Issue No. 08 - August (2006 vol. 39)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/MC.2006.282
David Alan Grier , The George Washington University
Last fall, while I was giving a talk to a local business group, I was interrupted by a question that was only loosely related to my presentation. Did I have an opinion, one member asked, on the suitability of women for scientific and technical careers? The issue had recently been raised by Harvard's then-president, Lawrence Summers, who suggested that innate differences were keeping women out of high-level positions in science and engineering. In an instant, the room became quiet. I sensed that the issue had initiated a debate within the group and drawn a battle line between two opposing camps.
"I am loyal to my grandmother," I began. I went on to explain that my mother's mother had been trained as a mathematician at the University of Michigan and graduated in the class of 1921. That year, half of the mathematics students had been women. I added that I also knew, from much personal experience, that my grandmother would have objected to the notion that women were not the equal of men.
"If she had been at dinner with the Harvard president when he made that statement," I said, "she would have immediately sent him to his room without dessert." I added that I had felt the firm hand of my grandmother's correction on many an occasion and could testify that Summers would have been denied dessert "until he had written an individual letter of apology to everyone that he had possibly offended."
Everyone in the room laughed, though at least a few voices carried a nervous tone. "Not good enough," said one man at the back. "What do you think?"
"I think that I am the loyal grandson of a woman scientist," I repeated, "and I believe there is nothing more I need say."
Family relationships—father to son, mother to daughter, grandparent to grandchild—are the fundamental associations that we have in modern society. We learn the nature of these relationships as children and use them as models for the relationships we have in the civic, commercial, and educational worlds. Many a company has tried to portray itself as a happy family. The male bosses are like fathers. Female office managers take the role of mothers. Our fellow workers, our peers, serve as the equivalent of brothers and sisters.
The computer industry has expanded and tested these comfortable family relationships. Bosses take to the road and become absent parents. Coworkers are distributed around the globe with no familial connection beyond an Internet link.
None of these relationships has been tested more than those offered to women: mother, daughter, sister. Over the past 30 years, women have found that computer technology offers new opportunities, new positions, new roles. At the same time, computer technology has shown how firmly our ideas about women are carved into the foundations of our society.
When I was barely two weeks out of college, I took my first job with a computer company, where I shared an office with a female coworker named Ro, which was short for Rosemary or Roxanne or Rosalind or some other pretty and delicate name that didn't quite suit her. Ro was brash, funny, confident. She had worked in the computer industry for more than a decade and had accumulated a great deal of both technical and managerial experience.
In a very short time, Ro and I settled into roles that resembled those of an older sister and a younger brother. She guided me though the lessons of a new employee, taught me how to act in front of our boss, and critiqued my presentations to the customers. Ro even suggested, at first gently but later with some force, that some of my clothes were inappropriate as business attire. She even warned me that a secretary in another division of the company was on the prowl for a husband and had identified me as a likely candidate, a warning that was repeated, weeks later, by one of my male colleagues.
Perhaps the most important advice Ro gave me came at a key moment in one project. As we were finishing some essential work for this project, one of the salespeople suggested that we might want to replace the performance numbers with more optimistic estimates. He contended that this was appropriate because the company hoped that our computer would soon be much faster.
I became angry at this request. I felt that the salesman was unethical and wanted to denounce him to my boss. I told Ro of my plans, undoubtedly adding a sheen of self-righteousness. I expected her to support me and was surprised when she didn't.
Ro pointed out that the salesman was quite senior to me, he would probably deny that he had asked me to falsify performance numbers, and he most likely would claim that I was being uncooperative. She told me to do what the salesman requested, but when I presented the numbers to the boss, she suggested that I call them "estimated performance numbers" and be ready to explain how I created those estimates.
As Ro had anticipated, our boss immediately wanted to know how the numbers had been calculated. I explained that the sales department wanted to know how the next generation of the machine would perform. He dismissed this idea with some phrase such as, "We have to sell the machine as it operates, not as we want it to operate." In a moment, the problem was resolved, no one was embarrassed, and we moved on to other issues.
An Invisible Metaphor
From time to time, I attempted to prove that I was not the kid brother, not the subservient sibling following in Ro's footsteps. Few of these episodes did me any good. Usually, they only confirmed the fact that Ro knew more about the technological and business worlds than did I.
One day something caused me to give a lengthy speech on how the computer would change the world for women. Ro listened with a look of long-suffering patience on her face. As I was ending my little diatribe, she gave a curt little laugh and started to tell me the story of the obscene operating system.
Some eight or 10 years before, Ro had taken a course on the company's operating system and was one of two women in the class. On the first day, after introducing himself, the instructor had hurried through the course's preliminaries. "He was clearly anxious from the start," Ro recalled.
Rushing into the first lecture, the instructor started fumbling with his words and speaking in vague generalities. As he tried to describe the operating system's basic structure, the source of his anxiety quickly became apparent. He was uncomfortable teaching a class in which there were women because the operating system in question was designed around a sexual metaphor.
The genesis of the sexual metaphor was an obvious and somewhat silly pun. When the operating system started a new process, it employed a concept known as "forking," a series of steps that copied the data of an existing process and established the new process as a branch of the first. At some point in the design of this system, a programmer noticed the similarities between the word "fork" and a common term for sexual intercourse.
Once he established that link, the programmer began creating other sexual puns. The data structure that held requests to start new processes was known as the "fork queue," an initialization procedure was called the "mother forker," and the first step of starting a new process took the name "unlocking the frock." From this beginning, it was only a short step until the code was filled with sexual imagery, puns, and double entendres.
Some people can discuss sexual imagery in front of a mixed audience without a shred of self-consciousness. Others can mitigate the sting of such words with a careful apology or a well-crafted dismissal. Ro's instructor had no such skills. He was never able to summon enough composure to move beyond the sexual ideas or to explain the operating system properly. The students, both men and women, left the class feeling that the instructor had failed them.
Ro and another member of the class complained about the sexual metaphor and its impact on the quality of the instruction. This complaint was passed through the company and produced a request to modify the names within the operating system.
At first, the operating system department resisted the modifications. "The developers just didn't understand the issue," Ro recalled. "This was a time when customers were regularly entertained at topless bars." They apparently argued that the system worked well, the metaphor was not visible to most customers, and any change would introduce errors into the code.
In the end, the operating system developers lost the fight. In the next release of the program, they used a new metaphor that was vaguely related to the British colonization of North America. Although it wasn't a very interesting metaphor, it didn't interfere with technical training or customer presentations.
Women in Computing
The story of the obscene operating system suggests an obvious conclusion, the idea that men dominate the computer business, as they dominate other business sectors, and use their power to exclude women. While this conclusion has much truth, it fails to capture the full dynamic of the computer industry. The industry has always needed more technical workers than it could find among the available group of men. In turning to women, it has had to make adjustments, though it has usually ceded to these adjustments on technical grounds.
"[C]omputing was a relatively welcoming field for women," concludes Janet Abbate, one of the leading scholars of gender and computation. "It offered them a rare chance to put their technical abilities to work."
The earliest computing projects, those carried out during World War II, employed women in technical positions. Grace Hopper worked with Howard Aiken on the Harvard Mark I; Clara Froelich assisted George Stibitz at Bell Laboratories; Kathleen McNulty and a dozen other women were part of the ENIAC project. In the years that followed, we find Ida Rhodes and Margaret Fox at the National Bureau of Standards, Gertrude Blanch at the Air Force Research Laboratories, Mildred Koss at Univac, Jean Sammet and Ann Hardy at IBM, Mary Lee Berners-Lee (the mother of World Wide Web inventor Tim Berners-Lee) at Ferranti, and Dame Stephanie Shirley at the head of her own software firm.
Most of these women worked in software development, a field that was not well-defined in their time. In the 1950s and 1960s, software development was a lump of soft clay, awaiting the print of the hand. It had no fixed categories of "programmer," "software developer," "systems analyst," or "software engineer."
Even without these labels, programmers and software developers found ready employment among the early builders and users of computer equipment. These companies found that they needed large programming staffs to keep their machines fully occupied. When these companies were unable to find male programmers, they opened their doors to women. When skilled women were reluctant to sign on, the companies offered generous benefits. "Because experienced programmers were in high demand [during the 1950s]," recalled Mildred Koss, "I was able to work part-time until my children were in school."
The unfulfilled demand for skilled programmers had also brought my office mate Ro into a career in computer science. In the late 1960s, she had started looking for a job after an injury forced her to drop out of college. Because she knew Fortran and some assembly language, she came to the attention of a local office of General Electric, which was then marketing a computer. They offered her a job, but she did not stay long. "The building had no women's bathrooms," she explained. After one frustrating day, she felt the need to tell her boss "exactly what it was like to be a woman."
A brief period of unemployment followed this exchange, but a second firm soon offered Ro an interview. At first, this company seemed to be no more promising than the first. The manager "was extremely uncomfortable," Ro recalled. "He clearly did not want to hire a woman, but he had little choice. He kept repeating that he had never hired a girl for a technical job." Even with such a dubious introduction, Ro took the job, quickly gained the attention of the firm's senior managers, and soon moved into a job with national responsibility.
When Ro shared an office with me, she was approaching the midpoint of her career and was starting to move into the company's leadership ranks. Her progress was neither as smooth nor as easy as it might have been for a man, but she was still able to find parts of the firm that appreciated her skills. She ended her career overseeing one of the company's research divisions, something that resembled an in-house venture capital office. "I would get to hire bright, off-the-wall young people with specialty skills that you have never heard of," she said. "I set up a little business incubator where I would try to leave these thinkers alone to develop their ideas."
Ro described this last job about a decade after she left the firm. As she talked about her position, she reviewed the basic models for the roles that women can follow in business. To the roles of mother, wife, and sister, she added a fourth—mistress, which recalled the challenges of the operating system class.
As she talked, Ro described how she worked with the young researchers. "I was the daughter in the fairy tale, who had to spin the straw into gold." She had to constantly remind her young charges that they had to produce something in return for the money they received from the company. She had to guide these researchers, protect them from other divisions of the company, and champion their ideas. "I had to reconcile differences, stop fights, and nudge ideas into products."
"Was it really the role of a daughter?" I asked. "Or were you more of a mother or sister?" I didn't really want to go to her last category. Ro paused and laughed, then she said, "I was a woman in the computer business, and that is all there is to say."
David Alan Grier is the editor in chief, IEEE Annals of the History of Computing, and the author of When Computers Were Human (Princeton University Press, 2005). Grier is an associate professor in the Center for International Science and Technology Policy at the George Washington University. Contact him at firstname.lastname@example.org.