In the last selection we found Ashby making what may strike us initially as a surprising inference. Starting from the assumption that “an essential function of as a regulator is that it shall block the transmission of variety from disturbance to essential variable” he draws the conclusion that “the regulator’s function is to block the flow of information”.
Ashby’s reasoning at this point caused me to do a double take, because I normally think of information as a resource for reducing variety, in other words, the dispersive quality of entropy. But a little reflection convinced me Ashby is making sense here, so long as we read him right.
Jack Ring’s suggestion, “Consider that the regulator blocks information that is detrimental to the system mission”, serves to point us in the right direction. Strictly speaking, though, it is not the information about temperature variation that is detrimental to the system’s mission but the temperature variation itself. The regulator acts in such a way as to block the information about variation, but solely as a side effect of damping the real variation.
But we need to keep one thing in mind. When we speak of the regulator blocking the flow of information, we are talking about the whole system as a “black box”, where the net information flow from input to output is as low as possible. When we turn to a finer-grained analysis of regulated systems we will see that all sorts of information has to be processed inside the system in order to achieve its mission.
- Ashby, W.R. (1956), An Introduction to Cybernetics, Chapman and Hall, London, UK. Republished by Methuen and Company, London, UK, 1964. Online.