Theorizing About Future Machines

Isn’t it incredible that human beings can conceive of things in the abstract years before being able to build them in the physical world? Sometimes, we come up with an idea for a machine years, decades, or even centuries before technology advances to the point that we can construct the required components to assemble them. The classic example of this is the computer. Turing came up with the Turing machine in 1936, but the first working general-purpose electronic computer wasn’t built until 1945 (it was called ENIAC). Von Neumann came up with a standard computer architecture in 1945, on which our current computers are very much based, but the first one wasn’t built until 1948. I was surprised (and then surprised that I was surprised) to find that the same thing has been going on in quantum computing. Hadamard gates and Toffoli gates are important building blocks in that field. The Hadamard transform was proposed by the French mathematician Jacques Hadamard in 1893. The use of the transform to propose a quantum logic gate occurred in the 1980s, but the first one wasn’t built until 1995. Similarly, the Toffoli gate was theorized in 1980 by Tommaso Toffoli, but the first one wasn’t built until 2002. Science is often presented as empirical, namely as a subject in which we empirically observe, then hypothesize and theorize, and then confirm through further experimental observations. Computer science often acts like mathematics, in which we theorize, conjecture some more, and then years later engage in experimentation and engineering. I happen to think that is amazing.