Abstract The notion that theories of information and computation can augment and even complete thermodynamics has proven too enticing for many to resist, even though careful analysis has long shown that the notion fails. In so far as the results of this information-computation theoretic literature succeed, they are merely tendentious relabeling of mundane thermodynamics. When they go beyond it, they fail. The difficulties include an unsustainable conflation by Landauer’s principle of the dynamic probabilities of thermalization with the static probabilities of memory devices. The most serious failure is an enduring neglect of the import of thermodynamic fluctuations.
John D. Norton (Fri,) studied this question.