When I was around 10 years old I would regularly wake up at 5 or 6 a.m. to play Nintendo before school. My brother, six years my senior, would sleep in as late as possible and remind me on a daily basis that I should do the same while I still could. I, however, thought being awake was simply more fun than being asleep. And the notion of living a full life by avoiding sleep stuck with me into adulthood.
Three years ago I started a company in San Francisco with some friends. I didn’t quit my day job, so this was an after-hours project. We set up an office and established a routine of working from 6 p.m. to midnight. After we started working for the new company full-time, I fell into a classic trap of San Francisco startup culture: I confused work hours with productivity. Somewhere in the back of my mind I knew that working smart was better than working hard, but I convinced myself that I was doing both.
I wasn’t. As a programmer, I averaged 10-12 hours in front of a computer every day and rarely went to bed before midnight. Eventually, I felt dull and unmotivated. When I took two weeks off to travel in Colombia, I spent the first few nights sleeping for 10 hours each. I realized that if there is such a thing as sleep debt, I had accumulated some. I made up my mind to correct my sleep patterns.
I began sifting through studies, articles, and anecdotes about programmers and sleep. I spent some time reading a Reddit thread on polyphasic sleep, where an alleged sleep scientist recommended a regular bedtime and no alarm clock. I was convinced after reading this Harvard Business Review piece and this New York Times Magazine article that sleep deficits impair work performance.
In the spring of 2011 I overhauled my approach to sleep. I bumped my bedtime to 10 p.m., which means I actually got in bed at 10:30 p.m. and probably fell asleep at 11 p.m. I stopped setting an alarm clock—programmers, especially the freelance type, can usually get away with this. I cut out what little caffeine I already drank, made sure my room was dark and quiet (with the help of a fan for white noise), and truly made sleep a priority.
For a almost a month I slept from about 11 p.m. until 8 or 9 a.m. It’s possible I was catching up on sleep debt I had accrued over the previous two years. I was much more motivated and happily doing more in less time. I found it easier to direct and hold my concentration: programming had my attention for two four-hour blocks per day, and I played music for a solid hour or two in the evenings. I will not go back to compromising sleep unless it’s absolutely necessary.
I’ve recently started experimenting further. This fall I read an article about Haruki Murakami. His schedule fascinates me: He goes to bed around 9pm and wakes up naturally at 4am. He writes from the time he wakes up until 9 or 10 a.m., and then goes about the rest of his day.
Recently, I was in the midst of an isolated software project for GOOD. A small team and I would spend three weeks offsite and come back with something to share. Most programmers will tell you that they solve their most difficult problems by distracting themselves with a break or a good night’s sleep. As my days grew longer, I decided to try Murakami’s schedule.
Each day, I would make a list of all of the problems I wanted to solve in the next 24 hours, then get in bed at 8 p.m. I would wake up somewhere between 4 and 6 a.m. with most of the solutions flooding through my head. Implementation took far less time, I made fewer mistakes, and even though I’d gotten up so early, I didn’t feel completely terrible because I had actually slept.
The intense work project came to an end, but I’ve started applying this somewhat insane schedule a few times a week with moderate success. I don’t find motivation to write music or stretch my programming skills in the evening, so I do it in the morning from 5 a.m. to 9 a.m. I get in bed between 8 p.m. and 10 p.m., but being 27 in Los Angeles and not yet a full-blown urban hermit means I don’t have as regular a bedtime as I would like. My sleep schedule is still more erratic than is probably healthy.
After only a year of prioritizing sleep, I feel I’ve already learned and produced more, relatively speaking. I’ve even been able to conquer that weird desire to stay up late despite having nothing to do. The dull feeling I had a few summers ago has not returned, and the tradeoff between longer, lower-quality days and shorter, higher-quality days has been worthwhile.
