I liked PG's essays better when they were more concretely rooted in the details of his work his work at his startup and at YC. This one feels like a sort of non-specific variant of Bulverism - diagnose a bunch of people without presenting evidence, then start psychoanalysing them (also without evidence).