[extropy-chat] System dynamics (Urban dynamics)

[Keith Henson]

At 11:28 AM 1/17/2007 +0100, Anders wrote:

>Keith Henson wrote:
> > "The model was, and is, very controversial, because it illustrates why
> > many
> > well-known urban policies are either ineffective or make urban problems
> > worse. Further, the model shows that counter-intuitive policies -- i.e.,
> > policies that appear at first glance to be incorrect, often yield
> > startlingly effective results. As an example, in the Urban Dynamics model,
> > a policy of building low income housing creates a poverty trap that helps
> > to stagnate a city, while a policy of tearing down low income housing
> > creates jobs and a rising standard of living for all of the city's
> > inhabitants."
>
>We might get a test of this in Beijing, as large parts of low income
>housing in the city were torn down for the olympic city.

It's almost dogma now.  Virtually all of the massive public housing has 
been torn down.  Where they do build new it is in scattered units all over 
the city.

> > I might add that system dynamics has a lot of ad hoc assumptions in the
> > models.  Most of them are reasonable ones, such as economics, birth rated
> > integrating to population levels and the like.
>
>To me it looks a lot like a model where all plausible variables and
>interactions are included, but the functional form of the interactions are
>empirically unknown and hence have to be approximated by even more
>plausible stuff.

True.

>And if there is anything we have learned from dynamical
>systems and computational biology it is that they matter tremendously.

Most of them were investigated and they didn't have a lot of effect on the 
model performance for small changes in the shapes of the interaction curves.

>The
>"counterintuitive dynamics" Forrester describes in
>http://web.mit.edu/sdg/www/D-4468-2.Counterintuitive.pdf
>seems to be fairly typical instability, but he cannot determine what
>causes it because the model is complex and somewhat arbitrary.
>
>That is why I love zero-parameter models. Too bad they are very hard to
>make :-)
>
> > But now that Anders got me to thinking about this, system dynamics lacks a
> > foundation to the level of biology.
> >
> > Another application for evolutionary psychology!
>
>I think one need to model agents in this kind of system, both individual
>humans but also the "agents" of parts of the cities. Many continous models
>give a far too smooth behavior and don't show creative emergence in the
>same way as when agents can select strategies and copy variants from each
>other.

That happens on two widely divergent time scales with genes and memes as 
agents.  Even genes like lactase take thousands of years to spread into a 
population where memes can spread into a population on a scale of weeks 
(the Danish Cartoon riots for example).

My EP theory of war amplifies up a slight change in the average gain of 
xenophobic memes to switching an entire population into supporting a 
war.  Certainly emergence at work.

Keith