Wheatley Dialogue LO10307

Benjamin Compton (bcompton@geocities.com)
Thu, 03 Oct 1996 02:30:14 -0700

Replying to LO10250 --

Richard Karash wrote:

> I have done a little engineering, and a lot of model building and
> statistical data analysis in my career. For many problems, it is extremely
> helpful to break them into parts. But, *good science* would be especially
> attentive to whether this is a satisfactory approximation, whether this is
> a good model, given the system being examined and the objective of the
> examination.
>
> I believe it would be foolish to stop breaking things into parts, not
> ever do it. We would lose too much progress.
>
> But, we must learn to see wholes and interconnections and become more wise
> about *when* to break into parts, when this is a satisfactory assumption.
> And, we must *always* be on the look-out for systemic considerations,
> always on the look-out for what we can only see if we look at the whole.
> We need to improve our skills, increase our ability to perceive these.
> Especially because of our cultural orientation not to.

[Host's Note: I trimmed Ben's quote of my previous msg...]

I like what Rick is saying here. While I'm not a trained scientist, I've
faced the practical problems of designing and supporting complex networks
and messaging systems. I learned very quickly I needed two see the
network/messaging system two ways:

1) As a whole

AND

2) The individual parts

Without a systemic view of a network/messaging system, it is difficult to
correctly identify the cause and effect of a problem. I've seen many
engineers literally spend hours chasing a problem down the wrong path,
because they couldn't see the entire system.

On the other hand, I've seen engineers who could see the whole but never
took the time to understand how each part functioned and connected to the
other parts. They could identify potential problems, but were unable to
correct it!

Thus, I agree with Rick, we need to learn to think systemically and use
reductionist logic when required. One complements the other. They are not
mutually exclusive.

--

Ben bcompton@geocities.com

Learning-org -- An Internet Dialog on Learning Organizations For info: <rkarash@karash.com> -or- <http://world.std.com/~lo/>