x
Humans interpreting humans — LessWrong