Music

In 1975, I was labbie for the Computer arts class at Rice Unversity, run first by Ed Hayes (who later went to the NSF) and then by M. Stuart Lynn (Editor of the ACM). I had taken the (junior level) class in the first semester of my freshman year. It was tough and the labbie left. Dr. Hayes then asked if I would be the labbie. I still remember the conversation. "Why would I want to do that?" I asked. He said "Good question. The reason is becaue at some point, you will need a reference. And if you do a good job for me, I'll write a good reference for you."

Howie was in my first class and had build a music synthesizer at age 14 out of discrete TTL. So I convinced him to build a new digital (TTL) music synthesizer. We controlled it by tapping into the 10 bit X and 10 bit Y registers of the class's Tektronix 4013 graphics terminal. This gave us 4 two octave voices. We then wrote two different music composition algorithms.

One based on Markov matrices, which generated variations on Bach's "Alleluia" (similar to the approached used by generative AI 50 years later). You can hear it in the first 3:49, and at 6:49 of the recording below.
And one based on diminished seventh chords (Howie wrote this one). Starting at 3:50 in the recording below. There is also a link below to his algorithm and a stylized flow chart.

Each algorithm pumped notes in real time over a 300 baud line to the terminal. The pauses you hear happen when the mainframe stops sending results. So the synthesizer just "hangs" on the last note sent. Howie hated this. I rather liked it because it was the only truely indeterminate part of the composition. Occasionally you will hear an odd sound, not part of the composition. (e.g. around 9:55, 11:24). This is the mainframe sending a "READY" prompt to the terminal, and our hardware dutifully rendered the positions as notes.

I cleaned this up and entered it in the ACM's Computer Art Contest. Howie was convinced it wasn't up to snuff and ducked out last minute (note below). But he had done so much. So I entered it under both our names . We won a prize. How? There were 10 prizes and only 6 entries. The guy in charge was going to cancel the contest but I got M. Stuart Lynn (Editor of the ACM) to argue that all 6 enties ought win something.

We did this work in the media center. The media center head, Geoff Winningham, didn't think much of it (or us). So a large part of my work was keeping the equipment repaired and keeping us from getting thrown out. Ed Hayes and M. Stuart Lynn were sympathetic to my plight and wrote great reference letters for me. In fact, when I first spoke to Joel Moses (creator of MACSYMA, the first symbolic math program) at an MIT party, he said "you're the guy the knows Ed Hayes, right?" I thought that was curious, but later realized that Ed (in the NSF) was responsible for funding MIT's research and Stuart (ACM Editor) was responsible for publishing their results.

My Dad (who paid full freight for my Rice Education) once told me that in addition to learning stuff, He had heard that the reason for going to college was to meet people. (Dad himself only made it through eighth grade). And Howie has been a fast friend now for 50 years. And although I was rejected as an MIT undergrad, Dr. Hayes and Dr. Lynn helped me get into grad school.

So Dad, Right again.

1975 Howies Diminished 7th Chords.pdf
1975Howie Bill ACM Submission.m4a
2018-12-05 23.26.55.jpg