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

Re: Interactive Drama: Why I've lost interest



When the player starts
angsting about "how many hexes away is that?" there's a problem
delivering story.

Stories would be much easier to tell with First Person Shooters because
the mechanics are much closer to drama.

I agree, and I think one of the problems with games today is that they're to.. physical. We have plenty of action movies. What we need are more dramas, and dramas are often based heavily on dialogue. Now, I don't expect us to suddenly develop advanced AI for having fully human-like conversations any time soon, but I do think that we can make some palpable progress on that front, that stylistically abstracted conversations could become central to a game the same way stylistically abstracted fighting is. The problem with conversation engines these days is that they're pretty much all human-scripted, which really limits you.


Essentially, conversation-wise, we're at the stage of early games where you could "FIGHT" "DUCK" or "RUN". I don't expect it to be like a real martial arts fight, but if we can get to the point of Street Fighter combos or even FPS tactics... in dialog form, we'd be getting somewhere.

There's been some work in gimmicky poetry generation and the like. My favorite is this:
http://www.elsewhere.org/cgi-bin/postmodern/
(reload to get a new randomly generated post-modernist essay every time!)


And my idea is that we can apply some of those research principles to generating dialogue that's more exciting than repeated "Welcome to Corneria!": http://www.nuklearpower.com/daily.php?date=010412

Again, simulation, not story.

I'm sorry if I wasn't clear. I'm not just interested in interactive narrative, but also in gaming as a communication medium for expressing ideas, and you can express ideas in more ways than just straight fiction. I guess you've caught me in that I am definitely biased toward simulations, and I'm actually rather interested in using story to serve simulation, in addition to vice versa.


People escape to dramatic human interaction for a reason, to
relieve the tedium of their daily lives.  A dramatist doesn't want an
open ended system, he wants some kinds of human interactions and not
others.  He wants the interactions to be related in a way that's
psychologically fulfilling.

Point taken. We don't have to simulate truly basic human life... we can simulate extreme human life...


One of my ideas on this front was to make a game set in a auto-generated film noir world, which would have both a simpler subset of human interactions and also exaggerated emotional context to make it more interested than brushing your teeth and going to the bathroom. (Though the latter seemed plenty interesting to all the people playing the Sims!)


Oh, so there's a good example: Imagine if the Sims could communication in a richer vocabulary than just symbols.


Not neat at all. You now have the full burden of a research scientist.
Many of those situations are going to be very, very boring. People
think it would be 'neat' because they don't sit down and actually try to
write the damn thing. They aren't dramatists.

I really need to go play the Sims at some point.. It seems to me that much of what you do in that game is quite banal, and yet, people love it...


Do you feel you have a literary background sufficient to describe
"emotional response" in dramatic terms?  If not, why would you kid
yourself that you can get an AI to do something you know little about?
If you do, why do you think "width" of responses is what we all need?
Sounds like 4 million boring stories to nowhere.

Well, to go back to your earlier point about extraordinary circumstances, I'm imagining doing this in interesting situations, so that you'd act more like a normal person, perhaps, but in a situation that wouldn't happen to you in real life.


Well, here's a chance for you to prove your mettle.  Recently some
Italian guys conjured up a pure hypertext storytelling thingamabob.  It
doesn't do anything, it's just a simple hypertext linkage thingy.
http://www.terradif.net/online/cyoa/interactive/storyland/en/index.php

Go there and write a hypertext story.  Keep expanding all the nonlinear
branches you want.  See how much work it is.

I did actually write a hypertext story once, for a class. It was a multiple-perspective time travel story, but I controlled the branching by making it closed-loop time travel. The thing about all these hypertext things is that I don't consider them "non-linear" at all. They're no different, fundamentally, than Choose-Your-Own-Adventure. They are extremely linear, just linear along several branches. I prefer to think of them as "branched linear". Truly non-linear storytelling would require game mechanics to generate story dynamically, not just statically.


So in case you're interested, my story is here:
http://narrative.mit.edu/~kenlu/ci/
Be warned that it's kind of long, and I had a lot of changes I wanted to make to it, but that I never got around to doing. :P (For one thing, I was going to remove a lot of illusion of non-linearity by removing most of the linkages between perspectives, and limiting them only to certain points.) (Also, I'm not a very good prose writer since I don't read enough, though I've been trying my hand at some comics in my spare time.)


My point is, the ridiculous effort required for branched-linear stories is precisely why I'm interested instead in having authors spend more time on dynamic generation. I totally agree with you that branched-linear is never really going to go anywhere truly different.

Have you explored how to write a regular story?

Sure. See above. Again, I realize it's not particularly good writing...


-ToastyKen

---
Kenneth Lu
kenlu@subjunctive.net
http://subjunctive.net/