Skip to content

Interactive Storytelling Spiel

July 29, 2011

Okay, I’ve been super busy on The Art of Words lately, and haven’t had a chance to get the next segment of Redefining Interactivity ready. However, while I was searching this afternoon for a document that it seems has disappeared without a trace (and which means I’ll be retyping pages of manuscripted rough work in the very near future *mutter grumble*), I stumbled across this rant post I made elsewhere that more or less lead right up to the creation of this blog. Probably nothing in here that you havent’ already gleaned from my other posts, but still, I figured it would be nice to post it here and remind folks that I’m still thinking about this stuff.

So, without further ado…


Something I’ve been arguing about for years (to little fanfare) is that the rules that govern good storytelling in passive mediums don’t really fulfill the potential of interactive storytelling.

We’ve learned, thanks to tabletop RPGs, that to some extent, what often makes a story game really immersive is the open collaboration between player and game master, once the basic parameters of the game have been defined.

In video games, this has been difficult to implement because most gaming AI are not up to the challenge of improvising plot developments that resonate and resolve satisfactorily. That really does require some careful planning, which tends to run counter to how people think improvisation works, and pretty much flatly contradicts general wisdom about the nature of good storytelling.

What is called for is a new form of writing; less fixated on the principle that it’s entirely up to the author to create the magic; and to find new ways to deliver plot and character development without losing your audience.

See, the old tools demand that you start with a skeleton and flesh it out. That’s exactly what you can’t do in a properly interactive story. The argument is that you can’t tell a story without one; at least, not an EFFECTIVE story. My counterargument is that it can be done, but not like this. When writers finally commit to the (hitherto undefined) principles of interactive storytelling, and force game developers to make it work correctly, what you will wind up with is a lot of little modular bits of story that add to or subtract from the total experience, and (hopefully) lead to a more harmonious and original, emergent storyline.

Now, I know this may sound like a lot of pie-in-the-sky stuff, but the thing is, this is what a lot of big game developers are looking at these days, from various directions. Whether they can or ultimately will go with it is another question. It may not yet look feasible, based on cost effectiveness and proven market demand. I figure it’s just a matter of time until developers figure out a way to make in-game choices contextual, with values assigned to each portion of ‘script’, and an AI that assesses what happened and how the player reacted, and then uses that data to mold the ‘plot’ using modules of available story to complete the adventure.

The first thing you need is a relatively egoless writer (or writers), who can handle the notion that all of hir ideas have to be tailored to a non-linear story construction model. The equations involved will have to do the actual conflict-escalation-resolution bit, which most writers aren’t really equipped to deal with. That’s the payoff for most writers: knowing they constructed this thing that evokes a strong response in the audience. No writer wants to think that their payoff is going to be credited to a mere software application.

And that’s really the biggest reason it doesn’t exist yet. Most writers haven’t cracked the code, because it’s counter to how storytelling is supposed to work. Experiments with so-called AI storytelling in the past were pretty awful. I read some. Trust me, it was horrid. If I didn’t believe it could be done much better, I would shut my mouth and speak not another word on the subject, for fear of foisting more of that insipid crap upon the Earth.

Particularly galling to authors of any stripe (myself included) is the notion that a computer could replace them… or more to the point, that some short-sighted publishers might elect to print computer stories instead of human-written stories, because they’re cheaper. That’s been a heated subject of debate for as long as computer programmers have been trying to program storytelling software, which is a surprisingly long time. The standard panacea for this bogeyman is to simply insist that computers will never have the imagination needed to dream up original scenarios. This may be true. Perhaps we never will achieve a truly sentient computer with the ability to intuit responses without any computational algorithms to guide them. Perhaps it’s just as well if we don’t.

But without getting into a ‘dangers of AI’ argument, what I’m talking about isn’t the same thing anyway. I don’t need a machine to do my dreaming for me. I merely need a machine to remember every scenario I conjure up, read what the audience’s reactions to these bits are, and calculate (and recalculate) paths for the audience to follow to the next decision gate. It’s probably a whole lot more work. It might never take off. It might even prove to be terribly unsatisfying, for the author if not the audience. But that’s the Holy Grail of Interactive Storytelling.

You have to build a tool box big enough and modular (I would aalmost say ‘granular’) enough for a computer to sew the bits together as the audience reacts. There will be rules, of course. There will be things the story simply won’t permit, much as there are things you cannot do just because you want to. Life is like that. Deal.

I’ll bet that any writers reading this have been shaking their heads vehemently, fully aware that storytelling cannot–should not–be done by formula. To them I say, good storytelling is always going to be about more than just plot and snappy dialogue. Every author worth hir salt dreads the day when they look up and realize they’ve become a hack writer. Inspiration requires a lot more than algorithms and data.

But there are a lot of components that can be planned for, and the basic building blocks of decision making can be anticipated, given a certain set of conditions. Oh yeah, someone is bound to get it wrong and try to go left when there is only a right turn. I’m not saying it’s easy to anticipate humans. But within certain parameters, given lightning quick analysis and evaluation of a player’s response, you can present them with a tailor-made variation on your basic story, the same way you can rework any story you’ve ever dreamed of, incorporating and discarding segments as it becomes clear to you that certain bits don’t work in a given context. The difference is, you’ll be telling the computer (via the programmer) how to handle those decisions.

So you see, in the end, it’s still authorial; just not to the control freak level writers are used to. In the end, you have to let the story go and let the audience make of it what they will, the same as always. Only, a little earlier than usual. For those authors who can live with the uncertainty, the ‘incompleteness’, it may just prove to be an exciting new medium that is virtually unexplored and waiting to have the bugs shaken out.


No comments yet

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: