October 22, 2003

That Darn Conundrum

by Andrew Stern · , 2:01 am

I find interactive drama to be a fascinating topic. It’s a fairly undefined and unproven thing, which makes it a lot of fun to think about, and attempt to build. Frustrating and humbling, too, of course.

Here’s a few rambling thoughts on the topic (many not new), partly motivated by a short essay about reality television that I recently came across while surfing.

While I’ve thought and read about interactive drama a lot (but am always finding more!), and with Michael have come up with an approach to it, a question I keep asking myself is: what exactly do you *do* in an interactive drama? What is it exactly? How does it operate, on a design / structural level?

The general concept of interactive drama is the following: a dramatic situation is happening in which you are free to say things and take actions that affect how the drama unfolds. You are the protagonist of a story being created in real-time.

Put another way, interactive drama is a simulation of the more interesting conflicts of real life, with the boring and meaningless bits taken out.

Sounds great, but that’s pretty high level; the devil’s in the details. What do you “do” exactly?

(This concept is related to but distinguished from adventure games, which are usually goal-oriented, where typically what you “do” is perform a quest, solve a mystery, solve a series of puzzles for their own sake, etc.)

Even ignoring the daunting technical hurdles for a moment, as we know there are problems that quickly come up when you sit down and try to flesh out this concept, namely the same-old, same-old, *sigh* same-old conundrum of interactive stories: if you are given the freedom to do what you want, how can a well-formed story be created?

I think it’s true that the purist concept described above — complete freedom at all times for the player, and the real-time creation a well-formed story — is, simply by definition of the terms involved, impossible, as much as I wish it weren’t. For example, if towards the end of an interactive drama, you suddenly start acting very differently than you did up till that point, that would likely ruin the well-formed-ness of the story as it had been unfolding so far.

There are several approaches to work around the conundrum. (In Facade, a project still in progress but to be released soon, our approach is do a little of each of these, in a particular balance.)
- constrain the player’s freedom at certain times (number, range and frequency of choices), in such a way that you can’t “ruin” the well-formed-ness of the story.
- diminish the player’s affect on the story (agency), such that their actions less directly influence the story.
- try for, but don’t guarantee a very well-formed story; allow the story to be less than well-formed. (The extreme case: don’t even bother trying to make a story at all; it’s the wrong question. Make games or simulations or worlds that don’t care about well-formed-story-ness; let events and patterns emerge on their own.)

It’s kind of like “cheap, fast, good”. Freedom, agency, story — pick two.

I’m with the ludologists that freedom and agency are essential to a satisfying and meaningful interactive experience. Those are the qualities that distinguish interactive experiences from non-interactive ones, and should be maximized. It’s when freedom and agency feel overly constrained relative to the context of the experience (as they are in so many interactive stories and games built to date!), that I feel straitjacketed, and lose interest in playing.

Following the criteria of high degrees of player freedom and agency, that seems to leave the story-ness requirement to be the one that gives.

The emergence approach, in vogue among game designers, is a very powerful one, that has and will continue to bear a lot of fruit.

But, for me, I worry emergence isn’t enough. Maybe it’s the atheist-who-wishes-it-weren’t-so in me, but I’m interested to think of ways to shape open-ended experiences, in real-time, in an attempt to give them some degree of well-formed-ness, while keeping a very high degree of freedom and agency. That is, as events are unfolding during an interactive experience, have the system regularly look at the experience as a whole, and attempt to direct it into something meaningful as a whole. Versus allowing things to happen haphazardly, such that the whole experience can end up an incoherent, fragmented mess (story of my life :-)

Several questions come up:

Why do we want or need well-formed-ness? or should we just let things be as open and unplanned as real life is? will we find plenty of meaning anyway, provided the virtual world is inherently rich enough with possibilities? i.e., could emergence alone be satisfying enough?

What does well-formed-ness mean? besides the satisfying forms that literature and drama can take (e.g., here are two dissertations on the topic of shaping interactive dramas), what are other satisfying cohesive, meaningful forms — especially ones that are achievable with high degrees of freedom and agency? have they been identified? (here are some essays that I know of that begin to offer answers to these questions, that relate interactive experiences to the pleasures of architecture, instruments, recurrence, and gardens; and Nick recently asked, how are they lyrical.)

It’s interesting — it’s like I want there to be purpose and meaning and well-formed-ness (pacing, efficiency, etc.), to have lots of influence and affect (without it feeling like work, a chore), but without any feeling of being goal-directed, without the pressure that there is something I’m supposed to do, things in fact I must do for the experience to operate. Am I wanting to have my cake and eat it too?

Living in Chicago the past two years, I saw some amazing improvised long-form sketches, mostly comedy but the occasional dramedy. Watching them perform, I totally loved the freedom, the possibility and the sheer excitement of it. Here’s a book on how they do it. It’s hard. Can’t imagine a computer program pulling this off anytime soon.

Things brings me to Reality TV, particularly those shows that follow real people around in their daily lives, which may provide some clues and ideas. Here’s an excerpt from an essay by Dyske Suematsu about the recent reality show The Restaurant:

“The Restaurant” is not a sitcom, it is not a fictional drama based on a true story, neither is it a documentary.
It is an “unscripted drama”, a controlled chaos, where reality helps them write a script.
A fictional script is written after the fact, inspired by what happened in reality.
It employs reality-based script writing.
It is a piece of reality-based fiction.
It has no ultimate objective like many goal-oriented reality TV programs have (e.g. “Survivor”, “Trading Spaces”, “Blind Date”, “American Idol”, “Queer Eye for the Straight Guy”, etc).
Its only goal is to write a dramatic script inspired by reality.
It is a format that has a lot of potential.

It’s late, I’ll stop here; enough food for thought.

Creative Commons License The text of this blog entry is licensed under a Creative Commons License.

9 Responses to “That Darn Conundrum”


  1. Sean Says:

    Freedom, agency, story — pick two.

    That is an awesome insight. (I may have seen some of the individual A+B means no C, but I don’t think I’ve seen the 2 of 3 before.)

  2. Chris Crawford Says:

    Good points! I too like the “freedom, agency, story — pick two” as a clean, elegant expression of the conundrum, although I do not believe that the correct value is 2.00. My guess is that it’s somewhere in the range 2.3 – 2.6. So my clunky restatement of your elegant version is “freedom, agency, story — pick 2.45± 0.15.”

    I’m writing a book on interactive storytelling and I intend to write a chapter on Facade once I figure it out. In the meantime, I really should be spending more time here.

    Chris

  3. Grand Text Auto » Some Joe Schmo Was First to Experience True Interactive Drama Says:

    [...] enacting this vision of interactive drama, at least once, for a single person. I’ve alluded to the similarities between reality TV and interactive drama in the past, [...]

  4. IF in the ACM literature, Part Two « Emily Short’s Interactive Fiction Says:

    [...] hit two of an important triad in game design: agency, freedom, story. (Grand Text Auto had a long discussion on the challenge of writing all three of these into a game. It& [...]

  5. Grand Text Auto » GTxA Symposium: Future Directions Says:

    [...] ama before that — in addition to that potential, digital fiction can be a very hard, very challenging, and therefore very frustrating and rewarding problem to work on. Ther [...]

  6. Further reading on narrative construction « Emily Short’s Interactive Fiction Says:

    [...] comes back to traditional issues we’ve seen before: people re-discovering, for instance, the conflict between freedom, agency, and story. A lot of the techniques suggested (sandboxes where the player has to make his own story; branching [...]

  7. What is interactive story? | Immersive Story Says:

    [...] - Andrew Stern [...]

  8. What is immersive story? | Immersive Story Says:

    [...] - Andrew Stern [...]

  9. Can we turn the art of story over to the players? — Justin Gibbs Says:

    [...] try to create one. Not only it is a technical and interface problem, it’s a logic problem (That Darn Conundrum). But beyond even that, as an artist you have no idea where to begin. You know how to move an [...]

Powered by WordPress