Confessions of a Failed Female Coder

I was supposed to be a software developer, if only because I looked and acted the part.

In my early teens, I was socially awkward, dressed funny (there was a cargo shorts phase), excelled academically across the board, stocked my bookshelves with science fiction novels, and spent nearly all my spare time in our school’s computer lab. I taught myself turn-of-the-millennium web development basics like CSS and CGI, and built sites on a volunteer basis to fulfill my school’s community service requirement. Plus, after being bullied by other girls in middle school, the idea of working in a field dominated by nerdy men who would understand all my X-Files references seemed thrilling.

Then I enrolled in our school’s standard advanced-placement computer science course and thought it was so un-engaging and isolating and boring that I dropped it before it could bring down my GPA.

It wasn’t my teacher’s fault; she tried hard to engage me and figure out why I wasn’t connecting well with the work. Nor was I facing discrimination or isolation based on my gender — it was an all-girls school, and every single other woman who took that computer science class that year became an engineer or scientist. Rather, I was frustrated because the course consisted of lone work on exercises that seemed to have little application to the real world, the expectation that coursework required coding and debugging for hours after school threatened my ability to excel in other classes or in varsity sports (in spite of being a textbook nerd, it turned out I was good enough at rowing for a few colleges to recruit me), or to spend time buried in random books. And then there was the GPA. At the end of the day, I was a worrisome perfectionist.

In short: Despite the cargo shorts and the loner attitude, I learned like a girl. And at the time, advanced computer science classes didn’t cater to how girls learn best, even in an all-girls school.

I ultimately went into technology anyway, first as a journalist and then on the business side at Google and the startup world. But my experience as a failed coder has been on my mind considerably more lately given recent years’ dialogue about how best to accommodate women in computer science, which in my observation has devolved considerably since the publication and backlash against the now-infamous “Google memo,” now-fired engineer James Damore’s exhaustive manifesto of why diversity-in-tech efforts aren’t working and why ideological diversity should be a priority instead. I’ll probably irritate a lot of people with this take, but it’s been on my mind since this whole affair surfaced, and I reached a point where I wanted to put words to print: Damore’s memo is symbolic of a lot of what’s wrong in Silicon Valley. But so is the backlash to it.

First, to be clear, there is a lot to dislike about Damore’s memo. He demonstrated, at a bare minimum, a profound lack of judgment and what appears to be very poor self-awareness. The fact that his ultimate argument continually referenced his beleaguered status as a Silicon Valley employee whose politics tilt to the right (plus the fact that after his firing he quickly accepted interviews and cultivated relationships with alt-right kooks) insinuated that he perhaps did have a more unsavory agenda than illustrating how a failure to accommodate gender differences in learning and work was part of the reason why multimillion-dollar diversity initiatives have been failing. He also draws conclusions that apply his “findings” to issues of diversity overall, not just gender diversity, which is a leap too far.

And, I’ll note, Google is a publicly traded company that has the right to terminate any employee who is making a distracting public spectacle. I also, as an ex-Googler, know that the company places a high value on internal open dialogue and personal expression — from fringe political ideologies to scathing criticism of its own executives — and that there likely are also aspects to the story that have not been made public, as is often the case with scandals involving a large tech company, so I refrain from judgment as to whether the company was in the right to fire Damore. (That said, his bizarre recent tweets about the KKK have not helped his case.)

But there are two points I want to make that I think that many critics of Damore’s memo are deliberately missing, and their refusal to acknowledge them are going to make the (very real) struggle harder rather than easier for women in computer science:

1. Denying research on population-level biological differences between how boys and girls learn, and how men and women work, risks discrediting the efforts that girls’ education advocates are making to help girls advance in STEM. Many of the same claims that Damore makes, and the research he cites, are used by advocacy groups like the National Coalition of Girls’ Schools in a decidedly empowering context — that girls and women prefer collaborative rather than solo work, that they are more engaged in “things” if they can see the “people” aspect to it, and that they are prone to perfectionism and a fear of failure.
2. When someone takes valid facts and observations and distorts (or appears to distort) them in a way that reaches a flawed conclusion — especially one that was perhaps informed by the agenda-driven pseudoscience that permeates the deepest dregs of Reddit and 4chan — shouting them down and exaggerating the negative impact of the opinion is rarely effective. Damore’s memo was not a “screed,” as headlines claimed it was. His words were not “violence” as some critics claimed they were, nor was he advocating for female engineers to be fired or suggesting that they were biologically unsuited to a career in computer science (he suggested that they were perhaps less interested in pursuing it, which is very different and which I’ll address shortly). I’m no advocate of engaging with attention-seeking nutjobs (or in popular parlance, “feeding the trolls”), but Damore’s memo was underpinned by tones of curiosity and a desire to engage in dialogue. The backlash to it, however, pushed him squarely toward the camp of the crazies, adding fuel to the fringe of alt-right and men’s-rights-activist arguments. Dissent can be productive; it became clear that this dissent became unproductive.

Let’s take a look at my first point. Differences in how boys and girls learn — and, by association, how men and women work — are not just evident, but well-documented to be rooted in biology, not just social expectations. Traditional American coed schooling has not figured out how to address sex differences in learning, which has led to failures for both boys and girls. Girls’ grades tend to excel in these learning environments, but girls’ confidence lags, especially in math and science; meanwhile, boys who are not among the upper tiers of achievement fall behind considerably, making up the vast majority of delinquencies, dropouts, and failures, and contributing to the result that men are now a considerable minority of college students in the U.S.

I was raised by academics, and biological differences in learning were not some set of edgy theories to my parents back in the ’80s and ’90s; they rather understood them as reality and consequently committed to sending us kids to single-sex schools as much as they possibly could. (There were no nearby affordable* options for all-boys education in our area past middle school, so my brother went to the local public high school.) But, at the time, many people were still sending their children to single-sex schools out of tradition rather than because it was considered the best way for them to learn. That’s changed. Since then, the high school from which I graduated, Stuart Country Day, has furthered its commitment to educating girls not simply out of tradition but in a manner deeply informed by research on how girls learn best.

STEM education has been the biggest beneficiary of this. Introductory computer science is now a mandatory class at Stuart, eliminating the misconception that it’s insurmountably difficult and meant only for the students pulling in A’s in calculus and physics (or, more tritely, that it’s only for nerds). It also means that computer science needs to be taught in a way that connects to girls who do not instantly find an affinity for it, which challenges teachers to look outside the confines of a traditional curriculum and provide their students with a more holistic approach to the discipline. Instead of a lonely computer lab, the school now has a vibrant “maker space” that encourages collaborative work in everything from coding to robotics. The result of this is that STEM education is taught in a way that gives girls an opportunity to collaborate, as well as allow them to access the broader ideas, people, and goals that bring coding off the screen and into the real world. In many ways, I regret that I wasn’t born ten years later, because had I been introduced to computer science in this environment, I think I would have loved it.

There’s no question that we need more female computer scientists, just as we need more computer scientists coming from other backgrounds and experiences that don’t fit the Silicon Valley norm of a white or Asian male who graduated from an elite engineering program at Stanford or MIT. It’s good business, plain and simple. With recent gains in the likes of virtual reality and artificial intelligence, what we think of as “computer science” has expanded far beyond the realm of code as we know it and into that of human interactions and experiences. Without a diverse tableau of professionals in this space we risk building technology that caters only to a sliver of humanity rather than the billions of us who will ultimately use it. (An easy example is virtual reality headsets giving women motion sickness at a far higher rate than men. How many women were involved in their design?) Perhaps the way to accomplish this when it comes to gender diversity is to change the practice of computer science in a way that fits the workstyles of women as well as men, rather than trying to bend the minds and lifestyles of women to fit architectures of computer science education and work that were never built to accommodate them in the first place.

This leads to my completely untested theory** as to why women stopped making gains in computer science in the ’80s, which I talked about on Stuart’s podcast recently (start at 19:20 if you’re interested in a listen). As is well-known from the much-needed push to acknowledge the neglected contributions of women in early computer science (and the excellent film Hidden Figures), coding and programming was dominated by women in everything from early business machines to the NASA manned space program, and they were on track to reach parity in computer science bachelor’s degrees when suddenly they began dropping off.

No one has come up with a definitive answer as to why. Some researchers have speculated that it was the rise of ’80s geek culture and the introduction of an iconography around the newly commoditized personal computer — Star Wars posters, arcade games, pizza and beer — that girls from childhood familiarization with computers and hence put them at a disadvantage when they got to entry-level computer science classes. I don’t entirely buy it, and not just because the most rabid Star Trek fans I know are women. (Trust me, as a girl who loved sci-fi, computers, and the arcade games at the local pizza place, the only people implying to me that this stuff wasn’t “for girls” were the female classmates who teased me relentlessly in junior high, and my social isolation actually pushed me more toward technology.)

The ’80s did see a masculinization of computer science as a field, but I think it’s simplistic to assert that it happened because pop culture pushed PCs as “for boys.” The ’80s were also the era of Bill Gates and Steve Jobs, the rise of the archetype of the lone, antisocial hacker in a garage spending all-nighters building machines that were at many times ambiguous and without a clear path to success. Without going into too much more detail because this essay is already longer than it needs to be, this can be seen as the linchpin in which computer science took a hard turn toward learning and work environments that either pushed many women away entirely or resulted in them feeling like misfit puzzle pieces when they did take on coding. It was a manner of work utterly unsuited to everything we know about how women thrive and excel. In no way does this imply that women are biologically unsuited to engineering work; it means that the framework of the profession does not suit them and as a result likely has led to a precipitous drop in interest. It’s not a problem with “geek culture”; it’s the architecture of creativity, learning, and work. (Does the idea of being a lone hacker appeal to some women? Sure. As many of James Damore’s defenders have emphasized, population-level differences cannot be used to judge the abilities or inclinations of any individual.)

We’re now in an era where computer scientists are no longer universally considered to be ‘80s-style loners, but the structure of workplaces is still often dominated by the manner of thinking that was created by them (see also: “move fast and break things”). Similarly, when it comes to the academic backgrounds that are almost necessary to work in computer science, it’s nearly impossible to enter a top college program without prior experience, and high school computer science classes are still largely failing girls because in large part they don’t cater to the ways that girls learn best. (The absence of quality computer science classes in disadvantaged and under-resourced high schools also impedes the progress students from those backgrounds can make, irrespective of gender. Shameless plug: I am on the board of Mouse, a nonprofit that emphasizes a purpose-driven, real-world-centric approach to technology education for inner-city students. Consider supporting them!)

This also extends to the reality that many women do want to adjust their career paths to raise families, and Silicon Valley’s marginalization of parents (and older workers in general) and inflexibility about work-life balance is well-documented. But it’s possible to denounce the tech industry’s hostility to women without indicting the individual men within it. This is the system they inherited, and likely the only one they know. We can point out how they need to change — and why they need to change, you know, before they waste millions of dollars building another VR headset that makes women puke — without denouncing them as bigots, sexists, Nazis, and so forth.

This brings me to the second point I summarized above. As much as I hate to admit it, as a woman in technology who vocally and wholeheartedly supports the advancement and success of other women, there was merit to quite a few of the points James Damore raised, and discrediting the research he cites (rather than simply disagreeing with his conclusions) will hurt rather than help women’s advancement in computer science. But I’ve been hesitant to assert this until now, due exclusively to the backlash I’d face in my own professional circles. Shedding light on the biological differences between how men and women work is not sexist, and it was disheartening for me to see so many smart people I know adopting a vicious, reactionary attitude toward the very same research that has caused my high school to completely overhaul its STEM programming to ensure that more girls succeed in computer science.

This, to me, is indicative of something more: The social-media-fueled breakdown in civilized discourse and our addiction to cycles of outrage and counter-outrage is dividing us further, not encouraging us to work together, and I take the Google memo fiasco to be yet another example of this. There is nothing deferential or conciliatory about being willing to engage in civil debate and disagreement. It’s mature, it’s professional, and it’s what makes progress. And when we reject civility, we further polarize those we disagree with. We retreat into our bubbles, and we send them into theirs. Our opposing opinions grow only more extreme with lack of access to the other side.

Will adopting this thinking — that computer science needs a framework that is suited to the ways in which girls and women have been overwhelmingly documented to work best — solve everything? Of course not. There’s rampant and well-documented sexism and sexual harassment in the technology industry, particularly when it comes to the venture capital sector and male investors both directing their funds toward entrepreneurs who match their perception of what a startup founder should “look like” as well as abusing their positions of power around female entrepreneurs in ways that range from inappropriately clueless to truly creepy. We can take every shred of this seriously while also considering the James Damores of the world to be opportunities for learning and dialogue rather than hashtag hangmen.

My lifelong immersion in all-girls education, first as a student and as a daughter of educators, then as a board member, volunteer, and alumnae advocate, has meant that I’ve spent my whole life around the idea that girls and women succeed when they are in learning and working environments that address the biological realities about how women operate best. But, clearly, there’s a lot of misunderstanding around what this means. The Google memo fracas could have been an opportunity for constructive change, not a political catalyst for the alt-right. And in my opinion, that’s a real shame.

*I would be remiss not to mention that at least in the U.S., single-sex schooling really isn’t scalable, and that most of the opportunities for it are found at elite and expensive prep schools that are not open to all American students. (That’s a story for another day.)

• *I’m not a scientist, a neurologist, or a psychologist. I have a bachelor’s degree in the history of science that I primarily used to research the ways in which fears of artificial intelligence were depicted in comic books from the 1930s-1960s. But am I going to keep my opinions and observations to myself due to lack of qualifications? Nah. This is the internet.