Unscientific Poll

[Jack Crenshaw]

Brian Wood wrote:

> Jack,
>
> Here's an interesting tidbit that a coworker at Agilent told me about. He
> said that many HP engineers built Heathkit color TVs because they could get
> them either free or at a huge discount as part of the GI Bill. It was an
> attempt by the government to foster the creation of more engineers after
> WWII (or maybe Korea; it was before my time).

Well, yes and no.  The kits came from one of the electronics training houses
(ICS?), that sold essentially programmed learning courses using electronic parts
that you built.   Anyone could take the courses -- I know _I_ wanted to -- but
you're quite right, they weren't cheap.  Virtually all the folks who took them,
took them under the GI Bill, which was passed after WWII to help train veterans
for a career other than shooting people (we already had enough of those <g>).
When I was in college (right after Korea), lots of my classmates were vets on
the GI bill.  But many elected not to go to college, but to use the ICS courses
instead.   They ended up as TV repairmen and such like.

AFAIK the GI Bill is still in place, and the ICS courses are, as well.  That's
precisely the kind of course that Heathkit teaches, today.

> Perhaps we need an equivalent
> "Bill" today to rekindle youngster's fascination with electronics. I can
> tell you that in our lab, there aren't many EEs left, and the ones that are
> still there are getting retrained in CS. It's to the point that if you want
> to design something, you might just have to subcontract it. Blecch.

Hm.  What lab is that?  AFAIK, there are still plenty of EE's coming out of
college, though admittedly they're more into computers than TV and analog
electronics.  I work in the embedded systems industry (aerospace-related), and
we get lots of our recruits from the ranks of the EE's.  As a matter of fact, we
find that EE's and physics grads make much better computer scientists than CS
grads <!>.

On the other hand, you might appreciate these stories:

(1 ) Back in 1983, I was working with GE, doing robotics stuff.  We used a VAX
as our development system, 8086's with bubble memories as the target hardware.
One of our team developed a high-speed "Bubble Blaster" that would do a fast
parallel transfer to load the bubble memories from the VAX. It involved using
one of DEC's parallel I/O boards.

When we tried it, however, we found that it was not reliable.  It was dropping
bits, even though our data rate was well within the stated spec.  Slowing down
the data rate didn't improve things, either.

I had a Ph.D. CS grad working for me.  I suggested that the problem might be
strobe timing, and suggested that he hook up a scope and look at the waveforms.

After about ten days, I realized I hadn't seen Scott in awhile.  I went into the
lab, and there he sat, staring at the scope as though mesmerized.  He had gotten
one of the EE's to hook up the scope to the right places, but didn't seem to
have a clue what to do next.  As nearly as I could tell, the guy had been
sitting there staring at those waveforms for a week, but had too much pride to
admit that he didn't know what they meant.  I don't think he even knew what the
knobs did.

The answer was staring him in the face.  They had hooked the strobe up with the
wrong polarity.  That was very obvious from the traces, since the leading edge
of the strobe was changing at the same time as the data.  A simple inverter
fixed the problem.  But I think that guy could have stared at those traces for
years, and never figured that out.

(2) I studied Physics at Auburn.  We were very proud of our physics program,
partly because it was very much oriented to applied physics.  In addition to the
physics majors, we taught sophomore physics classes to all the engineering
students.  The burden was high; classes tended to run as large as 160 students,
and handling the load was tough. Several folks suggested that we drop the labs,
since they not only required skilled instructors, but lots of hardware as well.
But the dept. head, Dr. Howard Carr, was adamant that kids needed to know how to
do lab work.  As a result, we were among the last universities to require lab
work of all our students.

Some years later, Dr. Carr was invited to visit one of the Ivy League colleges,
either Yale or Harvard, I forget which.  The physics dept. had some kind of
gov't contract to build some electronic widget.  Their department head proudly
showed Dr. Carr around the facility, ending in the calibration lab where final
test and cal was done on the widget.  There was a huge lab filled with young
folks in lab coats, who were busily probing and adjusting the widgets.  They'd
hook up a test signal, and adjust a couple of pots to get the right reading on a
meter.

Now, at Auburn, we always had a few physics candidates who weren't very good at
the math, but very good at tinkering with hardware.  We'd gently suggest that
maybe they'd find a career as a lab tech more to their liking, and we usually
retained a few to maintain our own lab gear.  Dr. Carr reasonably guesses that
he was seeing more of the same, but he was surprised by their numbers.  He asked
the dept. head if some of them might be moonlighting undergrads.

The dept. head replied, "No, actually, they're graduates."

Dr. Carr was somewhat taken aback.  "You mean that all these lab techs already
have their bachelor's degrees?"

"No," said the dept. head, "They all have their Masters."

Stunned, Dr. Carr said, "Let me get this straight.  You're using guys with
Master's degrees to work in a lab reading meters?"

"Yes," said the department head.

"Do they like that?"

"Not much," was the reply.

"Don't you have a high turnover?"

"Well, yes, we do.  That's one of our problems."

Dr. Carr asked, "So why do you do it?  Why don't you use undergraduates?"

The man answered, "Because our undergraduates don't know how to read a meter."

Jack

Listserver Subscription:listserv at listserv.tempe.gov - "subscribe heath 'name' 'call'"
Listserver Submissions: heath at listserv.tempe.gov
Listserver Unsubscribe: listserv at listserv.tempe.gov - -"signoff heath"