Coding your own

[M. Lang & S. Railsback]

Since such stimulating discussion resulted last week from the rude
reviewer comments I posted, I thought I'd try again with another
modeling-software interface issue. 

It has been my experience that modeling research proceeds much faster
when the modeler uses a programmer to write the code and spends his/her
own time (1) writing the model down on paper, (2) checking and testing
the code, (3) doing modeling experiments, publishing, and otherwise
conducting science. The benefits of working with the programmer are
essentially the same as the benefits of using Swarm: more time on
science instead of toolbuilding, lower costs, and better software.

It has also been my experience that when I suggest modelers not write
their own code many people react with shock and (to be polite)
disagreement. I suspect the reasons for this reaction include:

a. Using a programmer violates the culture of graduate research, where
writing your own code is one of the tests one must pass to earn a
degree. (This culture apparently persists even in areas like ecology
where students have little training in computer science or software
engineering.) Once trained in this way, scientists seem to have the
write-your-own culture ingrained and never consider alternatives. 

b. People think its expensive to hire a programmer. (It may be,
sometimes; but often not if you value your own time and consider how
much faster the work gets done.)

c. The modeler is afraid to give anyone else control of their code. I
think this is often the strongest reason why modelers don't use
programmers, even though it makes little sense to me. (When I lived in
Tennessee I used to tell the passengers in my car that because it is so
hilly there, I insist on doing my own brake work instead of letting a
mechanic do it.) 

d. Writing code is more fun than potentially painful tasks like testing
the model and submitting papers. 

So: I'm trying to decide how much to say about this issue in a paper I'm
preparing to submit. Feedback?

Steve
-- 
address@hidden
Lang, Railsback & Assoc.
250 California Ave., Arcata CA 95521
707-822-0453; Fax 822-1868


                  ==================================
   Swarm-Modelling is for discussion of Simulation and Modelling techniques
   esp. using Swarm.  For list administration needs (esp. [un]subscribing),
   please send a message to <address@hidden> with "help" in the
   body of the message.
                  ==================================