Have a read here and here. It's about a 'thought experiment' of Ken Levine. He's looking into breaking away from the fixed narratives in video games (including the BioShock series). He says
"I spend five years [working on a game] and 12 hours later the player is done with it, and that is heartbreaking. There are some fans who will replay it but you can't expect that from the average gamer because it won't be meaningfully different the second time, and that is an important challenge."
I get that. It's the crux of narrative driven games. They fundamentally are at odds with the core trait of video games being interactivity - the ability to change the outcome. How can you tell a story with meaning and message if at any moment the player has the power to drive off the rails you set for them?
Anyway, Levine is looking to break down stories into their most primitive components, similar to how constructs of LEGO can be broken down to individual bricks. It'll be interesting to see what Levine actually thinks those are. Stories consist of different components. Mainly you've got characters and events. Characters are put into an initial situation and have defined (but not static) personalities which ideally naturally develop and evolve over the course of the story. Events happen as direct consequences to actions that characters take (which are in turn based directly on their predefined personality) but they also happen outside their realm of influence. So looking at it from such a rough angle, it can be done.
What Levine is trying to do is to break up stories into these "story bricks" so that he can reassemble them in various combinations, ideally yielding ever changing and fresh narrative experiences for his games. These story bricks can be processed and using a systemic approach to story generation the stories in his games can become actually reactive to player input. The next challenge I think will then be to generate purpusefully dramatic or comedic stories. It then becomes a question of how to define genres of story telling and coming up with ways to systemically generate them.
I hope he'll be able to get it working, even if only in baby steps. Innovation in gaming is always welcome, especially if it's trying to tackle one of the biggest challenges for game design today.
(originally publish on my blog http://lostinthezone.wordpress.com)