Murphy’s Quest Ch 6: Perverse Incentives — LessWrong