David J. Waks
Recollections, 1957 - 1961
One of my business associates (another Cornell graduate) recently sent me a link to the piece by David Bessel on the "Oral Histories - Cornell Computing and Information Technology" website (http://www.cit.cornell.edu/computer/history/). I had heard about the "history of computing at Cornell" section on the Cornell website, and had read the oral history by Dick Lesser. But I had not previously seen the piece by Bessel--my senior year roommate in a third-floor apartment over Johnny's Big Red Grill. So I decided to add my recollections:
I was a Cornell undergrad from Sept. 57 until graduation in June 61, majoring at the beginning in Electrical Engineering. The second week of my freshman year, my University Halls roommate (another EE) said he had heard there was a computer somewhere in Phillips Hall, and suggested take a look. That Saturday morning, we found the Computing Center with its doors locked, but the lights were on. So we knocked until a graduate student opened the door and reluctantly let us in.
The university had an IBM 650 -- a very early computer based on drum memory, and the university's first real computer. We were told that the Cornell Computing Center was only for professors and grad students, not for undergrads and especially not for lowly freshmen. But we were persistent, so one of the grad students gave me a manual for the language they used--the Bell Labs Interpretive System or BLIS--and said we shouldn't come back until we had written a program.
I showed up early Monday morning with a program to do my math assignment--I like to say it was the only program I've ever written that ran correctly the first time. And I've been hooked on computers ever since.
I started hanging around the Computing Center, doing odd jobs for grad students, and soon Richard C. Lesser, the Director, started paying me to punch cards, wire boards and after a while to write programs. By the spring of my freshman year, Dick offered me a summer job as the overnight operator, which I accepted. My job was to set up and run the programs that professors and grad students had left as large "decks" (often cartons) of punched cards to be run through the computer overnight and over the weekend. For each job, I had to put appropriate wired boards in the card reader/punch and 402 printer, then feed the cards into the hopper, refilling the reader hopper and removing punched cards and printouts as needed.
From that point until graduation, I worked for the Computing Center full time during the summer (I spent all three summers in Ithaca) and part time during the school year. (As I recall, the University had a rule that full-time students could not work for the University full time. So I worked for the computing center "part time" -- often 50 or more hours a week.)
In the middle of my sophomore year, the University ordered a new computer, a Burroughs 220 - much more advanced than the 650. The 220 had core memory rather than a drum although it still used electronic tubes for the logic elements. At the same time, the Computing Center was planning its move from Phillips Hall to Rand Hall.
One of the main ways computers were commonly programmed in those days was in "assembly language": the "higher-level" language just above "machine language" (the native language of the computer). One of my early tasks for CCC was wiring the board for SOAP (the Symbolic Optimal Assembly Program) for the IBM 650. SOAP was one of the earliest assembly languages. IBM published the SOAP source code in the SOAP manual as an example of how to code in the language, and it was the first example I had seen of what later became called "language processors". I read the code closely, and wrote several major programs in SOAP for CCC.
I looked at the documentation for the assembly program Burroughs provided for the 220, and observed that it was very primitive. I told Dick Lesser that it looked like it would be harder to program in assembly language than in machine language. He asked if I thought there was a better way, and I said "sure, I could write a better assembly program". So in the second half of my sophomore year, I designed and programmed the Cornell Assembly Program (CAP). Since we had not received the 220 as yet, I also wrote a simulator for the 220 running on the IBM 650, and I successfully debugged CAP on the 650 before the 220 arrived.
CAP became one of the main programming languages for the 220, and many people learned how to use the 220 from the manual I wrote during the summer between my sophomore and junior years, the summer the 220 arrived and the Computing Center made the move to Rand Hall.
For the professors and graduate students, computers were a necessary tool to solve problems; for me, they were an endlessly-fascinating end in themselves. Although I never had the title, I was effectively the Computing Center's senior system programmer for the rest of my time at Cornell.
By the first semester of my junior year in 1959, it was clear that my career would be in digital computing. The Cornell EE school seemed far more interested in power engineering than communications, and in analog computing than digital; for the next three years, all the courses in front of me were in power engineering and other EE subjects bearing little relationship to my interests. At the same time, Robert C. Walker, the Chairman of the Math department, had started teaching a numerical analysis course using my CAP manual as one of the texts, and the Math department seemed more interested in digital computers than was Engineering. Also at the same time, the girl I was dating made it clear that she would be more interested in somebody better rounded than an engineer ("engineers hang on their tails from trees"). All three suggested a move from Engineering to Mathematics. Walker was willing to support the move--although he wouldn't let me take the course he was teaching from my manual--and I did so in the middle of my junior year.
I'm pretty passionate about my enthusiasms, and I brought a lot of other undergraduates into the Computing Center orbit, including Dave Bessel and Alfred Kromholz, who became my roommates my senior year.
I'm almost certain I was the very first undergraduate computer "geek" at Cornell. The university didn't start teaching real "computer science" courses until my senior year (the Math department's course used computing as a tool for numerical analysis) and I never took any--indeed, I came close to teaching one of the first courses. Even if my degree doesn't say it, I feel I graduated as Cornell's first computing science major.
My Cornell experience set me on the path I've followed my entire life. September 2005 will be 48 years since I showed up at the door of the Computing Center in Phillips Hall.
David J. (Dave) Waks
Cornell 1961 B.A. Mathematics
Dave Waks, July 22, 2005