Communications of the ACM

1. Youth: I recently finished my Ph.D. and still know lots of students who are in the midst of their Ph.D. struggles. Thus, I can speak more credibly about issues that younger people in our field face, such as navigating the social and political hierarchy of academia while at the very bottom.
2. Relatability: I am not a famous researcher or visionary leader like my fellow CACM bloggers; I’m just a guy who recently finished school and got a software engineering job. As evidenced by the responses to my Ph.D. memoir, people relate much better to someone like me than to all-stars who published over a dozen papers in grad school and earned prestigious top-tier faculty jobs. Similarly, most CACM blog readers have more in common with me than with these well-known outliers.

I will devote my first few CACM blog posts to Ph.D. student life issues, but my credibility will disappear as I get older and more distant from the grad student experience. I don’t want to be nearly forty years old and still writing (or drawing comics) about grad school. Instead, I plan to transition to blogging about topics such as CS education (the focus of my current work) and the relationship between industry and academia.

Now that I’m done with self promotion, on to the main event for this month ...

Main Grinds and Side Grinds

Some of my favorite memories from grad school came from talking with fellow Ph.D. students about their side projects. What typically happens is that after a student gives a research presentation at a lab group meeting or department symposium, the audience asks conventional questions about methodology, alternative approaches, related work, and future directions. Then comes a polite round of applause, and the professors scurry off to their next meetings. When only students are left in the room, we’re finally able to speak freely and ask the speaker, "I know what your main research grind is, but what’s your side grind?" I’ve seen even the most introverted students’ faces light up with excitement as they passionately recount the details of their side projects, in stark contrast to the dry PowerPoint-filled research presentations they had just recited.

Ph.D.-level grad school is a unique environment where smart, motivated, creative individuals are concentrated together in one place without being held professionally accountable on a daily basis. Unlike undergraduate students, our days aren’t structured with classes, exams, or short-term project deadlines. And unlike our peers working in the "real" world, we’re paid so little that we don’t feel indebted to our "employers." As Professor Margo Seltzer mentions in her advice for professors, "[Ph.D. students] don't actually work for you [...]. If you're a good advisor, they will work with you and they will align their goals with your goals, but they don't really work for you. They work for themselves (face it -- in the sciences and engineering, the stipend we pay them doesn't come close to a fair wage)." Although not all advisors live up to Margo’s ideal, grad students generally have the freedom to slack off for a few weeks at a time and work on a side project rather than doing research. In contrast, it's impossible to imagine our peers with "real jobs" slacking off for a month and not doing the work they're assigned to do. That's grounds for getting fired.

Prolonged procrastination is common for grad students, since there are inevitably times when we feel stuck or depressed about our research. Although some fall into a state of paralysis, others are motivated to make progress in different ways. Upon reading an early draft of this blog post, my friend responded, "Dude I felt this article -- doing research can create such an empty feeling that you are spurred to be productive through other channels." In my experience, some of the most creative output of grad school comes from these so-called downtimes when students aren’t making visible progress on their main research projects.

Types of Side Projects

Most Ph.D. student side projects are purely personal in nature and done for their own sake rather than with some greater goal in mind. For example, I have friends who created beautiful art pieces such as pottery, knitting, painting, music, film, and even a giant metal and glass art installation for the Burning Man festival. Many of them have reported that the time they spent working on personal art projects actually improved their research output, since it gave their minds the necessary downtime to let technical problems simmer before returning to them a few days or weeks later.

Other side projects are social in nature, uniting groups of students with diverse backgrounds but common passions. For example, I recently received the following email response to my Ph.D. memoir from an old college acquaintance who has been working as a software engineer for the past seven years:

I do dream of going back and getting a PhD some day.  The main reason I want to do it is social: to be surrounded again by a diverse group of smart intellectually curious people.  The main reason I don't want to do it is not wanting to play the whole publish / grant writing / catering interests to money/recognition game.

It's surprisingly difficult in the real world to collect a bunch of smart people with lots of different interests who have free time to talk about whatever.  You can kind of doing [sic] it at a big company, but usually the interests of those people is really narrow: everyone is [a computer enthusiast], etc.  In contrast, I subletted for a month from a friend in the Harvard Econ PhD program, and every night was a long conversation with a different friend, Political Science, Nuclear Engineering, whatever.  (She was in no rush to publish or graduate.)

Although these sorts of intellectual discussions are mostly just fun chit-chat, they can sometimes spark project ideas such as influential student organizations, community outreach initiatives, or even social and political movements.

I’ve also seen side projects morph into notable research contributions. One of my friends grew frustrated with some annoying quirks in the popular PostgreSQL database and spent his spare time hacking on potential fixes. This work organically grew more substantive over time, and he was eventually able to publish it in a top-tier computer systems conference (OSDI) and turn it into his Ph.D. dissertation (advised by a Turing Award winner nonetheless). Another friend loves intense hiking, so to support this hobby, he wrote software for rendering topological maps. The lessons he learned from creating that side project indirectly inspired his dissertation work and led to two papers that both won Best Paper Awards at a top-tier compilers conference (PLDI).

Finally, side projects can unexpectedly lead to careers wholly unrelated to one’s dissertation research area. A common archetype is a group of students developing, say, a software project for fun and then using that technology as the basis to form a startup company. Other paths are less conventional: My friend spent the past year volunteering as a tutor in a remedial math class at a local middle school. The flexible work hours granted to Ph.D. students enabled him to discover and pursue that passion, which had been dormant since childhood. Now that he has just defended his dissertation (on modern database theory), he is planning to dedicate his post-graduate career to creating technologies for teaching K12 mathematics.

Take-Home Message for Ph.D. Students

I’ve met very few students who passionately love their research; most enjoy their dissertation topic and appreciate the academic challenge, but not enough to override the impulse to take on side projects. So don’t be ashamed if side grinds call to you; you’re not alone. Embrace those callings, since you probably won’t have the freedom or energy to do so after you graduate. Don’t underestimate the unexpected wisdom and serendipity that can come from side grinds.

However, balance your time between doing what you love and doing what you need to do to graduate. If you’re determined to finish your Ph.D., then there is no substitute for buckling down in your final few years and uber-grinding on your research. In rare occasions, though, if your side grind becomes compelling enough, then consider dropping out of the Ph.D. program altogether. After all, your Ph.D. years are the perfect time to discover your true passions, and if they happen to be non-academic, then so be it.

Take-Home Message for Professors

You survived a grueling gauntlet to get to your current esteemed position; you must accept the fact that most of your Ph.D. students are far less competent than you as a researcher. They mostly just want to do a solid job on their dissertation and then move on to finding something they love for a career. Some will find their true calling in research and make you proud as an academic parent by remaining in academia; however, side projects are another natural way for them to discover their professional passions.

Your students’ sense of self worth is strongly tied to your perceptions of them, so you have a responsibility for their mental well-being during their years under your tutelage. Set high but realistic expectations for their research output. Take a genuine interest in what they like to do besides research; maybe even tell them stories of what interesting side grinds you had back in grad school. Finally, no matter what, don’t dismiss side projects as being for slackers or those "not serious about the Ph.D. program." Although some side projects are frivolous, in general they make your students happier and can even boost their productivity on research projects that are ultimately in your best interest.