Cluster-state quantum computing
Something I hear people discuss quite frequently is the question “What’s the big result of the last year or two in field X?”
My current answer to this question, in the field of quantum information science, is Raussendorf and Briegel’s cluster-state model of quantum computation.
This model tells you that, in order to quantum compute, it suffices to prepare a single quantum state (the “cluster state”), and then do local (i.e., single-qubit) measurements on that state.
Such measurements suffice to write in the initial state of the computation, the dynamical operations performed, and to read out the results of the computation!
In Debbie Leung’s memorable metaphor (memorable if you’re a Unix nut, anyway!), you simply “ping” Nature, and she computes. No “dynamics”, in the usual sense of the word, just ask questions of a suitably correlated state! Those questions don’t even have to be nonlocal; all the nonlocality is carried in a single, universal state.
Whatsmore, if you believe the Church-Turing-Deutsch thesis, you can efficiently simulate an arbitrary physical system this way.
I’ve spent most of the last two months thinking about this model, and the better I understand it, the more interesting and beautiful it seems.
(Disclaimer: I guess I should say that I’ve got a bit of a personal interest in advertising this work, since I’ve done some related work.)