Scott Alexander comments on Proofs, Implications, and Models