||[Dec. 7th, 2005|09:15 am]
More poking around with The School today. I'm still having trouble figuring out what to do next, despite the great help I've gotten, and after skimming some plot-related entries in one of the writing books I picked up yesterday, I think it's because Ariel doesn't have any investiture in the plot yet. She doesn't have a very strong motivation. In the words of the book, it's a situation, not a complication, because it doesn't affect Ariel strongly, doesn't get in the way of anything she finds important. The problem is, developing the situation without resorting to stupid forms of motivation, like "I know the bad guy is evil (because the author told me), and we have to get out."
I'm beginning to understand why people make the mistakes I've been reading about, and nodding along with 'wow, that's a horrible way to do it.' Horrible makes these things easy. By playing into stereotypes, you can bring an instant affiliation with most readers, and by making the main character have plot sense at convenient times, you can set them up against the right people and get them moving without too much effort. They know the bad guy is bad, even if he isn't doing any bad things (allowing you to preserve the characterization of your villain).
Just like designing a computer program, writing an outline/developing a plot is a delicate balancing act. You have a set of hard requirements that the result has to fulfill, but there are a lot of tradeoffs to be made which don't compromise the requirements. For example, diagnosability is important, but how much error logging is the right amount? On one side, adding too many error cases increases program size, length of time in programming, the size of the testing matrix, and can confuse the user to all hell. On the other, adding too few can compromise the hard requirements. The question is, where should the program fall? What's important?
Then, later on, you may find out that another approach to handling one of your hard requirements compromises your logging decisions. It becomes a huge freaking matrix with a lot of variables that can vary infinitely, and with complex, subjective equations that place requirements on what are valid sets of values.
Writing is very similar, though often there are no real hard requirements, only things that the writer wants, and what the reader will accept. Neither of these can be reduced to something mechanical.
I just need more thought, I suppose.