Brendan Long comments on LessWrong 2.0