An on-again off-again idea I've been having about computer modelling/simulation of stories. Either used for the creation of novels, or roleplaying campaigns. There was a recent discussion over on the Hero boards about the Morphological method of creating a story based on the method used by http://www.cbc.ca/news/viewpoint/vp_strauss/20050420.html Fran Striker a writer of 60k words per week. The story is built around the characters. It's broken up in to 4 random lists that he took things from - Characters, Objectives, Obstacles and Solutions: ~He listed 44 potential characters ranging from bums to jugglers and playboys, virgins to slaves and governors. These people were in possession of 11 traits, honour, courage and love among them. ~They might be seeking relief from 13 things GÇô for example, bad habits, temptation, a threat GÇô or be seeking revenge against 14 different protagonists. He listed 40-odd things that could stand in their way GÇô a contract, a tradition, blindness, money, lack of influence, for example GÇô then portrayed 40-odd different ways to resolve this, prayer, logic, threats and sacrificing a friend among them. This can be applied to any genre, you would just need to modify the lists. However - for a simulation, I would want more than that. Mainly character links/hooks to each other and to events in the story. Also, I'd want events other than obstacles, and to break down events into external and internal (e.g. character development). For linking I was thinking some kind of investment stat - how much a character cares about the outcome of any given event - or how much they care about the fortunes of another character. If you view a book as a timeline, you could also break it down into acts similar to plays, television or movies - defining those acts as either pertaining to a major event or pertaining to a major location/setting. Anyhow, more musing later, in the meantime - here are some links to things I've found on the internet that are relevant to this subject:
—- CategoryNarrative