It’s hard not to picture vacuum-tube experts circa 1940 laughing about transistors. Widely desired - theoretically possible - never practically manufactured. Even a decade later, production was halfway to alchemy.

A computer scientist should know it’s about science - not computers. We knew how to use quantum machines long before any existed. When we figure out how to juggle any serious number of qubits, we already know there’s a few weird problems that will become hilariously easy. No shit we’re not there yet. If you had to build a computer using the several transistors which existed by late 1948, of course that device would get walked by any room-sized, crew-serviced, tape-eating series of tubes. (Valves, for our Commonwealth readers.)

The difference here is that quantum computing has no obvious utility beyond these few weird problems.

Seriously. Matrix operations on video cards will prove to have greater impact on consumer and commercial computing than anything quantum. Forever is a long time to bet against, but I expect decades of head-scratching dead ends, after we build these things. Quantum algorithms are like hideous math proofs. If we get some kind of bin-packing decision problem and accidentally fake our way into P=NP, that’ll be awesome, but otherwise - eh.

«PQC…isn’t mathematics or engineering, it’s augury: ‘A great machine shall arise, and it will cast aside all existing cryptography, there shall be Famine, Plague, War, and a long arable field.’»

🤭