Grace Hopper

Welcome to the History of Computing Podcast, where we explore the history of information technology. Because by understanding the past, we’re able to be prepared for the innovations of the future! Today’s episode is on one of the finest minds in the history of computing: Grace Brewster Murray Hopper. Rear Admiral Hopper was born on December 9th, 1906 in New York City. She would go on to graduate from Vassar College in 1928, earn a master’s degree at Yale in 1930, and then a PhD from Yale in 1933, teaching at Vassar from 1931 until 1941. And her story might have ended there.

But then World War Two happened. Her great-grandfather was an admiral in the US Navy during the Civil War, and so Grace Hopper would try to enlist. But she was too old and a little too skinny. And she was, well, a she. So instead she went on to join the women’s branch of the United States Naval Reserve called WAVES, or Women Accepted for Volunteer Emergency Service, at the time. She graduated first in her class and was assigned to the Bureau of Ships project at Harvard as a Lieutenant where she was one of the original programmers of the IBM Automatic Sequence Controlled Calculator, better known as the Mark I.

The Mark I did what the analytical engine tried to do but using electromechanical components. Approved by the original IBM CEO Thomas Watson Sr, the project had begun in 1937 and was shipped to Harvard in 1944. If you can imagine, Hopper and the other programmers did conditional branching manually. Computers played a key role in the war effort and Hopper played a key role in the development of those computers. She co-authored three papers on the Mark I during those early days. She also found a moth in the Mark II in 1947, creating a term everyone in software uses today: debugging.

When peace came, she was offered a professorship at Vassar. But she had a much bigger destiny to fulfill. Hopper stayed on at Harvard working on Navy contracts because the Navy didn’t want her yet. Yet. She would leave Harvard to go into the private sector for a bit. At this point she could have ended up with Remington Rand designing electric razors (yes, that Remington), or working on the battery division, which would be sold to Rayovac decades later. But she ended up there as a package deal with the UNIVAC. And her destiny began to unfold.

You see, writing machine code sucks. She wanted to write software, not machine language. She wanted to write code in English that would then run as machine code. This was highly controversial at the time because programmers didn’t see the value in allow what was mainly mathematical notation for data processing to be available in a higher level language, which she proposed would be English statements. She published her first paper on what she called compilers in 1952.

There’s a lot to unpack about what compilers brought to computing. For starters, they opened up programming to people that would otherwise have seen a bunch of mathematical notations and run away. In her words: “I could say "Subtract income tax from pay" instead of trying to write that in octal code or using all kinds of symbols.” This opened the field up to the next generation of programmers. It also had a second consequence: the computer was no longer just there to do math. Because the Mark I had been based on the Analytical Engine, it was considered a huge and amazing calculator. But putting actual English words out there and then compiling (you can’t really call it converting because that’s an oversimplification) those into machine code meant the blinders started to come off and that next generation of programmers started to think of computers as... more.

The detractors had a couple of valid points. This was the early days of processing. The compiler created code that wasn’t as efficient as machine code developed by hand. Especially as there were more and more instructions you could compile. There’s really no way around that. But the detractors might not have realized how much faster processors would get. After all they were computing with gears just a few decades earlier. The compiler also opened up the industry to non-mathematicians. I’m pretty sure an objection was that some day someone would write a fart app. And they did. But Grace Hopper was right, the compiler transformed computing into the industry it is today. We still compile code and without the compiler we wouldn’t be close to having the industry we have today. In 1954 she basically became the first director of software development when she was promoted to the Director of Automatic Programming.

Feeling like an underachiever yet? She was still in the Navy Reserve and in 1957 was promoted to Commander. But she was hard at work at her day job as she and her team at Remington Rand developed a language called FLOW-MATIC the first natural language programming language. In 1959, a bunch of computer nerds were assembled in a conference c...