CHAPTER 8

"Hilarity Ensues"

p. 128

The program owes most to Diffie: PowerPoint has traditionally been attributed to Robert Gaskins and Dennis Austin Forethought Inc., before being bought by Microsoft. But in a New Yorker feature from May 28, 2001, Ian Parker reveals that it was in fact Diffie. This was like discovering "that Lenin had invented the stapler," he memorably noted.

129

a charge levelled by the French writer Franck Frommer: In How PowerPoint Makes You Stupid.

132

A similar convenience hack caused AOL's servers to fail: See Computer World, "AOL email hit by outage." I note that some countries, like Ethiopia, have their own calendars. The Ethiopian new year begins (believe this if you will) on the Gregorian calendar equivalent of September 11—except in the year before a leap year, when New Year's Day falls on September 12.

134

"And we haven't even left Australia yet": Here are those Australian time zones.

139

In 1984, women accounted for almost 40 percent of CS degrees: New York Times "The Secret History of Women in Coding."

Bureau of Labor Statistics figures from 1987: Figures from the Bureau of Labor Statistics, quoted in Caroline Clarke Hayes' essay, "Computer Science: The Incredible Shrinking Woman," in Misa, ibid (p. 33).

there would have been no women programmers by 2020: This extrapolation based on figures from 2015 and surrounding Stack Overflow surveys.

140

in 2013 it felt more like they were threatened and wanted to keep you out: Diane Chen is not an outlier here. My laptop contains a folder stuffed with articles and blog posts by women recounting everything from uncomfortable experiences to dispiriting exchanges; from subtle marginalization to deliberate exclusion and outright hostility, belittlement, patronization, unwanted sexual attentions and clear career discrimination, mostly recounted with far more grace and forbearance than I think I would manage in their positions.

141

An emergent literature has begun to unearth stories like those of the ENIAC women: For more examples see "The Secret History of Women in Coding," New York Times, ibid; Janet Abate's book Recoding Gender: Women's Changing Participation in Computing.

saved the first moon landing as a direct result of being forced to take her daughter to work: How so? Hamilton was Director of Apollo Flight Computer Programming at MIT during the lunar years and is often credited with having coined the term "software engineering" in an effort to get her profession taken more seriously by NASA managers. At the end of 1968, with a first landing on the horizon, she was 32 and often forced to take daughter Lauren to work on evenings and at weekends while her team toiled all hours on software to control the spaceships. Poor girl, we might think, before learning that as compensation the waiting child was allowed to play on the Apollo flight simulator, making her the only human to have learned to land a Lunar Module before she could drive a car.

Then one day Lauren crashed a simulated landing in mysterious circumstances. Her alarmed mother traced the problem to a mistakenly activated pre-launch program called "P01," which sprang to life mid-flight and gatecrashed the memory, overloading and then paralyzing the computer through a condition called executive overflow. Hamilton raised the issue with managers, who thought grownup astronauts too well trained to make the mistake Lauren had—until the crew of Apollo 8, the first lunar circumnavigatory flight, proved them wrong. Caution validated, Hamilton was allowed to insert a "BAILOUT" routine designed to warn astronauts when the Apollo Guidance Computer was at risk of overflow, restarting essential tasks and scheduling less important ones to run when memory became free.

The most dramatic moment of the first moon landing duly came when, at 30,000 feet and with the "Eagle" lunar lander descending rapidly, an alarm sounded in the lander's cabin and Buzz Aldrin saw the now infamous "1202" code flash onto his computer screen, which he monitored while Armstrong stood at the flight controls. For years non-experts assumed this to have been a bug in the software, but Margaret Hamilton has spent much of her life explaining that it was intentional—and probably saved the first landing. For chapter and verse here, see my book Moondust: In Search of the Men Who Fell to Earth or Andy Chaikin's superb A Man on the Moon. This was no small feat because the Apollo Guidance Computer had less computing power than a recordable greeting card today (see Ellen Stofan of the National Air and Space Museum, writing in Slate: "The Apollo Guidance Computer was a marvel: As Poppy Northcutt, who calculated Apollo's return-to-Earth trajectories, told me, the AGC had less computing power than the greeting cards today that record a personal message. Yet it worked.") Poppy Northcutt, incidentally, was the only woman in NASA's flight control room during the first landing.

As a demonstration of how easy it is to get this stuff wrong, with sometimes catastrophic consequences, G. J. Meyers' book Software Reliability: Principles and Practices offers the tale of how NASA lost their first Mariner space probe in 1962 to a comma accidently typed as a period in the guidance program, turning what should have been a loop into a variable assignment. By this account (quoted in Ensmenger, ibid), the FORTRAN statement "DO 3 1 = 1,3" became "DO 3 1 = 1.3". Hilarity did not ensue. Mariner 2 launched successfully not long after. More recent reporting suggests that the culprit may have been a hyphen, as reported contemporaneously in the New York Times, July 27, 1962, and quoted in a 1987 thread on Risks Digest, Vol. 5, Issue #66. Lovers of detailed computing arcana with love this. The 2001: A Space Odyssey author Arthur C. Clarke dubbed this particular shift devil "the most expensive hyphen in history" (in Arthur C. Clarke, The Promise of Space, 1968.)

Suddenly programmers became important: See Ensmenger, ibid, for more on this.

paternalistic mandarins chose to classify the work of pioneers like Alan Turing: from Marie Hicks, Programmed Inequality: How Britain Discarded Women Technologists and Lost Its Edge in Computing: "The British case is different: in Britain, gendered labor change was part of a topdown government initiative to computerize. Not only women but also the explicit structural discrimination against them played a crucial, formative role in the uptake of computers and in the ultimate failure of the British computing industry."

a small study of professionals: Walter McNamara and John Hughes studied 57 professional programmers with an average age of 25.

142

the entire population of Sing Sing prison in New York was offered the PAT test: Ensmenger, ibid (p. 72-3)

"10X" lore starts here: at H. Sackman, W. J. Erikson and E. E. Grant, Exploratory, Experimental Studies Comparing Online and Offline Programming Performance, System Development Corporation, Santa Monica, California, 1968

female membership of the ACM to collapse from 40 to 10 percent within a few years: Ensmenger, ibid, p239

PAT performance was no better a gauge of coding potential than eye color: Ensmenger, ibid. Also see Lawrence J. Mazlack, "Identifying Potential to Acquire Programming Skill," Communications of the ACM, Vol. 23, No. 1 (Jan 1980), p14-17.

143

Brandon and others insisted that PAT favored mathematically inclined males by default: Ample anecdotal evidence since suggests that people naturally inclined to the arts and humanities can make just as good programmers. The C programmer and veteran embedded software expert Jack Ganssle, author of The Embedded Muse blog, claims to have had good experiences with training English Lit grads, noting their facility with syntax and ability to express themselves clearly and concisely.

Among the sternest critics of the selection mechanisms was...author Gerald Weinberg: As quoted in Ensmenger, ibid, p213-4.

papers with titles like "Vocational Interests of Computer Programmers": Journal of Applied Psychology 51, vol. 1, 1967, "Vocational Interests of Computer Programmers"

Richard Brandon characterized the coding breed as "often egocentric, slightly neurotic" and "excessively independent": See Richard Brandon, "The Problem in Perspective," Proceedings of the 1968 23rd ACM National Conference. The respected computing industry analyst elucidates—a little hilariously—that:

The incidence of beards, sandals, and other symptoms of rugged individualism or non-conformity are notably greater among this demographic group. Stories about programmers and their attitudes and peculiarities are legion, and do not bear repeating here.

And there's more. The author-journalist Stephen Levy offers more detail in his classic study of code culture, Hackers, with special reference to key members of the now legendary MIT hacker cohort. Most, he avers, "had locked themselves into what would be called 'bachelor mode.'"

It was easy to fall into," he explains. "For one thing, many of the hackers were loners to begin with, socially uncomfortable. It was the predictability and controllability of a computer system—as opposed to the hopelessly random problems in a human relationship—which made hacking particularly attractive. But an even weightier factor was the hackers' impression that computing was much more important than getting involved in a romantic relationship. It was a question of priorities. Hacking had replaced sex in their lives...You would hack, and you would live by the Hacker Ethic, and you knew that horribly inefficient and wasteful things like women burned too many cycles, occupied too much memory space. 'Women, even today, are considered grossly unpredictable,' one PDP-6 hacker noted, almost two decades later. 'How can a hacker tolerate such an imperfect being?'

a gender-biased feedback cycle: In The Computer Boys Take Over (p. 78-9) Ensmenger makes a convincing case that the personality quirks of the early coders, who started out working in Machine Code and Assembly Language, were institutionalized by aptitude and personality tests that deliberately screened for people like them.

It is almost certainly the case that these profiles represented, at best, deeply flawed scientific methodology," the historian notes. "But they almost equally certainly created a gender-biased feedback cycle that ultimately selected for programmers with stereotypical masculine characteristics. The primary selection mechanism used by the industry selected for antisocial, mathematically inclined males, and therefore antisocial, mathematically inclined males were overrepresented in the programmer population; this in turn reinforced the popular perception that programmers ought to be antisocial and mathematically inclined (and therefore male), and so on ad infinitum. Combined with the often-explicit association of programming personnel with beards, sandals, and scruffiness, it is no wonder that women felt increasingly excluded from the center of the computing community.

144

managerial elegies to the unmanageability of programmers: This perception mattered too. While some of the tensions were trivial, others were not. In the September 1995 issue of Wired magazine, Nathan Myhrvold, a former chief technologist at Microsoft, revealed that, even in the early nineties, traditional corporate managements could be clueless about what coding involved.

There was a hilarious phase during our relationship with IBM," the CTO chuckled, "when we ran into a lot of difficulty because the key metric for programmer productivity at IBM was the number of lines of code produced. Our people would go in and reduce the number of lines of code, generating negative productivity in IBM's eyes. Bill Gates used to call this 'The race to build the world's heaviest airplane.'

from the early eighties to nineties Ada became the new field's rising star: And is still widely used in defense and safety-critical commercial applications, notably aerospace, where the Boeing 777 and Canadian and British air traffic control systems make use of it, as do French TGV high-speed trains and metro systems in New York, London, Paris and Hong Kong. For more detail on the rise and fall of languages, there is a wonderful, fluid graphic showing the rise and fall of languages on YouTube, called "Most Popular Programming Languages 1965 — 2022"

Apple Computers' 1984 Superbowl ad for the first Macintosh: See this on YouTube.

managerial elegies to the unmanageability of programmers: for detail see Ensmenger, ibid.

four hundred-plus languages involved in its sprawling array of Cold War projects: Wikipedia is good on this.

145

The word "algorithm" didn't appear in Webster's New World Dictionary until 1957: Knuth, The Art of Computer Programming: vol.1, Fundamental Algorithms, page 1. The estimable Merriam-Webster cites the first recorded use of the word as occurring in 1699.

Debate about the real-world utility of Curriculum '68: From Ensmenger, ibid (p. 133): "Not everyone agreed with the theoretical turn that computer science took in the late 1960s. For many occupational computer programmers, most of what was happening in theoretical computer science seemed irrelevant or even counter-productive, a 'sort of holier-than-thou intellectual sort of enterprise' divorced from practical concerns of commercial computing."

Dispute over the proper relationship between programming and math would never end: See Nathan Ensmenger, "The Question of Professionalism," EEE Annals of the History of Computing, 4(23):56-73, 2001

claiming a lineage to the Persian mathematician al-Khwarizmi (Latinized to Algoritmi): In his work A History of Modern Computing, the computer historian Paul Ceruzzi notes that "The objective study of algorithms as the fundamental unit of computing, on a par with the laws of gravity in physics or elements in chemistry, is a new idea." I'll be honest here in saying that I adore Donald Knuth. He is a stylish writer, full of play and humor; of care and empathy for his reader. Over time I will learn to appreciate his view that coding "can be an aesthetic experience much like composing poetry." He wrote a delightful novelette about numbers and a lovely, quirky book about the relationship between computing and his personal understanding of God (Things a Computer Scientist Rarely Talks About.) So it seems clear to me that Knuth has the soul of Aretha Franklin. However. My sense is that The Art of Computer Programming is the tech equivalent of Stephen Hawking's A Brief History of Time. I get pleasure from seeing it on my shelf, from knowing someone wrote it and that I would be a better person for finishing it, while suspecting that I won't. The truth is that I had fun with the first two chapters—which are highly mathematical—before putting it down, unable to see how it would help me be a better programmer in any direct way. I assumed this impression would change as my knowledge grew, until I mentioned my experience to various professionals, including a couple of rarified system administrators, only to be told they had done the same. This doesn't invalidate the work or make it uninteresting (the same could be said of works by James Joyce or David Foster-Wallace), but does make one wonder whether it ever made sense as a basis for the study of "computer science."

the displacement of women programmers in the United States was an implicit effect of the fight for professional status: Ensmenger elaborates that "This bias toward male programmers was not so much deliberate as it was convenient—a combination of laziness, ambiguity, and traditional male privilege. The fact that the use of lazy screening practices inadvertently excluded large numbers of potential female trainees was simply never considered. But the increasing assumption that the average programmer was also male did play a key role in the establishment of a highly masculine programming subculture."

there was a deliberate expulsion of women from programming in Britain: According to Hicks, this involved the creation of a new work grade for programmers, to which feminized lower grades were specifically denied access. She constructs a compelling case that, where the UK led the world in computing at the end of World War II, the paternalism of the country's technocratic mandarins had all but killed the domestic industry by the mid 1970s. In contrast to the American experience, there are few individual women's names we can point to in early British computing, because no one was acknowledging or keeping records of what they did. As the author notes: "Most women in this study did not make major contributions as individuals, but they were important as a class of workers on whose shoulders was laid incredible technological responsibility with little corresponding economic or social status." Hicks also sounds a warning for the US. "The experience of Britain in the 20th century has many similarities to the US context in the 21st century . . . The failure of Britain's thriving midcentury computer industry serves as an unhappy reminder of the ways in which technologies can rarely fix social or economic problems and how they instead often make real the limited and myopic goals of small but powerful segments of society." See Marie Hicks, Programmed Inequality: How Britain Discarded Women Technologists and Lost Its Edge in Computing. Also Gender Codes: Why Women are Leaving Computing, ed Thomas J. Misa (p. 16)

146

female undergraduate enrollment in his department had fallen to just 8 percent: Unlocking the Clubhouse: Women in Computing, Margolis & Fisher (p. 7)

147

most women drew motivation and meaning from the purpose that computing was going to be used for: More specifically, she said "It wasn't just hacking for hacking's sake. There was a real social context that gave them motivation and meaning" (as quoted in Scott Carlson, "Wanted: Female Computer-Science Students," Chronicle of Higher Education, Vol 52, No. 19, 13 January 2006.)

the most interesting electronic music of that time, including first wave New York hip-hop, was made with technology on the brink of obsolescence: For evidence of what I mean here, check out the Street Sounds compilations released by Street Sounds Records in the early-mid-eighties, or the fabulous Minimal Wave Tapes collections issued by Stones Throw Records in the 2010s.

148

Victorian intellectuals denied women access to university based on pseudo-medical claims that it would disrupt menstruation: As detailed in Francesca Wade's much-admired book Square Haunting, on the lives of five women writers in London's Mecklenburgh Square in Bloomsbury between the wars. Not only did male intellectuals fear the demise of menstruation and expansion of licentiousness: spinster teachers, it seemed, might peddle "oblique and distorted conceptions of love," with women's widespread employment further presaging "an apocalyptic war between the sexes, culminating in the ultimate extinction of the race." About which it would have been hard to complain, on the evidence.

James Damore . . . claiming women have evolved to be inferior at tasks like programming: For the record, this is demonstrably wrong at a neurologic level. As Professor Gina Rippon of Aston University in the UK notes, "We have yet to discover any kind of structural difference which will reliably distinguish the brains of men from the brains of women." See also "Brain scientists haven't been able to find major differences between women's and men's brains, despite over a century of searching" by Ari Berkowitz, Presidential Professor of Biology and Director of the Cellular & Behavioral Neurobiology Graduate Program at the University of Oklahoma.

150

women's work being accepted more often than men's when gender is not specified: "Gender differences and bias in open source: pull request acceptance of women versus men," May 1, 2017.

men skewing overconfident and women underconfident: Tons of research to choose from here, but a reasonable overview is given in "The Confidence Gap In Men And Women: Why It Matters And How To Overcome It," Forbes, April 8, 2018.

151

Anna Kiefer is now a full-stack programmer: Since we met, Anna has worked at a number of Bay Area companies, gained a Master's in Computer Science at California Polytechnic State University, and at time of going to press is a Senior Software Engineer at Orbital Insight, a geospatial analytics company based in Palo Alto. So her story turned out just fine.

152

IBM made a sustained—and successful—effort to bring more women and minorities into computing: McIlwain, Black Software, ibid (p. 38).

an industry in which 7 percent of coders are female and under 3 percent are Black: The U.S. Bureau of Labor Statistics ups these percentages in all cases, but from a tiny and unreliable sample. There is no way to know the precise composition of the coding population, but nearly all coders use and are committed to Stack Overflow, so it seems likely to provide as accurate a snapshot as we can get. Significantly, the SO survey also tallies with subjective experience and impressions. A little shockingly, the 2023 and 2024 surveys dropped all demographic categories apart from age—anf it's hard not to suspect this is because the fuller demographic picture was embarrassing to the industry.

close to a third have been on the job less than two years (etc): Figures taken from Stack Overflow surveys 2018-21. The most recent surveys available as I write here suggest a slow broadening on these fronts. The 2023 survey suggests 75% of pro coders to have fewer than 15 years' experience on the job, where 2024 gives the equivalent figure as 71%. In 2024, 60% of professional respondents were under 35 years old. It looks like people are sticking with the profession and growing into it more than they did. These more recent surveys dropped most of the demographic questions, including the one about dependents.

the relationship between race and technology: Just as the monumental documentary The Summer of Soul recently—and joyfully—made clear that while mostly white kids were grooving at Woodstock in 1969, the predominantly Black residents of Harlem in New York were doing likewise at the "Black Woodstock", the Harlem Cultural Festival (and anyone with even a passing interest in music will be glad to watch this, it's fantastic,) so recent books like Charlton D. McIlwain's Black Software: The Internet and Racial Justice, from the AfroNet to Black Lives Matter and Ruha Benjamin's Race After Technology have begun to excavate the limited but significant role Black people were able to claim for themselves in the annals of computing, while anatomizing the ways people of color were/are consciously or unconsciously disadvantaged by not having a proportionate seat at the software table in the U.S. It remains the case that the ratio of Black and Hispanic coders to their proportion of the population in the U.S., where the greatest volume of software is written, has been roughly equivalent to women's in recent years (according to Stack Overflow surveys from 2018 onward.) In 2018, fewer than three percent of professional coders identified as Black.

153

more dogs than people over forty: This was the coder and author Ellen Ullman.

"in 1980 my company had fifty programmers, all in their 20s and 30s, and only three were women": I was interested to see that in a talk at the 2020 Domain-Driven Design Europe Conference, the great Kent Beck described his early-career nerves upon giving a presentation watched by "all the heavy hitters of the day." It's hard not to notice that of the three "heaviest hitters" cited, two were women (Jeanette Wing and Adelle Goldberg.) For completists, the third was Grady Booch. Nothing about this seemed unnatural or unusual to Beck at the time.

Weizenbaum's description of "compulsive programmers": See Islands in the Cyberstream: Seeking Havens of Reason in a Programmed Society by Joseph Weizenbaum (p. 14)

154

One of the most important stats gathered by Stack Overflow: According to Stack overflow surveys from 2018 to 2022.

Stewart Butterfield...was originally named Dharma: Inc magazine, "Slack Is Our Company of the Year. Here's Why Everybody's Talking About It."

In January 2021, the firm was sold: MarketWatch, December 1, 2020.

75 percent of programmers have at least one parent who went to university: From the 2020 Stack Overflow survey, education section This is another demographic question that has since been dropped from the survey.

the top three feeder colleges to Silicon Valley as UC Berkeley, Stanford and Carnegie Mellon: HiringSolved report

17 percent of Stanford undergraduates to be drawn from families in the top 1 percent: New York Times, "The Upshot"

only one-fifth of Stanford students come from what researchers—with no apparent irony—dub the "bottom 60 percent": Neither are the figures quoted unique to America—or code. A few years ago, there was an outcry in the UK press over figures showing that top earners at the BBC in a broad range of categories included very few were women and almost no people of color. A reckoning ensued, which most reasonable people welcomed. Using available information from public records, however, I found something hidden in the figures that went all but unremarked: in every category, including pop music DJs at BBC Radio One, there existed a giant bias in favor of the 6% of Britons who are privately educated. In current affairs, this reached almost 80%, meaning 80% of top earners were drawn from the 6% of privately educated people. Among the few women in the top five earner lists, only one went to a public (state) school, while none of the people of color did. In other words, if you were Black and/or female and wanted to do well at the BBC, you better be privately educated. When I saw this, I alerted some editors I know, including at the BBC and The Guardian, but no one touched it. All, natch, had been privately educated. In other words, it will be perfectly possible for the BBC, or by extension for tech co's, to appear to fix their obvious race and gender imbalances without much improving diversity of experience among their staff. Add to this a 34 percent gender pay gap among people with postgraduate degrees by 2024, according to the United States Census Bureau—up from 34 percent a year earlier.

one-third of US adults have under-graduate degrees: United States Census figures for 2018. These numbers appear to be fairly constant.

155

the status quo at Carnegie Mellon is roughly the same: New York Times, "The Upshot" for Carnegie Mellon and for Berkeley.

William Shockley . . . a virulently racist and paranoid man: See Stanford Daily News, the "Eugenics on the Farm" series of articles by Ben Maldonado. In logic there is a category of argument called a reductio ad absurdum, which refers to a proposition that, in being stated, proves itself false (the classic being "I do not exist.") Shockley may have had a feel for materials, but like a lot of technologists he clearly struggled to weigh nondeterministic evidence with any degree of sophistication. His claim of superior intelligence on grounds of race is a clear case of reductio ad absurdum, given the preponderance of evidence to the contrary.

Nazis...citing California statues in their defense at Nuremberg: See "Eugenics and the Nazis—the California Connection" by Edwin Black, San Francisco Chronicle, November 9, 2003.

156

The point I want to make is not about fairness: My aim in highlighting these workplace imbalances is not to shame or devalue anyone. These are systemic problems with their roots in specific history; are seldom the fault of individuals, least of all individual coders, many of whom are working hard to open the profession up. Nor is this problem unique to America, or to code. US Department of Commerce figures from 2019 to 2024 show a 34-36% gender pay gap among people with graduate and postgraduate degrees across the economy as a whole.

Adolescent brains are constitutionally, physically different from the adult brains they will become: The prefrontal cortex and corpus callosum are especially significant here.

front-end development sits with academic research at the bottom of the status and pay stack: See any Stack Overflow survey report.

157

some early voice recognition and videoconferencing software struggled with vocal timbres typical of women: Unlocking the Clubhouse, Margolis & Fisher, ibid (p. 2)

a factor in the high incidence of young mass shooters: See The Washington Post, June 3, 2022, "Young men, guns and the prefrontal cortex: The Uvalde, Tex., shooter is part of a long list of male perpetrators of similar ages. Some experts think gun laws need to change to address that."

Uber used a sophisticated application known as "Greyball": New York Times, March 3, 2017, "How Uber Deceives the Authorities Worldwide"

158

a tool to help sales agents fraudulently obtain [their] licenses: Vanity Fair, Feb 9, 2017, "Zenefits, of Sex-in-Stairwells Fame, Fires Nearly Half Its Staff"

Volkswagen ran concealed software: BBC News, 10 December, 2015

Volkswagen's actions are likely to cost sixty US and 1,200 European citizens up to a decade of life: MIT News, March 3, 2017

a team of young, privileged, single "compulsive programmer" men: Joseph Weizenbaum's elucidation of his term "compulsive programmers" in his book Islands in the Cyberstream.