor, the smith is out of stock until he crafts another one. And where does the smith get the mithril from which to make the armor? Well, there's some guy down in Louisiana who has a Dwarf character who mines for a living, when he's not getting chased down by the Balrog....

So you can get entire "virtual economies" going in these game worlds. This is where it gets interesting: these virtual economies have attracted the attention of all kinds of social scientists, economists, epidemiologists....

So the first of the stories that caught my attention is this one. Obviously, if you've got a fully functioning virtual economy driven by thousands upon thousands of transactions, it's possible to have, well... recessions, currency crises, shortages, overproduction, underproduction, you name it. Apparently, in the game Second Life, they use a currency that the game's managers have pegged to the dollar. The dues that you pay (real dollars, here) when you play this game have become something of a currency reserve, and the virtual currency in the game therefore has real value and can be cashed out under certain circumstances.

Well, if your virtual currency has real value, then it's possible to gamble in this game; your player character can gamble with other player characters, and it becomes just another method to do online gambling--except it's more fun, because you're role-playing at the same time! So the managers of the game decided to shut down the gambling that was happening in their world...

...and started a major recession in their world, as the gambling ban shut off about half the economic activity. So, the CFO of Linden Lab (which produced Second Life) behaved like a Federal Reserve Chairman, and took steps to pump up the virtual currency, by changing the circumstances under which his company would trade the virtual currency for the real currency. It's a little like the way that China keeps the Yuan pegged to the Dollar, even though free-market economic forces should be causing the Yuan to appreciate.

The second story was here. This one affected the game World of Warcraft. Apparently back in 2005 a scenario was entered into the game in which an evil wizard named Hakkar the Soulflayer lived in a dungeon somewhere, where only the most advanced characters could hope to best him. Hakkar had the power to inflict a disease dubbed Corrupted Blood onto his challengers. This disease would gradually steal life points from players, until all but the strongest and most advanced characters would die. This disease was also contagious; one character could spread it to anyone else nearby.

To prevent this disease from breaking out into the wider world, a rule was entered into the code that would remove the disease from any player that left Hakkar's dungeon. But the programmers overlooked one problem: the players' animal pets could also get the disease; and they would still retain the disease if they were magically transported out of the dungeon.

The result became the World of Warcraft's answer to the Black Death, with entire virtual cities becoming depopulated as the population got wiped out. But the interesting part of this is how the players behaved: they acted very, very similarly to the way real people behave when real epidemics hit. There were those that heard rumors of the Corrupted Blood plague, and went to investigate, and got sick, and spread the disease. There were those who rushed in to help their neighbors (presumably to give them food and other things that would boost their dwindling life points), and got infected. There were a lot of players who sent their characters away from the large cities to the countryside to get away from the plague. There were attempts at setting up quarantines, that failed. There were infected players who decided to go out with a bang, by deliberately infecting as many other players as possible before expiring.

Ultimately the problem was fixed in the software; the Corrupted Blood curse has been changed so that it's no longer contagious. But the whole experience has caught the eye of the journal Lancet Infectious Diseases, and a bunch of researchers at various universities.

What people have started to notice from these two (and other) examples, is that one can do some serious social science work by studying these games. Things like economies, and epidemics, are notoriously hard to model. So here we have two examples of online games, which were designed for no reason other than to entertain the masses, outperforming the best theoretical models the academicians can come up with--because each character in each game is controlled by a person with a person's intelligence.

One of the articles to which I linked draws out this point a little further. In the social sciences, it is common for debates to go on for decades over which theory is better. (e.g. Is it better to use fiscal policy or monetary policy to control inflation? That was a big one for years, until Reagan and Volker apparently settled the question in favor of the latter.) The trouble is that it's really, really hard to set up controlled experiments on real populations to determine which of two competing theories works better in real life, and until now it's been impossible to check how close to reality the theoretical models have been. It appears that the advent of these online games is giving social scientists a powerful new tool for checking how well their theories actually work.

I find that highly fascinating.

Full disclosure: I do not play any MMORPGs. You think I'd have the time to write 25+ blog posts in the last three weeks if I did? :)

1 comment:

Anonymous said...

That really IS fascinating! So all that wasted time the kids spend online might not be for naught afer all!

Great post, and great blog. I really enjoyed reading it.