And of course there's the venerable http://www.fourmilab.ch/hotbits/ Getting back to the arguments between uncertainty in physics and undecidability in mathematics (assuming this topic is also FUN): To differ a bit: I also used to readily discount this alleged connection as mystical, but lately I'm more sympathetic to a further dialog. Don't we only believe these hardware generators to be "truly" random to the extent that we believe that the empirical evidence supports the theory that says they "truly" are? And, how is an experimental outcome being "truly" unpredictable, given all physical observables and applicable laws, different, in principle, from a theorem being "truly" undecidable, given all axiomatic derivables? A possible (and now conventional) model of reality essentially consists of a Platonic universe in which axiomatic inevitables reside (such as the mathematical laws of physics) plus this bizarre additional inscrutable mechanism that arbitrarily squirts out "truly" random bits that then determine (via the laws) all the otherwise theoretically-uncertain physically observed actualities. But it would seem another possible model is that there is only the one kind of universe, and this supposedly "truly" random extra component is merely akin to the inevitable additional "sporadic facts" that we know must appear in any derivational system. Until the relationship between these apparently different models is satisfactorily clarified (say by showing them either equivalent or experimentally differentiable) there will remain an uneasy urge to shave off that protruding random bit spigot with Occam's razor. So I feel that in its introductory motivation the paper may implicitly exaggerate the plausibility of the random-spigot-free model of reality somewhat, but the result about the structural similarity of the Platonic and physical uncertainty laws is independently interesting and, to my limited understanding, at minimum contributes something to this dialog.