No disagreement there either, Alan.  Indeed, I hold that for policy
purposes narrative scenario analysis (because it is richer) should take
pride of place over (but be informed by) model outputs.  The importance
of empirical data gathering comes under the heading of validation but,
as my thoughts on that have been widely published for many years now, I
won't pursue that issue here.

My best wishes to all for a good weekend.

scott


On Fri, 2007-01-19 at 15:04 +0000, Alan Penn wrote:
> Scott,
> 
> Completely accepted :-) those are also very good reasons to do simulation.
> However, I might suggest that the use of any technology (simulation may be a
> particularly useful one) in this role of disambiguation is a bit like a prop
> in a conversation. The real work lies in the conversations that surround the
> prop, the way it guides the flow of knowledge between you and the
> stakeholders (in both directions) and the way it stimulates you finding a
> common language and understanding. Other similar props involve empirical
> data gathering and presentation which then form a basis for discussion.
> 
> Alan Penn
> Professor of Architectural and Urban Computing
> The Bartlett School of Graduate Studies
> University College London
> Gower Street
> London WC1E 6BT
> +44 (0)20 7679 5919
> [log in to unmask]
> www.vr.ucl.ac.uk
> www.spacesyntax.org
>  
> 
> > 
> > On Fri, 2007-01-19 at 12:31 +0000, Alan Penn wrote:
> > > Are we even close to giving our software agents the cognitive
> > > power of an ant or a mouse? If not then what we are looking for are
> > > 'created phenomena' out of our simulations to help us test whether our
> > > 'high road' theories and philosophies are fatally flawed, or perhaps
> > > more to the point, to stimulate new theories.
> > 
> > I disagree.
> > 
> > What I am looking for is a means of reducing ambiguity in the
> > understanding of decision makers.  One consequence of the modelling
> > exercise in such cases (i.e. involving stakeholders as participants) is
> > that they come to recognise which of their own assumptions and verbal
> > formulations are key elements in the expectations they form and
> > sometimes which elements of the problems they face are being ignored or
> > misstated in order to support the conclusions they want.
> > 
> > That is one purpose of social simulation.
> > 
> > Another purpose is to explore theoretical concepts. Here, too, the
> > function of modelling is to reduce ambiguity.  One of my PhD students,
> > for example, is investigating whether the term "asset specificity",
> > widely used by economists and management researchers has (or can have)
> > any definite meaning.
> > 
> > Personally, I think it would be a useful exercise to produce models in
> > which some of the terms used in the discussion so far would have to be
> > made precise. :-)
> > 
> > regards.
> > --
> > Scott Moss
> > Professor of Social Simulation
> > Centre for Policy Modelling
> > Manchester Metropolitan University
> > Aytoun Building
> > Manchester M1 3GH
> > 
> > http://cfpm.org/~scott
> > 
> > (t) +44 (0)161 247 3886
> > (f) +44 (0)161 247 6802
> > (m) +44 (0)7740 942564