It's the data. Stupid!

Sometimes you come across two-thought provoking articles that overlap in a single day: One is an article that has gone around the open source world over the last few days: Tim O'Reilly pointiing out that applications, operating systems aren't where it's at any more. Shared, distributed, data-rich, market-creating: these are the new buzzwords. The application and OS used to build them are as relevant as the CPU running the app is to the individual user. The second one is Greg Costik pointing out that the games industry is having a problem : an exponential curve - quantity of content needed for a complete game - meeting a linear one - the increase in game sales.

Serveral solutions are posited. One is more algorithimic, procedural content: this is an approach that as a programmer I'm intensely interested in: it's possible to evolve landscapes, textures, faces, plants, dungeons, and possibly gaits via an genetic programming / evolutionary approach. This can only reduce costs to some extent. Content still has to be shaped and coordinated by talented artists. Another point is that of the mod scene: the fact that given the right tools and the right framework, people will generate content for the hell of it, for free. Polycount, Desert Combat, the extensive Half-Life mods are all cited as examples. EA has tied this in with the Amazon distribution model that O'Reilly is talking about.

I imagine a peer to peer game platform based on a componentized engine that lets the users generate and swap content with ease, adjust the ruleset in their own area of the network, changing the bias in their area to storytelling, combat, exploration, whatever..but that's just probably a crazy idea. There isn't any money in it. Unless content can be sold within the network.

0 Comments:

Post a Comment