Computer science taught me that books weren’t important. It was wrong.
I’ve been reading a lot of books lately. Books by social scientists, books by teachers, books by practitioners, books by writers. I’ve been spending an immense amount of time reading, including 30 minutes each morning during time, and then even more in the evenings and weekends. There’s just so much I want to learn deeply, and—surprise—most of the internet isn’t a place for deep insights.
In a way, this is a return to an earlier phase of my life in high school, when I’d fallen in love with fiction and devoured dozens of American classics. It was my senior English teacher in high school who cultivated my love of books. She had us read Michael Ondaatje’s The English Patient, a 1992 novel about four very different people encountering each other during World War II. It’s hazy, pastel story of recent trauma, in a setting removed from war, was a shocking voyeurism to my sheltered, reclusive adolescence. I asked her if she had other recommendations, and she gave me the reading list she was using for AP English (which I wasn’t in, because no one told me I could be a book lover, or that the class existed). The list had twenty books on it, for a forty week school year, and I read all of them for fun, while my peers read them for credit.
Only a year later, however, I stopped reading books entirely. Ironically, it was college that did it. No one made me read any. Not in English, not in social sciences, and certainly not in the natural sciences. Not even history! Maybe there were a few research papers, a chapter from a book, and of course several textbooks. But there was just no expectation that I’d take the time to read fiction, non-fiction, or any other lengthy writing. In a way, this was a shock, as most of my K-12 experience had been defined by books and book reports.
But it wasn’t just that I wasn’t asked to read any books. As a computer science major, I regularly received explicit signals from CS faculty that not only were books irrelevant, but that they were to be explicitly devalued. There were no books in the curriculum, aside from textbooks, and textbooks were often only used for the problem sets they contained, rather than the ideas or arguments they presented. Faculty who wrote textbooks were derided for “only” writing books about computing, rather than advance it. And as I began to learn about and engage in CS research, many faculty mentors explicit said to me: “books don’t count.”
There was one exception to this. My wonderful mentor, Margaret Burnett, had me read Bonnie Nardi’s A Small Matter of Programming when I first joined her lab. She told me that if I were to understand anything happening in the lab (which was studying end-user programming at the time), I needed to read this book, as it was essentially the “bible” of all of the problems of programming that the lab worked on. Inside it, I found a rich collection of ethnographic work about people attempting to harness the power of code to facilitate their information work, and of people relying heavily on the expertise of coworkers, who’d stumbled through acquiring some programming skills. It painted a picture in my mind of worlds I had not seen, worlds which I regularly returned to in order to imagine new possibilities for how that world might be changed. It was like the fictional worlds I’d fallen in love with in high school, but real.
Of course, once I was done with the book, it was clear that research still wasn’t about writing books, but at best, about consuming and citing them. There was no paradigm in CS for producing the kind of knowledge that I’d found in that book, there was no apparent expertise for how to write books, and there was no apparent value given to what I had learned in Nardi’s book, aside from supporting a claim at the beginning of a research paper. Whether my faculty mentors had intended it or not, what I learned from them was that books are nice to have, but they ultimately don’t matter. At least not in computer science.
And so I gave up books. I gave my attention instead to reading and writing conference papers, the occasional journal article, and a dissertation. I did that for a decade. Maybe I’d pick up a book for pleasure for a time, to distract myself with fiction, or to check my understanding on something. My world of knowledge, while it felt like it was deepening greatly in my area of expertise, was also narrowing, both in the phenomena I learned about, but also in the form of knowledge I was learning.
It wasn’t until I joined The Information School at the University of Washington that all of these false ideas about books became clear. The first crack in the book-resistant shell I’d grown was when I was on the academic job market in 2008, when I interviewed for a faculty position at UC Irvine. There, I actually met Bonnie Nardi, who had written the book I had so adored a decade earlier. Here was a person who had not only written a profoundly impactful book to my undergraduate research, but also had built an academic career on writing books, along with other research genres. And she could be my colleague! When I eventually chose the University of Washington as my home, it was partly because it was full of multiple faculty who had such dramatically different relationships to books than myself. Most of my new colleagues had deep, lifelong relationships to books. I had colleagues that wrote fiction as a hobby. I had colleagues who wrote books as part core part of their scholarship. I had colleagues who wrote prestigious book reviews of books, which was expected for tenure in their field. I had colleagues who were experts on book binding. I had colleagues who taught classes about book indexing, book metadata, and book history to our library and information science students. Many of my faculty colleagues, to start conversation at a social event, would start by ask me, “What are you reading?”, as if the only natural state of being as a scholar was to be engrossed in a book. Suddenly, the disregard of books I’d learned in computer science was no longer normal, but deviant.
After a few hard years of pre-tenure life, and a few hard years post-tenure in a startup, I resolved to finally reconnect with books, personally and professionally. Back in 2018, I committed to always be reading something. I set aside 30 minutes of reading time each morning after email. Sometimes my reading would be a conference our journal article on my reading list; other times it would be a paper I was reviewing. But most mornings, what I’ve really wanted to read is the next chapter or section of a book. This habit has become essential in so many ways, not only in shaping my scholarly ideas, and in broadening my knowledge, but also in structuring my day. Books are that precious, quiet time for thinking, the kind of thinking that most people imagine is a professor’s entire day outside of class, but in reality, is largely absent, unless ruthlessly cultivated.
What became immediately clear after settling into this habit was how wrong computer science was about books. The following will seem utterly naive to anyone the humanities or social sciences, obvious to anyone who loves books, but perhaps radical to my many peers in computer science, many of whom tell me they haven’t read a book in years:
- Books are profoundly more capable of shaping thought than conference or journal papers. There’s little more powerful then engaging ideas, arguments, stories, and theories of a great writer for a dozen hours.
- Books are perhaps the only genre that can articulate and defend grand arguments. The ten thousand words in a conference or journal paper is simply not enough for our most potent ideas.
- Books, when read carefully, can be far better at teaching complex ideas than most other media or methods I’ve encountered. Better than lectures, better than educational technologies, better than many active learning methods. They have the capacity to create nearly anything in the reader’s mind and sustain dialog about it. And, they have a far better user interface than most information technologies.
- Books are harder to write well than conference or journal papers. The ability to coherently extend an argument across tens of thousands, perhaps hundreds of thousands of words, requires a profound mastery over the nuances of an idea, and over language. Computer scientists that believe that books are “soft” and computing is “hard” don’t know what they’re talking about.
None of this is to discount the challenges or necessity of conference and journal papers. I still read them, I still write them. I just more deeply appreciate what books can do that other media can’t. And I’ve become more deeply critical of the way that computer science dismisses and disregards books as a valuable form of knowledge to read or write.
I do wonder what effect the ignorance of books in computer science has on CS itself. After all, many of the most powerful ideas I’ve encountered have only been adequately articulated in books. What ideas are computer scientists missing because they don’t have well-developed habits for reading books? There’s an entire world of social science and philosophy simply inaccessible to computer science, except all but the most reductive forms, such as a summary on Wikipedia, or a hopelessly incomplete summary like the kind I write on this blog.
If you’re in a discipline that writes books, I’m sorry for ignoring your work for so long. And if you’re in computer science, it’s time to start reading. To get you started, here are a few of the books I’ve read in the past year:
- Race After Technology, Ruha Benjamin
- Geek Heresy, Kentaro Toyama
- Saving Capitalism, Robert Reich
- Design Justice, Sasha Costanza-Chock
- Tomorrow Will Be Different, Sarah McBride
- Dan Rather, What Unites Us
- Weapons of Math Destruction, Cathy O’Neil
- Reading Lolita in Tehran, Azar Nafisi
- The Organized Mind, Daniel Levitin
- Pedagogy of the Oppressed, Paulo Freire
- Talking to Strangers, Malcom Gladwell
- The Devil’s Highway, Luis Alberto Urrea
- Whipping Girl, Julia Serano
- Dark Matters, Simone Browne
- Black, White, and Jewish, Rebecca Walker
Some of these I chose; many were recommended. All were transformative to how I think about the world and my research. If you haven’t read a book in a while, what better time to start than now?
