@akkartik i think your prescriptions have to be guided by ones (conditional) prescriptions. i’d not see a doctor who’d prescribe a drug she wouldn’t predict would meet a risk-benefit threshold. 1/

@akkartik being the change you want to see in the world is great, but not immune. lots of time there’s limited harm / cost / risk in it, so even if your predictions are pessimistic, a small chance of contributing to a virtuous network effect is worth it. especially if there’s a community that might make the same calculation. what complicates social choices is social outcomes are very difficult to predict, and small initiatives unpredictably yield avalanches! 2/

@akkartik so i’m definitely kind of with you in practice. i devote much of my time to enterprises i’m predictively pessimistic about that are actually very costly to me. but it’s precisely because i predict the predictiveness of my pessimism is limited that renders the practice quasi-justifiable! (it strikes me as quite analogous to to the famous irrational overconfidence of entrepreneurs.) /fin