Broad Band

by Claire L. Evans

  It irked the ENIAC women to read newspaper articles claiming the machine itself was clever; they knew better than anyone that it was just a room full of steel and wire. “The amount of work that had to be done before you could ever get to a machine that was really doing any thinking to me just staggered the mind,” complained Betty Jean, and “I found this very annoying.” It was more than annoying; it effectively erased her. The 1946 New York Times story about the ENIAC demonstration breathlessly reported that “the ENIAC was . . . told to solve a difficult problem that would have required several weeks’ work by a trained man. The ENIAC did it in exactly fifteen seconds.”

  As historian Jennifer S. Light points out, that claim ignores two essential factors: first, that the “several weeks’ work” would never have been done by a man in the first place. It would have been done by a female computer working long hours at the Moore School. Second, the claim that the ENIAC solved the problem in “exactly fifteen seconds” completely disregards, through ignorance or willful dismissal, the weeks of work, again conducted by women, that went into programming the problem before it was even put on the computer. As far as the press was concerned, nothing outside of those fifteen magical seconds—not the hours of coding and debugging, not the labor of programmers and maintenance workers and operators—counted. Light writes, “The press conference and follow-up coverage rendered invisible both the skilled labor required to set up the demonstration and the gender of the skilled women who did it.”

  After the ENIAC demonstration, once the glad-handing and photo-ops were over, the university hosted a big celebration dinner. Judging from a menu for the event, no expense was spared. Military brass and members of the scientific community ate lobster bisque, filet mignon, and “fancy cakes.” None of the ENIAC Six were invited—not even the Bettys, who had created the demonstration the dinner was held to celebrate. They’d helped introduce the century to the machine that would come to define it, and nobody congratulated them. The Goldstines snubbed them completely. Even their supporters, John Mauchly and J. Presper Eckert, were too caught up in the excitement of the day to comment on the demonstration program. On February 15, just as they had late the night before, Betty Jean Jennings and Betty Snyder went home dejected. It was cold, and they were exhausted. “It felt like history had been made that day,” Betty Jean wrote in her autobiography decades later, “and then it had run over us and left us in its tracks.”

  History would run them over again and again. Neither Betty nor Betty Jean would be credited for writing the ENIAC demonstration program until they began to tell their own stories over fifty years later. Herman Goldstine, in his influential history of the ENIAC, wrote that he and Adele had programmed the February 15 demonstration themselves, a move of revisionist history that Betty Jean Jennings, late in life, called a “boldface lie.” In subsequent retellings, the women were skipped over repeatedly. In some historical images, the ENIAC Six are captioned as models, if pictured at all. “I wasn’t photogenic,” said Betty Snyder. “I wasn’t included on any of the pictures of the entire stupid thing.” When the army used a War Department publicity shot of the ENIAC for a recruitment ad, they cropped out the three women in the picture entirely. The War Department’s own press releases about the ENIAC cited a vague, genderless “group of experts” responsible for the machine’s operation, and mention by name only John Mauchly, J. Presper Eckert, and Herman Goldstine.

  It’s tempting to look at the historical association between women and software and assume some inherent affinity: that women appreciate the mutable, language-oriented aspects of programming, while men are drawn to the practical, hands-on nature of hardware. Some might posit as much from the Babbage-Lovelace partnership, Howard Aiken and Grace Hopper’s testy relationship a generation later, or from the gendered division of labor between male hardware engineers and female operators on the ENIAC. But in all of these instances, women have fallen on the software side not because the work was somehow more suited to them but because software, still inextricably entangled with hardware, wasn’t yet a category with its own value. As far as anyone understood it, software—the writing of code, the patching of cables—was really just the manipulation of hardware, and the title of “programmer” wasn’t yet distinct from the more menial “operator,” a rote job that leaned female because of a long female history of secretarial work. Further, the hiring of women to run computers like the ENIAC reflected a long tradition of women as computers themselves, laboring over applied mathematics in university and research settings. Women had been doing the math for as long as anyone could remember.

  “If the ENIAC’s administrators had known how crucial programming would be to the functioning of the electronic computer and how complex it would prove to be,” Betty Jean Jennings eventually determined, “they might have been more hesitant to give such an important role to women.” At the time, it was difficult to perceive of programming as an occupation distinct from simply operating a computer and, indeed, the ENIAC women’s jobs were officially classified as “subprofessional, a kind of clerical work.” It would be years before those who approached computers began to define themselves in relation to them, as programmers or computer scientists, rather than as operators or electrical engineers. It would take even longer before a vision of programming as an art form with the potential to reshape the modern world came into focus.

  Chapter Three

  THE SALAD DAYS

  World War II was a war of technology. The relationships between government, private industry, and academia forged to gain a national edge on the Axis gave us the military-industrial complex and bankrolled a generation of technological innovators in the process. It’s unlikely computing would have developed as a field or an industry as quickly as it did without the complex calculations demanded by the war machine. The war made risks worth taking and accelerated everything. It also let women into the process. Men may have dropped bombs, but it was women who told them where to do it.

  It’s unsettling to think of any war, especially such an ugly one, as an opportunity. But working on military calculations during World War II allowed Betty Jean Jennings, Betty Snyder, Grace Hopper, and their peers to do more with their lives than teach, marry, or be secretaries. It opened an entirely new technical field to women, one whose importance would become evident only after they showed what remarkable things could be done at the confluence of people and computing machines. But change is never so simple. As easily as war gave these women a ticket out of potentially desultory marriages and dead-end secretarial careers, peace threatened to take it all away.

  After the war, as military funding dried up and authority over computational projects transitioned back to civilian hands, Grace Hopper found herself at a crossroads. In a short time, she’d become an expert in a nascent field, but she’d made sacrifices. Although she and Vincent had been separated since the beginning of the war, they only divorced in 1945. He promptly remarried—to a friend of Grace’s, who had been a bridesmaid at their wedding. It can’t have been easy. She was forty-three, and after what she’d been through, she couldn’t dream of going back to teaching calculus to undergraduates and being the comfortable small-town college professor once again. But as soldiers coming home from the front reclaimed their place in American life, many women around Grace were returning to their prewar roles. Even a few of the ENIAC Six went domestic. Being someone who’d left everything behind to stake a career at the doubly male-dominated intersection of military and academic life, Grace was nervous.

  Howard Aiken parlayed the Computation Laboratory’s wartime computing work into a longer appointment at Harvard, and he kept Grace, who was indispensable to the operation of the lab, on as a three-year research associate. In her new role, Grace no longer did any programming. Instead, she assisted Aiken and wrote a user’s manual for the Mark II, for which she—as with the first manual—received no byline. She led tours of the facility, explaining to distinguished visitors the many applications o
f the once-secret computing machine. Aiken tried to maintain the Computation Laboratory’s sense of urgency, but things weren’t the same; wartime atmosphere in a time of hard-won peace lent a lightly sadistic air to the everyday. Grace had always been a social drinker, but watching her role dwindle, she drank on the job, keeping a flask in her desk. In 1949, Harvard’s contract with the navy expired. Aiken, safely tenure-tracked, stayed on, but the university didn’t offer promotions to women, so Grace—by an order of magnitude the preeminent national expert on computer programming—lost her job in the lab she’d managed for years. “My time was up,” she said.

  The ENIAC lab at Penn was gone, too. The computer itself was moved to the Aberdeen Proving Ground—they’d had to tear out an entire wall of the Moore School building to remove it—and eventually seven more women would work on the ENIAC at Aberdeen, where it ran until being decommissioned in 1955. The ENIAC Six, having moved on to the next phases of their lives, would never program their namesake again.

  During the war, it hadn’t mattered who owned the ENIAC. It was built with the army’s dollar, and it worked in the army’s service. With the conflict over, however, the computer became contestable. Eckert and Mauchly wanted to patent their invention, but the university demanded all manner of licenses and sublicenses for their design, as well as rights to whatever machine they planned to build next. Penn offered them tenure in exchange for a patent release, but the two engineers didn’t want to be beholden to the university for the rest of their careers. Pres and John chose to write off the ENIAC, cut their losses, and strike out on their own.

  By 1947, they’d founded their own company, the Electronic Control Company, later renamed the Eckert-Mauchly Computer Corporation. Their offices on Ridge Avenue in Philadelphia were nothing fancy: an old hosiery mill with no air-conditioning and a few rows of wooden desks someone picked up at a secondhand store. Right across the street was a junkyard. On the other side of the building was the Mount Vernon Cemetery. If the computer didn’t run, Grace Hopper joked, they’d throw it out one window into the junkyard and jump out the other.

  The narrative of enterprising young technology innovators striking out on their own to take risks and change the world is familiar to us now. Every garage-to-riches Silicon Valley story seems to follow the same approximate arc. But the Eckert-Mauchly Computer Corporation was the first commercial computer company in the world. Until EMCC, computers had been one-off machines custom built for the calculations of war: ballistics, code-breaking, and the fluid dynamics of nuclear bombs. But computers had a huge prospective market in academia and in calculation-intensive industries like aviation, and they could make quick work of accounting and payroll problems for large corporations. Pres and John weren’t the only people to realize the massive commercial potential of computers as business machines, but they were among the first to give the standardized production of computers a try. As architects of the now-world-famous ENIAC, they had a running start.

  EMCC scored its key contracts early: the Census Bureau, the National Bureau of Standards, Northrop Aviation, and the Army Map Service all wanted their own version of the ENIAC. Each installation would have to be customized to the client’s specific computing needs, bundled with systems, services, and support. Software wasn’t an off-the-shelf product yet—or a word, for that matter—so the Eckert-Mauchly Computer Corporation would need programmers. Fortunately, the best in the world were their former colleagues at the Moore School—Betty Snyder, Betty Jean Jennings, and Kathleen “Kay” McNulty, and the grand dame of code herself, Grace Hopper, who was looking for a path forward after losing her place at Harvard. Pres and John were smart enough to hire them all.

  For its first few years, the Eckert-Mauchly Computer Company was an anachronism. It was a tech startup among corporate giants, building the most advanced computers in the world, all of which ran custom code written by the talented female programmers leading its software development teams. Betty Jean Jennings remembered it as a magical time. “I loved it,” she wrote, “and I never again felt so alive. We worked constantly. We arrived early. We arranged meetings during coffee breaks and lunch. We worked late.” There were no formal titles, and departments were porous. People worked directly with one another when projects called for it; there was no chain of command, no bureaucracy. Everyone tackled the problems in front of them, taking initiative wherever it led. “The fact is,” Betty Snyder said, “we all believed so much in what we were doing that we worked together, that’s all.”

  The founders were widely liked and admired by their female staff, because they were good teachers, they listened, and they had vision. Meeting with clients, John Mauchly could rattle off new applications for their computers on the spot. Grace was motivated to join EMCC because she knew she’d learn something from John, whom she remembered as “a very delightful person to be with and fun to work with. He was just as excited about it as you were and he was right in the thick of it and thoroughly enjoying it and what I’d call a real good boss.” Pres, too, was a “real good boss,” albeit an odd one. While deep in conversation, he’d often lose track of himself, wandering along hallways, down into the basement, and back upstairs. He’d sit on tables. Once, he climbed up onto a file cabinet while talking to someone without even noticing. In all the years she worked at EMCC, Betty Jean never knew where Pres’s office was: she never saw him in it. He was always among them, moving from group to group with questions and solutions. The women joked that Pres liked talking so much that he’d stop the janitor from mopping just to have someone to bounce ideas around with. Despite these idiosyncrasies, he was theirs. “We all accepted Pres as he was,” Betty Jean recalled.

  During a time when many companies were firing competent women to make room for returning GIs, Eckert-Mauchly Computer Company not only employed women but also valued their contributions, giving them even more significant programming responsibilities than they’d had during the war. Both Betty Snyder, by then Betty Holberton, and Betty Jean Jennings managed major projects: Betty wrote the instruction code and Betty Jean did the logical design for the company’s marquee product, the Universal Automatic Computer, or UNIVAC. The early years were lean—Betty, along with many engineers, worked for nothing at first—but everyone was treated well once the company got its legs. A year into her employment at EMCC, Betty Jean discovered that a recent male hire was making four hundred dollars more than her a year, and when she took her grievance to John Mauchly, he heard her out and immediately raised her salary to two hundred dollars above the man. EMCC even held company-wide programming courses every spring, in the hopes of training fresh blood from elsewhere in the ranks. “That’s how so many secretaries got to be programmers before we were through,” Grace Hopper recalled. That was the company ethos. “A gal who was a good secretary was bound to become a programmer, meticulous, careful about getting things right. Step-by-step attitude. The things that made them good secretaries were the very things that made them good programmers.”

  Grace had been struggling with alcoholism during her final years in Howard Aiken’s thankless Computation Laboratory, and her friends hoped that the job at EMCC would help her stop drinking and regain her confidence. Even if she wasn’t valued at Harvard, her talents were in high demand in the burgeoning computer industry: in addition to the offer from Eckert and Mauchly, in 1949 she considered a job in ballistics research at Aberdeen, making cryptological machines for Engineering Research Associates, and with the Office of Naval Research. But only EMCC had working machines available to program within the year, and Grace didn’t want to wait any longer. At EMCC, she could return right away to her true passion: writing code, and surrounded by brilliant women to boot. She was grateful. Much as she’d reveled in the transition from civilian to military life back in Midshipmen’s School, she “slipped into UNIVAC like duck soup,” as though she’d “acquired all of the freedom and all of the pleasures of the world.”

  The woman Grace worked most closely with at EMCC
was undoubtedly Betty Holberton, the ENIAC whiz who solved programming problems in her sleep. This was only one aspect of Betty’s unique mind. From a young age, she’d been keen on puzzles. At fifteen, when her family’s calculating machine was on the fritz, she took the whole thing apart, labeled every component, and laid them all on the table, precisely diagrammed. She liked to think of machines that way: as a concert of small pieces working together in a logical whole, and found that she thought “like a radar screen,” going “around and around the problem” until she had it all in mind. After these radar sweeps, she was able to see errors almost instantly. Her talent was so potent that she was often brought in, “if the programmer and the engineer couldn’t resolve what the problem was,” to debug other people’s projects. Later in her career, she’d insist on not being called in until after the engineers had been stuck for at least four hours. Otherwise, it was a waste of her time.

  Betty was EMCC’s secret weapon, and by the time Grace Hopper arrived in Philadelphia, Betty had already been with the company for two years, working one-on-one with John Mauchly, at his living room table before they rented the warehouse on Ridge Street, developing C-10, the operational code for the UNIVAC. Although she claimed not to have been “photogenic” enough to be pictured with the ENIAC, there was no denying the singular beauty of her code. C-10 was the elemental instruction set for the computer, and Betty was extremely thoughtful about writing it. Instead of relying only on numbers and symbols, she included letters as shorthand for commonly used operations: s for “subtract,” a for “addition.” This was a first, and it made the UNIVAC more user-friendly to operators without a mathematical background. Grace thought the C-10 code was brilliant—beautiful, even. She loved anything that made programming more approachable. And she thought Betty was the best programmer she’d ever met. Betty was more modest. According to their colleague Betty Jean Jennings—who didn’t always get along with Grace—Betty Holberton would say of Grace that “if anyone could do something she couldn’t do, that person obviously had to be the best in the world at it, because only the best could outdo Grace!”