On the Resolution of Frightening Paradoxes and Inferences — LessWrong