An Attempt at Logical Uncertainty — LessWrong