Heuristics and Biases in Charity — LessWrong