At the MSR Social Computing Symposium, a good part of the conversation was massively multiplayer online games such as World of Warcraft. Researchers have issues accessing data and setting up proper control and test environments require $10M. Because of the high level of user adoption and engagement, there seems to be great promise for game engines to support learning. Most approaches suggest developing new content on top of game engines. In effect, creating new games.
But when discussing this it dawned on me there are not only prohibitive costs, but risks for adoption. Meanwhile, there are platforms like WoW that have sunk cost and proven adoption. Perhaps someone can build a learning environment on top of WoW?
For example, WoW is all about gold and leveling. You know, math. AddOns like Auctioneer give you an advantage in the game by helping you price auction items. Other interface AddOns help display statistics to coordinate attacks across players. What if someone leveraged the simple scripting language to create an Auction add on that gradually introduced mathematical concepts towards building full fledged arbitrage models? Or a UI mod that presented problems that, if solved, provided you an in-game advantage?
Judith Donath pointed out to me the need to frame and relate such in-game experiences to that of the everday world. This seems to be the key -- explaining not only how solving a problem or building a model gives you advantage in a virtual world -- but the real world.
I would LOVE to see some prototypes in this direction.