Exprimental Gameplay

For years I have swallowed the myth espoused by the game industry that it's not possible to make commercial fun anymore without a huge stable of developers and cutting edge machines and graphics cards. But as I play with different simulation tools on the job, without the high-end machinery around them, I'm revising my thinking slowly but surely.

I am firming up my latest theory on certain Things that Are Fun on the computer: There's some element of chance involved (that mimics expectations about the "real world"), a system to interact with, and a well-presented result of the interaction between the two that's both surprising and predictable in some measure. The more the system is complex and hidden, the more it can look like "chance" and remain fun as discovery occurs. Here's a site that I think agrees with me, that I want to spend some time playing with:

Exprimental Gameplay, at CMU.