Symposium on Complex Systems Engineering/Schedule/Thursday 2b comments
From CSWiki
M Kuras here:
As regards Kreitman's excellent paper...
No human individual has the “requisite variety” (in Ashby’s sense) to directly control a sufficiently rich organization (of people in the real world) regardless of what feedback loops are available and employed. This is “control of the first kind.” I wholly agree with Kreitman. (And I wish that I had seen his paper ten years ago!)
Many of his prescriptive measures to implement his “control of the second kind” align with the regimen of cSE. His measures are cast entirely in terms that are (on the surface) independent of the processes of evolution. (There are some interesting potential areas of divergence. This is wonderful.)
We need to investigate (in detail?) the alignment of his prescriptions and the regimen of complex-system engineering.
The regimen is based on an understanding of the processes of natural evolution. We need to consider whether or not such an understanding is necessary. Or whether a foundation based solely on “control theory” is sufficient. My opinion is that control theory (even as elaborated by Ashby) is entirely a single scale articulation. It is not sufficient to provide a sound “theoretical” foundation for the engineering methods Kreitman and I both seem to espouse. (This is so regardless of the soundness of the motivation for a second kind of control that Kreitman provides, which is based on an investigation of the implications that can be drawn from Ashby’s control theory.)
There is an amazing congruence between Kreitman’s work and that of Browning and that of the regimen of cSE. Great minds think alike – and fools seldom differ. We need to figure out which it is in this case.
I would hope that we follow Doug Norman’s suggestion – and examine whether or not evolutionary processes must be understood first, and how we can parse problems that merit “control of the first kind” and those that merit “control of the second kind.”
If the remaining papers are as rich as Kreitman's I will be intellectually exhausted. (I already am, actually...)
