They were considered outsiders, both skilled in performing mathematical equations quickly, yet neither of them received a warm reception when they arrived at the laboratory. One was human and one a machine. But both Janez Lawson and the IBM she programmed were known as computers.
In 1952, Lawson had just completed her bachelor’s degree in chemical engineering at the University of California, Los Angeles. The president of her sorority and a straight-A student from a prestigious university, everyone expected greatness from the 21-year-old African American woman. Yet as Lawson perused the job board on campus, there wasn’t a single engineering position open to someone of her race and sex, no matter her qualifications. When she applied to the Jet Propulsion Laboratory in Pasadena, she saw the job of a “computer”—that is, a person responsible for the lab’s calculations—as a way to work in the field, even if she couldn’t have the coveted title of engineer.
At the same time that she was entering the workforce, so was the IBM 701. Thomas Watson, IBM’s president back then, called the machine “the most advanced, most flexible high-speed computer in the world.” It was the first IBM computer available commercially and 19 of the new-fangled machines were making their way to laboratories across the country. Lawson commuted the 30 miles from her home in Santa Monica to Pasadena by car on crowded freeways, still under construction, and then down the dirt road that led into the lab. The IBM 701, on the other hand, arrived by airfreight, its 20,516 pounds of computing power traveling nearly 3,000 miles to reach its new home.
For an industry focused on advancing new technology, aeronautical laboratories in the 1950s were surprisingly slow to embrace digital computers—and their reluctance was rooted in the societal implications placed on women in the workplace. A pervasive mistrust of automation paired with the view that computer programming was mere women’s work had a lasting impact on American space exploration.
* * *
In the first half of the 20th century, American laboratories frequently hired large numbers of human computers to perform the calculations needed to run their experiments. The laboratories that would one day become NASA, including JPL, started in much the same way, with a mix of mathematically minded men and women performing needed computations. U.S. involvement in World War II, however, shifted the dynamics, bringing women to the forefront of computing at academic and government research centers across the country. These computers analyzed data from thousands of tests, their work forming a critical component in designing early missiles, rockets, and military aircraft.
Military research installations and defense contractors were the first to order the new IBM machine. As an army facility, although operated by the California Institute of Technology, JPL took advantage of their access to the 701 with the hope that it could speed up their calculations, now made urgent by the start of the Korean War in 1950. But still, when the IBM 701 arrived, the administrators felt they were taking a gamble. They couldn’t be certain how the machine would shape their operation, even though they were willing to pay a monthly rental price of $11,900—about twice the annual salary of an engineer and four times that of a female computer.
As soon as the IBM was unpacked and installed in its own room with a square footage rivaled that of the average American home, it instantly drew suspicion. Despite the computer’s reported 16,000 operations per second, the laboratory engineers wanted little to do with the machine. They dismissed programming the IBM, and preferred to put their trust in computers made not of vacuum tubes and steel wire, but of flesh and blood.
Even with its high price tag, the IBM came with no instruction manual. In order to program the machine, employees had to attend an IBM training school. Lawson was one of the first sent from JPL to learn how to work on the behemoth IBM computer. During her three weeks in the school, she became skilled in programming. She found her college calculus courses helpful as she learned Speedcoding, a computer language specifically built for the IBM 701 that was innovative in its use of “floating point” arithmetic—a strategy that allowed engineers and scientists to include very large and very small numbers in calculations and is now ubiquitous.
To do her work, Lawson wrote her programs with paper and pencil and then brought her notebook to a keypunch, a device that looked like a small typewriter and that translated Lawson’s handwritten programs into a series of holes punched into rectangular, cream-colored cards. Lawson then fed her cards into the 701. The output came out neatly in the computer’s included printer. She made sure to keep the original notebooks handy. She could always run the cards again, of course, but her handwritten notes were the real source code
Despite the new technology, Lawson and her colleagues still did most of their work by hand. Part of this was rooted in the limits of the computer itself. A far cry from the modern silicon chip, the IBM 701 contained tens of thousands of germanium diode tubes that routed data from input to output. Each time a tube burned out, it took down the entire operation with it. Between flare-ups and a general distrust of technology, the expensive IBM 701 often sat unused.
Soon, another machine entered the women’s midst: the Burroughs E101. The desk-sized computer was even less popular than the 701. Unlike the IBM’s punch cards, the E101 used pinboard programming. Margie Behrens was just a teenager when she first tackled the E101, placing pins in the holes in the pinboards to run her numbers. They were eight pinboards and each one could hold 16 instructions. The computer could perform 4 multiplications or 20 additions per second. Burroughs boasted that the machine “saves 95 percent of manual computation time!” but Behrens found its operation painfully slow. For the women, machine was clumsy and prone to breakdowns.
Although both the IBM and the Burroughs were attempting to transform computing, neither of the machines were used the night America entered the Space Race. Instead, it was the human computers who were responsible for the calculations. On January 31, 1958, Behrens entered mission control. She was a 19-year-old high school graduate but her skills were more valuable than the high-powered machines the laboratory had spent a fortune obtaining. She and the other women in the room that night used mechanical pencils and graph paper to calculate the path of Explorer 1, America’s first satellite, as it tore through the sky and entered Earth’s orbit.
Once satellites were revolving overhead, everything at JPL changed. The laboratory was now dedicated to space exploration, becoming part of a brand-new space agency: NASA. With an expanded budget came new technology, and the 701 and Burroughs were considered merely relics as they were carted off, making way for a cutting-edge IBM 1620. The 1620 had its own air-conditioned room next to the women’s offices with a name on the door reading, “Core Storage.” This wouldn’t do, the group agreed, so they renamed the machine “Cora.”
While the engineers remained wary of the 1620, many still believing that computer programming was beneath them, the women quickly realized how valuable the machine was to calculations—a lesson that wasn’t always learned easily. In the summer of 1962, while working long hours on the first missions to the moon and planets, for instance, they launched an un-crewed spacecraft toward Venus that started traveling wildly off course only minutes after the launch. With fear that the rocket might crash in a residential neighborhood, a safety officer made the decision to self-destruct the rocket. As upsetting as it was to see their work strewn across the Florida coast, it was even more disturbing to learn that the mistake could be traced to a single transcription error: A superscript bar accidentally had been left off when the handwritten guidance program was transcribed in Florida. That one mistake had spelled disaster and gave the women new respect for computers’ promise of accuracy.
* * *
Around that time, Sue Finley, one of the women computers at JPL, was fresh from her training in a new computer language, FORTRAN, designed to translate mathematic equations into code, when a male engineer told her, “Your jobs will be gone soon.” His intimidating words reflecting a rising fear of technology in the 1960s. The threat of machines could be seen firsthand among those that worked at the telephone switchboard. With the advent of automated operations, the number of switchboard operators fell 43 percent between 1947 and 1960. It was enough to make anyone working with computers nervous.
When I first learned about the early women computers at JPL, I assumed their work would follow a similar trajectory to those at other NASA centers, characterized by short careers, and with but a peripheral involvement in the science they performed. Yet while the technology kept evolving, these women remained constant. The more things changed, the more their expertise was needed to bridge the gap between a revolving door of digital computers, each more sophisticated than the last, and our exploration of space.
On August 27, 1962, the JPL launched Mariner 2, a second attempt at sending a spacecraft towards Venus. The women watched in the control room as the ship escaped Earth’s atmosphere and made its way to the neighboring planet, following a trajectory they had designed with both pencil and paper and their favorite digital computer, Cora. Four months later, the ship passed Venus, revealing the planet’s hot atmosphere, roughly 900-degrees Fahrenheit. It was, of course, just the beginning. Next, the women would help send the first lunar spacecraft, paving the way for Apollo. When man walked on the moon in 1969, it was, in part, thanks to the calculations of women.
Their relationship with technology continues to today. It developed as the group of women used early computer animation to guide the Voyagers through the solar system in 1977, marveled over their first laptops in the mid-1980s that weighed a mere 13 pounds, and in the late 1990s, wrote programs in C and C++ to send off the first Mars rovers, one of which, Mars Opportunity, still roams the red planet.
Perhaps no single employee demonstrates the longevity of this connection better than Sue Finley, whose job certainly did not soon disappear like she was told. Finley was hired in 1958, eight months before congress and President Eisenhower officially created NASA. Today, at 79 years old, she is NASA’s longest serving female employee. Over her 58-year career there is scarcely a mission her work has not touched.
Still, hurdles remain. In 2004, the space agency degraded Sue’s employment, taking away her title of engineer and reducing her salary to an hourly rate because she doesn’t hold a college degree. Finley loves her work, regardless. She is currently working on NASA’s mission to Jupiter, and keeps a stash of aging graph paper in her office for hand-plotting trajectories, her long experience with computers prompting her to keep back-ups handy.
It’s difficult to imagine a time when computers didn’t dominate laboratories, yet the women who heralded their revolution have largely been forgotten. Lawson finally became a chemical engineer, and while her career far outlasted that of the IBM 701, the programs they created together form the basis for technology today. The programs the pioneers of JPL built have traveled across the universe—even, in one case, leaving it altogether. In 2014, Voyager 1 became the first man-made object to leave our solar system. But it wasn’t just made by men. Stashed aboard the spacecraft’s 40 kilobytes, soaring through the space dust, are the relics of a relationship between women and machines that span half a century.
We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.