I Code Like a Girl

I Code Like a Girl

Programmers Betty Jean Jennings (left) and Fran Bilas (right) operate ENIAC's main control panel at the Moore School of Electrical Engineering. (U.S. Army photo from the archives of the ARL Technical Library)

Programmers Betty Jean Jennings (left) and Fran Bilas (right) operate ENIAC’s main control panel at the Moore School of Electrical Engineering. (U.S. Army photo from the archives of the ARL Technical Library)

The American computer industry seems dominated by people like me – males. But it was not always this way, in fact I stand on the shoulders of women.

While writing about SAGE in previous posts, I learned that at the beginning of the computer age, most programmers were women. My skills, including programming languages and tools, have a foundation built on the contributions of those who came before me. It turns out that I code like a girl, using concepts pioneered by women.

Computers: Human Women or Machine?

In the 1940’s and 50’s, there was a peak in women working as computers. Mathematically intensive industries like nuclear research, ballistics, and engineering often employed women to perform computations.

Do you remember “When Computers Were Women?” The article reminds us that a “computer” was actually a human being until around 1945. Afterward, a computer was a machine and humans were called operators.


In 1946, the first electronic general-purpose computer called ENIAC (Electronic Numerical Integrator And Computer) was built. This began a process of reducing and automating manual calculations.

A select number of women operators and former “computers” were enlisted to become programmers for ENIAC. Notably, the first ENIAC programming team was comprised entirely of women: Kay McNulty, Betty Jennings, Betty Snyder, Marlyn Wescoff, Fran Bilas, and Ruth Lichterman.

Women programmer School of Electrical Engineering. (U.S. Army photo from the archives of the ARL Technical Library)

School of Electrical Engineering. (U.S. Army photo from the archives of the ARL Technical Library)

At the time, there was a strong division between the male domain of hardware and the female sphere of software. Male electronic engineers built the ENAIC system. But since software design and programming were considered clerical work, women instructed the the 27-ton hand-built mass of wires and vacuum tubes to perform calculations in sequence.

In true programmer fashion, the women learned by doing. The hardware engineers dropped the blueprints and wiring documents on them and said, “Here figure out how the machine works and then figure out how to program it.” So the women crawled around the massive frame and learned how each component worked. They successfully understood the interplay between hardware and software and how the computer’s behavior could be traced to a hardware or a software issue.

Unfortunately, all the credit for creating ENIAC went to the men who conceived it and built the hardware. The media covered the debut of ENIAC in February 1946, which showed off the centerpiece calculation of a trajectory. The program created by Betty Snyder and Betty Jennings impressed the VIPs because it allowed the computer to calculate faster than the projectile itself. But the women were not mentioned, seen in pictures of the event, nor invited to the press lunch with the men. In the end, the computer was the star of the show, depicted as an autonomous brain.

Gender Code

Words powerfully describe gender roles. What is now considered a male-dominated field, was once defined as “women’s work.” In the days of the first computers, the norms were as follows:

Male Female
Hard sciences Soft sciences
Engineering Programming
Hardware Software

Society was keen on recognizing men’s contributions, while neglecting those of women. In the book, _Recoding Gender: Women’s Changing Participation in Computing_, Janet Abbate found that publicity materials for ENIAC state that the machine reduced 25 man-months of human computer time to two hours on the ENIAC. However, it fails to mention that most of the human computers were really women. The materials also neglect to highlight the years of labor by both men (on the hardware) and women (writing software) to create the system. The only human labor noted in the press was the initial design of the machine, which was performed by men.

But even women of the time seemed to define their computer jobs as gender specific. Elise Shutt was a programmer on a later version of ENIAC called ORDVAC. When she was hired by Raytheon in 1953, she said, “It really amazed me that these men were programmers, because I thought it was women’s work.”

In another example, Grace Hopper compared programming to tasks like sewing clothes, making a recipe, and the work of a mother teaching a child. Thus, she defined programming as a female occupation. But this seems to have been lost on her supervisor, Howard Aiken who said in praise of Grace, “Grace was a good man.

Recruiting materials were also used to attract women to programming with various metaphors and generalizations. In the 1940s, MIT had a shortage of men and highlighted skills such as needlework and knitting as characteristics useful for programmers. Others noted that female pursuits like crosswords and puzzles would make good programmers.


The first women who pioneered programming on the ENIAC finally gained the recognition they deserved in 1997 when they were inducted into the Women in Technology Hall of Fame in 1997 and IEEE in 1997 and 2008.

It seems that we continue to struggle with metaphors and defining skills used to train and attract the next generation to computer work. From recent statistics, we are finding a wider gender gap in the computer industry. The reasons for this are inconclusive and give us a reason for self-evaluation and consideration of language used to hire and promote, treatment of women, and how skills are evaluated.

It becomes increasingly important to value each member of a programming team regardless of gender, age, race, or creed to attract and keep the best minds to build our future software. There seems to be no end to the amount of programming work needed. Code on!


Check out more of our work at Volume Integration and follow us on Twitter.

How I Learned to Stop Worrying and Love the Computer

How I Learned to Stop Worrying and Love the Computer

In the 1950s, the United States was worried about the threat of a Russian nuclear attack. But an Air Force computer helped allay those fears.


My last post about oscilloscopes mentioned the Semi-Automatic Ground Environment (SAGE), which was the technological savior of its time. SAGE was a system built in the 1950s to warn of incoming Russian bombers with nuclear bombs.

SAGE was composed of over 24 computers, and each contained a primary and backup CPU to provide a 99.6% reliability rate. But you have to keep in mind that each CPU took up 10,000 square feet. They were spread across the US plus one in Canada and connected by the first modems tied to standard AT&T phone lines.

In 1959, IBM created the 7090, a faster and smaller solid-state enhancement of the work done with the SAGE system. This computer appeared in the 1964 movie Dr. Strangelove or: How I Learned to Stop Worrying and Love the Bomb. The movie would have been more realistic with the original SAGE system, but it was in use protecting the country.

Peter Sellers as Captain Lionel Mandrake surrounded by an IBM 7090

Computing Innovators

SAGE kicked computer research and the creation of programming methods into high gear. This machine was responsible for creating multiple companies and was the starting point for researchers who went on to create the forerunner of the Internet.

The info graphic below shows one branch of the tree that leads to present day companies that continue to provide research and development in the computer industry today.

Legacy of US Government Funded-Computer Industry (IBM, MITRE, Noblis, L-3 Communications, Lockheed Martin, Unisys, and MIT Lincoln Laboratory are all trademarks of their respective companies)

More Problems…More Innovation

Of course, it was not that easy to deter the Russians. Unfortunately, once the entire network of SAGE sites were in place around 1964, the threat of attack by bombers had been replaced by the threat of intercontinental ballistic missiles (ICBM). But the SAGE system was not fast enough to track incoming missiles from space. In fact, the system was not even fast enough to prevent a large-scale attack against radar jamming technologies that were in use. (A recently declassified Air Force training video confirms this.)

These new threats pushed the Air Force to continue building computers using successive generations of computer and software companies. These companies and their knowledge have spawned an increasing array of computers, software, and networks that are in use today.