Rear Admiral Grace Hopper joined the US Navy and in 1944 was posted to the Bureau of Ordnance Computation Project at Harvard University, where she began work on the IBM Automatic Sequence-Controlled Calculator, known as the Mark I. She retired as a Naval Commander in 1966 and is credited with popularising the idea of machine-independent programming languages.
How did you learn to program?
The idea of the ‘program’ hadn’t come over from England yet – we ran ‘problems’ instead. Nobody trained. All I had was a code book and there was no time to talk or to be taught.
It was wartime, and we were operating on just one idea: win that war. We hadn’t had rockets for long and we had no firing tables for them. We also had new torpedoes and nobody knew what they would do. All of these had to be computed. We were in there 24 hours a day, for the number of days a problem was running.
One of the many things that you are famous for is inventing the word ‘bug’ to describe a computer glitch. How did that come about?
I didn’t invent the ‘bug’. Engineers had been using ‘bug’ to mean a fault or problem for at least a century before we picked it up, but we did have plenty of opportunity to use the word.
The Mark I was the largest electromechanical calculator built, 51 feet long and eight feet high, and made of panels of small gears, counters, relays, vacuum tubes, rotating shafts and clutches. We fed it instructions on 24-channel punched tape but if the tape got on the floor, and there were holes on the floor, those things would get back into position and stay there. Frequently we found bugs that were nothing more than a punched-out hole that had been filled in again.
We once found a dead moth in Relay #70 – that was our bug – so we stuck it in the log book. I think that was the first case of an actual bug being found.
After the Mark I, you moved on to write the first compiler. What is a compiler and why did you create one?
Nowadays, programs are written in languages that are easy for humans to work with, like C++. A compiler takes those human-readable instructions and translates them into a sequence of numbers that the computer chip can understand.
My compiler was much simpler, because we didn’t have high-level programming languages back in the early 1950s. I was working on the UNIVAC, the Universal Automatic Computer. We were writing pieces of code, subroutines, which I collected and put on punch tape, each with its own identifying ‘call number’. All I had to do was to write down a set of call numbers, let the computer find them, and do the mathematics. I was lazy, and didn’t want to have to repeat anything that I had already done!
Did that kind of approach make life easier for everyone?
Not for two years or so. I had a running compiler but nobody would touch it. They told me that computers could only do arithmetic; they could not do programs. As with any new idea, I had to get out there and sell it because people are allergic to change.
Tell me about COBOL, the Common Business-Oriented Language that you were so heavily involved in.
The idea of using computers in business, in insurance, was one that I first heard from my father, Walter Fletcher Murray, in 1944. Fifteen years later, a team of about seven of us created COBOL. The idea was to put together a standardised, universal computer language that could run on any machine. If it was going to work in a business environment, it had to be able to run on anything.
You’ve received at least 37 honorary degrees, seven medals and numerous awards. Which of your achievements was the most important?
I’ve received many honours and I’m grateful for them, but the most important was not the development of the compiler or COBOL, but all the young people I trained. We must give young people the positive leadership they’re looking for.
Suw Charman Anderson is the founder of Ada Lovelace Day. findingada.com