Re: [Phys-l] Simulated Labs for General Physics Course

[John Denker <jsd@av8n.com>]

On 01/21/2011 02:30 PM, Donald Polvani wrote:
> 3) Several feel that the ideal beginning general physics course would have
> three components:
> 	- theory (with math)
> 	- simulations/visualizations
> 	- traditional labs
>
> With the simulations/visualizations mostly supporting theory in helping
> conceptual understanding but all three components mutually interacting and
> reinforcing one another.  According to one member, Princeton is already
> doing No. 3, and it could be even more suitable for community college open
> enrollment students.
>
> I myself like No. 3 and had even suggested a small scale (pilot) version of
> it to the full time faculty last semester.  I suggested that we try it with
> just one of the conventional labs (projectile launching) to see how it
> worked.  I believe I will make the suggestion again.

There is some confusion here.  A certain amount of confusion is 
understandable, because if you go to 10 different schools you will 
get at least 15 different definitions of what "integrated science"
means.  So let me be clear about what I am talking about:

What Princeton and some other institutions are doing is 100 or 1000
times more ambitious than the "suggestion" mentioned above.
 -- It is not confined to a single lab project.
 -- It is not confined to a single course.
 -- It is not confined to a single academic department.

I am talking about a revolutionary restructuring of the curriculum, 
tearing down the traditional vertical silos and replacing them with 
horizontal integration.

I'm not any kind of expert on this.  I haven't thought about it all
that much.  But I know guys who have thought about it, and they 
decided that starting small was *not* a good idea.  Their motto was
the old proverb: Between two stools one sits in the ashes.  The
modern equivalent is "Go big or go home."  The rationale for this
is simple:  If you start teaching computer simulation techniques
in the physics class, that takes time and thereby weakens the physics
course, and it duplicates what is covered in the computer science
course, so the students are strictly worse off than before.

The only way (so far as I know) to win this game is to get the 
computer science guys to use some of *their* time to teach physics-
related modeling techniques, and the same with biology et cetera.
This changes the whole story.  It becomes a win/win proposition 
for the students rather than a lose/lose proposition.

This is however very tricky to set up.  If you do it wrong, the
result is a turf war that looks like a replay of the battle of
Stalingrad.  The Princeton guys were clever enough to figure out
in advance that they didn't want a war, and instead they wanted
everybody on the same side.  They worked out the details in advance
and got the six or seven relevant departments to sign off on the
proposal, not just tolerating it but deeply invested in it and
actively supporting it.  I reckon the task force wore out several
pairs of shoes walking from office to office hashing out the details.
The main points were worked out before any of the courses were offered 
for the first time.

Here's the acid test:  Does the computer science department recognize
the first-year integrated science class as fulfilling the prerequisites
for the second-year courses in their department?  Ditto for biology,
chemistry, physics, engineering, math, et cetera.

As a general rule of thumb, other things being equal, I am in favor of 
starting small, doing lots of pilot projects, not letting the perfect 
be the enemy of the good, et cetera ... but there are limits.  If you
plan to build a steel mill, it is good to build a pilot plant first,
but if you make the pilot plant too small it won't work at all.  There
is a certain minimum size.  I don't know what is the minimum size for
an integrated science program of the sort we are talking about here,
but it's not zero.

I am not the expert, but I reckon a lower bound on the sort
of thing I am talking about would be something like this:  Pick
a physics course, a biology course, and a computer science course.
Get the three professors together and say:  The plan is to make
all three of these courses strict corequisites of each other ...
really just one big course.  The exact same students will be 
enrolled in all three courses, with no exceptions.  So, then,
what can each of us do to make the other guys' jobs easier?  What 
can we do to make things better for the students?  As a framework 
for answering /part/ of the question, 
 -- Suppose you were teaching the computer science course; 
 -- Suppose the computer guy were teaching the biology course; 
 -- Suppose the biology guy were teaching the physics course;  
... what would each of you do differently?

If you can get these guys to sit in on each others' classes this
semester, at least intermittently, that helps you figure out
what needs to happen in order to offer an integrated course next
semester.  Some professors think it is beneath their dignity to
sit in on an undergraduate course ... to which I say:  Feynman
did it.

Princeton is not the only place offering an integrated science 
program.  UMass just got through putting up an entire integrated 
sciences *building* and the first integrated science classes 
started this week.  Chicago has had an "integrative science" 
building for 5 years, although this does not yet seem to have 
had much effect on the undergraduate curriculum.  Caltech has 
made a virtue of interdisciplinary studies since Day One;  
Millikan didn't want there to be academic departments at all, 
lest they become turf-conscious fiefdoms, to the detriment of
the students and the school as a whole.  They started offering
undergraduate courses in computational biophysics 25 or 30 years
ago.  There must be hundreds of other programs I don't know about.

We should remind ourselves that mqny or these schools are solving
an easier problem than the one you face:  They are offering their 
programs to science majors, not to the wider "open admissions" 
community.

At the non-major extreme, we note that integrated instruction has 
been in widespread use since the pleistocene ... in grade school.
It's called "making a unit".  In grade school one single teacher
is expected to cover all the subjects, so you can integrate the
subjects without worrying about personality issues and departmental
turf issues.  Using "units" is a widespread practice if/when grade-
school teaching is a profession.  By definition, professionals 
are empowered to decide /how/ to solve a given problem.  Nowadays 
in some districts the teaching is so heavily scripted from On
High that teaching essentially becomes blue-collar job, not a
profession, and good teachers don't have the flexibility to make 
units like they used to ... but that is a topic for a separate 
discussion.

As another buzzword for the same thing, integrated science can
be seen as a greatly expanded version of the /spiral approach/.

As yet another way of looking at it, this is my answer to the
"physics first" question:  You don't want physics first, except
insofar as it is first, second, third, fourth, and fifth ...
and biology is also first, second, third, fourth, and fifth,
and the same for chemistry and computer science et cetera.  You 
want to spiral around and around and around, using each to 
reinforce the other.  I have a multi-legged version of "itsy
bitsy spider" that I do with my fingers when talking about this.

As a physicist trying to set up an integrated sciences program, 
the biologists are your most natural allies.  By and large, the
biology community figured out a decade ago that the future of
biology (including medicine) will not even remotely resemble 
the past.  It will be much more interdisciplinary.  Specifically, 
it will be more quantitative, more mathematical, and more 
computational.  Their motto is "evolve or die".  The old advising 
practice is dead, namely the practice whereby a student who liked 
science but didn't like math would be advised to major in biology.  
I have no idea what such students /should/ do now, but they have 
(almost) no future in biology.

Having a couple of biologists and a couple of physicists on board
might make it relatively easier to get folks from other departments 
to hop on.  I say /relatively/ easier because I reckon that in 
absolute terms it will be tremendously tricky and labor-intensive to 
get such a program started.  Once started, it should be easier and 
better for everybody including the students -- which is the point -- 
but there's going to be a tremendous start-up transient.

I reckon community college students have more to gain from 
an integrated science program, even more than the Ivy League 
students do.