Will_Newsome comments on MathOverflow as an example for LessWrong