Not in my understanding.
My understanding of a complex system is that it is inherently
unpredictable because a new order emerges out of the many, many
interactions (read: countless positive and negative feedback loops) that
happen. This MUST lead to some order, but what it will be like (or: will
it be good or bad?) is unpredictable. Also WHEN it will arise is
unpredictable. But, inherently unpredictable.
Sometimes you can predict roughly in what SPACE it may emerge (usually
called an attractor) but in my understanding that is quite a
mathematical approach to it.

I usually look at human organisations. We interact in many different
ways. Some 'patterning' may happen in this interaction. I would suggest
to look at this patterning as self-organising.
Thus, the 'pattern' is continuously emerging out of those interactions.
But WHAT pattern (and whether we like it!) is inherently (sic!)
unpredictable.

My sense is that via models people try to get a feel for what influences
the emergence of this new order. I do believe that, by its very nature,
it will always be IMPOSSIBLE to predict the order that will emerge.
Until it HAS emerged that is, but we cannot call that 'prediction' can
we?! :-)

Regards,

Frank Smits
Symphoenix Ltd
Tel: +44 (0)1732 450 495
Mobile: +44 (0)7715 423 150
E-mail: [log in to unmask]
Website: www.symphoenix.net


-----Original Message-----
From: Complexity and chaos theories applied to primary medical and
social care [mailto:[log in to unmask]] On Behalf Of
Roger Walters and Fiona Shackley
Sent: 27 November 2002 16:49
To: [log in to unmask]
Subject: Re: One for the geeks

Question?

Do complex systems have elements that are intrinsically unpredictable or
just not predictable at the current level of knowledge.

If a computer system becomes so complex it is very difficult to predict
does it move to being a complex system?

Might it be helpful to think of a continium here?

First post by previous lurker

Roger Walters
SpR in Child Psychiatry
Manchester


At 11:08 27/11/02 -0500, you wrote:
>Complex systems are more than the sum of their parts cannot be
replicated
>precisely, and exhibit behaviour that is unpredictable from an analysis
of
>the sum of their parts.
>
>Complicated systems are no more than the sum of their parts and can be
>replicated and exhibit predictable behaviour.
>
>It is my contention that computer based simulations do not exhibit
complex
>behaviour because they are rule based and inherently predictable.
>
>I've chundered on about this in the past, and Chris and I have opposing
>views on it.
>
>Blizzard anybody?
>
>Ceri