The author's objection to proof-of-stake seems to be based entirely on some ostensibly-inherent vulnerability to the nothing-at-stake problem, but at least one

The author's objection to proof-of-stake seems to be based entirely on some ostensibly-inherent vulnerability to the nothing-at-stake problem, but at least one consensus protocol¹ has had explicit mitigations against that vulnerability (and numerous others) for almost half a decade now, and I'd be very surprised if other protocols haven't adopted any mitigations at all. ¹: https://eprint.iacr.org/2016/889.pdf https://eprint.iacr.org/2016/889.pdf