Re: [Phys-l] Simulations and Computer Homework Problems for Freshman and Sophomores

[John Denker <jsd@av8n.com>]

On 12/10/2011 08:46 AM, Donald Polvani wrote:

> I am an adjunct physics instructor teaching at a community college.  I have
> been trying to get the full time faculty to add computer simulations to
> their lab courses and computer programming to their homework assignments.

That's obviously necessary and important.

> The lab simulations need not replace the physical lab but could supplement
> the existing lab by pre (or post) showing the students how typical results
> "should" turn out. 

I agree with the sentiment, but I would go much farther than
that, and I wouldn't express it in quite those terms.

For one thing, I see simulation as more closely related to 
theory than to experimentation:

                 science
                /       \
               /         \
         theory           experiment
          /  \
         /    \
   algebra   simulation


Saying "we're not going to use simulation" would make about as 
much sense as saying "we're not going to use algebra".

I wrote a 1100-word analysis of this a few months ago, and the
same analysis applies today.  I am tempted to cut-and-paste the 
whole thing here, but instead I will just link to it:
  https://carnot.physics.buffalo.edu/archives/2011/1_2011/msg00402.html

> Assigning a few such problems, over the semester, as homework (or even a
> full lab session ) seems worthwhile to me.

It needs to be more than "a few" and it shouldn't come out of
the "lab session" time budget.

> While the full time instructors acknowledge that such computer
> simulations/homework may be useful,  they maintain that such work
> (particularly, the lab simulations) belongs in junior/senior year level
> courses

That's nuts.

The most important issue concerns budgeting the classroom
time and homework time necessary to learn about computers.
In an ideal world, or even a halfway-reasonable world, this
time would not come out of the physics budget, for the same
reason that learning algebra (ideally) doesn't come out of
the physics budget.

  Non-ideally speaking, we know that a great deal of the
  physics time is spent "reminding" students of algebra and 
  other things they supposedly already know, but that's not
  what I want to discuss at the moment.

The point is that all these things need to be /integrated/.
Princeton has for years not offered introductory courses in
the narrow topics of physics, chemistry, biology, math, or
computation.  Instead they offer one big tightly-woven 
tapestry that can be called a "mathematically sophisticated
course in computational biophysics".  Students seem to like
it.

A possible starting point is to take all the "physics hours" 
plus all the "computer science" hours and integrate them.

Getting this done requires a lot of heavy lifting.  We are
talking about significantly restructuring of the curriculum
... and restructuring the institution itself, in the sense
that the physics faculty and the computer faculty need to
cooperate.  This is not easy to set up, and adjunct faculty 
cannot be expected to lead the charge.  Still, it needs to 
be done.  It is necessary for the good of the students, and
therefore for the good of the institution as a whole.

>  and the freshman/sophomore courses we exclusively teach (as a
> community college) should only contain real physical measurements. 

We agree that's not the right answer ... but what's worse, 
it's not even the right question.  Do not allow them to
frame the discussion in those terms.

Modern science, since Day One, has been about /integrating/
theory and experiment.  Galileo is the called the father of 
modern science because he showed how experiment /reinforces/
theory and vice versa.  Each builds on the other.

  He did not "substitute" theory for experiment or vice versa,
  not even a little bit.

  Computer simulation does not "substitute" for experimentation,
  not even a little bit.

Perhaps it will help to call it _computer modeling_ rather than
"simulation".  _Modeling_ is a trendy word that seems to cover 
all of theory and indeed most of teaching and learning.

> I'd appreciate your inputs on your experiences on either using such
> simulations/homework in freshman/sophomore courses or why you have not done
> so.

Computer modeling should start at the pre-high-school level.

The freshman/sophomore level is not too early;  it is too late.