Re: Business orgs and Re: Decentralization

[Rick Riolo <rlr@maria.physics.lsa.umich.edu> (by way of vanessa@media.mit.edu(Vanessa Stevens Colella))]

I believe Kevin Kelley describes this group-airplane piloting
in his book "Out of Control."  I don't recall if he
directly addresses the questions you ask, but the whole
book is an attempt to raise just these kinds of questions
in the context of many "complex systems."

  - r

Rick Riolo                       rlriolo@umich.edu
Program for Study of Complex Systems (PSCS)
4068 Randall Lab     University of Michigan
Ann Arbor MI 48109-1120
http://pscs.physics.lsa.umich.edu/PEOPLE/rlr-home.html

On Wed, 17 Sep 1997, Miklos Gyorgy wrote:

> Date: Wed, 17 Sep 1997 08:03:49 -0400
> From: Miklos Gyorgy <MIKLOS@ttt-atm.ttt.bme.hu>
> To: starlogo-users@media.mit.edu
> Subject: Re: Business orgs and Re: Decentralization
>
> Hej,
>
> > I would like to extend this discussion to business organisations
> > as well. I saw an MIT demo (presented by Professor Tom Malone, Sloane)
> > of an audience - that acted as decentralised pilotes - guiding a virtual
> > airplane through a target landscape . Each person in the audience could
> > guide either up-down or left-right and this was then fed into navigation
> > of the virtual airplane. The emergent behaviour - of course - was that the
> > virtual aircraft was safely guided through most of the  landscape. I liked
> > this analogy and it worked well as a popular demo of decentralised control.
>
> I do not know too much about business organization structures, but the
> airplane analogy makes me ask a question.
> (Can we read about it somewhere?)
>
> To me, it is not at all obvious that the emergent behaviour of the virtual
> aircraft is safe. What happens if there is a mountain which must be
> avoided either from the right or from the left? Isn't it too slow to wait
> for the audience to decide on a solution through some kind of positive
> feedback?
>
> And similarly in a business organization, there could often be situations
> where there are many good solutions, but the org must decide firmly on one
> of them and drop the others. Doesn't a decentralized strategy perform worse
> in this case because it is more close to trying to find a common
> denominator?
>
> Gyorgy
>
>
>