REPLY The Death of System Dynamics? (SD6224)

[System Dynamics Mailing List]

Posted by  Tom Fiddaman <tom at ventanasystems.com>

I'm sympathetic to Richard Stevenson's worries about the field. However, I'm unsure e
xactly what his prescription is. "Get back to roots" doesn't seem like enough. Was SD 
growing much faster in the Industrial Dynamics days? Does the spread of systems thinking 
slow the progress of system dynamics modeling? Does emphasis on business consulting push 
the right positive loops to accelerate the growth of the field?

> Posted by  Richard Stevenson <richard at cognitus.co.uk>
>
> ... recently  the SDS seems to me to have become a self-
> serving and introspective  club that resists change and is entirely blind to new
> opportunities  and problems in the real corporate world.


Maybe SD, like other disciplines, has a lot of inertia, but I'm curious - which changes 
and opportunities are going unseen? More importantly, how could we direct attention toward them?

> Astonishingly, after 50 years of SD, the Society can now pride itself  on engaging
> just 400 people at its annual conference!  That's just  not good enough.

Three's some data on this in Mike Radzicki's 2006 conference address:
http://www.systemdynamics.org/newsletters/President'sAddress2006.pdf
See slides 12 & 13. The Society's doubling time is about 9 years. Mike presents a dynamic 
hypothesis for field growth (slide 27). I think the key question is, what doubling time 
could we aspire to, and what loops would help to achieve it? (Given the time it takes to 
become a good practitioner, I think we might hope for 5 years but not 2.) Also, are the 
society and the conference good measures of the dissemination of our ideas?

> This is, I admit, a challenging statement.  I'm not going to back  away from it,
> however.  Jay Forrester's original (and exciting!)  vision of "designing
> organisations" has been completely lost by the  SD community.

I don't think the fact that people are working on other problems means that the community 
has lost Jay's original thought. Many non-intervention activities, such as behavioral 
experiments, are aimed at understanding pieces of the decision-making system that are 
absolutely critical to the solution of the biggest problems. They also build bridges to 
other fields, and thus help to attract new practitioners and spread our ideas.

The organizations in greatest need of redesign are governments. Governments are not easily 
influenced in a consulting mode. Their agencies frequently can't use the answers to the 
biggest problems because they have no statutory basis to implement them. To change the 
government one must change the minds of people and politicians, and that requires 
nontraditional SD.

I do see a lot of projects that tackle small problems where they could just as easily 
tackle big ones. There are a lot of models that are simply collections of things changing 
over time, with little feedback and many exogenous inputs. There are many projects with 
no clear path to implementation. These are all things we could work on, but they strike 
me as a nuisance more than a core malady. Many of the changes to the conference format 
have been directed at such problems.

> The problem with SD today is its lack of business focus and a  complete absence of
> self-discipline.  That freedom has given  academics a "right to roam" and probably
> stimulated original abstract  thinking over the years.  But in the real, corporate
> world, SD's  impact has been minimal compared to its potential - and far from
> getting stronger, in fact it is disappearing.

Perhaps this is due to something about the nature of business, not the nature of system 
dynamics. I don't think we can lay the woes of the SD consulting industry at the door 
of the Society. Some of those woes may be merely cyclical (the low point for me after 
the dot.com crash was when someone approached me with an $8,000,000,000 problem and a 
$5,000 budget). If SD is really failing to thrive in the corporate world, I think we 
should be looking for the internal explanation - what loop are we missing? Is it 
something about the personality of people attracted to the field? Are we selling a 
product (truth) no one wants? Do our projects have a poor risk profile (high cost, 
low probability of big success)? Have we consulting firms choked off our own supply 
of talent by stealing too many PhDs from academia?

> Perhaps the most dispiriting aspect is the "bottom up" approach that  has
> characterised SD since "Industrial Dynamics".  Causal loops,  archetypes, stocks and
> flows....all true but managers just don't have  the time.  It's like saying "I know
> you all speak English but I want  you to learn Esperanto".  Unfortunately, using
> Esperanto imperfectly,  most managers just tend to talk gibberish to each other.

This sounds dangerously close to, "It's hard and time consuming to teach managers to 
fish, and most of them are too stupid to bait the hook, so let's just give them fish." 
To the extent that it's true, it may explain why the most successful technologies in 
management (embedded IT systems) require little thinking. That may be a model to exploit, 
but I don't think it's one to which SD should aspire.

The problem with the English-Esperanto analogy is that managers are speaking 
Staticlinearish while living in Nonlineardynamica. Sure, they can muddle through 
in the short run, but in the long run they'd do better to learn some Nonlineardynamic. 
Managers are already speaking gibberish to each other because, no matter how 
grammatical, their Staticlinearish is not a very expressive language (nine words 
for "not my job", none for "side effect"). Naturally the best way to learn a language 
is to use it, so it makes sense to teach Nonlineardynamic using real problems (rather 
than in the abstract for its own sake), but there's no shortage of real problems.

It's important to remember that not everyone is like us. Martha Miller has run 
the Meyers-Briggs personality inventory at several gatherings of SD practitioners. 
Most of us lie in one corner that represents just 4% of the population, as I 
recall. It's not really surprising that we need more than one language.

> I am also often astonished by the poor quality of much of what at  passes for
> commentary on the SD forum.  There seems to be little or no  distinction between
> crazy student rambling, learned tablets of stone,  detailed technical enquiry and
> academic philosophical meandering.   It's completely random. Who's really in charge?

No one is in charge. The same criticism could be leveled at every forum I've ever seen, 
whether the topic is SD, economics, climate, or photography. I think it reflects on the
medium more than the field. However, if you look at the annual totals on the Ventana UK 
archive of the mailing list,
(at http://www.ventanasystems.co.uk/forum/forumdisplay.php?fid=34 )

it does appear that traffic was down in 2006. There's now a lot more competition from 
more-accessible and attractive media (blogs with RSS feeds and Google Reader, Yahoo 
groups and other online fora with better threading and monitoring, spam-free Groove 
messages supplanting email, etc.). There are strong positive feedbacks to quality list 
participation, so it's easy to imagine blather driving out intelligent conversation, 
and any exogenous effect of improved competition amplified by participant defections. 
Maybe this is a sign that it's time to move this discussion to newer technology. I 
wouldn't expect it to raise the quality of the average post, but it should make it 
easier to keep track of and engage in the good ones. Also, I would like to see more 
student input, not less - the field should be investing in growth, not shutting off 
the inflow because it's inconvenient in the short term. I see no evidence that student 
ramblings are more prevalent or crazier than others.

Tom

****************************************************
Tom Fiddaman
Ventana Systems, Inc.
Posted by  Tom Fiddaman <tom at ventanasystems.com>
posting date  Thu, 01 Feb 2007 12:42:02 -0700