Scenario analysis: a parody — LessWrong