Edsger Dijkstra’s One-Day Workweek

Within my particular subfield of theoretical computer science there’s perhaps no individual more celebrated than Edsger Dijkstra. His career spanned half-a-century, beginning with a young Dijkstra formulating and solving the now classic shortest paths problem while working as a computer programmer at the Mathematical Center in Amsterdam, and ending with him as a renowned full professor holding a prestigious chair in the computer science department of the University of Texas at Austin.

During this period, Dijkstra introduced some of the biggest ideas in distributed and concurrent computing, from semaphores and deadlock, to nondeterminacy and fairness. In 2003, the year after his death, the annual award given by the top conference in my field was renamed The Dijkstra Prize in his honor.

This is all to say that I was intrigued when an alert reader recently pointed my attention to a fascinating observation about Dijkstra’s career. In 1973, fresh off winning a Turing Award, the highest prize in all of computer science, Dijkstra accepted a research fellow position that the Burroughs Corporation created specifically for him. As his colleagues later recalled:

“[Dijkstra’s] duties consisted of visiting some of the company’s research centers a few times a year and carrying on his own research, which he did in the smallest Burroughs research facility, namely, his study on the second floor of his house in Nuenen.”

Dijkstra maintained an academic appointment during this period, but ramped down his involvement with his university so that he only visited campus one day per week, on Tuesdays, during which he would gather likeminded colleagues to read papers and discuss ideas. He even pulled back on the time-consuming task of preparing papers for peer-reviewed publication, capturing more of his ideas directly in hand-written, sequentially-numbered reports that he called “EWDs”, referencing his initials.

At this point, Dijkstra had become the opposite of busy. He spent almost all of his time thinking and recording his ideas. He only came to campus on Tuesdays. And yet, as Dijkstra’s colleagues noted:

“The Burroughs years saw him at his most prolific in output of research articles. He wrote nearly 500 documents in the EWD series.”

In this specific case study we see hints of a general observation about slow productivity. Busyness is not the engine of production. It can, in many cases, instead be the obstacle to accomplishing your best work.

#####

As you may have noticed, this newsletter took a little break over the summer, during which time I’ve been serving as a Montgomery Fellow up here at Dartmouth College in Hanover, New Hampshire. As the summer quarter winds down, and me and my family are preparing to move back from the idyllic Montgomery House here on campus to our home in Takoma Park, I’ve now restarted the newsletter, which should return to something like its normal rhythm of 2+ essays per month.

A couple quick administrative notes to share:

• The second edition of my Time Block Planner is launching on August 15th. I’ll probably post more about it closer to that date, but I’ll mention now that the new edition has spiral binding (!), a beautiful new grade of paper, and an extra month’s worth of planning pages. If you’re already thinking about ordering one, you might consider pre-ordering it, as my editor warned me that once we sell out the first printing it might take a minute until we receive the next one (due to supply chain nonsense).
• We recently revealed the cover for my upcoming book on Slow Productivity, which is scheduled to come out in March. I made a conscious choice with this design to separate myself from the standard vernacular of business and advice guides and instead emphasize this book’s aspirational focus on crafting a more humane and sustainable life. I’ll of course be talking a lot more about all of this as we get closer to the release date next year.