LO & the New Sciences LO5356

Kevin Dooley (kdooley@maroon.tc.umn.edu)
Mon, 5 Feb 96 07:20:16 -0600

Replying to LO5331 --

In message <823402275.14062.0@kbddean.demon.co.uk> writes:
> To further the application of Stu Kauffman's research to organisational
> design challenges, I'd like to add a piece.
>
> To the extent that the problem or challenge being faced is simple, to the
> extent that the space of possibility to be explored is small, to that
> extent it works to have singular, central control or a few very large
> teams.
>
> To the extent that conditions are the opposite, the independent actions of
> intelligent agents are needed. In these cases, the best solution is many
> small and independent teams working on the same problem or bits of the
> problem without much connection between teams.
>
> This seems counterintuitive for management. If the problem is difficult,
> the tendency is to centralise control. The research of Stu and SFI
> indicates that the opposite is the case.

I think this interpretation is correct... however I might challenge the
use of the word "simple" in the second paragraph.

I think the question is whether the issue at hand can be dealt with in a
reductionist fashion; if so, a mechanistic approach is appropriate, hence,
centralize.

To the extent that the issue cannot be dealt with reductionistically, an
organic, decentralized, agent approach is warranted.

--
..............................................................................
Geek Ratio = $$ value of computer/$$ value of car (>10.0--> Geek City)
..............................................................................
Kevin Dooley...U Minnesota...Industrial Engineering...kdooley@maroon.tc.umn.edu
New address! : http://www.me.umn.edu:70/1/home/kdooley

-Info: learning-org-approval@world.std.com or <http://world.std.com/~lo/>