Every now and then, somebody asks me “What is interactive storytelling?” There are three answers to this question: what interactive storytelling is NOT; a general answer and one that is specific to my own work.
What Interactive Storytelling is NOT
Let us begin by dispensing with some of the common misconceptions about interactive storytelling. First, interactive storytelling is not a static story grafted onto a game. Most games that include stories take this form. You play the interactive game for a while, then you get a snippet of non-interactive story, then you get some more game, then some more story, in the hope that somehow the combination of the two constitutes interactive storytelling. This is simply a variation on the concept of the thaumatrope:
Interactive storytelling is NOT a “choose your own adventure” (more technically known as a branching tree). For 25 years now I have been explaining that this approach can never provide anything near the richness of genuine interactive storytelling. Branching trees are toy cars compared to the real thing: they may look like it at first, but they certainly don’t perform like the real thing.
Interactive Storytelling In General
The general answer is that interactive storytelling is a form of computer entertainment in which the player interacts in a dramatically significant way with artificial actors. The key concept here is dramatic significance. You can interact with a refrigerator by opening and closing the door and trying to catch the light going on and off — but that’s not dramatically significant interaction. Genuine interactive storytelling requires you to talk to the actors and have them talk back to you.
My Storytron Technology
My own technology sits in an odd location: it falls far short of the goal of true interactive storytelling, but leaps far ahead of the other technologies that I have seen. Yes, it does allow you to converse with computer actors — but the conversations are in Deikto, a language system I have developed that compresses language down to a shadow of its true self. The player doesn’t speak or type in sentences; instead, the player builds sentences word by word under strict constraints that insure that the results are dramatically meaningful and make sense. The resulting conversation has a stilted feel to it, making it a linguistically claustrophobic experience. Instead of wandering freely across a huge open domain of dramatic possibilities, the player wanders through a huge mansion of a thousand doors; it’s too big to explore in many hours, but the player can never leave the mansion. By contrast, branching stories place the player in a much smaller house with a few dozen doors.
Here’s a hypothetical example of what might happen in a storyworld created with my technology:
Joe greets Mary politely.
Mary greets Joe warily.
Joe compliments Mary extravagently.
Mary thanks Joe weakly.
Joe tells Mary that Tom is taking Jane to the Prom.
Mary tells Joe that Tom is not very handsome.
Joe tells Mary that Tom is definitely not handsome.
Mary tells Joe that she likes Jane a lot.
Joe tells Mary that he likes Jane moderately.
Mary asks Joe if he has a date for the Prom.
Joe answers Mary negatively.
Mary says, “I see”.
Joe asks Mary if she has a date for the Prom.
Mary answers Joe negatively.
Joe asks Mary who she wants for the Prom.
Mary answers Joe evasively.
Joe invites Mary to the Prom.
Mary answers Joe evasively.
Joe asks Mary how much she likes him.
Mary says goodbye to Joe hurriedly.
Joe says goodbye to Mary unhappily.
On the one hand, you can see that this is obviously a strained language system with limited dramatic elbow room. On the other hand, I think you’ll agree that this kind of dramatic interaction is much better than anything we’ve seen before.
So why haven’t I conquered the world with this technology? I think that I am in the position that Charles Babbage was in nearly 200 years ago. He had designed the first programmagle computer, using thousands of gears, levers, and other mechanical contrivances, with RAM of 5KB, and could multiply, divide, add, and subtract; it was programmed in something like modern assembly language. It was never built because it was simply too big and too complicated.
I am in much the same position with Storytron. The technology is extremely complicated. Unlike Babbages Analyitical Engine, I have actually built a complete working system along with all a powerful editing system for creating storyworlds. However, the system remains too complicated to be programmed by mere humans. I myself have struggled with the creation of Storytron storyworlds; I have built four incomplete storyworlds, all of them devoid of entertainment value.
Recognizing the problem, I have now begun work on a stripped-down version of the technology. That project is called Siboot, and if you are determined enough to plough through the hundreds of pages in the design diary, you can see that even this stripped-down version is horrendously complicated to design.
I have lost hope that I will ever build a good storyworld with the original Storytron technology; but I have some hope that Siboot will show the way to the future, and that by making my technology open source — when people are ready for it — I might at least help get the ball rolling.