Today in my quantum cryptography class I was drawing parallels between the development of classical computers and their quantum counterparts. I am of the opinion that there will need to be, in the language of Thomas Kuhn, some sort of paradigm shift that brings about some discovery that is a bit like the quantum computing analogue of the transistor (not necessarily in the specific logic-related sense, but rather in the scalability sense). Apparently a group in Britain is inching closer to that goal (though they’re not quite there yet) having now developed an optical quantum controlled-NOT gate for processing individual photons that resides on a chip of silicon a few millimeters across. The gate consists of six parallel silicon waveguides that are brought close enough together at certain points (less than a wavelength of the light passing through the guides) such that the photons can leak out in a process known as evanescence, thereby entangling two of the photons.
Previous optical quantum CNOT gates have required large lab benches consisting of beam splitters, mirrors, etc. As such, this is one small but crucial step in the quest for scalable quantum computing. There is some work to be done, of course, even on this particular setup. For example, the photons only become entangled about 1/9 of the time. Nonetheless, additional qubits can be used in an error correction scheme in order to clean up the results.
While I don’t want to get overly excited, we need to remember points such as this. As an occasional historian of science, I would like to remind people that things like this often turn out to be crucial and it is fascinating to be able to say sometime down the road, “I remember when _____ happened back in…”