The USC summer — a PDP-8, a laser, and Brownian motion

I had programmed a PDP-8 computer when I spent a summer at USC's Aerospace Engineering Department after sophomore year in high school, using it to control the laser we used for illumination for tracking particle transport in fluid boundary layers. The professor I worked for was running an experiment with a large rectangular water flume made of clear plastic or glass that established laminar flow of water, and they would lower aerodynamic shapes simulating aircraft fuselages, wings, submarines or torpedo-shaped submersibles into the water, and then inject tiny particles into the flow, and take high speed videos of the particles flowing around them to see if they maintained laminar flow, or if they became turbulent, and where. There was a laser that swept back and forth rapidly so you could take sort of a point in time view of the particle stream, and visualize the flow.

The PDP-8 was a vertical rack computer — you could push it around on little wheels, and it had a keyboard and monitor halfway up the rack so you could program it. Lots of cables and wires also. All this to control a little laser. It was massive overkill — they just had it sitting around, so my job was to program it, assemble the laser and the little motors that swept the silver-covered mirrors (to ensure no distortion). I once touched the silver-covered mirror with the tip of my finger, which left my fingerprint embedded in the silver — totally messed up that run of the experiment, and it was a giant pain to fix that little mirror.

Add a link to Einstein for this one, because the fourth (and least significant in cosmological terms) of his Miracle Year papers was on Brownian Motion, which is the basis for what we were studying with particle transport in fluid boundary layers. That was the first real scientific experiment I ever did, and it was based on Einstein's work.

This summer program was sponsored by USC, which was trying to recruit the best and brightest high school students — they started on me years before I graduated. We lived in the dorms on campus, which were mostly empty in the summer. The only people on campus other than faculty and grad students were football players and cheerleaders, and the USC marching band, because they all practiced for a couple of months in the summer.

I used to hang out in the movie theater that George Lucas donated to USC. No food allowed, it was all red — heavy curtains to prevent echoes, luxurious seating, a giant screen, and great air conditioning. That's where I used to read a lot. I never saw a movie there, and I would like to.

A boat trip out to Avalon on Santa Catalina island, about 30 miles out to sea from Long Beach. This was great, as I was the youngest member ever of the Sea Scouts at age 12 — I was 'the Cabin Boy'. I could navigate in a pre-GPS age — RDF, LORAN, at night using sextants, by tracking the sun's arc and an accurate clock. Sea Scouts' mission was to train future merchant mariners and naval officers. It was literally like being in the navy.

Avalon was nothing like that — a little Mediterranean style town in a cove, beautiful weather, beautiful people, and you either walked, biked, or rode around in little tiny resort cars. It was created as a resort in the 1920s, back before jet setting was a thing. I realized that being rich is better than being poor. My only experience of this in the past was in books (Agatha Christie novels) and movies (James Bond), and from reading Richie Rich comic books when I was a child living on a clothing-optional Marxist-Leninist commune. This seemed a lot better, and I decided I would like to own an island some day, and a yacht, and speedboats. And a seaplane.

On another trip we went to see a water pump power storage system out on the way to Tehachapi Pass — they pumped water up to a reservoir at the top of a mountain when excess power was available, and then used it to power turbines to generate power during peak demand. We visited the power generation plant, the electrical grid substations, and studied the various ways of managing a power grid — it's extremely complicated and very important. All this was very useful when I worked at Impell Corporation in the nuclear power industry after my sophomore year at UC Berkeley, back when the United States led the world in nuclear power and energy systems in general.

The AI revolution is driven by computation, and most people think about Nvidia and TSMC, but the smartest people realize that it is really power that is the constraint — Moore's law is much weaker in power than it is in silicon.