Personal Reflections

A Personal Friend

Ralph Griswold is not exactly a household name, but he was one of the most important people in the history of Computing, and certainly the only one that I would call a “personal friend”. While at Bell Labs in the 1960’s, Ralph developed the programming language SNOBOL, refining his concepts in the 1980s in Icon, an extremely powerful C-like programming language that runs on nearly every computing platform known to Man. Sadly, he was retired and suffering from pancreatic cancer before he had a chance to see the world catch up to him with Industry now beginning to accept Open Source software. Dr. Griswold had kept both SNOBOL and Icon in the Public Domain, arguing tirelessly that the Computer Science Research community could provide better support for software “products” than commercial vendors. In my experience, he proved it.

I attended two conferences with Ralph in the early ’90s, and got to know him and his wife Madge quite well. We kept in touch until well after his retirement in 1995, and much of our discussion was about the IBM mainframe version of Icon. He was one of those rare Human Beings with a mixture of personal Humility and Passion for his projects. I miss him already.

written in my weekly team report, 18 hours after I learned of his death, and three weeks before my retirement

I Owe my Degree to Ralph Griswold

In my fourth and final year of Computer Science at UBC (1973-74), I made two major mistakes:

  1. Taking nothing but Computing courses (overall, I had taken twice as many as required);
  2. Taking a Compiler Design course.

UBC was non-semestered, so courses lasted the full academic year. It was clear that, unless I gave up sleeping, I was not going to finish my Compiler Design projects by the time I was scheduled to graduate with a Bachelor of Science degree. I finally made an appointment and visited my professor’s office.

We struck a bargain. If all I wanted was the 50% mark I would need in Compiler Design to graduate, successfully completing the two remaining projects was all that was required. Unfortunately, I knew that Algol-W was not the programming language for the task, not just because of the volume of coding required, but also because programs that big tended to fail on obscure memory-related problems.

I proposed writing them in SNOBOL4. But my professor was unfamiliar with the language, so we agreed that I would comment each line of code with enough detail that any programmer unfamiliar with SNOBOL4 would understand it.

I did my part, and so did my professor, and I graduated on schedule, in April 1974.

Thank you, Ralph Griswold.

Jon Pearkins