Feedback-forward

[Ian Griffiths <100326.1264@Compu]

In response to Paul Atkins comments on SD0268:

>>> I just wonder whether, in this case, anything is gained by looking at it at
this level when
>>> the molecular level explains the phenomenon perfectly.

I think it is a question of abstraction, simplicity and ease of getting useful
results.

You sometimes get valuable new insights by thinking about things at a very
detailed level, eg motion of individual molecules. For some purposes, thinking
(and modeling) at this level of detail is vital. But it is a different game: not
system dynamics.

Useful system dynamics modeling seems to involve rather high levels of
abstraction, and a focus on issues where loops of influence matter a lot. This
is not ALWAYS the best way to look at things. But it is often pretty powerful.

Ian Griffiths
100326.1264@compuserve.com
tel: (44) 01383 872937

["Robert M. Kane"]

Ian Griffiths observation points out an interesting feature of discussions
Ive been watching:

> You sometimes get valuable new insights by thinking about things at a
> very detailed level, eg motion of individual molecules. For some
> purposes, thinking (and modeling) at this level of detail is vital. But
> it is a different game: not system dynamics.
>
> Useful system dynamics modeling seems to involve rather high levels
> of abstraction, and a focus on issues where loops of influence matter
> a lot. This is not ALWAYS the best way to look at things. But it is
> often pretty powerful.

A problem is a problem and the effort to resolve a given level of complexity
is the same. Looking for problems isnt the right way to allocate energy to
solutions. The right way to select a problem is to think through the
potential pay-out and invest energy in the best potential return.

Just previously Bill Harris responded to Budimans reference to socio-economic
problems:

>Id inject the word faster here (***). Weve used simulation in such
>systems to great advantage in terms of doing experiments much more rapidly
>than one could try them on the real system (not to mention the clearer
>insights which a good model can sometimes give).

This is a problem of high level and potentially high pay-out. To know where
to invest in social structures to bring about desirable ends is of ultimate
value in human terms.

Finally, Geert Nijland follows with:

>Our department wants to develop a course in designing new sustainable
>agricultural systems on an ecological base.

Now its all spelled out (in my opinion) where the highest pay-out may be
attained. SD should apply methods of social and ecological simulation (if
social simulations can be reliably designed and Bill thinks they can) to guide
our race toward economically and ecologically sustainable system of living.

I also suggest to Mr. Nijland that very early in your work you should target a
practical application that will provide feedback to assumptions. Assumptions
(wrapped in simulations) left too long untested become paradigms of confusion.

And, yes, I do have a suggestion.

Bob Kane, North American Coordinator
Institute for Sustainable Agriculture in the Tropics (SIAT)
Jaen, Peru
robert_kane@corp.cubic.com
http://www.ran.org
an
an_campaigns/amazonia/coffee_html