Thanks for the reference to Kauffman... but what issue of Sci American was it?
>In a conversation with Ruthens (published in The Scientific American)
>Stuart Kauffman speaks about a system bringing itself to the edge of chaos
>1) by tuning its ties to other systems in the environment, and 2) by
>tuning the richness of the ties between its subsystems.
These principles by the way are the same as those espoused by Rodney
Brooks at M.I.T. where he was building mobile robots. Instead of using
the traditional AI (artificial intelligence) techniques of trying to
build-in intelligence from the top-down understanding of the designers, he
took a bottom-up approach using those two principles: rich interactions
amongst the components, and a " healthy " interaction with the
environment. He used a rich network of fairly simple machines which were
very responsive to the environment.
You wrote about its relevance to the Learning organization:
> " ... which I see as an adaptive complex system functioning at the edge
of chaos. This leads to the idea that such an organization needs to tune
the coupling between its subsystems and also its couplings with other
organizations in its environment so that it can function adaptively at the
edge of chaos (i.e. with the right mixture of order and chaos). "
And went on to emphasize " customer closeness and sub-system autonomy " as
being important ways to implement these two principles...
I agree with this emphasis. In our consulting work in Australia, the
tuning with the envirnoment is reflected in a number of ways... i)
detecting and utilizing demand variation (sufficiently real-time within an
appropriate time interval) in order that the dynamic impact upon the
business ability to absorb and respond can be handled. [conventional
approaches based upon accounting data and most management accounting
approaches completely blurs such dynamic variation, and delivers the data
to decision-makers with time lags which are not very useful.], ii)
emphasizing the dynamic coupling with suppliers and significant
clients.... this has led us into extensive work with trans-corporate
applications in coal export chains, grain export chains and others.., and
iii) a pro-active planning approach to managemnet which attempts to change
management from being so reactive (i.e. fighting fires.. chasing JIT where
it shouldnt be used, etc.) to being more anticipative and hence more
supportive to colleagues.
The autonomy issue within the business has of course opened up
considerably with the power of PCs growing so much. In fact, we often
encounter situations where the whole IT function (when it is stuck in
mainframes and control issues) being sidestepped. I dont think that this
is the best direction in the midterm, but it certainly emphasizes the role
of autonomy. There are also the emphasis upon matrix approaches to
management, and the emergence of companies structured more around a
networks rather than hierarchies, of which our own company is an example
[I have heard suggestions that there are more and more such entities
emerging in North America, but have not seen anything written about them.]
... Our main approach to this with our clients has been an approach we
call " Trusted Linkages " which stresses the integrity and alignment
between operational measurements and board level financial information.
Conventional approaches wipe out much of this integrity through
inappropriate numerical techniques for the sake of simple overviews [the
overviews dont have to go, just the way they are obtained and handled.]
For me, my earlier postings about the avalanche/implosion effects of
changing connectivity speak as well to these two principles of Kauffman
and Brooks. The rich connectivity and autonomy speaks to the connectivity
between components (without autonomy, the information flows are stunted
with little feedback). The tuned relationship with the environment is in
a sense a reflection of a healthy and wholistic tuning of communication
within the components of the system (overloading (too many connections)
produces chaos, and underconnected yields little adaptation).
The principle concerning the environment coupling raises the issue for me
that in making a heavy distinction between the business entity and its
environment (like the species and the ecosystem), we sometimes forget or
keep out of our awareness, the essential co-existence between the two
(from the Taoist perspective... every form has its formless background.)
Do these ideas make sense to anyone in the context of the two principles
from Kauffman?
Uri, I will address your questions in a subsequent posting.
-- Dr. Doug Seeley: 100433.133@compuserve.com " Are there any places or times where networks dont exist?