EPLY Some questions Re:accreditation

[Tom Fiddaman]

>Is SD a Software ?

No. But SD software makes it a lot easier to build and use most types of
feedback models.

>Could SD be taught/mastered in a 5 days course?

No. Its possible to take someone whos already proficient with a related
discipline like control theory and make them technically competent in a few
days, but theres still a large body of useful formulations and process
knowledge that must be learned by apprenticeship, trial-and-error, self-study,
etc.

>How would be defined a competent SDist ?

Many professional societies have a code of ethics even if they dont accredit
members. Perhaps this would be a useful intermediate step between anarchy and
accreditation. Of course, what to put in the code is a tough question. I think
there could be two components:

The first would be some minimal standards for model construction and testing.
An analogy would be building codes. A lot of this knowledge is pretty
concrete,
and just needs to be made widely available:
- thou shalt test for integration error by reducing DT
- thou shalt not have physical stocks without first-order negative feedback on
the outflow
- thou shalt look upon the behavior of each variable in the model at least
once
in a while
- thou shalt perform parameter change experiments one at a time
- thou shalt enforce dimensional consistency
- ...
Some of this can now be codified and automated with Vensims Reality Check
language. Do we need periodic inspection to establish conformance with model
building codes? Probably not. Would model building codes become arcane and
restrictive, like real building codes? I doubt it.

The hard part for me is defining codes for intervention. How do we know a
technically excellent model is relevant to a problem? How do we know when a
diagramming exercise is missing some critical insight that would only come
from
moving on to a model? I think this is the bulk of the problem. Im not
going to
make enemies by creating a list here.

For both elements, it would help a lot if the consumer were better informed.
Perhaps the SD Society could create a set of questions prospective clients
might ask, along the lines of Sterman, J. D. (1988). A Skeptics Guide to
Computer Models. Foresight and National Decisions. Lanham, MD, University
Press
of America: 133-169.

- Tom

****************************************************
Thomas Fiddaman, Ph.D.
Ventana Systems http://www.vensim.com
34025 Mann Road Tel (360) 793-0903
Sultan, WA 98294 Fax (360) 793-2911
Tom@Vensim.com http://home1.gte.net/tomfid/
****************************************************