I halted work on interactive storytelling at the beginning of this year. I was exhausted, stressed, and making little progress. I felt that the best course of action was to take a break, step way back, and reconsider the entire project from an appropriate height. Six months have passed now and I have figured things out. I have decided to end my work on interactive storytelling.
Interactive storytelling has been my life’s work; I have said before that, should I fail in this endeavor, then my life is a failure. I am not so sure of that now, but I do know that I can spend my time better on related efforts. But first I must explain how I came to this conclusion.
My goal all along has not been to create a single great piece of interactive storytelling; it has been to ignite a new industry, a community devoted to the furtherance of interactive storytelling as an art form. Suppose that I could indeed finish Siboot to my own satisfaction. What would happen?
First, few people would appreciate it. They would judge it by the only standards with which they are familiar: the standards of current game design. They would conclude that Siboot is a lousy game — and they would be right. Siboot was never intended to be a game; it’s interactive storytelling. The emphasis is on character interaction, and it already offers interesting dramatic character interaction. But people aren’t looking for interesting dramatic character interaction; they’re looking for the things that make great games: challenge, a smooth learning curve, impressive graphics, catchy little tunes to accompany their play. Above all, a game must be winnable. Yet stories aren’t necessarily about winning and losing; they’re about drama.
No matter how good Siboot turned out to be, it would not create the splash I had hoped for. It would not go viral and trigger lots of tweets and viral videos on YouTube. It would certainly attract a small comradeship of people who recognize its importance. Everybody else would be unimpressed.
When the first moving pictures came out in the 1890s, most people dismissed them as vulgar wastes of time. They were too short to show anything of significance: trains pulling into stations, two people kissing, and so on. There simply wasn’t anything of value in those first moving pictures. Most people were unimpressed. However, some people saw the potential of the medium and plunged into it. For the first ten or twenty years, moving pictures remained pathetic exercises. Nobody knew how to use the medium to advantage. The movies really were junk.
But the devotees remained adamant that the medium was capable of great things. More important, there was a clear path forward: make movies longer, keep experimenting, and keep learning how to do a better job.
Interactive storytelling faces exactly the same problems that cinema faced more than a hundred years ago, but it has one fatal handicap that cinema didn’t have: the steep learning curve. Anybody with a camera could fiddle around and make simple movies. But interactive storytelling requires a lot more than a camera: you also need to understand how to express dramatic ideas in algorithmic form. That whooshes right over most people’s heads.
I had long felt that sufficiently motivated people would teach themselves how to apply algorithm design to drama. In lectures, I pointed to the Renaissance artists who, needing detailed knowledge of human anatomy, attended dissections of corpses in Italian medical schools. In Italy’s hot summers, those corpses stank something horrible, but if you wanted to know the muscles, you had to examine each muscle, bone, and tendon closely. Those artists were willing to endure ghastly experiences in order to master their art. Surely, I said, modern artists would rise to the challenge and learn the mathematics behind algorithms.
I was wrong. In nearly thirty years of orating, writing, preaching, and goading, I have encountered just a handful of people who have truly risen to the challenge. Most are mere patrons of virtue, to use Thoreau’s phrase.
It took over 2,500 years for the basic concept of rationalism to finally flower into its fullest expression, the computer. (To learn about how this happened, read the lengthy material beginning here and especially on the material here.) Despite having 2,500 years of experience with logical thinking, we humans are still really weak at it. The jump from logical thinking to algorithmic thinking is a smaller jump than that from pattern-based thinking to logical thinking, but it’s still a big jump — and it will take centuries for civilization to embrace the concept. Until that happens, interactive storytelling will just have to wait. Without artistically inspired algorithms, you just can’t make interactive storytelling.
Distractions
One experience that led me to this conclusion was a story about a new software product called Florence. This app for the Apple iOS won several awards, so I thought that it might be worth looking at. What I found was nothing new; it’s the same-old, same-old stuff that we’ve been seeing for decades. It’s basically a predefined story with a few little interactive buttons thrown in. This is exactly what people attempt when they first get started on interactive storytelling. They build their story-with-a-veneer-of-interactivity, get good reviews, then try something deeper, run into problems, and give up. I’ve seen this happen so many times that I don’t spend much time looking at new software — it’s always the same old thing.
The problem with these products is that they define excellence in interactive storytelling. Since these are the products that are extolled as award-winners, nobody aims any higher. Why bother to aim higher when you can do so well with a pedestrian design?
The Coming Dead End
Even worse is in store for us: deep learning AI. It is only a matter of time before somebody comes up with the brilliant idea of applying deep learning AI to storytelling. After all, deep learning AI is the hottest, sexiest, get-rich-quickiest technology on the planet.
Every deep learning AI system requires vast amounts of input data to get started. You feed in millions of pictures of clowns to get a clown-recognizing AI, or millions of songs to get a song-recognizing AI, or whatever. In this case, the designers could tap into any of three huge databases of fiction: murder mysteries, action thrillers, or gothic romances. We’ve got a zillion of these, and if you just feed them all into a large enough deep learning AI system, then it can figure out the entire pattern. You need only turn it around to have it spit out new versions of the same old stuff. You could probably make it interactive by having it sequentially find the best fit pattern response to the player’s actions.
There’s no question in my mind that such software could generate reasonably good murder mysteries, action thrillers, or gothic romances. After all, even the authors of such works will tell you that they are formulaic. If there’s a formula in there, a deep learning AI system will figure it out.
Therein lies the fatal flaw: the output will be formulaic. Most important, the output won’t have any artistic content at all. You will NEVER see anything like literature coming out of deep learning AI. You’ll see plenty of potboilers pouring forth, but you can’t make art without an artist.
This stuff will be hailed as the next great revolution in entertainment. We’ll see lots of prizes awarded, fulsome reviews, thick layers of praise heaped on, and nobody will see any need to work on the real thing. That will stop us dead in our tracks for a few decades.
Conclusion
I have been trying to teach people about interactivity, process intensity, algorithm design, and interactive storytelling for nearly thirty years. While there are a great many people who admire my efforts and would very much like to see them advanced, there simply isn’t the cultural oomph to make any of it work. That’s why I am stopping work on interactive storytelling…
…but
while writing this, an idea occurred to me for a way to perhaps get the ball rolling again with other people. I’ll give it a lot of thought. Like most ideas, it is most probably junk. If it passes initial review, I’ll post the idea here.
Appendix
The other day I was discussing some of this with my old colleague Laura Mixon. Laura has worked with me for years on interactive storytelling. While I have been the driving force behind the underlying software engine, she has been the primary storyworld designer. It dawned upon me that there is a remarkable historical parallel. I am reprising the role played by Charles Babbage; Storytron corresponds to his Analytical Engine; and Laura Mixon fits the role of Ada Lovelace quite nicely.
