Analogies and General Priors on Intelligence — LessWrong