2011/6/20 Smári McCarthy <[email protected]>:
This is a common understanding of self-regulation, but really it's doublespeak.
A system is self-regulating if and only if it possesses the capacity to control its state in entirety.
I contend that, as an extension to Ashby's law, a system can only self-regulate if its regulatory overhead is zero.
Ashby's law roughly states that only variety consumes variety, or, in order to control a system with a certain amount of variety, the control system must have at least as much variety. For system X to control system Y, V(X) >= V(Y).
However, this reading is somewhat naïve, as the system X actually "spends" some of its variety on regulating system Y. We can call this the "regulatory overhead", and it is roughly equal to V(Y) in the trivial case: Unless the regulation of a system is the intent of the system, its purpose will be less well served by the regulation.
So, it stands to reason that if a system X is self-regulating, then one of these will follow:
- X's sole purpose is self-regulation (i.e., it's a trivial self-regulating system) - X does not manage to capture its entire variety, meaning that it is only partially self-regulating. - X's regulatory overhead is zero.
What does it mean, when the regulatory overhead is zero, then?
It means that either V(X) = 0, or, that all of the regulatory work produces beneficial byproducts. i.e., the system, by virtue of self-regulation, achieves other goals as well.
Whether this is possible, I do not know.
I have not yet seen a self-regulating system. The universe expands and cools, the human body whithers and dies. However, a town can be governed by a state, which can be governed by the laws of physics. There is sufficient variety in the Universe that it cools slowly - its velocity of deregulation is low.
Self-regulating markets then either have accomplished zero regulatory overhead, or, they yield no nontrivial results (zero surprisal; 0 bits of information output), or, they are leaking variety, in the sense that they are not actually stable, but are deregulating at some rate.
d(R(X))/dX > 0
Just thinking out loud...
- Smári
I think you are right, Smári. There are no fully "self-regulating" systems. Even where corrupt organizations have imposed influence on governments, the activity itself exists in an environment that is in dynamic state. The dynamics may simply take longer to make themselves apparent. For instance: A popular business man establishes a perception that natural gas is the answer for the United States to impending oil shortages http://en.wikipedia.org/wiki/T._Boone_Pickens#Natural_gas A company or companies may push the government to allow for http://en.wikipedia.org/wiki/Hydraulic_fracturing to harvest natural gas from shale deposits. The company perhaps succeeds in influencing the government to de-regulate this type of activity. http://www.huffingtonpost.com/2011/01/03/fracking-pollution-in-wat_n_803737.... However, the activity rapidly ends up affecting people not directly involved with either the regulating parts of the government, nor the corporations. Namely, the affected are the people who live where the Hydraulic fracturing is taking place. Their ground water (drinking water) becomes unusable. Now, the company and government may find themselves dealing with a set of factors that possibly were not foreseen, and are not easy to control, as people who live in the region become aware of what is happening and work to resist it in effective ways. http://adoptresistance.blogspot.com/2011/02/buffalo-bans-fracking-in-groundb... -- -- Sam Rose Hollymead Capital Partners, LLC Cel: +1-(517)-974-6451 email: [email protected] http://hollymeadcapital.com http://p2pfoundation.net http://futureforwardinstitute.com http://socialmediaclassroom.com "The universe is not required to be in perfect harmony with human ambition." - Carl Sagan