I'm also looking forward to part 3. A good counterpoint to some of the concerns in the series is https://cacr.uwaterloo.ca/techreports/2015/cacr2015-01.pdf htt

I'm also looking forward to part 3. A good counterpoint to some of the concerns in the series is https://cacr.uwaterloo.ca/techreports/2015/cacr2015-01.pdf https://cacr.uwaterloo.ca/techreports/2015/cacr2015-01.pdf. This article shows "several examples of attempts to avoid random oracles that have led to protocols that have security weaknesses". Of course, that article is only focused on more classical constructions and not on the newer SNARKS/SNARG/STARK and other constructions in the zero knowledge zoo. So there isn't really a disagreement, but we probably shouldn't ditch random oracle based constructs in use today.