the everyday let down of interactivity

It starts somewhere, it doesn’t matter where, as long as it starts. In this case it starts here, a brief post from the Le Monde-hosted blog Playtime, which describes the first games created by a machine along a set of rules of game grammar as explained in this pdf doc from Michael Cook, a PhD student in computer science at Imperial College London (the English-version and original publication about Angelina is here, from The New Scientist) — and by the way the whole area, new to me, of “procedural content generation”, which seems to be already a reality of game designing, though in isolated functions of games, ought to be something I look into in the near future, not just because of the promise of its use for an entire game design, as Cook and his colleagues at Imperial College point out, but also for movie production. I wonder how long it’s going to take before we have films with AI-generated plots (and I mean more than movies as computing code, though that’s a (funny) start) ?

“If this is what you call enhancement, you’ve got a gift for understatement” (Star Trek)

Anyhow, to go back to the point, today that’s where it started. Down that Le Monde post, the game Façade, a 2005 self-described “one-act interactive drama”, gets mentioned. Not just another video-game, me-thinks, but “interactive drama” — gentlemen, the holodeck has arrived !

Well, not quite. Next stop is a visit to the greatly named interactivestory.net where the vision behind the game is alluring:

The dream of interactive drama, perhaps best envisioned by the Star Trek Holodeck, has players interacting with compelling, psychologically complex characters, and through these interactions having a real influence on a dynamically evolving storyline. Motivated by our belief that a fully-realized interactive drama has not yet been built, we embarked on a five year effort to integrate believable characters, natural language conversation, and dynamic storyline, into a small but complete, playable, publicly-released experience. Façade is the result of this effort.

Roll over, Činčera, “global agency”, not just multiple-choice questions, has come to the audience of fiction – -the cinema audience, mind you, even more than the gamer audience. As Michael Mateas (UC Santa Cruz) and Andrew Stern, the game creators, explain:

Rather than focusing on the traditional gamer market, we are interested in interactive experiences that appeal to the adult, non-computer geek, movie-and-theater-going public.

Thanks to AI, the description goes, the player/audience becomes a character in a story no one knows the end of:

Interaction is seamless.

Intriguing to say the least, and there’s a whole literature on the question of AI-driven narratives, or when “ludology meets narratology” to use part of the title of a paper by Mateas and Stern on this issue and it seems to me that looking into the promise of AI for interactive story-telling is vital and fascinating research. But has it arrived, really, yet ?

Since I could not download the game (does not work under Mac OS lion), I had to rely on youtube and video captures of the game. Didn’t take long to find the limits of the Façade game. This chap, in particular, thrives on finding out what happens when the player takes the promise of interactivity, well, literally:

My favorites here: not saying anything or moving at all, something the game fails to recognize entirely; or stealing objects from your friends’ apartment, again and again; or trying to discuss interior decoration in an empty apartment, not sure what happened there. What this means is that there  seems to be little room in Façade for sheer perversity, making me wonder how much human perversity AI can actually simulate (try romancing Trip’s girlfriend, as several users seem to have — unsurprisingly? — done, and all you get, apparently, is a total blank face). Interactive seamless drama this is not — at least, not more than your regular movie-fan relationship with his or her movies: think an audience repeating lines back to the screen at cult shows of cult movies.

And so it ends, with the disappointment of a promise unfulfilled: the promise of virtuality becoming real, the revelation that virtuality must remain a game, with its own (limiting) sets of rules. Procedural content generation indeed promises a future where rules may evolve as the game is played, and thus a virtual experience that may come very, very close to our experience of our live, ever-changing real world — and down the line it is not hard to see that fiction-making in cinema is about to be transformed by the expansion of AI-induced virtuality. If, and when, it does, it will happen because the aesthetic enjoyment of virtuality is what audiences of film have been grappling with since cinema came along, an enjoyment that relies, partly, on that continual unfulfillment, on that tantalizing never-achieved border where our dreams do come true.As the lost, thirst-crazed heroine of a great Hollywood film about Hollywood film-making, Souls for Sale (Rupert Hughes, 1923, for the Goldwyn studios), puts it as she spots a sheik on a camel behind a sand-dune:

Are you real, or a mirage ?

The actor, well-versed in Hollywood’s double-speak about reality and fiction, and the virtuality of the film worlds that he is helping create, has the answer, simple and seductive, down pat:

neither.

Full disclosure: I’ve just wasted spent half an hour hitting a combination of arrow keys and space bar jumping from one block to another, playing an Angelina-designed game — which felt pretty much like any other arcade game, but I guess that’s a compliment to the machine — and you too can do the same and give it a go here…

Full disclosure #2: as this 2007 news story makes clear, the Façade game is supposed to be more than a computer game. Its true objective is to develop into an Augmented Reality game in which the player will evolve as naturally as in a real living-room, once (s)he dons on Virtual Reality goggles. Not clear how that works out without goggles and with merely a cellphone, as the article mentions, but there’s no doubt that as an immersive experience it would work much better — not because the dramatic options would then be truly limitless as in real life (the false pursuit of realism in art, to have works of art seemingly as undecided as life), but because the whole fun would be in exploring this I am here/this world is fiction distance that Virtual Reality interactivity keeps on promising.