REPLY SD Impact on National Government Policies (SD6161)

[System Dynamics Mailing List]

Posted by  "McConnell, George \(SELEX Comms\) \(UK\)" <george.mcconnell at selex-comm.com>

As always, Bill Harris has clearly articulated (some of) the issues that can 
too easily be missed when considering "why don't (the management) use (the 
technique) more/better".  This thread has Government Policies as its focus, 
but the same issues appear at almost alll levels of interaction - whenever 
the things that we (as individuals) know about are not utilised to what we 
see as their optimum advantage.

In this we ourselves fall into the trap of not thinking "systems"-like - we 
preach that we should take into account feedback, different perspectives, "a 
wider view" and other good systems practices while simultaneously suggesting 
that we have the 'only' answer.

I don't make a good preacher - I find it far too easy to see another's point 
of view.  What I hope I can achieve is that by applying SD (and other 
techniques) others will be educated to step outside of their own comfort zone.  
The problem with this is that it is very difficult to programme - so answering 
the question "how do we get SD (substitute what you like for SD) used more" 
becomes impossible as stated.

Douglas Mckelvie made a good point regarding the size of the 'typical' 
modelling effort - certainly my 'customers' are reluctant to spend big bucks 
on any modelling task - no that is not quite right - they will spend on something 
complicated, flashy, difficult to understand because they perceive that "big is 
beautiful" when it comes to a model.  Synthetic Environments they like - why, I 
suspect because it looks almost like the real thing!  The problem, of course is 
that the more like the real thing it looks, the more like the real thing it is to 
understand - anyone want to model the model?

SD, and similar, doesn't look like the real thing to the majority of people (at 
least not at first glance) and therefore faces an immediate hurdle.  There is an 
immediate distrust of any model built in a short time - it must be too simple!!  
The fact that one of the big skills of the modeller is to abstract the problem to 
an appropriate level is lost on most (including some modellers).

Personally, I think that point 6 of Bill's contribution is perhaps the most 
important.  It is incumbent on each and every modeller to make their 'customer' 
feel that (whether it is true or not) they know much more and are much smarter 
than the person doing the modelling.  The only thing the modeller is allowed to 
be smarter at is creating and manipulating a model.  I see too many examples of 
the modeller assuming that they know best and 'forcing' particular representations 
of the problem simply because that is what the modeller is comfortable with.

What is clear is that there is no easy answer to this.  Systems thinking is not 
for everyone - but it also is not (as Jack Ring already pointed out) only the 
Systems Dynamics Society who should be promoting it. 

regards
	George
Posted by  "McConnell, George \(SELEX Comms\) \(UK\)" <george.mcconnell at selex-comm.com>
posting date  Mon, 8 Jan 2007 10:40:00 -0000