I’ve already finished an admittedly short — but nevertheless quite fascinating — book called “A Theory of Fun for Game Design” by Raph Koster. The author was one of the creative masterminds behind Ultima Online and Star Wars Galaxies (popular massively multiplayer online role-playing games).

My friend Duane lent the book to me on Friday. It’s a short read, and the ideas presented weren’t quite what I expected.

The five-cent summary of Koster’s thesis is that games have a proper social function, and the best games are those that teach some sort of real-world skill (e.g., mathematical reasoning or complex pattern recognition) before they become too boring and lose a player’s interest. Effective game design entails keeping the brain sufficiently interested (through varying of the patterns and complexities of play) until the essential lesson or real-world skill is taught.

I recommend the book even to those who have no interest in game design, since Koster spends a fair amount of time discussing psychology and, at times, even evolutionary biology. Good stuff, and applicable far beyond the gaming industry.
I was struck by Koster’s take on the role of multiplayer online games. He made the point, sagely I think, that most games featuring human interactivity — even those explicitly designed to elicit cooperation as a preferred gaming strategy — will work only until there is a critical mass of players who reject the cooperation paradigm. When that happens, a “competition” paradigm logically follows, until the “cooperatists” are forced out (or lose the game). The more competitive types are often motivated by victory-at-any-cost thinking, which explains the prevalence of hacks and cheats for most games.

Koster explains this in terms of the role of games in teaching essential skills, which is rooted in evolutionary biology. Part of being “successful” in a long-term genetic sense is passing along one’s genes, and those mostly likely to do that are those males who are socially dominant and driven by a sense of competition.
Hence, “competition” usually becomes the dominant paradigm, even in games designed to minimize it. This is demonstrated by the ultimate collapse of Ultima Online, and by the tendency in World of Warcraft for high-end players to spend most of their energy fighting other human players. Humans are essentially tribal, and that essence usually bubbles to the surface eventually.
It occurred to me that Koster’s ideas are applicable to other “virtual” hobbies, too. In particular, to online political simulations. I’ve grown frustrated over the years with the inevitable death-spiral of sims: They start full of promise, they work well for a while, then they begin to dissolve when “cheaters” take positions of influence (cheating, loosely defined, as actively resisting the impulse to engage in non-superficial compromise). From there, tribalism asserts itself and the sim last for months (or, in one case, years) of decline until eventually there are no more messages posted and people filter away.

Something else to ponder.