Re: Lerning from simulations

[Ludwik Kowalski <kowalskiL@Mail.Montclair.edu>]

> Hestenes said,
> > The greatest deficiency of simulation software like
> > INTERACTIVE PHYSICS is that it fails to explicate
> > (i.e., make explicit) the structure of the underlying
> > model that generates the computer animations.

When I started using Interactive Physics it was not "magic"
to me because I was already familiar with the "underlying"
mathematical model of its algorithm. I learned it by writing
my own simulation programs in Fortran and True Basic.
The "accuracy" command in "world" menu, for example,
would not be clear without this experience. Programming
(and mathematical language in general) should still be part
of learning physics. That is how I interpret David's concern.
It is not a defficiency of IP, it is a potential defficiency of using
it by teachers.

And there are philosophical aspects as well. We believe data
collected in a lab because they are real, we believe simulated
data because we have confidence in a program. If my prediction
agrees with that generated by a program I know that my way of
thinking agrees with the model used by the programmer. This
is not the same, at least in principle, as testing my predictions
against real data. This philosophical argument is probably not
very important in practice but it is something to think about.

By using IP (or Maple, Mathematica, etc. ) we benefit from work
of others; we can even "solve problems" without knowing the
underlying physics (or mathematics). In the past I often felt
guilty about using programs written by other people "for the
sake of efficiency". The argument of "not reinventing the wheel"
does not necessarily applies to teaching. Models of thinking
should be reinvented by each student, as much as possible.

Ludwik Kowalski