Scott Aaronson I think himself would agree, as do I. He has some legitimate counterpoints though too on NP-Hard problems being solved. He points out the canonic

Scott Aaronson I think himself would agree, as do I. He has some legitimate counterpoints though too on NP-Hard problems being solved. He points out the canonical instance of such problems "folding protein" (chemical dynamics) can not really count, as proteins that misfold (pryons) are in extremely dangerous molecules: evolution has heavily selected to find proteins that it can fold quickly and reliably. He'll also point out that rocks don't find global minimums - that there are in fact a plethora of these examples. He tests soap bubbles and finds many 'easy' instances where it fail. Namely, he's responded to that criticism countless times - once with an actual physical experiment (the paper itself announces). It very likely is interesting to see to what degree other physical processes can approximately compute otherwise hard problems. It's just not fair to pick Aaronson to do it; he himself would be the first to admit that it isn't his area and that you'd want someone like Champagne to do it. In summary, yeah definitely his work is on a very theoretical and abstract level. He proudly admits this. Nobody would be more excited about others performing experiments than he would be.