A code pioneer

Last month, J.N. Patterson Hume was inducted into the Canadian Information Productivity Awards (CIPA) Hall of Fame. Although born in Brooklyn, N.Y., Hume moved to Canada in the early 1950s to take a job as an assistant professor, where he continued to make a name for himself a the University of Toronto, eventually helping to found the school’s computer science department.

Hume’s list of accomplishments is long. In 1953, working with one of the first computers in Canada, he helped to invent one of the first programming languages in the world. He was also a pioneer in the field of science/IT education – along with a colleague, Donald Ivey, he founded the television show The Nature of Things in 1960. And with the help of a university colleague, Calvin Gotlieb, Hume wrote a textbook called High-Speed Data Processing, the first book ever published on the use of computer applications in business. Long since retired, Hume continues to write books, specifically on the Java programming language.

ComputerWorld Canada editor Michael MacMillan had a chance to sit down with Hume shortly before he was inducted into the CIPA Hall of Fame.

CWC: Can you explain how you became involved with computers at a time when they were still in their infancy?

Hume: I had been trained as a physicist, doing calculations of atomic wave functions. This involved an enormous amount of calculations and the only kind of equipment we had available were mechanical calculators, not even electrified. So I was subjected to this as a grad student.

After I was at Rutgers University for a year doing my PhD, I was asked if I wanted to come back to Toronto and become the assistant professor of physics. So I came back and happened to be assigned to an office with Calvin Gotlieb. He was in charge of the computation centre, which had a lot of punch-card equipment, and they even had a fancy thing called a card-programming calculator. So I thought I would try to use this to do my dirty work for me. Bit by bit, I got more and more interested in this computing equipment.

Then in 1952, the university acquired the first ever-electric computer in Canada. It was a Ferranti, and was also the second computer in the world ever sold. Having gotten this machine, I thought to myself, ‘Well, I’ll use it.’

CWC: So how did the programming language come about?

Hume: There weren’t people who knew how to program the thing, so I learned how to program it, and that required an operating system. So I created this operating system for it, and then it needed debuggers, so I did that. And still with all our attempts to train scientists, very few people knew how to do it except people paid by agencies like defence research boards or national research councils. No one else picked it up, it was too hard. So I said, ‘this is ridiculous, why can’t the machine translate the fairly simple language we use into machine code?’

And so with the help of Beatrice Worsely, who was an assistant professor at the time, we set about and created a thing called Transcode, which absolutely revolutionized the activity that could go on the computer. This reduced the amount of learning time to program to just a few hours, and anyone could be taught to do this because it was so straightforward and simple, and not only that, they didn’t make mistakes – most people who tried machine code got into a heck of a lot of problems with errors.

CWC: What was the reaction to Transcode?

Hume: Grace Hopper at Univac wrote us a letter saying that she was encouraged to hear of our success in this area because it was very hard to convince people who knew how to program in machine code, that they shouldn’t be doing that, they should be using some simpler language.

Of course the whole language thing took off after Fortran. But this wasn’t exactly a full blown programming language, and it was also very specific to our computer, so it was not transferable. But in those days the notion of 10 or 15 copies of the same machine was nonsense.

CWC: How has the task of educating students in the art of programming changed over the years?

Hume: There’s an awful lot more to the whole business of computer science, the notion of artificial intelligence, the theory of computing and things that are behind data structures and so on, building systems that underlie the actual individual things that are going to happen.

At one stage there was a very important development in computing and that was the recognition of what’s called structured programming. We were producing vast programs and they were not easily understood except by the person who wrote them, so they had to start over again. It was the constant reinvention of the wheel. This changed with structured programming, the notion that programs could be written in a way that anyone could pick them up and say, ‘oh I can see what you’re doing here and here.’ It’s just moved gradually away from somebody who’s a real high priest of programming to now, where everyone can program.

CWC: I recently read a report that called 2002 the worst ever year for IT. Do you agree?

Hume: Yes, this is terrible. I worry that we’ve had a good ride in computer science for a number of years. I don’t think, on the whole, computer science has the cachet of law and medicine as far as prestige. We have been attracting students because there were jobs at the end of the rainbow and work after a bachelor degree. Now that jobs are much harder to come by, I don’t know whether it is going to turn off prospective students.

CWC: Professionally, what are you most proud of?

Hume: I’ve always done just what I wanted to do. I just followed it from interest to interest, so I suppose I’m not much of a specialist. But I don’t want to be. I have been retired for 15 years. I’ve produced a number of books on the Java language that didn’t exist when I retired.

I also think teaching, passing things along to young people is terribly important, and making science seem like an accessible thing rather than some ‘other activity.’