This is a really good presentation! The idea that there are emergent details at the large scale not capable of reduction is a big one. The point about counterfactuals not being valid was equally big.