[gb22]

Hi Scott

1. What do you think is the potential for emergence of a third computing model, apart from classical and quantum, which is sufficiently different in its fundamental computing paradigm from these? Do you believe biological substrates offer opportunity for computing which classical/QC cannot replicate?

2. If we achieve quantum supremacy, who do you think will be the biggest winners and losers, in terms of professional skill set and in terms of business models for private companies? Will physicists become much more influential if their ability to model and deliver new evidence (and thereby theory) and convert them into commercial products is advanced?

3. Would it be fair to say that d-wave machine is more an analog computer than a quantum computer; And do you think there is possibly a class of computing problems that might be better/more efficiently solved by analog computers than digital, and without requiring QC?

[ScottAaronson]

1. Quantum computing is the most powerful model of computation we have based on currently known physics---in the sense that anything more powerful would need to be based on new physics. From the standpoint of theoretical computer science, a biological computer is "just" a different way to implement classical computation, typically with very slow speed but very enormous parallelism. It might someday have practical advantages, but unless we're totally mistaken about biology, it's not going to solve any problems efficiently that are outside BPP (i.e., classical probabilistic polynomial time). For that you need a quantum computer.

2. Yes, if quantum computing is successful, I imagine that might be good for the careers of many of the people involved in quantum computing.

3. Regarding D-Wave, there's this weird tendency to get fixated on words and definitions (but is it "really" QC or "really" analog?) even after you've explained the reality of the situation. See https://news.ycombinator.com/item?id=17426023 for my current summary.

Again, I'm skeptical that analog computers will ever be able to solve any problems outside BPP---for that I think you need a QC. (I.e., if not for quantum mechanics, I would have believed in the Extended Church-Turing Thesis. :-) ) Whether analog computers will ever again compete with digital ones on the constant factors, leaving aside the asymptotics, for problems that people actually care about, is a harder and more complicated question to which I don't know the answer.