1

Breaking Codes and Finding tajectories:

Women at the Dawn of the Digital Age

I was a Colossus operator, which we considered to be the cröme de la cröme.

\üe felt we were "at the sharp end," where there was a great tension and flow of
adrenaline . . . operating those incredible machines.
Beech, Colossus operatorl
-Jean
I don't know if you can picture how exciting the ENIAC was to all of us. And we
didn't talk socially or any other time about anything else. It was-we discussed
it almost all the time.
ENIAC programmer'
-Jean Jennings,

During the Second World'W'ar, women like Jean Beech and Jean Jennings
ventured into a wholly new field-digital computing. They were neither
intimidated by the noisy, room-sized machines that they encountered nor
deterred by any sense that they were entering a masculine domain. In-
stead, the accounts of women who worked on wartime computer projects
convey a sense of exaitement, fun, and pride at mastering a challenging
task and contributing to the Allies' eventual victory.
This sense of the importance, high status, and sheer joy of working
with computers still motivates women to choose careers in software or
computer science today. Yet throughout the history of computing, the
ability of women (and men) to enter and flourish in these careers has
been strongly influenced by gender stereotypes and unequal structures
of opportunity. This chapter uses examples from the'\X/orld'War II era
to demonstrate how assumptions about the gendered nature of techni-
cal skill, about women's place in the workforce, and about the nature of
computing could constrain women's options and lead to an undervalua-
tion of their contributions. Over time, however, both gender roles and the
 12 Cbapter 1
Breaking Codes and FindingTrajectories 13

meaning of computing itself-what computers are good for and what it machine, completed in 194L, was a program-controlled electromechani-

takes to make them useful-have changed, and women have played an cal computer.T Zuse's work was little known outside of his immediate
active role in defining both. These twin themes are explored throughout circle, however, and the Z3 was destroyed in the war. Most subsequent
this book. developments were made in the United States and Great Britain. John
Atanasoff and Clifford Berry at Iowa State University built an operational
Wartime Missions: Colossus and ENIAC computer in 1942,which was a special-purpose (nonprogrammable) elec-
tronic machine. In the UK, a series of ten electronic computers called "Co-
The pressures of !7orld
'Süar
II drove a wave of technical innovation. Lives lossus" were built at the top-secret codebreaking facility at Bletchley Park
were saved or lost by developments such as radar, the atomic bomb, and near London, with the first one operating in February 1944.The Colossi
the mass production of penicillin. Less known among these wartime proj- were theoretically programmable, but in practice they were configured
ects are the first electronic digital computers.3 Before the 1950s, the tasks for a single codebreaking task. In March 1944, the IBM Automatic se-
that we associate today with computers were done by an array of special- quence controlled calculator, also known as the Mark I, came online at
ized machines. Mechanical desk calculators were used to perform arith- Harvard University. Designed by Howard Aiken, the Mark I's program-
metic, and the people who did these calculations-often women-were ming staff included Grace Murray Hopper, soon to become a major figure
called computers; this term did not yet refer to a machine.a For business in computing. The first fully electronic programmable digital computer
and government record keeping, there were electromechanical tabulators was built at the university of Pennsylvaniaby J. Presper Eckert and John
that sorted and tallied information that was stored on punched paper Mauchly beginning in 1943.This machine, the Electronic Numerical Inte-
cards. Punched-card machines had been invented by Hermann Hollerith grator and Computer (ENIAC), was completed in November 1945.8 After
to tabulate the 1890 U.S. Census'in 1911, Hollerith's firm became the the war, computer designers took the final step to a truly general-purpose
core of a new company called IBM, which quickly dominated the busi- computer by creating machines that stored their programs in software,
ness.s Scientists sometimes did complex calculations using analog com- resulting in a processor that was functionally similar to those used today.
puters, which used physical motion, such as the rotation of a cylinder, The first operational stored-program computer was the EDSAC, built at
to represent mathematical functions or model physical processes. Analog Cambridge University in 1949.e
computers had been introduced in the 1930s, most notably the Differen- This chapter focuses on two of these pioneering projects, Colossus in
tial Analyzer invented by Vannevar Bush at the Massachusetts Institute Britain and ENIAC in America. Both made important technical advances,
of Technology. A general-purpose computer that could handle all of these both influenced the development of computers after the war, and both
diverse applications did not exist.6 In addition, the prewar machines op- employed teams of women to operate the machines. After introducing
erated relatively slowly because they relied on mechanical or electrome- each machine, I examine how and why women were drawn into these
chanical components. Not until the wartime emergency did computer de- projects, what the perceptions and realities were of the jobs they per-
signers dare to build machines whose circuits used thousands of vacuum formed, and what women's options were for pursuing a postwar career
tubes-components that were considered unreliable and impractical-to in computing. These historical episodes illustrate how technical innova-
operate at electronic speed. tions and social expectations framed opportunities in the new computing
A brief survey of the invention of the digital computer shows how a field. I also highlight women's agency-their motivations and constraints
series of experimental rnachines gradually advanced the state of the art, in choosing computer work and their accomplishments and pleasures on
eventually producing a modern computer that was fully electronic (rather the lob. Programming and operating computers was hard and important
than electromechanical) and general-purpose (programmable). One of work, but it was not always recognized or rewarded as such when the
the first digital computers was built in Germany by Konrad Zuse;his Z3 workers were women.
 14 Chapter 1
Breaking Codes and Pinding Traiectories 15

Colossus wheels. Flowers built this machine, which used 2,500 electronic valves
It is regretted that it is not possible to give an adequate idea of the fascination of (vacuum tubes), with a team of engineers at the Post Office research sta-
a colossus at work: its sheer bulk and apparent complexity; the fantastic speed tion at Dollis Hill. Coded messages were input using punched paper tape
of thin paper tape round the glittering pulleys, . . . the wizardry of purely mechan-
at a rapid rate of 5,000 characters per second. The room-filling com-
ical decoding.
puter was dubbed Colossus, and eventually a series of ten Colossi were
Report on Tunny, 194510
-General produced, differing mainly in size (figure 1.1). The first Colossus was
assembled at Bletchley Park in December 'l'943 and began decoding mes-
colossus was part of the Allied effort to read secret communications sent
sages in January 1944.
by Germany and Japan. Codebreaking w-as an essential element of war-
Like hundreds of other workers at Bletchley Park, the Colossus oper-
time intelligence: if the Allies could decode Axis messages, then they could 'Women's
ators were chosen from members of the Royal Naval Service,
anticipate attacks and take appropriate measures. Such eavesdropping
known as'Wrens. New S7rens recruits were given assignments based on
also gave the Allies feedback on whether their own plans had become
immediate needs and interviewers' assessments of their abilities. Those sent
known to the enemy. one of the mosr importanr Allied codebreaking op-
to Bletchley Park were not told the nature of the work until they arrived;
erations was located at Bletchley Park, about i0 miles north of London.l1
to maintain secrecy, the Bletchley Park'S7rens were officially assigned to
The challenge for Bletchley Park was that the Germans used a series
of highly sophisticated encryption devices that defied the usual methods the fictitious ship HMS Pembroke V. At the peak of Colossus opera-
tions, there were approximately sixfy operators, working in three shifts or
of decoding. The first of these German decoding devices was called the
"watches."13
Enigma machine. rn 1940, the brilliant mathematician Alan Turing built
an electromechanical device called the bombe that could decode messages
prepared by Enigma. By that time, however, Hitler and his high command
had adopted an even more advanced technology, the Lorenz machine. This
machine had ten wheels, each of which had the letters of the alphabet ar-
ranged around the rims. By changing the positions of these wheels at the
start of each message, the sender could vary the way that the messages were
encrypted. To reverse engineer this process, the codebreakers at Bletchley
Park had to work out the logic of the machine and then derermine, for each
individual message, the starting positions of the ten wheels. The first step
in mechanizing this process was a special-purpose electromechanical com- ,!

puter, built in 1942, that could simulate the actions of the Lorenz machine. I

Since the German cipher was code-na med Fish,the Bletchley park machine
that mimicked it was whimsically named Tunny. Given correct starting
positions for the ten wheels, Tunny could take an encoded message as input
and quickly produce unencrypted German rext as output. The question
now was how to figure out the initial wheel settings.l2
Mathematician Max Newman, drawing on the ideas of Aran Turing,
worked with British Post office electrical engineer Tommy Flowers to de- Figure 1.1
Colossus operated by Dorothy Du Boisson (left) and an unidentified Wren. Photo
sign a computer that could search for the starting positions of the Lorenz
courtesy of the National Archives, United Kingdom.
 16 Breaking Codes and FindingTraiectories 17
Chapter 1

vacuum tubes-18,000 of them-to perform calculations, making

it the
Colossus was a boon to the Allies, allorving 13,508 messages to be de-
coded by war's end. The second Colossus machine was built in a rush to first fully electronic computer in the united States (figure 1.2).To program
pro-
have ready in time for D-Day. Put in operarion on June 1, I944,it allowed the ENIAC, the army recruited a team of six women' The original
Fran-
Bletchley Park staff to decode Adolf Hitler's battle orders to Field Marshall grammers were Jean Jennings (Bartik), Betty Snyder (Holberton)'
Erwin Rommel. These revealed that Hitler had been fooled by the Allied ces Bilas (Spence), Kay McNulty (Mauchly Antonelli), Marlyn'Wescoff
ruse to draw attention away from their planned Normandy invasion. Hit- (Meltzer), and Ruth Lichterman (Teitelbaum).17 several additional wom-
en worked on the ENIAC after it was transferred to Aberdeen
in 1947 '
Ier ordered Rommel not to shift his forces to Normandy, leaving the way
open for General Dwight D. Eisenhower to attack on June 6. As the battle The ENIAC was completed and ran its first program in November
progressed, further details of Hitler's plans were revealed through Colos- 1945-too late to help in the war effort but just in time to influence
the postwar computer industry. one of its major tasks while still
at the
sus, hastening the Allied defeat of the German army. Eisenhower is said to
have estimated that the intelligence from Bietchley Park shortened the war Moore School was a set of calculations that modeled thermonuclear
by at least two years and saved hundreds of thousands of lives.la bombs in 1946.It was then moved to Aberdeen, where it began opera-
tion again in July 7947 and,continued to be used until 1955.18 Eckert and
ENIAC tvtao.hly left Penn after the war to start their own company, developing
'!7ell, a series of government and commercial computers including the highly
how d'you like this new knick-knack,
The Süonderful Answer:ing ENIAC? successful UNIVAC.
!7ith thirty tons of mechanical brains,
It multiplies, adds, divides, explains.
to the ENIAC,' 79461s
-"Ode
On the other side of the Atlantic, the U.S. Army was exploring the use
of machines to tackle a very different problem-ballistics calculations.
Faculty at the University of Pennsylvania's Moore School of Electrical En-
gineering had been assisting the army's Ballistics Research Laboratory, at
Aberdeen Proving Ground in Maryland, to determine the traiectories of
projectiles fired from various weapons. Some calculations were done on
the Moore School's Differential Analyze4'ryhile others were done by two
hundred female computers who were trained in mathematics and used
desk calculators. The goal was to compure firing tables, which showed
gunners how to aim their weapons to hit a target at a particular range.
The calculations needed for such a table were complex. To create a single
firing table required a month of continuous work for either the Differen-
tial Analyzer or a team of a hundred women.16
Moore School professor.|ohn Mauchly proposed to speed up rhis pro-
cess by building an electronic digital compurer to do the calculations. In
1943, Mauchly and his colleague John Presper Eckert received a con- Figure 1.2
tract from the army to build the ENIAC. Like Colossus, the ENIAC used firreC compurer, 7946.photo courtesy of University of Pennsylvania Archives
 18 Chapter I Breaking Codes and FindingTraiectories 19
F
i
Computer Work and the Gendered Division of Labor the U.S. Army), the WAVES (Women Accepted for Voluntary Emergency
Service of the U.S. Navy), and the \7RNS (Süomen's Royal Naval Service
81'the time I reached ny third year of college, I started looking around for some of the UK Navy, popularly known as "'Wrens").'Sfomen were accepted
type of occupation that could usc a math major. I dicn't want teaching. Insurance in "men's" production yobs only with the understanding that they would
comparries' actuarial positions required a master's degree (and they seldom hired
vacate these positions after the war to restore the accustomed gendered
women, I later found out). . . . .Just after graduation, I happened to see an ad in
the daily paper. The Army was looking for wc,men with a degree in mathemat- division of labor.
ics-right here in Philadelphia. Yet the many constraints on women's options did not prevent some
McNulty', ENIAC progranrmerl' from actively choosing computer jobs, whether as a career path, a brief
-Kay
adventure, or a patriotic duty. It is worth asking what made computer
How did won'ren come to staff the world's first electronic digital com- work seem possible and desirable for these women. In other words' what
puters, a seemingly masculine engineering domain? The short answer is defined this type of work as a good opportunity rather than something
that tl,ey u.ere actively recruited during a time of urgent need and scarce unimaginable or unappealing? This question resonates with today's ef-
nrale labor. The contributions of Rosie the Riveter and her like are well forts to recruit more women into computing, as it has become clear that
known: the Americarr and British governments exhorted women to take the mere existence of computer jobs and the lack of explicit gender bar-
wartime productiorr jobs, and millions of women filled blue-collar f actory riers do not automatically translate into having women regard these jobs
positions that formerly were tlre do'-rain of men.l0 Just as women were as attractive opportunities. The historical cases suggest that important
recruited for jobs in the aircraft indusrry, so they were also channeled into factors include whether the skills that women already possess are seen
computer jobs that might otherwise have gone to men. as relevant for computing and how computer jobs compare to more fa-
But invoking wartime labor scarcity begs the question because labor miliar alternatives for women. In the'S7orld \War II context' one factor that
scarcity is not a giverr fact, it is a social construction that depends on encouraged women to consider technical work was government efforts
assumptions about what types of people are suitable to fill which to recruit them into factory jobs. Billboards and slogans recast techni-
rypes
of positions. c)ther solutions ro the problem of supplying both fighters cal work-formerly off-limits as a masculine area-as appropriate and
and workers were certairrly possible.'$fhy not allow women ro serve in manageable for women, assuring them, "If you've used an electric mixer
combat so thar men could keep their existing jobs? vhy not recruit black in your kitchen, you can learn to run a drill press."23 At the same time,
men for skilled wartinre jobs instead of confining them to the lowest- the limited careers that were open to women in more traditional fields
paid work? These alternatives would have been unappealing, even un- made computing säem a promising alternative. Before the war, American
thinkable, to most political and business leaders (and wornen themselves) and British women with a college degree or other advanced preparation
because the sanre cultural attitudes, priorities, and power relations that in math found themselves tracked into low-paid jobs with little hope of
had excluded women and minorities from certain occupations before the advancement. Typically, they were hired either to teach math in secondary
war were still in play.2l As hisrorian Jennifer s. Light has argued, .,the school or to do tedious repetitive calculations for insurance companies
meaning of 'wartime labor sl.rortage' was circumscribed even as it came or other information-intensive businesses. Many women, like McNulty
into being"; it did not lcad to having large of women promoted quoted above, found neither choice appealing. For them, a job in scientific
into the "mc)re 'important' technical and 'umbers
classified work', reserved for computing presented one of the more exciting and well-paid opportunities
men.22 At midcentury, the segregation of labor markets by sex was an that a mathematically inclined woman could hope for in the mid-1940s.
open, legal, and ubiquitous fact of Iife in the united States and united As in other areas, women's roles in wartime computing projects were
Kingdom. x7omen wlro wanted to join tl-re military were shunted into circumscribed by the existing gendered division of labor. Designing com-
auxiliary noncombat groups such as the'wAC (\7omen's Army corps of puter hardware was done solely by men, since women were excluded
 20 Chapter 1 Breaking Codes and FindingTrajectories 21

from gaining the engineering education, experience, and authority

that of Colossus, "ln those deeply sexist times no woman rose to a senlor
might have allowed rhem ro participate. v/omen's only i*volvement in management position in the Newmanry [the Colossus area, named after
building the ENIAC hardware was pertbrming factory-like assembly of Max Newmanl, and none achieved the status of cryptographer, despite
conrponents.2aorr the other hand, operating and programming compur- the fact that nine per cent of Newmanry Wrens had received a university
ers could more easily be constructed as women'swork. since the jobs of education" (and this at a time when only 4 percent of British men had
con.lputer operator and programmer did not exist before the war, they earned college degrees).2s The wartime computing environment presented
were not alreadl, stereotyped as masculne. Moreover, women had their women with a complex mixture of new opportunities and old gender
olvn prewar history in conrputing that gave them entry-within limits- boundaries.
to ncw compurer projects.
In the scierrtific world, worlren had b,een employed in large numbers Women's Entry into Wartime Computer Proiects: Motivations and
to do calculations by hand. These human ccmputers generally, although Experiences
not always, had sorne training in mathematics, and it was fron-r
the ranks The Bletchley Park '$7rens were not necessarily looking for technical
of such mathemarically trarned women that the ENIAC project drew its work. Indeed, they had no way of knowing about the computers' whose
first corps of progranrmers. However, e'en women with college degrees existence was kept secret. The women had volunteered for a variety of
were relegared to subprofessional job classificarions. As Kay personal reasons, including a desire to help with the war effort or a wish
McNulty
recalled, "Tlre girls were told that only 'men' could get professio to leave the confines of home or a dull 1ob. Yet becoming a'wren was
nal rat-
The ENIAC programmers were not promoted from computer
ir.rgs."25
to still a significant choice. .|oining up meant committing oneself to leave
mathematician-a professional rating-until 1946.26 familiar surroundings, submit to military discipline, and work under gru-
January
In the separate world of administrative data processing, women I eling conditions.2n For women already willing to take on these challenges,
typically held the low-level job of key-punch operaror (the person masterin€l an unfamiliar machine may have seemed more intriguing than
who
punched holes in the paper cards used fo: input) or: the I
intimidating. Eleanor Ireland, a newly recruited Wren in September 1943,
slightry more ad-
vanced job of tabulating machine operarcrr (the person who ,
recalled being eager for a chance to operate Colossus after she caught a
loaded decks !
of cards into the machine and initrated its operation). The existence glimpse of one: "l was transfixed by it. All these whirring tapes ' ' ' and
of
female tabulator operators may have pror-ided a precedent
for employing I
the noise of it all. . . .I was fascinated by it' . . .I thought,now that's
women as computer operators at Bletchley park. Because where I'd like to be, not doing the other things. I wanted to be on that
of the particular
role that the computer played at Bretchley park, the \7rens machine."3o As ä Wren, however, she had no real say in her assignment'
were not cailed
on to become programmers as rhe ENIAC women were. Although other than by doing her best on the placement test that was used to pick
the
Colossi could, in theory, be programmed for a variety which women would operate the most complex machines.
of tasks, in practice
they were devoted to codebreaking, and Bletchley park The work environment at Bletchley Park was both exciting and harsh.
had no prog."--
mers as such' Instead, the work of programming was Material conditions were often unpleasant due to wartime shortages;
divided between a
mathematician, who decided what operations the machine many rü/rens complained about the sweltering heat of the machine rooms
srrourd per-
form to solve a given decoding probrem, and an operator, in summer and the bitter cold of the dormitories in winter. "The food was
who used the
matl-rematician's instrucrions to set up and run these prerty grim," recalls Ireland; "To be faced with cold congealed liver and
operations on the
computer. Dividing the task this way was meant prunes in the middle of a night watch was not very good." The workers
to conserve the scarce
labor of trained mathernaticians. It also reproduced were also subjected to military discipline-although as with many aspects
a gender hierarchy:
all of the cryptographers working with Colossus were men, of life at Bletchley Park, there was a gendered double standard. Male
and all of
the operators were worren'2t As B.
Jack copeland notes in his history mathematicians recruited for the project from Oxford or Cambridge
 22 Chaptcr 'l Breaking Codes and Fmding Trajectories 23

"remained in civvies, independent of the demands of army life, thus hav- In contrast to the'sürens, the ENIAC women in Philadelphia made an
ing freedorn for conscientious research and overtime, a deciding factor explicit choice to work with a computer-although neither they nor any-
in favor of original thought and good w,frk."31 While men's supposedly one else at that time had a clear idea of what that would involve. Some of
more creative work excused them from military exercises, Colossus op- them had already been working for tl're military doing hand calculations;
erator Catherine Caughey described hovr the \X/rens were forced to do others came straight from school, having found out about the openings
drills and marches after worl<ing all night on the machine: "They made us at the university of Pennsylvania through college advisers or friends. one
do an hour's squad-drill on the gravel in front of the Abbey every morn- thing that they had in common was a dearth of better options. Kay Mc-
ing. . . .We were all being killed by the consrant changes of shift. Church Nulty (quoted above) was painfully aware of the limited prospects for
Parade on a Sunday was compulsory. On my first Sunday, wl-ren I was on female mathematicians. Marlyn'lrescoff, one of two ENIAC women with
nights, I had to assemble with the others. We had to march two miles each a nonmathematical college degree, found the job market even worse. she
way to the village church, along an icy r,tad."32 The arduous and seem- had graduated from Temple university in June 1942 as an education ma-
ingly unnecessary demands placed on the Colossus operarors reveal their jor specializing in social studies, English, and business but could find no
superiors' unspokerr presumption that women's work is by nature mun- teaching jobs in rhose areas. Hoping to make the most of what skills she
dane and does not require one's full energies. Such gender stereotyping had,'$Tescoff sought out a position at the Moore school: "Late in August
can be destructively self-fulfilling. \X/ith their work seen as less imporrant, of that year a friend told me they were hiring at the Moore School and
women are loaded down with extra chores such as military drills (or in if I knew how to run a calculator, that would stand me in good stead. I
the present day, housework or low-level a,iministrative tasks). These bur- made an appointment for an interview with John Mauchly, but he quickly
dens, in turn, make it harder for women t,t perform up to their potential, turned me over to his wife, Mary, who hired me when informed that I
reinforcing the idea that they are less capable rhan men. In the Bletchley could operate an adding machine,"3a Jean Jennings, who had been doing
Park case, there was a somewhat lrappier ending: those in charge of the desk calculations for the army at the Moore School, recalled how she
Colossi eventually noticed the exlraustion of their operators, and tl.rey jumped at the chance to train as an ENIAC programmer instead: "It was
told the Wrerr officers to stop requiring the two-mile march to church..r3 something new! I mean, I didn't know what it was, but I knew that it
Like all personnel at Bletchley Park, the Wrens had to observe strict was boring sitting there and pounding the calculator, and [ENIACI was
secrecy, given the highly sensitive nature of their work. New recruits to something new. And I always believed that I could do anything as well as
Bletchley Park had to sign the official secrers Ac and were threatened anybody else, if I were on an even footing."is Jennings's reaction reflects
with dire consequences should they ever L'treathe a word about the code- an awareness tha't the "computer" positions traditionally held by women
breaking operation. All personnel were forbidden to talk about their were dead-end lobs with little chance for promotion. Since she planned to
work, to keep diaries, or to tell their families what they were doing- continue working after the war, she hoped to get in on the ground floor
either during the war or lorrg afterward. Despite all tl.ris, many women of a new profession where she could be "on an even footing" with men.
fou'd the social environment one of freedom, adventure, and indepe'- The ENIAC project had a much more relaxed atmosphere than Bletch-
dence. They had a chance ro escape parental supervision, go to London ley Park. Security was still a factor: there were restricted areas, and ap-
on their days off, mingle with people of other social classes, and find plications such as the hydrogen bomb calculations were kept secret from
romance; many women married men they met ar Bletchley park. And the the women. But the programmers were allowed to talk about their work
work, however tedious, was made meaningful by their awareness that among themselves and with the engineers, and they were encouraged to
thel' were performine a vital (if .rysterious) service for their country. For learn as much about the machine as possible and to devise new techniques
the colossus operators, there was also the thrill of working with cutting- for programming it. Like the colossus women, the ENIAC women also
edge machines. found adventure and independence in their wartime work. vomen from
 24 Chapter 1 Breaking Codes and Pinding Traiectories 25

small towns got ro see Philadelphia, New York, and Sfashington, and Initial Assumptions about Computing Work
several of them married men they met at Penn or Aberdeen. In addition, The selection criteria and lob training for Colossus and ENIAC indicate
they enjoyed the intellectual challenge of the project, which gave them a that these tasks were believed to require at least some mathematical abil-
chance to use their mathematical and logical ingenuity to the fullest. ity and that the best available female workers were assigned to the com-
puters.'Wrens were placed in Max Newman's mechanical codebreaking
Constructions of Skill in Women's Wartime Computing Work section based on an eclectic mix of technical and social qualifications:
"'Wrens were chosen by interview. . . . No fixed qualifications were re-
occasionally, the six of us programmers all got together to discuss how we quired, though a pass in mathematics in School Certificate or apparently
thought the machine worked. If this sounds haphazard,it was. The biggest advan- 'good social recommendations' was normally considered essential. . . .
tage of learning the ENIAC from the diagrams ''vas :har we began to understand
what it could and what it could not do. As a r:sulr we could diagnose troubles None had studied mathematics at the university."3T Newman gave pro-
almost down to the individual vacuum tube. Sirrce r,r'e knew both the application spective operators a two-week course covering binary math, the teletype
and the machine, we learned to diagnose troubles as well as, if not betier than, alphabet code, and machine operation, after which they were given a
the engineer.
written exam. Those with the highest scores were given the most chal-
ENIAC ptogrammer3o
-.fean .fennings, lenging assignments, including Colossus.38 The first Colossus operators
were presumably trained by Newman himself to use the machines, but
unlike prior women's work, such as hand calcuiations, programming and
after that experienced operators took on the responsibility of training
operating computers were completely new tasks with unknown demands.
new recruits, adding another dimension to their duties.
No one had ever operated an electronic digital computer before colossus.
Qualifications were higher for ENIAC. The six programmers all had
Although programming of the Harvard university Mark I had started in
college degrees, four of them in mathematics, and the women passed
May 1 944, the ENIAC women had no knowledge of this project, and in
through two levels of selection-first as hand computers and then as pro-
any case the procedures for programming ENIAC were completely dif-
grammers. The level of skill involved in hand computing can be gleaned
ferent. Even the men who built and oversaw the computers did not have
from a description of the Moore School's 1.942 training course for those
a clear idea of what would be involved in using them. since the role of
who were working on ballistics problems. Students were expected to
programmers and operators was not well understood in the early years of
begin at a fairly high level: the course was aimed at college graduates
computing, gender stereotypes partially filled this vacuum, leading many
with at least a year,of college math (although later this requirement was
people to downplay the skill level of women's work and its inrportance to
dropped), and prospective students had to pass an aptitude test and go
the computing enterprise.
through an interview. The course ran for thirteen weeks, forry hours a
A review of the colossus and ENIAC projects reveals that in both cases,
week, including three lectures per day as well as five hours of "super-
project leaders underesrimated the level of training needed and skill exer-
vised study, the latter to include 4 hours per week of supervised labora-
cised by their female staffs. But even though gender bias played a large role
tory work with computing machines." Topics covered included algebra,
in shaping women's responsibilities and rewards in the workplace, this role
trigonometry, analytical geometry, calculus, interpolation, exterior bal-
was mediated by the open-endedness of the technology itself. computers
listics, and least squares.3e Much of the teaching was handled by Adele
were new and complex enough that the work of programming them could
Goldstine, a trained mathematician and the wife of Herman Goldstine,
be organized in many possible ways and their personnel could extend the
the army liaison to the ENIAC project. Marlyn Wescoff (who, as noted
boundaries of their jobs to incorporate new skills. The interaction of evolv-
above, had been hired on the strength of her familiarity with adding
ing cultural constructions of both computing work and gender could result
machines) described the class as challenging but rewarding: "I had not
in either a familiar patern of sex-typing (as happened at Bletchley park) or
taken mathematics in college, so all of this was wonderful, strange, dif-
an unexpeced opportunity (for the ENIAC women).
ficrrlt- end ercifins."40 There rvas e second rnrrnd nf qelertinn for FNTAC
 Breaking Codes and FhtdingTraiectories 27
26 Chapter 1

across the room, endangering the unfortunate operator and bringing

op-
programmers, with Hernran Goldstine picking "six of the best comput-
ers" for thc job.al Tlrose chosen were seni to Aberdeen in June 1945 to I erations to a halt.
he trained on the IBM inputloutput equipment to be used with ENIAC
j'
Keeping the decoding process on track therefore depended on the
("tabulator, sorter, reader, reproducer; and punch").42 I Wrens' mechanical ingenuity' and as Ireland noted, they keenly felt this
.,That was a tricky operation, getting the tape to the right
Despite all this preparation, the ENIAC women were not given any
I
L
responsibility:
tension. . . .'Sre were terrified of the tape breaking should the tension
be
specific training in programming-a telling omission. As noted earlier,
wrong. Breaks meant that valuable time was lost. . . . It was instilled into
gender bias was certainly at work in the official classification of the {

ENIAC women's jobs as "sub-professional." However, I suggest that us that time was of the very essence. \7e knew we were working against
The
Mauchl,v and Eckert themselves did not devalue programming so much the clock and that people's lives depended on what we were doing."a6
r
as overlool< it, consumed as they were with the challenge of building a I operators also had to devise methods for keeping the tapes in good repair'
as Dorothy Du Boisson recalled: "Each tape . . . had to be
joined into a
working machine in time to aid the war efJort. Rather than having a clear t.

loop. It was difficult to get the join right, and if we didn't the tape might
i

image of programming as merely clerical, as some historians have argued, I

up to the speed of the machine. After many experiments, we

I

the ENIAC designers seem to havc had only the vaguest notion of what not stand
prograrnming might involve.a3 As Mauchly explained in a 1973 inter- found that special glue, a warm clamp, and French chalk produced a good
view, programming was an afterthought: "!7e didn't think that we should joint-the art was using iust the right amount of glue.''47
spend our tinre worrying about figurir-rg out programming methods and The operators did not need advanced mathematics, but they had to
programmine aids and all that. We felt that if rn'e had the machine capable be able to do basic arithmetic quickly and accurately in their heads. To
of doing all these things . . . there would be time enough to worry about check whether the decoding was successful, the operator set switches on
those things later."44 For the ENIAC women, the undefined nature of their the front of the machine to program it to count the frequency of certain
jol"r was challengirrg, but it also gave them the freedom to define program- letters. These counts were printed out on a nearby typewriter. If the cryp-
i

ming-through their actions and achievements-as a wide-ranging and l tographer chose the wrong wheel pattern, the results had no particular
intellectually fulfil ling task. pattern of letter usage, but if the correct wheel pattern had been found,
the output lrad a nonrandom pattern that was characteristic of the lan-
i

Actual Challenges of Operating and Programming the First Computers guage being used (in rhis case, German military language).as The operator
Colossus was used for a sinelc purpose-to iiscover the starting wheel might be asked to check whether the letter frequencies that were recorded
patterns that the Germans used to encode particular messages on the Lo- on the typewritei deviated significantly from a given norm, signaling a
i

renz machine. To start a decoding cycle, the male cryptographer (seated at I possible SucceSS. Ireland recalled, "sometimes I was given a nofm' and as
a desk) proposed a wlreel pattern to try, and the female operator (standing each figure came up on the typewriter, I did a calculation and wrote down
at the n.rachine) prograrnmed that patter:t into the machine by insert- i, against the figure how much above or below the norm it was' I became

ing pins into a grid on the l"rack. Slre then prepared to run the machine i very good at mental arithmetic."ae
I
b1'threading a tape containing an encoded message onto the computer's To be a successful Colossus operator therefore required physical endur-
i
ance, mechanical aptitude, attention to detail, and the ability to memorize
I
tape reader, whiclr was mounted on a large metal frame nicknamed "the I

bedstead"a5 (see figure l.1). Thc tape reacer had multiple pulleys to en- ! codes and do mental arithmetic. Female operators were also entrusted
able it to handle message tapes of different lengths, and the operator had I
I
with a certain amount of initiative. They did some runs without supervi-
to adjust the tension on these pulleys to keep the tape running smoothly. i sion, evaluating the results to determine if the correct wheel setting had
Once the computer was turned on, rhe tape sped through the reader at been found.50 They were also trained by the engineers to do routine test-
I
40 feet per second-which meant that if the tape broke, ir went whipping ing of the machines, and they solved some minor mechanical problems,
I
 28 Chapter 1
Breaking Codes and Findhg Traiectories 29

such as keeping the typewriter from shaking out of position due .When
to the work and that of a \7ren was overdone. . . . the section grew to such an
machinet vibrations.s I
extent that there was a shortage of males[,] some of the '!ürens did wheelbreak-
Despite rhe various physical and menral abilities that they ing jobs and other tasks requiring technical knowledge which were previously
emplol,ed,
thought possible only by a man. A programme of training Wrens in theory and the
\vrens were not treated as skilled workers or expected to understand the
work began in earnest fin Ianuary 1945,] but if this change in policy had started
logic behind the machine's operations-at least at first. The organization earlier greater dividends would have been noticed in the results.sa
of work at Bletchley Park made a sharp distinction between men, who
Vergine's forthright critique highlights how security concerns rein-
were rotated among tasks to give them a broad knowledge of the code-
forced a social hierarchy based on gender to restrict women's roles to a
breaking operation, and women, who -were confined to narrow tasks
on degree that even their male supervisors found detrimental. It also high-
the assumption that they had only limited technical abilities. As
tlie official lights the difference in approach between the Colossus and ENIAC proj-
1945 report nored, "wrens (unlike men) were organised in fixed watches
ects. Perhaps the Colossus operators could have developed a comparable
and given fixed jobs in which they could become technicalry proficient.,,52
level of autonomy and expertise as the ENIAC programmers if they had
But the assumption thar women's work was by definition
unskilled did received similar encouragement to learn about the machine and to experi-
not reflect the reality of the work . By 1944, it had become evident
that ment with programming techniques.
tl-re colossus operators' tasks were more challenging
than their male su- Like Colossus, ENIAC was set up by plugging in wires and setting
pervisors had originally thought. As trre \üy'rens were asked
to perform switches (figure 1.3). The various components of ENIAC had to be labo-
more complex operations and to work unsupervised, they
needed more riously reconfigured for each calculation, in effect building a new special-
theoretical knowledge, and in.fanuary 1945 Newman's section
set up an purpose computer each time. As with Colossus, operating the machine
Education comrnittee (composed mainry of Nfrens) to arrange
technical was very much a hands-on, physical activity-but unlike Colossus, this
seminars. The report conceded, "rn practice it became increasingly hard was only the last stage of the women's programming process. It began
for wrerrs to ger a comprete picture of an organisation i' which
they with numerical analysis-taking a mathematical problem and converting
might have only done one job. Moreo'er the mathematical styre
of the it to a form that could be solved using the arithmetical and logical capa-
Research Logs made them unreadable fc,r Wrens, and
before they (or new bilities of the computer. Then the program had to be broken down into
men) undertook chi-breaking and col.ssus-setting on rheir
own, some a sequence of detailed steps, and finally these steps had to be converted
other introduction to tlre tlreoretical side was needed." The
colossus op- into instructions for setting up the machine. Only then did the women do
erators' seminars were evidently fairly advanced, as the report
contrasts the physical setting up and finally run the program. Since most programs
them with "the less rnathematical general lectures.',i.]
do not run perfectly the first time, the programmers usually needed to
At least one conrenlporary wit'ess berieved that gender bias had
red to debug the instructions and repeat the cycle until they achieved the desired
an underestimarion of the comprexity of the women's
work. several u.S. results.
Army signal intelligence officers were srationed at Bretchley park,
and the The first problem run on the machine was a calculation for the hy-
first of them to worl< in Newman,s section, George H. Vergine,
wrote a drogen bomb. It was devised by physicists Stanley Frankel and Nicholas
section of the 1 945 report. Vergine explicitry critiqued
the gendered divi- Metropolis, who came to Philadelphia from Los Alamos for this purpose.
sion of labor on colossus, arguing that stereotyped
assumptions about For this problem, Herman Goldstine taught Frankel and Metropolis how
skill had delayed making full use of the \frens, talenrs:
to specify a program for ENIAC, and the female programmers used these
The structure of the organisation was built o' the
concept that the male members specifications to set up the program on the machine. For subsequent proj-
should carry the responsibirity for any decisic,ns and
the *ork shorlJ ü" r.au..a ects, however-including the ballistics trajectory calculations for which
to a routine so that the wrens were able to do it. . . . perhaps
the rnain faurt of
the organisation was that the primary principle the machine had been built-the women performed the mathematical
of thedifferen.. u.r*..n
".nn,, analysis and logical design of the programs as well as the physical setup
30 Chapter 'l Breaking Codes and Finding Traiectories 31

In Jennings's account, the women continued to take the initiative in for-

mulating programming methods' with Mauchly simply answering ques-
tions that they brought to him'
As noted earlier, the intellectual sophistication of programming was
belatedly recognized in early 1'946,when the ENIAC women were pro-
moted to professional status in the u.S. civil service. Yet the assumption
that women,s work must be less scientific or complex persisted for many
years afterward. In 7971., as part of a patent dispute between the ENIAC
creators and rival inventor John Atanasoff, the ENIAC women gave de-
positions in which they were asked to characterize their work as pro-
grammers. These documents provide a valuable glimpse into the women's
own understanding of their technical role at a time when the ENIAC was
still fairly obscure and its programmers were not self-consciously claim-
ing a place for themselves in history. Their experience of the work as
novel and intellectually demanding contrasts with the male interviewer's
evident assumption that the women had merely been rote operators.
Asked to define their iob, the ENIAC women's responses made a clear
Figure 1.3 distinction between the physical "setting up" or "plugging in" (what we
ENIAC with Frances Bilas and Jean fennings. Photo courtesy of University of would now call operating) and the more intellectual work of numerical
Pennsylvania Archives.
analysis and program design. Kay McNulty replied, "I should say that the
programmer's principal job is to reduce the equations to forms that can
and subsequent debugging. This diverse set of tasks required advanced ENIAC and to plan the setup. By virtue of the nature of
be used by the
mathematics and a detailed mastery of the machine's logical design and
the ENIAC itself-no other machine is like this-we had to do this physi-
its physical layout, as well as creativity, initiative, and physical endurance.
cal labor, which certainly wasn't programming, the physical setting up."58
As noted above, the ENIAC women were offered no instruction in pro-
Y/hen the interviewer tried to characterize the work as merely plugging in
gramming (indeed, no one was qualified to teach it), and no manuals were
cords, Ruth Lichterman was quick to correct him:
available until Adele Goldstine produced one in June 1946, well after the
'When you say you programmed the machine, did that mean physically that
original programmers had trained themselves.ss The women learned by Q:
ylo,, took these plugs from otte place and plugged them into another place?
trial and error, discussing ideas among themselves with little more than You got a
[Ruth Licbterman]: \üell, now, Program means several things' ' ' '
engineering drawings to guide them.56 Jean Jennings recalled that only problem, and you started with pencil and paper, and you decided how you were
after she and Betty Snyder had met Mauchly by accident were they able going to do this problem and which numbers went where. . . . [Y]ou drew a dia-
i."-".rf all this stuff and then you actually went on the machine, and we call that
to enlist his help: ;plugging in" rather than "program."5e
\üe were sitting there, trying to figure out how the accumulator worked, and this
Lichterman's response describes the programmer handling the entire
man walked in. . . . He said, "I'm John Mauchly." We said, "Oh my God, are we
glad to meet you!" [She laughs.] 'We'd never met him before.. . . So he came over spectrum of tasks that were required to go from a problem statement to
and helped us figure it out. He said, "\üüell, look, my office is right next door. a logically worked out program that was ready to be "plugged in" to the
'!7hen
you have any problems, you can come in, and I'11 help you." So we began to machine. This work was complex and varied enough that the women spe-
save up our problems, and in the afternoon, if he was there, we would go in and
c'ralized i n,dj ff erent area s, accord t"*.,: t"*:
ask him all our questions. So that's how we learned how to program the ENIAC.TT ? i:-:::t Y l.:ll:1"::::'
 .i2 Chapter 1
Breaking Codes and FindingTraiectories 33

Kay was often more creative, suggesting clever ways to reuse and reduce I felt the tremendous pressure to make this thing work. Everybody was
the total size of the program. Marlyn and Ruth agreed to generate a test counting on us: the BRL, the Moore School, the ENIAC design group'
trajectory, calculating itexactly the way the ENIAC was to do it so we the Goldstines. \fle both loved the attention and knew we could do it!"6'
could check the detailed steps once it was on the ENIAC."60 By the night before the demonstration, there was only one major bug
Designing programs for the ENIAC to,tk considerable creativity, since left, but it was an embarrassing one: the trajectory program kept going
every prograrnming technique had to be worked out for the first time. after the projectile was supposed to have hit the ground. As Jennings re-
The women made a number of innovations in programming methods. called, "During the night it came to Betty what was wrong. She came in
Betty Snyder is credited with inventing break points, a debugging tech- the next morning and flipped one switch on the master programmer and
nique that involves stopping the machine in the middle of a program ro the problem was solved. Actually, Betty could do more logical reasoning
check intermediate results.6l Break points ha'ze since become a standard while she was asleep than most people can do awake."66 But when it came
debugging tool. The women also developed a sysrem of programming to getting credit for their accomplishment, the women were in for a rude
notation tl-rat allowed them to visualize the multiple simultaneous opera- awakening.
tions that the computer perfonned at each step of executing a program.
\Xiith this notation, according to .|ennings, "rve could keep track of the Constructing Opportunities for Women in Computing at War's End
timing of program pulses and digital operations. The ENIAC was a paral-
lel machine, so the programmer had to keep track of everything, whether We had to sign the Secrets Act when we left, and we had to take all these ma-
chines down. My parents never knew what I did, and neither did my husband. I
interdependent or independent."62 The programmers also developed new
never told anybody at all. None of us ever did.
methods in numerical analysis.6s For examplg a fundamental concern in
Ireland, Colossus operatort-
applied mathematics is that for many problems, the computer calculates -Eleanor
an approximate rather than exact solution, which can lead to significant
What opportunities were open to women in computing after the extraor-
errors. The ENIAC women responded to this challenge by creating ex-
dinary circumstances of war gave way to the social norms of peacetime?
perimental programs to generate knowledge about the magnitude of these
Historians such as Alice Kessler-Harris, Ruth Milkman, and Penny Sum-
errors and ways to minimize them. As Je:rnings described in L971,,,Ve
merfield have described how British and American women who took
ran a number of problems to find out how to use the machine, to test
"male" jobs in factories during the war were expected to give them up af-
what interval of integration gave us the most accuracy if you considered
terward to men u.dro were returning from military service.6s Many defense
round-off error and truncation error, and we were trying to determine
projects-including hand-computing tasks that employed hundreds of
what interval we could use in order to minimize the effects of trunca-
women-were closed down at the end of the war, so the jobs themselves
tion error without building up the round-off error. . . . The view was to
vanished.6e This was not unexpected. Sühen the Moore School recruited
establish sorne procedure that gave us the most overall accuracy."6a As
women to do ballistics calculations in 1.942, its advertisements warned,
this example illustrates, the ENIAC women took the initiative to creare
"lt should be understood by each prospective employee that the appoint-
pr:ogramming tools as well as programs.
ment is an emergency war service appointment, which in all probability
In addition to the various programming methods that the women in-
will terminate with the end of hostilities."t0 Compared with most women
vented, Snyder and.|ennings created an important software product-the
in wartime industries, however, the ENIAC programmers had consider-
program to calculate ballistics trajectories, which had been the original
ably more opportunity to stay in the field after the war. There was no
goal of the ENIAC and would be the cenrerpiece of its public debut in
preexisting male workforce to push them out, and as digital comput-
1946. .lennings described the excitement of being responsible for such ers became commercialized after the war, the demand for programmers
a high-stakes project: "\7e worked nights and weekends. . . . Betty and
 34 Chapter 1
Breaking Codes and Finding Trajectories 3.t

increased. In this labor market, women with computer experience could

potentially find well-paid and interesting jobs.
women in both the ENIAC and colossus projects demonsrrated that
they had the interest and ability ro excel in computing work, but their
outcomes were very different. The wrens' jobs abruptly ceased with the
end of hostilities, and none of the colossus women seem to have pursued
postwar civilian computing careers.Tl In contrast, the ENIAC women had
the option to continue working, and some of them went on to have distin-
guished and influential careers in computing. Äs these different outcomes
suggest, gender bias was one only factor affecting women's postwar op-
portunities. several other characteristics of the American and British
projects-including secrecy, divisions of labor, and training practices-af-
fected women's chances for making their wartime work the basis for a
long-term career in computing.
For all its importance, colossus and the rest of the Bretchley park op-
erations remained unknown to the public for many decades. At war's end,
winston churchill-fearing that the technology could be used by an en- Figure 1.4
emy in a future war-ordered the colossus computers to be destroyed.T2 Former operator Eleanor Ireland with rebuilt Colossus at Bletchley Park, 2001
Photo by author.
only many years later, when the publication of some declassified docu-
ments revealed the secrers of Bletchley Park, did the machines and their
inventors and operators begin to receive credit for their crucial wartime with Colossus. Newman went to Manchester University, where he helped
role. Today Bletchley Park is a museum and features a rebuilt, working design the first Manchester computer, known as the "Baby."7s As noted
Colossus (figure "l .4).7i academics, these men possessed the intellectual authority to lead comput-

The extreme secrecy of the colossus p:oject hindered both men and ing projects even without credit for their wartime work, and they were
women from reaping the full benefits of their work. The wrens were pro- able to apply knowledge that they had acquired from the Colossus proj-
hibited from even mentioning ro a prospective emproyer that they had ect without having to specify the source of their inspiration.

had experience with computers. Tommy Flowers and his engineers at Limited training also blocked women from moving on from Bletchley
Dollis Hill received neither credit for their ertraordinary contribution Park to computing careers. Few of the women who operated Colossus
nor support for building the next generation of computers at the post had mathematical backgrounds, and the gendered division of labor and
office, which they might have been granted had their wartime achieve- security-driven compartmentalization of knowledge at Bletchley Park
ments been known.Ta In contrast, the men working on the more theoreti- kept them from gaining theoretical knowledge of how the computers
cal side of computing-Max Newman and Alan Turing-were able to put worked. Also, there was no direct commercial spinoff from the Colossus
their experience to immediate use following the war. Alan Turing went machines (as there was from ENIAC), so the operators could not translate
to work on the Automatic computing Engine (ACE) project at the Na- their machine-specific hands-on skills to a peacetime job.
tional Physical Laboratory, drawing up plans for a stored-program elec- For the Moore School women, all of these factors were reversed. They
tronic computer in late 1945. These plans vrere much more complete and had been encouraged to learn every aspect of the machine's design and
specific than the largely abstrac EDVAC Report authored by operation and had been given mathematical training. ENIAC itself was
John von
Neumann a few months earlier, reflecting Turing's firsthand experience not destroyed like the Colossi. Instead, it was transferred in January 1947
 .36 Chapter 1
Breaking Codes and Finding Traiectories 37

78 yet Snyder and Jennings received no attention or official recogni-

to the Ballistics Research Laboratory at Aberdeen, where it was used until
1955. At least seven additional women became ENIAC programmers at tion for their contributions, and the programmers were not mentioned
Aberdeen.'6 Most important,in March 1,946Eckertand Mauchly formed the publicity for the event-an omission some of them felt keenly^ "I
the Electronic Control Company (soon renamed the Eckert-Mauchly remember not being invited to the luncheon" for the press,
Jennings recalled in 197'l . Snyder complained, "I
wasn't photogenic' and
Computing Corporation) to commercialize their invention. ENIAC pro-
grammers Betty Snyder and Jean Jennings irnmediately joined the new wasn't included on any of the pictures of the entire stupid thing." KaY
company, which was bought by Remington Rand in 1950 and produced noted ruefully, "None of us girls were ever introduced [at the
the very successful UNIVAC computer. conferencel; we were just programmers."Te It was not until the fifti-
The ENIAC women who did not stay in programming were constrained anniversary of the ENIAC that historians rediscovered and began to
more by social convention than by the job market. Most remained in their the ENIAC women's contribution.s0
jobs for a year or two after the war but then left voluntarily to marry or Historian Jennifer Light has argued that the erasure of the ENIAC
have children. Marlyn Wescoff left the group in December 1,946 to be stemmed from a more general devaluing of women's labor
married. Frances Bilas, Ruth Lichterman, and Kay McNulty continued to American sociery concluding that "the press conference and follow up
work with the ENIAC after it was moved ro the Ballistics Research Labo- rendered invisible both the skilled labor required to set up the
ratory. Bilas left in 1948 when she was expecting her first child, McNulty and the gender of the skilled workers who did it."81 This
left the same year to marry John Mauchlg and Lichterman left after two undoubtedly true, but the public image of ENIAC also reflected a view
years to get married. None of the four returned to full-time computing 'of computing that privileged machines over people (of either sex)' As Di-

work, though McNulty assisted informally w-ith her husband's business anne Martin points out, press coverage of ENIAC depicted the computer
ventures. Although women of the 1940s felt some pressure to conform an autonomous "brain" that performed calculations with little reliance
to the role of full-time wife and mother, marriage was not necessarily a humans.82 This distorted image was not simply the fault of ignorant
bar to continued work as a programmer. Snyder married John Holberton but also stemmed from the way that ENIAC was presented'
(who had been put in charge of ENIAC operations in June 1945) but either the female programmers nor the many male engineers who as-
continued to work as a programmer and software researcher until 1983. sisted Eckert and Mauchly received credit at the press conference. Yet
Jennings followed a different path, pursuing full-time work and family emphasized by Barkeley Fritz, who was one of the first male ENIAC
serially: she remained a programmer until 1,951,,\eft. to raise a famlly, programmers to join Aberdeen in 7948, the machine would have been
returned to computin g in 1967 , and worked until 1985 as a project man- worthless without the work of the programmers and engineers who came
ager and editor of technical publications.77 Their technical training and before him: "success was achieved during this first time period largely
experience gave the ENIAC women some room to negotiate a balance because of the availability of a dedicated and often brilliant group of men
between traditional gender roles and new opportunities in computing. and women who were available to make use of the new tool. These people
One common experience for the Colossus and ENIAC women was understood the problems they had to solve. They studied and learned the
that neither group received public recognition for their work, either dur- characteristics of the new tool at their disposal, working around the clock
ing the war or for many decades afterward. At Bletchley Park, there was when necessary. \Tithout these skilled, dedicated people, ENIAC could
complete secrecy: not even the'$7rens' families found out what they had not have been the initial success that it was."83
been doing until the 1980s. At Penn, horvever, the ENIAC was given a Publicity materials prepared by the Moore school demonstrate how
showy debut in February 1946 that was rvell covered by the media. The human labor of both sexes was subtly erased. For example, one of the
centerpiece of the event was the ballistics program created by Snyder and "Basic Items for Publicity Concerning ENIAC' stated that ballistics cal-
Jennings, and the assembled reporters and VIPs were greatly impressed culations had been reduced from "25 man-months of expert computers'
that a trajectory could be calculated faster than the flight of the projectile rime" to two hours on the ENIAC.84 But those two hours included neither
38 Chapter 1

the "feminine" work of preparing programs nor the "masculine" work

of machine maintenance (which was substantial, as component failures Seeking the Perfect Programmer: Gender and
were frequent in the early years of computing). The only work that is
Skill in Early Data Processing
given visibility and value in the press reports is the initial machine design.
Of course, to exalt hardware design as the main site of skill, creativify,
and importance in computing is also to reinforce the status of what was
and remains a male-dominated area. Gender stereotypes and assumptions
about the nature of computers were inextricably entwined.

Conclusion
qualifications for
[Electronic data processing managers'] lack of definition of the
has yet been estab-
The ENIAC and Colossus stories illustrate a number of themes that run irogr"rnnl.r, is iemarkabie. No set standard of prerequisites
iirn.a i" most companies, and the selection of programmers is often based on the
through this book. They explain how and why women became involved
EDP manager's intuition.
in computing from the earliest days, hinting at the appeal and rewards
Russakoff Hoos, "'When the Computer Takes over the Office," 19601
that were offered by these new occupations. They also show how attribu- -Ida
tions of technical expertise and the status and authority that flow from
In the early decades of digital computing, programming was seen as a me-
them have tended to follow social categories-specifically, gender-rath-
thodical process based on scientific principles, yet the practice of hiring
er than necessarily reflecting the demands of the job. At Bletchley Park,
seemed just the opposite. From the 1950s into the 7970s,
the organization of labor reinforced gender stereotypes even when this
employers and industry pundits complained a bout the difficulty of iden-
detracted from the work itself. At the Moore School, assumptions about
talented applicants. The comments quoted above, written by busi-
the nature of computing-the primacy of the machine and relative in-
ness sociologist Ida Russakoff Hoos, scolded the managerial readers of
significance of the humans directing it-obscured the important role of
the Haruard Business Reuiew for their lack of personnel standards for
the programmer and influenced how work and credit would be divided
electronic data processing (EDP), as computing was commonly known in
befween women and men.
the business world. They were echoed by Elmer Kubie, who had cofound-
The pioneering women of Colossus anC ENIAC made the most of the
ed one of the world's first software firms in 1955. In a 1,963 interview,
opportunities that they were given, despite a culture of unequal expecta-
Kubie identified "the principal problem in the industry today" as "the
tions. Some of the obstacles that they faced have since disappeared. Yet
matching of talent with growth," explaining that "The field has grown so
gender as a set of beliefs, roles, and practices has continued to be a factor
rapidly that the problem of selection and training of personnel has been
shaping the computing professions, even though the meanings of mascu-
a rather haphazard process."2 Surveying hiring practices in the United
line and feminine have varied over time and across cultures. Likewise, the
Kingdom, a manager for Rolls-Royce grumbled that "There is not a great
cultural meaning and power of computers, programming, and computer
deal of evidence to suggest that the selection procedures currently in use
science have been constantly in flux, appropriated and contested by vari-
in the majority of organisations in the United Kingdom are particularly
ous actors. This uncertain identity has sometimes been a source of frustra-
effectiver" adding that many managers simply chose staff "on hunch'"3
tion for computer professionals, as the following chapters illustrate. Yet
the fluidity of these two terms, gender and cotnputing,has also left open
L 7969 survey of EDP trade schools-70O of which had sprung up in
the United States to meet the surging demand for personnel-voiced a
a space for women to create their own meanings.
, common complaint among schools that "industry won't spell out exactly
 182 No/es Notes 183

Chapter 1 ITar II Cryptoanalytic Center," Annals of the History of Computing 26, no. 2
(2004): 86-89.
1. Quoted in text accompanying exhibit on 'Wrens, Bletchley Park museum, 12. Tony Sale provides a lucid, illustrated explanation of the Lorenz machine at
viewed September 2001. http://www.codesandciphers.org. uk /virtualbp/fish/fish.htm.
2. Archives of the University of Pennsylvania, ENIAC Papers, box 12, ff 11, De- 13. Given the limitations of surviving records, I have been unable to determine
position of Jean J. Bartik, May 72, 1971, 32. the exact number of operators. I arrived at my estimate as follows: There were, at
3. I am not concerned here with priority claims, bur there is considerable de- the peak, twenfy women per shift (two at each of ten Colossus computers). There
bate in the historical literature about which machine deserves to be called the were three shifts per day, for a total of sixty women. Because there was some turn-
first digital computer. The verdict depends on which aspects of the machine are over of personnel, the total number of women who worked on Colossus over the
considered essential to the definition (for example, fully electronic rather than course of the war is probably somewhat higher. By April 1945, the total number
'Wrens
electromechanical, stored program or reading programs from external media, of working in Max Newman's section, which included the Colossus and
general purpose or specialized). Good overviews of the history of computing ma- Tunny operations and a tape registry, was 273. Good' Michie, and Timms, "Gen-
chines include Paul E. Ceruzzi, A History of Modern Computing (Cambridge: eral Report onTunny," 276.
MIT Press, 7998), and Martin Campbell-Kelly and William Aspray, Computer: A 14. Thomas H, Flowers, "D-Day at Bletchley Park," in Colossus: The Secrets of
History of the Information Machine, Sloan Technology Series (New Yorkr Basic Bletchley Park's Code-Breahing Computers, ed. B. Jack Copeland (Oxford: Ox-
Books,1996). ford University Press,2005), 81. An assessment of the overall contribution of the
4. David Alan Grier, "The Math Tables Project of the \fork Projects Administra- British codebreaking activities can be found in F. H. Hinsleg "The Influence of
tion: The Reluctant Start of the ComputingEra," Annals of the History of Com- Ultra in the Second'$forld War," in Codebreakers: The Inside Story of Bletchley
puting 20, no. 3 (1998 ): 33-50; David Alan Grier,When Computers'Were Human Park,ed. F. H. Hinsley and Alan Stripp (Oxford: Oxford University Press, 1.993).
(Princeton: Princeton University Press, 2005); Mary Croarken, "Mary Edwards: 15. "Ode to the Eniac," Chicago Daily Tribune, February '18,1946,72.
Computing for a Living in Eighteenth-Century England," Annals of the History 15. Campbell-Kelly and Aspray, Computer: A History of the Information Ma-
of Computing 25, no. 4 (2003):9-15. chine, Sl-83.
5. Campbell-Kelly and AspraS Computer: A Historl.t of tbe Information Ma- 17. Surnames in parentheses are later married names. For consistencSI refer to
chine, 47-5'1.. the women throughout by the maiden names they had during the project.
6. rbid. 18. Herman H. Goldstine, The Computer from Pascal to Von Neumann (Princ-
7. For Konrad Zuse, seeJ. A. N. Lee's history of computing site, http://ei.cs eton: Princeton University Press, 1972), 234-235.
.vt.edu/-historylZuse.html, which includes the transcript of a talk by Zuse on 19. Quoted in uü. Barkley Fritz, "The'Women of ENIAC," Annals of the History
computer design, and the site created by computer scientist Horst Zuse (Konrad's of Computing 18, no. 3 (1996\: 75.
son) at http://user.cs.tu-berlin.del-zuse/konrad-zuse.html.
20. David M. Kennedy, Freedom from Fear: The American People in Depression
8. See Raul Rojas and Ulf Hashagen, eds.,The First Computers: History and and'War 1929-1945 (New York: Oxford University Press,1.9991,778.
Architectures (Cambridge: MIT Press, 2002), f.or more information on the first-
21. I am indebted to Amy Slaton for insights on this point. For the racial division
generation digital computers.
of technical labor, see Amy Slaton, Race, Rigor, and Selectiuity in U.S. Engineer'
9. The origins of the stored-program concepr are the subject of continuing de- ing: Tbe History of an Occupational Color Line (Cambridge: Harvard Universiry
bate. See rü(/illiam AspraS "The Stored Program Concept," IEEE Spectrum 27 Press,2010).
(1,990):51, for an overview. 'Were 'Women,"
22. Jenniler S. Light, "'When Computers Technology and Cuhure
10. Jack Good, Donald Michie, and Geoffrey Timms, "General Report on Tun- 40, no. 3 (t999\ 461.
ny," Bletchley Park, 1945. The Tunny report, which describes Colossus and other
codebreaking machines at Bletchley Park, was ciassified until 2000. It is now
23. Quoted in Doris Kearns Goodwin, No Ordinary Time: Franklin and Elea'
available online at the Alan Turing archive, http://www.alanturing.net/turing_
nor Rooseuelt: The Home Front in World'War II (New York: Simon & Schuster,

archive/archive/index/tunnyreportindex. html.
1,994),474.

11. Parallel codebreaking activities in the United States included a navy-run oper- 24. Fritz, "The'Women of ENIAC," L4.
ation at Arlington Hall near \üTashington, D.C., that focused on cracking Japanese 25. Quoted in \üü. Barkley Fritz, "ENIAC: A Problem Solver," Annals of the His-
encryption. As at Bletchley Park, women played a key role operaring codebreak- tory of Computing 1.6,no.7 (1,994):28.
ing "bombes." Laurie Robertson, "Arlington Hall Station: The U.S. Army's l7orld
 184 Notes Notes 185

25. Duringmy 2001 interview with Jean Jennings Bartik, she showed me the pro- Material (date approx. 1'940-1950), folder "DAF, Early Computer Aberdeen
motion letter that officially changed her job title from computer (subprofessional Course," "Course in Mathematics for Ballistics Computations."
grade 6) to mathematician (professional grade 2) in January 1"946. Ensmenger 40. Quoted in Fritz, "The Women of ENIAC," 23.
states that the ENIAC women were called coders,but this seems to be based solely
41. Goldstine, interview with Jennifer Light, t994, quoted in Light, "'When Com-
on a publication by Goldstine and von Neumann and does not match the usage
puters'Were Women," 469.
that I found in other published sources or interviews. Nathan Ensmenger, The
Computer Boys Take Ouer: Computers, Programnters, and the Politics of Techni- 42. Fritz,"The Women of ENIAC," 19.
cal Expertise (Cambridge: MIT Press, 2010), 35. 43. Ensmenger calls the ENIAC programmers "glorified clerical workers." Ens-
27. There were a few female mathematicians at Bletchley Park, but apparently menger, The Computer Boys Take Ouer,35; see also Light, "When Computers
'Were'!(/omen."
none was assigned to the Colossus area, which was a relatively small part of the
mathematicians' work. 44. John R. Mauchlg "Interview by Henry Tropp," January 1'0,1973, Computer
28. B. Jack Copeland et al.,"Mr. Newman's Section," in Colossus: The Secrets of Oral History Collection, Smithsonian Institution National Museum of American
Bletchley Park's Code-Breaking Computers, ed. B. Jack Copeland (Oxford: Ox- History, #796,box 13,f13, n.p. (7th page from end of ms.).
ford University Press, 20061, 159 . About 4 percent of British men received college 45. Good, Michie, and Timms, "General Report on Tunny," 334.
degrees in 1950, the first year after 1,939 for which statistics are available. Central
45. Copeland et al., "Mr. Newman's Section," 763-I64.Ireland pointed out that
Statistical Office (UK), Annual Abstract of Statistics, vol. 97 (London: Her Maj- when a Colossus was rebuilt in 1,996 for the new Bletchley Park museum, the
estyt Stationery Office, 1960). male engineers discovered just how tricky it was to thread the tapes: "It took
29. My description of the experiences of Bletchley Park'Sürens draws on Gwen- them three hours to get the tape right, and we all thought that was very amusing!"
doline Page, ed., We Kept tbe Secret: Noaa It Can Be Told-Some Memories of Eleanor Ireland, interview by Janet Abbate, April 23, 2001.
Pembroke V'Wrens (Vymondham, UK: Reeve, 2002); lvlarionHill, Btetchley Park 47. tbid.,763.
People: Churchill's " Geese That Neuer Cackled" (Stroud, UK: Sutton, 2OO4); and
B. Jack Copeland, "Colossus and the Rise of the Modern Computer," in Colossus:
48. Flowers, "D-Day at Bletchley Park," 98.
The Secrets of Bletchlq Park's Code-Breaking Compuiers, ed. B. Jack Copeland 49. Copeland et al., "Mr. Newman's Section," 164.
(Oxford: Oxford University Press, 2006), 101-115. 50. Ibid., 163,
30. Eleanor lreland, interview by Janet Abbate, April 23, 2001. The compurer 51. Copeland, "Colossus and the Rise of the Modern Computer," 1.59,164-765.
that Ireland saw was presumably the Heath Robinson, a protorype of the Colos-
52. Good, Michie, and Timms, "General Report on Tunng" 278.
sus that had two tape drives instead of one. In the Colossus, the second tape drive
was replaced by a set of vacuum tubes. The Heath Robinson was installed at 53. Ibid., 279. "Chi-breaking" meant discovering the settings of the first five
Bletchley Park in June 1943, and the Colossus in December 1943. wheels on the Lorenz machine.

31. James K. Lively, "Technical History of 6813th Signals Security Detachment," 54. Livelg "Technical History of 6813th Signals Security Detachment," 20.
1,945, 1 5 . 55. Adele K. Goldstine, "A Report on the ENIAC (Electronic Numerical Inte-
32. Copeland, "Colossus and the Rise of the Modern Computer," 156. grator and Computer)," University of Philadelphia, Moore School of Electrical
Engineering, Philadelphia, 1 945).
33. Eleanor Ireland, interview by Ianet Abbate, April 23, 2001. Thanks to an
anonymous reviewer for suggesting the parallel with housework. 55. Later, both Herman and Adele Goldstine claimed credit for preparing the
ENIAC demonstration program. Light, "'$fhen Computers'Were'Women," 472,
34. Quoted in Fritz, "The'Women of ENIAC," 22.The other programmer with-
478. But the programmers themselves denied that the Goldstines had either
out a math degree was Betty Snyder.
taught them programming or created the program in question. See Archives of the
35. Jean Jennings Bartik, interview by Janet Abbate, August 3,Z0OI. University of Pennsylvania, ENIAC Papers, box 15, ff 5, "Deposition of Kathleen
36. Quoted in Fritz, "The !7omen of ENIAC," 20. Rita Mauchly," ca. 7977, 322-324; and Jean Jennings Bartik's description of pre-
37. Good, Michie, and Timms, "General Report on Tunny,,'278. paring the program in Fritz, "The'Women of ENIAC."

38. Eleanor Ireland, interview by Janet Abbate, April 23, 2001; Copeland et al., 57 . Jean Jennings Bartik, interview by Janet Abbate, August 3,201't.
"Mr. Newman's Secrion," 762; Good, Michie, and Timms, ..General Report on 58. "Deposition of Kathleen Rita Mauchly," 322.
Tunnyr" 278. 59. Archives of the University of Pennsylvania, ENIAC Papers, box 16, ff 3, "De-
39. Archives of the university of Pennsylvania, UPD 8.4, Moore school of Erec- position of Ruth L. Teitelbaum," ca.797L,54-55.
trical Engineering, Office of the Director Records, 1931-7948,box 46: ENIAC
 186 Notes Notes 187

60. Quoted in Fritz, "The Women of ENIAC," 20. construct a stored-program computer; it was at Bletchley that they learned how
61. Archives of the University of Pennsylvania, ENIAC Papers, box 14, ff 7,,,De- it might be done. Large-scale electronics was the secret." Copeland, "Colossus
position of Frances D. Holberton," 1971,70-71: also "Deposition of Kathleen and the Rise of the Modern Computer," 105. Copeland claims that Newman
Rita Mauchly," 144. (and later Turing) taught engineers Tom Kilburn and F. C. \üilliams, who built the
Manchester machines, how to design an electronic computer (112).
52. Quoted in Fritz, "The S7omen of ENIAC," 20.
76. Fritz, "ENIAC: A Problem Solver," mentions Gloria Gordon Bolotsky, Lila
63. Fritz, "ENIAC: A Problem Solver," 25;Fritz, "The'Süomen of ENIAC," 14.
Todd Butler, Ester Gersten, l7inifred (I7ink) Smith Jonas, Marie Malone, Helen
64. Archives of the University of Pennsylvania, ENIAC Papers, box 72, ff 1!, Mark, and Hom6 McAllister Reitwiesner (30).
"Deposition of fean J. Bartik," ca. 7977,34.
77. McNulty obituarS May 4, 2006, http:llwww.chestnuthilllocal.com/is-
65. Quoted in Fritz, "The rü(/omen of ENIAC," 21. BRL is the Ballistics Research sues/2006.05.O4/obituaries.html; Fritz, "The'Women of ENIAC"; Jean Jennings
Laboratory at Aberdeen. Bartik, interview by Janet Abbate, August 3,207t.
66. Quoted in Fritz, "The Iü(/omen of ENIAC," 21. 78. C. Dianne Martin, "ENIAC: Press Conference That Shook the Vorld," IEEE
67. Ruth Milkman, Gender at'l(/ork: The Dynamics cf Job Segregation by Sex Technology and Society Magazine 14, no.4 (1995): 3-10.
during'world-war II (chicago: university of Illinois Press, 1987). Alice Kessler- 79. "Deposition of Jean f. Bartik," 65; "Deposition of Frances D. Holbenon," 81;
Harris (7982) nores that American women's postwar employment level was not Kathy Kleiman,The Computers:The Untold Story of the RemarkableWomen'Who
significantly higher than it would have been without the war (277-278).penny P r o grammed tb e EN I AC (Documentary Preview), 20 0 7, http:l I eniacprogrammers
Summerfield, 'women workers in the second world war: Production and patri- .org. On the other hand, McNulty recalled being invited to the press lunch. "De-
archy in Conflict (London; Routledge, 1989), argues that in the United Kingdom, position of Kathleen Rita Mauchly," 707-702. The lack of public acclaim did not
"the implementation of official policy [women's mobilization] during the war mean that the programmers went completely unrecognized, For example, Jean Jen-
did little to alter but rather reinforced the unequal position of women in society" nings and Adele Goldstine created a program ln fall 1.946 to model the reflection
(185).
of shock waves for A. H. Täub of the University of rü(/ashington, and Jennings noted
58. Eleanor Ireland, interview by Janet Abbate, April 23, 2001. that "I did receive a letter of commendation for the significant help we had given."
69. Grier, Vhen Computers Were Human,288; Light. .,When Computers rü7ere "Deposition of Jean J. Bartik," 49; see also Fritz, "The'Women of ENIAC," 21.
'Women," 479. 80. See Mitchell Marcus and Atsushi Akera, "Exploring the Architecture of an
70. "Course in Mathematics for Ballistics Computations." Early Machine: The Historical Relevance of the ENIAC Machine Architecture,"
Annals of th e History of Computing 1 8, no. 1 (199 6) z t7 -24; Fritz, "The'Women
71., Eleanor Ireland suggested rhat some of the'sfrens who worked on colossus
of ENIAC." Kathy Kleiman wrote one of the first papers on the ENIAC women
went on to work with computers for the British navy in ceylon, but I have been
in 1986 as an undergraduate in sociology. She later shot documentary footage of
unable to verify this.
many of the programmers, although a finished documentary has not been pro-
72. Two colossus machines were kept for the use of Government communica- duced as of this writing. See http://eniacprogrammers.org.
tions Headquarters (the newly renamed codebreaking agency). They were actively
81. Light, "Whin Computers Were'$ü'omen," 474.
used for fifteen more years until 1960, and their exact use is still classified. cope-
land, "Colossus and the Rise of the Modern Computer," 173. 82. Martin, "ENIAC: Press Conference That Shook the'S7orld."
73. For a description of the colossus rebuild, see http://www.codesandciphers 83. Fritz, "ENIAC: A Problem Solver," 31.
.org.uk/lorenz/rebuild. htm. 84. Archives of the University of Pennsylvania, UPD 8,4, Moore School of Elec-
74. Harry Fensom, "How Colossus Vas Built ani Operated,', in Colossus: The trical Engineering, Office of the Director Records, 1937-7948, box 48, folder
secrets of Bletchley Park's code-Breaking computers,ed. B. Jack copeland (ox- "PX - Publiciry - ENIAC," "Basic Items for Publicity Concerning ENIAC," De-
ford: Oxford University Press,2006), 302. cember 27,7945.
75, Copeland, "Colossus and the Rise of the Modern Compurer,', 10g, 1g6. Co-
peland argues that colossus was the "missing link" that showed Turing and New- Chapter 2
man how to build a stored-program compurer: "In 1936 [the year ihat Turing
wrote on computable Numbersl the universal Turing machine existed only as an 1. Ida Russakoff Hoos, "\ühen the Computer Takes over the Office," Haruard
idea. Right from the start Turing was interested in the possibility of building such Business Reuiew 38 (1950): 1.04. Hoos worked for the Berkeley Insritute for In-
a machine, as to some extent was Newman. But it was not until their Bletchley dustrial Relations.
Park days that the dream of building a miraculously fast ail-purpose compuring
machine took hold of them. Before the war rhey knew of no practical way to
History of Computing Recoding Gender
Villiam Aspray, editor
'Women's Changing Participation in Computing
.Janet Abbate, Recoding Gender: Women's Changing Participation in Computing
John Agar, The Gouernment Machine: A Reuolutionary History of tbe Computer
'William
Aspray and Paul E. Ceruzzi, The Internet anC American Business
'William
Aspray,.lohn uon Neumann and the Origins of Modern Computing
Charles J. Bashe, Lyle R. Johnson,.fohn H. Palmer, and Emerson lf. Pugh, lBMt
Ear\,Computers
Martin Campbell-Kelly,From Airline Reseruations to Sonic the Hedgehog: A His-
tory of the Softutare Industry
Janet Abbate
Paul E. Ceruzzi, A History of Modern Computing
I. Bernard Cohen, Howard Aiken: Portrait of a Computer Pioneer
I. Bernard Cohen and Gregory rW. \üelch, editors, Makin' Numbers: Howard Ai-
ken and tbe Computer
John Hendry, lnnouating for Failure: Gouernment Pclicy and tbe Early British
Computer Industry
Michael Lindgren, Glory and Failure: The Difference Engines of Johann Müller,
Charles Babbage, and Georg and Eduard Scbeutz
David E. Lundstrom, A Few Good Men from Uniuac
Ren6 Moreau, The Computer Comes of Age: The People, the Hardtuare, and the
Softu.,are
Arthur L. Norberg, Computers and Commerce: A Study oiTechnology and Man-
agement at Eckert-Mauchly Computer Company, Engineering Research Associ-
ates, and Remington Rand, 1945-19.17
Emerson W. Pugh, Building IBM: Shaping an Industry and lts Technology
Emerson V. Pugh, Memories That Shaped an Industry
Emerson rJf. Pugh, Lyle R. Johnson, and John H. Palmer, IBM's 360 and Early
370 Systems
Kent C. Redmond and Thomas M. Smith, From Whirlt,ind to MITRE: The RüD
Story of the SAGE Air Defense Computer
Alex Roland with Philip Shiman, Strategic Computing: DARPA and the Quest for
Mach ine Intelligence, 1 9 8 3-1 9 9 3
Raril Rojas and Ulf Hashagen, editors, The First Computers: History and
Architectures
Dorothy Stein, Ada: A Life and a Legacy
John Vardalas , The Computer Reuolution in Canada: Buil"ding National Techno-
Io gical Contp etence, 1 94.1 -19 8 0
The MIT Press
Maurice V. \filkes, Memoirs of a Computer Pioneer
Cambridge, Massachusetts
London, England
First MIT Press new paperck edition,20'!7 For Tom and'Wendy
@ 2072 Massachusetts Instirute of Technology
All rights reserved. No part of this book may be reproduced in any form by any electronic
or mechanical means (including photocopying, recording, or information storage and re-
trieval) without permission in writing from the publishe:

This book was set in sabon by the MIT Press. Printed and bound in the united States of
America.

Lifuary of Congress Cataloging-in-Publication Data

Abbate, Janet.
Recoding gender : women's changing participation in computing / Janet Abbate.
p. cm,- (History of computing)
Includes bibliographical references and indei.'
ISBN 978-0-252-07806-7 (hardcover : alk. paper), 97 84-262-5 3453-6 (pb.)
1. rffomen in computer science. 2. Computer industry. L Title.
QA76.9.I[65A33 2012 '..,. . er-
004.082-dc23
20L200989I ' .*,. .. ; . '
'i"
i0987654
Contents

Acknowledgments ix

Introduction: Rediscovering Women's History in Computing 1

1 Breaking codes and Finding Tiaiectories: women at the Dawn of

the Digital Age 1l
2 Seeking the Perfect Programmer: Gender and skill in Early Data Pro-
cessing 39

3 Software crisis or Identity crisis? Gender, Labor, and Programming

Methods 73

4 Female Entrepreneurs: Reimagining Software as a Business 113

5 Gender in Academic Computing; Alternative Career Paths and

Norms 145
Appendix: Oral History Interviews Conducted for This Project 777
Notes 1.79
Bibliography 225
Index 243