Matt_Simpson comments on The self-unfooling problem