Subject: Scratch and System Dynamics

Posted by George Richardson on 12/10/2010

Message:

Pappert, Forrester, and Beer go together really well. Which is to say, LEGO, system dynamics, and VSM can instructively be interrelated. In fact, Beer knew a lot about Forrester's work (and tried to build a system dynamics model for Chile, reported in Platform for Change), and I'm quite certain Jay knows a good deal about Pappert and Beer.

But it is very important to *not* characterize system dynamics as computer programming. There's a big, important difference between a "program" and a "model."

The purpose of computer programming is to get a computer to do something clever, or neat, or hard, or whatever, any way the programmer can make it do it, the quicker or cleverer or more efficient the better.

The purpose of system dynamics is helping people think about their world(s) by defining problems dynamically, thinking hard about system boundaries (temporal and conceptual), and thinking endogenously in stocks-and-flows and feedback loops.

System dynamics practitioners use computer simulation because endogenous thinking about feedback loops gets complicated very fast and gets to the point where people can't figure out the dynamic implications of structure that's in their heads.

So SDers do try to build computer models of real world structure and dynamics, but they aren't as free as computer programmers. The structures SDers formalize are constrained by the ways people think the real world actually behaves (we call that "operational logic"), what actors in the system actually know, how they really respond to pressures, and how the system endogenously changes active structure over time (that's what nonlinearities do).

There are computer programs that can play chess at the level of a grandmaster, but their creators make no claim that they are playing the ways grandmasters actually play. In contrast, it does us no good at all to build a system dynamics model that does something clever in a way unlike what we know about the real world. Such a model would be useless in our efforts to help people think about how the world really works and how people might make it better.

So the best way LEGO, VSM, and system dynamics can come together is not in thinking about how they use computers but in combining their various strengths in a collaborative effort to understand real system behavior.

George P. Richardson

Follow Ups:

Scratch and System Dynamics - Sang-Don Lee 12/10/2010 
Scratch and System Dynamics - Pedro D. Almaguer Prado 12/10/2010