eugene@eos.UUCP (Eugene Miya) (05/04/89)
We have touched on some interesting (difficult) topics in this
discussion. I have a question about .edu all by itself (i.e. teaching).
Why is it (rhetorically) we separate ourselves into teacher and "doers?"
There is the old joke about those who can't do, teach. Those who
can't teach, administrate. 8) Essentially why does the public look
down upon the practically of teaching? You have been to school, now it
is time to really learn. Computers have upset this balance.
Schools are sort of regarded as learning centers (classrooms).
It allows one to study a problem in isolation, and frequently,
this is a good way to solve problems. It justifies the use of ivory
towers. It's an aspect of specialization. You do chemistry, and I
do physics, and we won't meet for anything by lunch. But similarly,
many problems aren't solved in isolation, they require integrated solutions.
I think if there is any one thing which is needed to change the educational
system it is to prepare for this integration. One attempt, at the
Wang Institute, failed. There are other reasons as well.
The problem is particularly bad in software. We have people who come
in with particular mindsets based on their everyday experience. They,
in some cases, have particularly bad backgrounds when it comes to the
physical sciences. A real problem because we deal with "exotic"
environments. The equally bad alternative is to have some physical scientist
come in an write code (Oh, this looks easy, I'll just...) and have
chaos ensue. Our scientists have the problem that they largely use
computers as simulation engines, when there are other applications.
Simulations are fairly safe. They are the product of the emphasis
of non-destructive testing.
But WHY do we joke when it comes to reality? "Oh, I took 40 jumps,
38 of them simulated" evokes a laugh or sudden loss of confidence.
Would your trust a robot to do brain surgery on YOU? Will there
ever become a time when our models, our simulations give us as much
confidence as "the real thing?"
Returning from OSU once, on the long drive back I thought of two
crazy ways to teach a class on "expert systems." Call it non-destructive
casualty learning. Expert systems (not a term that I like) are kind of
important because we have too few real experts and too many applications.
The first was for a class to write an expert system to defuse a bomb.
So you send in the entire class. Boom. Oh well, the entire class fails,
they all die. A smarter class sends it best programmer. He makes
one mistake (well how many perfect programmers are there?) and it
costs him his "life." He is placed in a "penalty box" where he can't
comment, but can watch. The other programmers have to do a real post
mortem.
Well the best programmer is gone, so it is up to the less than best to
solve these disarmament problems. Maybe a collectively smart class
sends in a not so brilliant programmer, etc. The tension mounts.
Your brightest programmer learns with the experience of the not quite
as sharp, and perhaps she (or he) learns a key clue to write such a system.
Eventually, a system gets written. You lose a few. You learn to
partition your resouces.
Exercise 2:
We actually have (not in my division) a requirement here for ESs like
the above. The NASP (National Aero Space Plane) which will fly out
of Edwards AFB/NASA Dryden Flight Research Facility will be fueled
by liquid Hydrogen. There are very few experts on the handling
of lots of LH2. They are mostly based at the Kennedy Space Center
where they are busy.
LH2 is very tricky stuff. It freezes just about anything. Rubber
breaks down, grease does not work, pumps freeze, the slightlest bit of
water is trouble. So its tricky. But these are the kinds of things
you experience everyday. 8) So you give the students an assignment to
help write an ES for LH2 (Pu, what ever). Well, you have a whole bunch of
assumptions. Your direct concern is handling LH2, but it turns out
that one object you frequently encounter when handling LH2 is liquid
Helium, the coldest substance. LHe, which isn't of direct concern,
is really bizzarre stuff. It doesn't behave like conventional fluids,
it helps make things superconductors, it creeps up the walls of things,
etc. It, like zero-G, has properties which are not in the normal
realm of human experience. But you don't tell the students about LHe,
that's learning. Forces you to deal with the chemists and physicists.
Take a bizarre briain to think this exercise up, but that's what
real life ;) does to you.
Longish signature follows "Type 'n' now"
Another gross generalization from
--eugene miya, NASA Ames Research Center, eugene@aurora.arc.nasa.gov
resident cynic at the Rock of Ages Home for Retired Hackers:
"You trust the `reply' command with all those different mailers out there?"
"If my mail does not reach you, please accept my apology."
{ncar,decwrl,hplabs,uunet}!ames!eugene
Live free or die.