Early Programmers at Point Mugu
Programmers by definition are perverse.
In the very early days, the Navy had at Point Mugu in California
a computer named RAYDAC (designed by Dick Bloch, of whom more
later at GE in Phoenix), but more commonly called "The White
Elephant". It was a big monster, and one could walk inside it.
The task was to compute trajectories for shell fire. When the
program neared completion, it was noted that the range was given
in miles. This wouldn't do for the Navy, and the programmers
were ordered to put the ranges in nautical miles.
Back they came the next day, with range (as ordered) indeed in
nautical miles, but with altitude in "negative fathoms"!
Early Word Processing(?) at Marquardt
In 1953 my computer installation at Marquardt Aircraft had an IBM 607
and a CPC (Card-Programmed Calculator). We did a lot of data reduction
on ramjet engine test beds. This required the services of many
keypunchers, mostly girls. You must remember that these were big
machines, and required a lot of stable power. If that was interrupted,
you could not keep on running on batteries like a PC can. So the
machine room, right in the center of the building, had power supplied by
a motor-generator unit.
The calculating units in both computers were over five feet tall, about
two feet wide, and three to four feet deep. The arithmetic unit had two
groups of tube assemblies driving driving display banks of neon tubes
four deep, to represent 1, 2, 4, and 8. The groups were named Factor
Storage and General Storage. Any or all of the neons could be switched
on, if you knew how to trick the machine.
We had one Dave Hemmes as a programmer/operator, whom I had acquired
somehow from the RAND Corporation. In my entire life I have never seen
anyone that was such a character. And he did know how to trick it.
One day all the power in the building went off except for the MG in the
machine room, which was driving the little flashing neons merrily.
Otherwise, complete darkness. Dave Hemmes punched a special card, wired
a board quickly, and soon had the 607 doing a special operation --
namely, spelling out two words -- one in Factor Storage and one in
General Storage. Blinking 100 times a minute, at card reader speed.
The second word was "YOU"; the first word started with "F".
It was a little fast, so Hemmes put in a delay wire to cut it to 50 times
a minute. Then he called in all the keypunch girls to see how beautiful
the little flashing lights were in the darkness. They agreed, and duly
exclaimed about it until gradually they began to get the message.
"What's that?", they cried in horror.
Dave's reply -- "Why, that's a calculated insult".
Hemmes Impressing the US Air Force
Somehow the Air Force, a major Marquardt customer, wanted to film the
computer operation we ran for them. The only thing moving and
photogenic was the printer, which always ran at top speed of 150 lines
Everything seemed to be going well until Hemmes strolled into camera
view, peering over the answers, and thumbing an abacus as though to
check the results.
Hemmes in Leisure Hours
Dave loved to go hunting rabbits, and he was a very sporting type.
That is, he didn't just stand in one place and fire. He ran after
the rabbits, firing as he ran. Gave them a better chance, he said.
He invited me to go with him one weekend, but I had committed to a
party fishing boat on the far side of Catalina Island. There the
Captain announced that we might catch "jungle bunny", but would not
explain what he meant. "You'll know it when you see it", he said.
Sure enough, I caught a large one. Turned out to be a variety of
sheepshead, in stark red, white and black bands. I kept it.
I didn't see Dave when I returned, but there was a paper sack on my
doorstep. In it was a dead rabbit. Suspicious as always of Dave,
I checked the back of the neck, finding a big lump. Tularemia!
It remained only to get even. I got a 9" square cake pan, put the
fish head in it, filled it with plaster, inscribed it "To My Friend
Dave". A coat or two of spray lacquer, and off to his house. He
was much embarrassed at my turning the other cheek, and apologized.
He closed the garage door, and we went into his house to get some beer to
celebrate. Apparently he had no need to go into the garage for several
days. When he did, his nose found my embalming job to be very
Hemmes at IBM
When I decided to do the FORTRANSIT product at IBM [1. 2], a major
problem was the wiring board the 650 used for input/output -- reading
and punching. We needed alphabetic input for the FORTRAN statements, and
the IBM sales types wanted to sell a special version of the 650 for just
that. I wanted the broadest possible usage of FORTRAN, and did not want
to have users turned off by higher costs. What I needed was a board
Hemmes was the obvious choice among my acquaintances. He had worked for
me at Marquardt Aircraft, and then followed me to Lockheed, Missiles and
Space Division, where he was when I thought of him, But I had a hard
time persuading him to move to New York, He was really just a
After we finished that project we moved into plusher quarters at 425
Park Avenue in New York City. Forget which floor, but Dr. Charles
deCarlo, who headed the Applied Science Department, had the office at
one far end, to the left facing the avenue. My office was at the very
opposite end. I guess I reported to deCarlo, for he had to approve my
first trip abroad, to the ICIP Conference in Paris in the summer of
1959. But I also sort of worked for John McPherson, engineering-type
Vice President with roving portfolio. We were so busy that I did not
have time to get confused by this.
Our group was far enough removed physically from deCarlo that they
thought they had a license for pranks. Among them were:
All this must have been beneficial, for the computer business today
owes a large debt to what we created under these zany conditions.
How about half of COBOL, via Commercial Translator, for example?
Back to History Index
Back to Home Page
- R.W.Bemer, "Nearly 650 memories of the 650",
Annals of the History of Computing 8, No. 1, 66-69, 1986 Jan
- D.A.Hemmes, "FORTRANSIT Recollections",
Annals of the History of Computing 8, No. 1, 70-73, 1986 Jan