Pac Man is a mostly ridiculous metaphor for scholarly reading. Mostly.

Finding time to read in academia

Andy J. Ko
Jun 2 · 5 min read

One of the first research papers I remember reading was in college. My sociology professor had printed it out for us and challenged us to read it closely. It was about dual-income households in the United States and their affect on families and society. I remember it distinctly mostly because of it’s form. It was dozens of pages, it was a PDF, and it was very hard to read.

But buried in that document was full of all kinds of dense, fascinating ideas about families, parents, capitalism, gender, children, stress, and inequality. It showed me that research papers are powerful, potent containers for ideas, including the ideas they build upon, and the ideas they contribute. This first paper was one of many that inspired me to pursue a career where I could read more of these fascinating documents, and write them.

As I proceeded in my career, however, reading became less important. In academic computer science, I encountered a culture focused on the future, rather than the past. It often dismissed prior work as old and irrelevant, always emphasizing what was new. When I made time to read in graduate school, I encountered countless mentors and peers that subtly shamed me for reading, viewing it unproductive. It was something to do minimally, to check the “literature review box” necessary for a publication.

When I started as a professor, it became even harder to find time to read. Not only was I doing research, but I was teaching, and doing all kinds of service. None of them seemed particularly in need of me reading and learning, and so I didn’t make the time. And by many measures, it didn’t seem to hurt: I published a lot, my students learned, and by attending conferences, I learned some of the big ideas through presentations rather than reading.

But over the years, those initial ideas about reading kept surfacing. Good research papers shaped my ideas. They gave me tools with which to accelerate and deepen my discovery. They showed me perspectives I didn’t have, which helped me approach phenomena in new and novel ways. By not reading, I was making a bit more time for a quantity of discovery, but at the expense of the quality of my discoveries.

Since my last sabbatical, I’ve been trying to make time for reading by following a relatively simple process:

  • Gather. I’ve set up a large number of feeds that push potential reading to me, rather than me having to seek it. Google Scholar suggests reading to me, I hear about interesting work from collaborators, I scan the proceedings of conferences I care about, I subscribe to notifications about new journals, and when I read, I mine the references of the paper for particularly interesting papers. I’ve even started to encounter interesting work from other fields by listening to great podcasts that cover academic work, and great science journalism.
  • Triage. I triage all of these potential readings by looking on three types of reading: 1) discoveries I should know to maintain my expertise in my fields, 2) discoveries I should know because they affect my research, and 3) discourses I should know because they affect all of academia. Anything that doesn’t match one of these three I ignore.
  • Store. If I find a reading that meets one of the needs above, I download its PDF and store it in a file with the author, date, and title. I’ve found over the years that PDF managers, and the powerful metadata facilities they offer, just don’t offer much return on investment. Moreover, the software decays; what’s the leading bibliographic tool one year will be obsolete in 5 years, and then all of that metadata is lost. Basic file management will likely never go away.
  • Select. On my calendar, I block off 1 hour to read twice a week, and put the reading on my to do list. When it’s reading time, I go to my set of potential reads, skim some titles and abstracts, and choose whatever looks most interesting at the moment. I don’t worry too much about its relevance to my work; I just follow my curiosity or my emotions. For example, when I’m doing peer review for a conference or journal, I might be stressed out about getting a review done, and slot it in my reading time. Whatever I choose, I sit down on my office couch and enjoy a quiet, focused read. Sometimes, I’ll realize pretty quickly that I don’t want to finish reading the paper. If that happens early in the 1 hour, I’ll choose something else.
  • Read. If I do find the paper interesting, I’ll finish it. I don’t rush it; I really try to understand the ideas, and leave space for deep understanding, just like I did the first time I read that sociology paper. After all, the point of reading isn’t just to finish a paper, or learn the gist of what it discovered, but to have the work change how I think.
  • Archive or discard. After I’m done, I do one of two things with the PDF. If I found it interesting, but don’t think I’ll ever refer to the document again, I’ll delete the PDF, and just trust my memory (and academic search engines) to be enough of an index. If I do think I’ll refer to the document, then I have a basic file structure to archive the readings that tag them by their most distinguishing features. My “Impact” folder contains papers that were impactful to my thinking, but might not be considered seminal. My “Seminal” folder contains works that shaped (or might shape) an entire community’s thinking. My “Method” folder contains works about research methods and epistemology, which are good guidelines for conducting research. My “Survey” folder contains literature reviews that are helpful references for fields. If the paper doesn’t belong in one of those, then it probably wasn’t that important, and I delete it.

I think what makes this process work for me is that I already do a lot of the triage to decide what’s interesting to read as part of my normal research process. I’m just doing it in a way that makes it easy to quickly grab something from my reading list and then actually read.

I know I’ll never get through my list. And that’s fine. If I pass over certain papers over and over for decades, they probably weren’t that important anyway. And there are times when I’m just so hungry for more knowledge, I might read two papers in an hour, or spend more than an hour reading.

Of course, this isn’t the only time I read. When I’m starting a new project, I do extensive literature reviews and read a lot all at once outside of this process. And rather than archive all I’ve read in my PDF folders, I archive it in the form of a grant proposal or research paper, which cites all of the relevant works from the review as a reference list. All of my grant proposals, research papers (accepted and rejected), and my folders of PDFs, constitute my true bibliography.

In one sense, these two hours a week, and the occasional epic literature review I do, are my classroom, and the process above is my curriculum. This is how I keep learning, how I let others shape my perspectives, just as I do for students when I teach.

How do you make time for reading? And if you don’t, what do you think you’re missing?

Bits and Behavior

This is the blog for the Code & Cognition lab, directed by professor Andy Ko, Ph.D. at the University of Washington. Here we reflect on what software is, what effects it's having on the world, and our role as public intellectuals in help civilization make sense of code.

Andy J. Ko

Written by

Associate Professor @UW_iSchool, Chief Scientist+Co-Founder @answerdash. Parent, feminist, scientist, teacher, inventor, programmer, human.

Bits and Behavior

This is the blog for the Code & Cognition lab, directed by professor Andy Ko, Ph.D. at the University of Washington. Here we reflect on what software is, what effects it's having on the world, and our role as public intellectuals in help civilization make sense of code.