Collected below are the conclusions Chris has drawn from his years of work with interactive storytelling, interwoven as lessons through the 2nd edition of Chris Crawford on Interactive Storytelling.
- Stories are complex structures that must met many hard-to-specify requirements.
- People, not things!
- Stories are about conflict, mostly commonly social conflict. Violent conflict is for simple stories.
- Puzzles are not a necessary component of stories.
- Spectacle does not make stories.
- Visual thinking should not dominate storytelling.
- Stories take place on stages, not maps.
- The overall quality of interactivity (human-with-human or human-with-computer) depends on the product, rather than the sum, of the individual quantities of the three steps. You must have good listening and good thinking and good speaking to have good interaction.
- Your designs should aspire to the ideal of metaphorically having sex with your users.
- Fast turnaround is always better than slow turnaround.
- The overall quality of an interaction depends on its depth as well as its speed.
- Combining non-interactive story with non-dramatic interaction does not produce interactive storytelling (was originally “Interactive storytelling systems are not ‘games with stories.'”)
- There is no conflict between process-driven narrative and interactivity!
- A storyworld does not offer the player a few momentous decisions; instead, it offers many small decisions.
- A story presents a singe, picture-perfect view of the truth; a storyworld presents truth in three dimensions, including the less elegant angles.
- When you can’t work through a problem, go over its head.
- Interactivity requires verb thinking.
- You must embrace the drooling monster of mathematics if you are to succeed in interactive storytelling.
- Don’t view interactive storytelling through old-media colored glasses.
- Branching trees suffer from geometric growth that always requires more work than designers anticipate.
- Numeric calculations are superior to Boolean calculations.
- Use bounded numbers in your models whenever possible.
- Use only blend and bounded addition when calculating with bounded numbers.
- The personality model must cover the behavioral range of your storyworld.
- Keep the personality model as small as possible.
- Achieve conciseness through orthogonality.
- The personality model must be derived from the behavioral universe of the storyworld.
- Don’t create personality variables specific to individual verbs.
- Build your personality model for utility in your storyworld, not accuracy in the real world.
- In storyworlds, the reward is the applause, not victory.
- Use scoring systems to reward players rather than using mandates and prohibitions that constrain them.
- Storyworlds need an absolute minimum of a hundred verbs.
- Constrain your storyworld to as narrow a scope as possible; that’s the only way you’ll ever be able to include everything that it needs.
Any one of these lessons could serve as a jumping off point for a blog post.
If anyone here hasn’t read Chris’ book I highly recommend it (extra credit to those who also read the 1st edition and create a concordance of differences).
Update: I just realized that the lesson list changed between editions. You can find the first list from the 1st edition on this page (guess I get the extra credit inadvertently).