I guess I've been blurring the distinction between storytelling and gameplay. While admitttedly that it's my goal to combine the two, it can easily become confusing.

I'm trying to define the set of actions that the player can make with respect to affecting the story. These do not have to be available at all times to the player (in fact, it wouldn't make a lot of sense if done that way). I see them more as a set of logical conditions; if the world is in story state X, then the player has the option of turning the world into story state Y. I'm not too interested in all the little things that don't have an effect on the story; simple things like generally moving around are too insignificant to be considered.

The problem I've had is aiming too high to begin with; such as designing an RPG with multiple potential themes. In this case there would be many rules that could be enacted at once, and the whole thing becomes intractable (at least with my present knowledge).

Maybe my definition of rules is best explained with an example. Lets consider a simple story, such as the theme in fighting games where the player must defeat a series of contestants in order to win some ultimate prize. In this case, at any point in time there's only two possible story related outcomes; either the player loses the fight or they win. If the player loses the fight, the story is finished with a big "YOU LOSE!" screen, and if the player wins the story continues to the next stage. This is obviously not a very interactive story [smile]. Note that little individual actions, such as when the player's avatar throws a punch or blocks an attack don't have any affect on the story, hence they are irrelevant from the perspective of the storytelling system.

However, maybe this could be extended by adding in some other choices. The fighting game could be extended to have a concept of fighting honourably, for example. In this case, there's four possible story options; win the fight with honour, win the fight by dirty tricks, lose the fight with honour, and lose the fight while playing dirty. The story system could then be extended to possibly giving the player another chance if they lose with honour against an honourable foe, or making subsequent fights easier if the player performs with honour by allowing NPCs to join them.

I admit this example is a bit hacked together (off the top of my head), but that's the kind of thing I'm thinking about here; defining a really simple set of rules that could influence the story, but with enough scope that there's enough there to provide interactivity with respect to the narrative.

Why the whole world?

Are you going to have a ton of different endings based on seemingly important choices the player makes early on in the game?

Maybe think smaller. Like, on your fighting game example. Consider who your opponent is. If he's an amoral bastard, he is not going to care about these rules, and hence, they don't have to apply. Moreover, there may be a larger choice in the future that negates your reputation, whether you were honorable or dishonorable. Not every change has, or should, be a world-shaker. Reducing the narrative scope and gameplay impact of your choices does not necessarily reduce the impact they have on the player's perception of the game.

Some of the most effective "choices" in games, are ones where, regardless of what the player decides, the outcome is fixed and doesn't actually change the state of the game at all.

You've been talking about rules that apply to the story. How about specific rules that apply to specific situations in the story?

I think there should be a not relatively but _completely_ constant set of actions available throughout the game(s) generated by the engine. But of course they wouldn't all be available at any time because they would be dependent on what object you were using as a tool or trying to act upon. For example, you can't shoot anything if you don't have a gun. You can't unlock if you don't have a lock.

Quote:

Original post by Beige Are you going to have a ton of different endings based on seemingly important choices the player makes early on in the game?

No, not really. I don't see the need for having a different ending for every possible combination of choices. sunandshadow's wedge example is still a good workable approach; if you are exploring a single dimension of character (such as selfishness over selflessness) you could have three possible endings (one for selfishness, one for selflessness, one in-between) and steer towards the ending that's most appropriate.

Besides which, I haven't yet decided whether my story domain will even have an ending...

The change in the story state doesn't have to be severe; I'm sure an abstraction such as a sliding variable to represent story themes (such as the "amount" of selfishness) can work. I'm just exploring various different approaches.

Quote:

Maybe think smaller. Like, on your fighting game example. Consider who your opponent is. If he's an amoral bastard, he is not going to care about these rules, and hence, they don't have to apply. Moreover, there may be a larger choice in the future that negates your reputation, whether you were honorable or dishonorable. Not every change has, or should, be a world-shaker. Reducing the narrative scope and gameplay impact of your choices does not necessarily reduce the impact they have on the player's perception of the game.

I agree that not every change has to be a "world-shaker". But every choice has to affect the story in some way. If it doesn't, then I'm considering it irrelevant for my purposes and I won't be factoring it into my model.

Quote:

Some of the most effective "choices" in games, are ones where, regardless of what the player decides, the outcome is fixed and doesn't actually change the state of the game at all.

Nearly every case that I've encountered this in games I consider more along the lines of "cheating". I can see how some clever smoke-and-mirrors can mask the fact that player choice is irrelevant to the story, and I've seen it done quite well in a handful of games (Baldur's Gate II springs to mind), but you can't build an interactive storytelling system without choices.

Quote:

You've been talking about rules that apply to the story. How about specific rules that apply to specific situations in the story?

I'm sorry, but I don't understand what you mean. The design of specific rules for specific situations is pretty much what a logic ruleset would be based on. Besides, I think the problem with my approach is more deciding on the general framework than specific cases; it's quite easy to think of specific rules that apply in only one situation, but much harder to apply that to an enitre story framework.

I agree that general theory would help us more at this point than specific methods - i.e. the difference between strategy and Tactics. (Knew I could get a Fallout reference in there ;) )

Oddly enough I think the solution is comes from the writers. One of my stumbling blocks in writing is that I have trouble conceptualizing "the end"

I really liked the computer-chess analogy earlier. If we call each move in a chess game a "story-state" (defined as the actions causing and the resulting description of environment after them) and checkmate parallel the end of the story with checkmate, I think this is a great model for an interactive story.

Technically in chess, there are hundreds or thousands of ways to get to checkmate, but their does exist a clear ending. What's more, you have a built-in character arc! The closer you get to checkmate, the more important each move gets.

Take a fork move for example. This is when you use one piece to attack two opponents. Can't this be likened to a plot fork? Taken literally, say you run up on two enemies in game which triggers certain game responses - both fight, both run, one of the two runs - and then there are consequences based on which actually occurs.

Perhaps Im going overboard, but if you can come up with the inter-locking "story-states" that put value to actions and their resulting states (after which NPCs can make decisions) I think you can make interactive stories work.

Quote:

Original post by Trapper ZoidI agree that not every change has to be a "world-shaker". But every choice has to affect the story in some way. If it doesn't, then I'm considering it irrelevant for my purposes and I won't be factoring it into my model.

This is what I'm trying to say. These changes do affect the narrative, the player's story, regardless whatever system you've put behind the scenes, because the player's actions are affected by the context you've created. You may be doing yourself a disservice by considering that sort of thing irrelevant.

Quote:

Nearly every case that I've encountered this in games I consider more along the lines of "cheating". I can see how some clever smoke-and-mirrors can mask the fact that player choice is irrelevant to the story, and I've seen it done quite well in a handful of games (Baldur's Gate II springs to mind), but you can't build an interactive storytelling system without choices.

Yeah, I know what you mean, but are you seeing this as cheating from the perspective of a gamer that's played a lot of RPG's and knows how "the system" works? As cheating from the standpoint of a game designer who's analyzing a work? A game writer that there's less going on than meets the eye?

What about the player who's willing to suspend disbelief because of the strength of your narrative, and who won't even notice that you're bending the established rules you've created?

Maybe they will notice, and appreciate the twist. The Metal Gear Solid series does stuff like this, though there's so much fourth-wall breaking in that game that I think the effect gets lost in the muddle.

Quote:

I'm sorry, but I don't understand what you mean. The design of specific rules for specific situations is pretty much what a logic ruleset would be based on. Besides, I think the problem with my approach is more deciding on the general framework than specific cases; it's quite easy to think of specific rules that apply in only one situation, but much harder to apply that to an enitre story framework.

Why do you need to apply this to the entire story framework? It may make it more interactive, but does it really make it any more effective?

Quote:

Original post by Beige Why do you need to apply this to the entire story framework? It may make it more interactive, but does it really make it any more effective?

If you are thinking in terms of writing a program to generate interactive stories, it's actually more difficult to aim for inconsistency than to aim for consistency.

What don't you like about the idea of a standardized theoretical framework for interactive story gameplay?

Quote:

Original post by Beige Why do you need to apply this to the entire story framework? It may make it more interactive, but does it really make it any more effective?

As far as I know, this is kind of what the makers of Facade did. The created little chunks of "plot" and then specified the connections between them. The result is 800MB of game data and years of work for a game that lasts for only about 15 minutes. Admitedly there is sigificant replayability, but this is the downfall of not creating an entire framework. What they created an efficient method of enumerating plot points while removing duplicates and near duplicates, and then proceeded to write all those story pieces.

Façade boasts 20,000 lines of dialogue, and "each line of dialogue and each story beat can be performed in multiple ways, in multiple emotional tones, and in multiple tension levels."

Not to say that there is anything wrong with Facade or that their method is stupid, or anything like that, but that's a lot of content creation for a 15 minute story unless you can replay it a lot of times.

Quote:

Original post by Vanquish ...

What you're saying is pretty general, but it depends how you generate the "story-states". Facade's method was to hand-craft each one. Personally, I want a method that works more like the chess example, where each state follows from the previous according to some rules.

tj963

Gamers will either play these sorts of games to experience everything the game has to offer, to become immersed in the experience, or to perform flawlessly.

Those that pursue the first and last path are going to end up trying to see through the systems in order to achieve the optimum outcome. That's perfectly natural, but a standardized framework is going to suffer from being... standardized, and a framework. I'm concerned that the single-minded focus on overarching interactivity systems will generate games that end up just being predictable and not well-suited to conveying anything but the most cliche narrative.

Wombah, I'm afraid I'll have to quote you on this one.

Quote:

Quote: What is the dramatic purpose of interactive storytelling? I'm not sure I understand your question Beige. But I have this creeping feeling that if I can't come up with the correct answer it means the system is worth squat in reality... My goal with this kind of interactive story system is to be able to recreate and surpass the multilinear games seen so far (at least in terms of interactivity) with several magnitudes less work. Not sure how that relates to the dramatic value though.

Look, I think the field is quite interesting and needs to be explored, but in a way that's conscious of a story's individuality. While I'm sure that's what some of you had in mind anyway, it's rare that it's actually said in these conversations, and rarer that the conversation makes it self-evident. This leads me to believe that a lot of the people working on interactive storytelling, an idea so tightly wound into dramatic narrative, aren't even considering it from that perspective.

[Edited by - Beige on October 25, 2005 3:52:24 PM]

One of the things I'm seeing here is that most of the discussion seems to be going on for single player type games. While interactive storytelling is all good and well for single player games I think the true power is in multi player, particularly MMORPG games.

The reason behind this idea is there usually is no particular end to a MMO. So why not have the players input affect the current state of the world. To this affect, you can have the NPCs all have various desires and they can buy the help of the players to meet these ends...