I was a Ph.D. student at Caltech from 1974 to 1977, and the Univ. Libre de Bruxelles from 1977-1978. The summer prior to arriving at Caltech, I read Jaynes’ then-little-known but now classic paper “Information Theory and Statistical Mechanics”, and was enamored. After starting graduate school, I saw how Jaynes’ insights could be generalized to quantum mechanics, and could give a physical basis for resolving a conundrum of classical physics, why the electrons in an atom do not collapse onto the nucleus. I developed this into a Ph.D. thesis on “Information Theory and Quantum Mechanics”. My thesis was presented to Caltech in 1978, but not accepted, and the papers which flowed from that were consistently rejected for publication. With the advent of the LANL e-print server these can now be made available. I've rewritten the first three papers, and posted them there:
The Boltzmann/Shannon entropy as a measure of correlation
Non-linear connections on phase space and the Lorentz force law
More are in-the-works, and should appear as time permits.
Also, I have placed locally a small footnote from the introduction to my thesis:
What Laplace really said - the myth of Laplacian determination
Robert Parr (U. North Carolina) and his student Stephen Sears also explored the relationship between information theory and quantum mechanics, in the late 1970’s, independently and shortly after my work. Now this area has become very fashionable, with a Gordon Conference, a journal, and even an NSF-supported institute at, of all places, Caltech.
The Economist magazine had a contest in 2000, to describe ‘The World in 2050’. I entered that contest with the following essay, which did not win and has never appeared in print.