Dissolving Confusion around Functional Decision Theory — LessWrong