Greetings from MIT!

November 21, 2008

I'm here in Cambridge, Ma, to speak on a panel at MIT's Futures of Entertainment 3. It is a conference about the future of entertainment, of course --- but the subject is a lot deeper that I thought about before.

The conference officially starts in a a few minutes, but there was some pre-conference entertainment yesterday. I arrived on MIT's campus last night at about 5 Pm to catch a lecture on Artificial Intelligence in Narrative Storytelling. It was one of the most brilliant things I've ever seen.

The presentation was given in a classroom that hosts mathematics classes, and the presenter was Michael Mateas of University of California, Santa Cruz. He was the head of the Expressive Intelligence Studio at UC Santa Cruz (where Donna Haraway, one of the founders of Cyborg Anthropology, teaches).

When I walked into the room, Mateas was playing what looked like a video of two animated characters  having a conversation in an newly decorated apartment. Upon closer inspection, I realized that he was interacting with the characters and influencing their interactions towards each other via keyboard. He'd type something, and the characters would react to what he types in varying ways. He could influence conflict, happiness, or randomness.

He said that he and his group based some of these interactive games off of a book published in the 1970's called "Games People Play". Having read the book, I understood exactaly why he'd chosen to use the book for social interaction studies. Games People Play is a book about social games people play with each other, and the non-verbal and verbal cues that seem to mean something on the surfance but really mean something else underneath.

For instance, in the  "Hot-Button Game", a player can push character 'hot-buttons' (eg. sex, marriage) to make the characters react in a certain way. In the "Affinity Game", the player can take sides in character disagreements. This often leads to hilarious results, where the character who is being supported will begin to like the player more than the character who is not being supported.

The level of interactivity in the game was pretty good. A lot better than many of the other interactive narrative games I've experienced. It was different from The Sims in the fact that keyboard commands were used in place of mouse clicks.

Then Mateas launched into virtual programming and the idea of a social atom of interaction. He said that he based a lot of the interactivity between the characters off of screenwriting, where the term "beat" is used as an element of social interaction or pause. He also said that he tried to work with a screenwriter to get the screenwriter to write social narratives for AI, but that it was very difficult for screenwriting for AI to be handled if one did not know both art and programming.

That was one of the main issues Mateas faced. The fact that it is difficult to find people who are both artists and programmers. He said that they were a rare breed -- but that he'd tried to teach programming for artists, and art to programmers. He said that it was the only way that we'd be able to have truly realistic and beautiful AI narrative games.

After the lecture, the group headed to the house of Henry Jenkins...but more on that later. More updates to come as the conference progresses. Thanks for tuning in!