I've been thinking about the ideas that Laura and I came up with last week in Colorado. I am now looking askance at the background displays, for a variety of reasons. First, the existing "Customizer" dialog boxes do all the essential work that the background displays would do. The background displays bring only two additional features to the party: the ability to take notes on characters, and the proper use of scroll bars instead of those handicapped things I now use. In return, we'd lose the ability to tab through the fields of the display; I think that's a significant factor. The notetaking facility would be nice, but I don't see it as essential, and this feature would consume lots of my time, which is in particularly short supply just now. I can always add this feature to the dialog box if we decide we need it.
This leaves me a tad embarassed. Laura and I had discussed this at great length and we went through a great many possibilities. I am convinced that additional help is needed for the beginning storybuilder, but I don't see this as the way. However, there was another element that we agreed upon that I would definitely like to flesh out. This was a provision for defining the basic web. Our idea was that the storybuilder would first establish a theme, then imagine an ensemble of stories on that theme. Then the storybuilder would write down two lists: a list of "launching storyatoms" and a list of "climax storyatoms". Then the trick is to challenge the storybuilder to imagine many different ways to connect the items on these two lists.
So far, the exercise I have described can be carried out on paper just as easily as on the computer. We need two more elements here: a provision for increasing detail, and something that takes us directly into the New Verb option.
I am reluctant to create some sort of fill-in-the-blanks screen that then transfers the results to a New Verb option. This strikes me as something that just begs for bugs. What happens if the beginner doesn't fill in all the details? Does the Erasmotron discard his input? Is it not likely that the beginner will change his mind? What happens in that case?
We're taking the wrong approach here. We don't want to replicate the functions of the Erasmotron in an easier-to-use format; if there's an easier way to do it, it should be in the primary architecture, not in some training-wheels version. The point and purpose of this design effort is more along the lines of an outliner, something that allows the beginner to see the structure of their work from a distance.
OK, what if we dispense with the notion that the information entered by the beginner will automatically be transferred into the main display. Instead, let's think of this facility as something like a note-taking facility, something that allows the storybuilder to keep track of their basic plan. We want to provide some sort of organizational structure here; if the computer can't do it, then we should just let them do it on paper.
A digression: multithreading is now standard on all PCs. There's no reason why a storybuilder couldn't keep track of things using a separate program and then just context-switch as needed. For example, if there were a use for elaborate word processing capabilities, I'd be a fool to try to build a word processor into the Erasmotron. Better to let a genuine word processor do word processing. The value of building data-editing into the Erasmotron comes when that data can be directly integrated into the larger structure of the program. Stick to the knitting.
So, getting back to the main track, how can we come up with a storyweb outliner? If I can't answer this question, I think that I might as well give up on this project, because if I can't explain the creative process of storybuilding in a rational fashion, nobody else will be able to do it.
I'm going to walk through this process by example rather than abstractly, because I am unsure as to the specifics. Perhaps by doing so I can learn something. Here we go:
My topic is adultery.
Next comes my theme. The theme can't be too narrow. For example, the theme "Adultery is a bad thing" is too narrow. The theme should be about a collision of desires, and every collision has repercussions that bounce both ways. Thus, in the forgoing example, the positive illustration of this would be the spouse who rejects temptation and lives happily ever after. If this were the actuality, then adultery would not be so vexing an issue. What makes it a collision is the fact that the spouse who rejects temptation will always wonder at the opportunity that was missed. Sexual tension between the partners might well rise as the saintly spouse realizes that the marriage relationship lacks the sexual passion that beckoned in the affair. Stories are about hard choices, not easy ones; if a choice is hard, then there must be negative repercussions with either path. The real point of any story might be reduced to: "Kid, someday you're gonna have to choose between A and B, and it'll look like a tough choice, but you really better take path B. Yes, path B has its downside, but path A's downside is worse in the long run."
So now my theme has become something like "Adultery represents a collision between marital duty and sexual desire". No, we need to reformulate this in more colliding terms: "Adultery represents a collision between marital duty and sexual growth." No, how about this: "Adultery pits love for spouse against sexual learning." The idea here is to delineate the internal strife represented by the choice.
OK, I like that theme. Now, how do I get to the network from here? It seems natural that the next step is to specify some climaxes that represent varying degrees of success or failure for each of the two competing drives. Here's my stab at the adultery case:
Sexual satisfaction, happy marriage retained at price of self-delusion
Sexual satisfaction, secrecy maintained, slight tinge on marriage
Sexual satisfaction, but marriage undergoes painful crisis that makes it stronger.
Sexual satisfaction, but marriage slowly disintegrates
Sexual satisfaction, but marriage suddenly collapses
Sexual satisfaction, but severe guilt triggers internal crisis
Sexual satisfaction, but illicit partner demands that adulterer get a divorce
Sexual satisfaction, but spouse retaliates with counter-affair
Partial sexual satisfaction plus all of the above possibilities
Unrewarding sexual experience, plus all of the above possibilities
Slippery slope to near adultery, plus many of the above possibilities
Nonadulterous dalliance, plus some of the possibilities above and below
Strict fidelity, but sexual resentment destroys marriage
Strict fidelity, but sexual resentment weakens marriage
Strict fidelity triggers sexual demands that lead either way
Strict fidelity followed by counter-affair by spouse
Strict fidelity, rewards from grateful spouse
This is a pretty good group of results, and I don't think that I've explored the possibilities deeply. There's plenty more where this came from. So where do we go from here?
The next step, I think, is to list some openings. These will not be as dramatically important as the resolutions, but they provide a useful intermediate step in the outlining process. All the openings are essentially the same: guy meets gal. The only variations possible are in the context of the meeting. Guy meets gal in bar, at the bowling alley, in the supermarket, at work, etc.
From there forward it's just a matter of tracing the many possible paths of seduction, and here I see gobs of opportunity for storyatoms. But what's the general process that can be used in all this? I think it's a matter of drawing links from beginnings to endings. How many different ways can you get from those beginnings to those endings? There's initial flirtation, dalliance, getting-to-know-you, exploratory behaviors, preliminary behaviors, greater intimacy, leading up to actual adultery.
There must also be a group of storyatoms dealing with the sexual frustration that drives the player toward adultery. These cover the player's relationship with his/her spouse, its daily grind and boredom, its lack of passion. These develop even as the potential affair brews.
Wait a minute! The idea here is NOT to draw lines connecting beginnings with endings. The idea here is to imagine intermediate states that could conceivably exist between any pairing of beginning point with endpoint. In other words, we don't ask, "How can we get from GuyMeetsGalInBar to RetaliatoryAffair?" Instead, we ask, "What are interesting moments between the two?" Thus, we certainly must have that critical moment where the player makes a decision. Even so, such a critical moment need not be confined to a single event; there are a thousand tiny steps to infidelity. Can we list such steps? Fantasizing about illicit affairs, serious eye contact, dancing together, private but nonsexual meetings, being together in the presence of a bed, and so forth.
I think that I have gone beyond my target. My goal in this exercise was to step through the creative process by which a storybuilder creates a web, and I think that I can now see the "Adultery" web clearly. The crucial steps were:
1. establishing a topic.
2. determining a theme, expressed in terms of the internal conflict that the player is meant to experience.
3. listing many possible outcomes of such a conflict.
4. listing many possible openings for such a conflict.
5. listing many possible intermediate points.
6. flesh out the web
In other words, we ask the storybuilder to create a bunch of endings, a bunch of beginnings, and a bunch of middles. Hey, interactive storytelling doesn't dispense with beginning, middle, and end -- it has lots and lots of them! Once all those beginnings, middles and ends have been created, the storybuilder starts linking them up, including additional twists and variations.
Now really, what could be simpler? And what can the computer bring to this exercise? Right now, this exercise looks easier done on paper than on the computer.