The Kilogirls
In 1945 one of the world’s first programmable, electronic general-purpose computers, the ENIAC, was unveiled. However, there was one small problem — there was no one to program it. In theory, the computer could be used to solve any problem with its vacuum tubes, function tables and accumulators. However, the creators didn’t bother to write a manual. Instead, they left that task to a group of women who would be replaced by the ENIAC.
Before modern computing, a “computer” referred to a person. In the same way an accountant is a person who does accounting, or a scientist is a person who does science, a computer was a person who computes. In fact, the word “kilo-girl” was used as a tongue in cheek way to refer to the relative computing power of new machines — 1 kilo-girl meant the machine had the same computing capacity as 1000 women.
Around the end of the second world war at the University of Pensylvania, a group of six women were being introduced to the ENIAC — a machine designed to calculate ballistics trajectories. Plucked from pools of manual computers, they were tasked to operate the hulking giant. The women jumped at the chance, since the only other career options open to a woman with a technical mind were teaching or doing calculations for insurance companies.
Kay McNulty, Jean Bartik, Betty Holberton, Marlyn Meltzer, Fran Bilas, and Ruth Lichterman were the first programmers of the ENIAC, and almost all of them were never recognised for their contribution during their lifetimes. At the time, programming was considered “clerical work”, even though the women had to have in-depth knowledge of mathematics and engineering to operate the machines.
“They sent it via telegram, they said I was hired, and they said ‘come immediately!’… And guess what? I was hired as an SP6-Subprofessional. They did not give professional titles to women at Aberdeen, even if you had a PhD.” — Jean Bartik
The machine weighed 27 tons, occupied 167m2 and consumed 150kW of electricity. Programming the ENIAC consisted of first mapping the problem to something that could be understood by the ENIAC’s architecture, and then physically translating the map to a working program on the machine by manipulating its switches and tables.
Kay McNulty, who ended up marrying John Mauchly, during an interview in 1977 tells the story that someone handed them a stack of blueprints. They were wiring diagrams for all of ENIAC’s panels. And he said, “here you can figure out how the machine works, and then figure out how to program it”. The women learned by doing and debugged problems by crawling into the machine to find faulty joints and tubes.
“The ENIAC was a bitch to program” — Jean Bartik
When asked if Eckert and Mauchly (the creators of the ENIAC) could program the ENIAC as well as Jean and the other women, she replied “well no, of course not. Don’t get me wrong, they knew how the ENIAC worked completely, but that’s not what they did.”
The women were tasked to write a program to be displayed to the press. They worked tirelessly over a 2 week period wrangling the machine in order to solve a bullet trajectory calculation — the main purpose of the ENIAC at the time. At the end of their work, the output from the calculation was displayed proudly to the press, but to the women’s dismay, they weren’t mentioned in any of the articles. They were often mistaken for models, showing off the machine by standing next to it.
“The day ENIAC was introduced to the world was one of the most exciting days of my life. The demonstration was fabulous. ENIAC calculated the trajectory faster than it took the bullet to travel. We handed out copies of the calculations as they were run. ENIAC was 1,000 times faster than any machine that existed prior to that time. With its flashing lights, it also was an impressive machine illustrating graphically how fast it was actually computing.” — Jean Bartik
The women went on to work on future projects with Eckert and Mauchly, and other companies formed in the early years of computing. These women were at the forefront of computer science, and arguably invented the practice of modern software engineering.
Kay McNulty invented the subroutine in order to increase the computing capability of the ENIAC. Betty Holberton invented the first generator program (a program that creates another program), the SORT-MERGE generator. Betty also invented breakpoints, specifying a point at which the computer should pause so that the programmer can inspect the state of the program to find a bug.
“We would narrow our problem down by pulling the circuit. Well, we had to have wires to do that, and that’s what we broke…. And so we can narrow it down by running it at full speed and stopping at various points along the way until we actually saw what was happening. That’s really the first chance I had at taking a large mass of equipment and determining at what point things went wrong.” — Betty Holberton
Betty was also responsible for another important decision in computing history — she convinced the engineers to make the UNIVAC beige, instead of the black of the ENIAC. That’s right, she was responsible for this trend:
The women who were the first programmers of the ENIAC went on to make a big impact on the history of computer science, that was all but forgotten for decades. In her incredible book, Broad Band — The Untold Story of the Women Who Made the Internet, Claire Evans notes that “Men may have dropped bombs, but it was women who told them where to do it.”
When Jean Bartik was asked what she advice she would give to young women in technology today, she said:
“Don’t ever let anyone tell you that you can’t do something because they think you can’t. You can do anything, achieve anything, if you think you can and you educate yourself to succeed.” — Jean Bartik
This article is part of a series of stories about the history of women in tech I’m writing in preparation for my talk at DevConf 2019. See my previous post in the series.
