[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: process vs. data (Bob's comments)

>        For my own interests I am closer to simulation, and hope to see
>the world provide sub-plots among the population around the player due
>to an object oriented interaction between player and actors and actors
>upon other actors. This would not result in a structured plot for the
>player in a first person sense, but rather something more akin to the
>wandering of Faust.

This is almost certainly the most productive and interesting approach at 
present. In general outline, what you're describing sounds similar to 
Erasmatron and also a bit similar to Joe Bates' Oz/Zoesis worlds.

As a consumer of interactive entertainment, I'm more interested in worlds 
than in stories. I'm starting to believe that in the interactive domain, the 
story exists to enhance the world, not the other way around. Interactive or 
non-interactive story can enhance an interactive world, though in different 

Which, by the way, brings me back the question this thread started on (Dan's 
re-interpretation of Ken's original two questions): How can interaction make 
a story more compelling? In my world-centric viewpoint, that becomes a 
secondary issue. My answer is that an interactive story is better than a 
non-interactive story at making an interactive world more compelling. An 
interactive story is more amenable than a linear story to being told via the 
medium of an interactive world.

Creating interactive worlds, in my opinion, is an artistic end in itself. 
Experiencing an interactive world engages human faculties and emotions in a 
different way than any other medium. Interactive worlds are thus a 
distinctive art form, not to be regarded as an extension of or a derivation 
from or an "improvement on" any other form.

>... in favor of yet another limited approach...

There's nothing wrong with limited approaches. Every approach that's actually 
put into practice has to be limited. All I want is for creators to keep the 
limitations of their chosen approaches, and where those limitations fit 
relative to the whole issue, in a corner of their mind. The difference 
between a limited approach and a dead end is whether or not you remain 
interested in ideas from other areas.

I'm working on data-oriented approaches because that area seems to be 
neglected currently, and I have a few ideas that may prove applicable to it. 
The main goal is to invent ways of breaking up a story into smaller units of 
data while retaining as much "story-ness" as possible in each piece. 
Chronologically contiguous segments (of any length or any degree of 
abstraction) of conventional narrative action just won't do; they lose their 
story-ness long before they get short and simple enough to be manipulated by 
an interactive process. (Although Doug Sharp's Dramaton approach should not 
be discounted.) So I'm looking for other ways besides the narrative stream 
itself to encode stories as data.

>        I suppose an ideal experience would be stories experienced in a
>truly first person mode, aspiring to the emotional power of Poe. But,
>how can a player's actions result in this without persistent direction?

I don't think anyone knows. This is the problem of "how to make a simulation 
more narrative" in a nutshell. An equivalent idea, "narrative guidance," was 
a buzz word among the Media Lab folks a few years ago, but I never saw 
anything workable come out of it. 

>        As I mentioned in a previous post, it seems to me that
>storytelling is a cooperative creative effort. If computer-aided,
>interactive storytelling is going to give the reader/player more
>control, perhaps it will be necessary to also confer some of the
>writer's responsibility. Perhaps, using a system founded in simulation,
>it will be necessary for a player/reader to create self-plot in the
>'storyworld.' This would be a system not of direction but of cooperative
>volition; the player who fails to write her own story will yet fail to
>find it, but may still enjoy the simulation.

I've suggested an approach in which the world maintains a database of "open 
conflicts" or "open subplots" and monitors for opportunities to resolve them. 
Ideally, multiple (but finite) possible resolutions would exist per open 
conflict. The resolutions of each conflict are authored in advance, so there 
can be at least some narrative thought/data embodied in each of them. The 
player is not directed but if she happens to create the conditions for a 
resolution of an open conflict through volitional actions or even just random 
mucking about, a resolution event kicks in. The conflict object might also 
incorporate mechanisms designed to nudge the world state (not the player!) 
toward possible resolutions. I tried to convince Chris to include an object 
type suitable for authoring "open conflict" objects in Erasmatron v2 but the 
technical overhead would be enormous for such an unproven concept. I admit I 
haven't done anything further with this idea since.

Some games have already been evolving in this direction. Zelda 64, for 
example, appears to advance its story lines in this manner, the _big_ 
difference being that there is only one possible resolution per subplot.

- Walt