As everyone has heard by now, John Glenn flies again tomorrow, 36 years after
his first flight, in 1962, although that isn't exactly the reason I'm writing.

John Glenn isn't the only thing that has gotten older in 36 years. While no
one could begin to count the number of computers that will be used on
tomorrow's flight, the number that were used on his first is well known. The
total count was zero.

The only exception to that statement is perhaps the mechanical cam-and-
microswitch-based timer that was the primary sequencer for the Altas-Mercury
vehicle. It was just a tiny thing, about a foot long, with a precision DC
motor that drove a series of gears to greatly slow the rotation of the primary
camshaft. Attached to that camshaft were a series of ramps, not unlike the
fingers that operate an old player piano, that caused microswitches to switch
on and off. From the information derived from these switches, the vehicle was
launched, pitched and rolled into orbit. When John Glenn said, "The clock has
started," he was speaking of this timer.

It's interesting to remember how much we used to do without digital computers.
Indeed, in 1962, the jargon was even different. Digital transmissions, such as
this Sunday's HDTV, were called PCM (pulse-code modulation), and although not
rare, neither were they common. Virtually all telemetry and data transfer was
analog, using some form of FM or PM modulation.

The first integrated circuits were only synthesized in 1954, almost
simultaneously, by Fairchild Semiconductors and Texas Instruments. By chance,
both initial circuits only had seven components on them. TTL, the first truly
successful small-scale integration series, didn't appear until about
1964-1965, and these circuits only had a few dozen components per chip, and
yet TTL circuits prospered well into the 1980's (and you can still buy them).

I don't know if you remember the HP9100 calculator that was introduced in
1968, but I do quite well (http://aics-research.com/history3.html). It was an
extraordinary piece of engineering, well ahead of its time. And it was a heavy
little beast, its casing made out of cast aluminum. In the base of the
calculator were four large circuit boards, sandwiched together as tightly as
HP could reliably put them, each completely filled with resistors and diodes.
Those four boards were the calculator's program. In 1968, ROMs didn't exist.

Quite similarly to the HP9100, the "computer" that went to the Moon on the
Apollo spacecraft was built in precisely the same way. It was called the
DISKEY (display/keyboard) and used a simple noun/verb programming language. It
was built by graduate students at MIT, by hand, and then encased in Lucite so
that it was completely vibration insensitive. Each flight to the Moon required
a new program be built by hand.

However, the flight to the Moon could have been accomplished wholly without
the use of electronic calculators. It could have been done just using the cam-
and-switch sequencers that John Glenn originally flew (or were used by the
Germans seventeen years earlier at Peenemunde).

The moral perhaps is: these damn things just aren't that necessary :-).

Wirt Atmar