Before Kathi Fisler and I took our first sabbatical, we agreed on the following principle:

Don’t do on sabbatical what you can do in your office.

For us, we were already successfully publishing papers, writing grants, and so on; and that’s true of just about anyone who earns a sabbatical. So we decided that we would make sure we do not-those things.

I think this is the single most important point in this article. In some ways, if you agree with this, you may find useful things below; if you don’t, you’ll probably disagree with everything that follows, too.

Many institutions offer a “junior sabbatical”, half-way to tenure. If you think you absolutely need this to make tenure, then you probably know exactly what you need to do.

But for many people, tenure will follow in the natural course of events. If that’s you, then consider deferring that sabbatical. If you take it, you can be sure you’ll focus primarily or solely on your tenure-related activities. In contrast, taking a post-tenure sabbatical is liberating! You’re free to once again—or perhaps even for the first time in your career—to truly think big thoughts and make big plans. A sabbatical is a great time for that kind of reflection. Coming to the office at 9am to handle your usual mound of work is not.

I didn’t take a junior sabbatical. It was a terrific decision for me: when I finally did take a sabbatical, I was ready to really make the most out of it.

This is a matter that may have many driving forces: e.g., you’re visiting a close collaborator, or you have to put a child in school for the year. In that case, your decision is already made.

But if you have some freedom, I’d say keep your visit taut. I know “taut” is an odd word to use here, but I really do mean it. A pattern I have seen over and over (from others) is that they get less done in a year than others do in a semester, who in turn may accomplish less than what others do in a month, and none of it surprises me.

Suppose you visit someone for a year. When you show up, they’ve got a full plate already. There’s something to distract them that first week: a class, a deadline, a guest lecture. That’s okay, you’re still catching your breath. Something else will probably come up in the second week: some papers to review, say. No worries, you’ve got a whole year!

In the third week, you’ve got something going on: a student back home defending, a colleague who wants to write a new grant. In the fourth week you’ve both got something critical. Hey, no worries, you’ve still got a whole year! (Well, actually, a notable chunk of a year less.)

And on it goes. Consider that people go half-way around the world to attend conferences—and spend much of the day staring at their inbox instead. I’ve talked to people who have confessed that their nine-month sabbatical stint was effectively 81 of those three-day conference events strung out.

Being taut means seeing the end from the beginning. You and your host need to both feel a sense of urgency. Ideally, your host views your visit as one of their highest priorities during your visit. If you’re in a position to do so, ask hard questions, and schedule your trip accordingly.

When I went to visit Philip Wadler in Edinburgh, I had only a month. Our flight landed at about 8:30am on a Friday, we dropped off our bags with the reception of our apartment, and I went over to see Philip. He opened with, “You’re only here a month!”, and we were sketching things on a board by about 10am.Philip is a fantastic host, by the way!

Yeah: enable serendipity.

You can’t make unusual things happen. But, as they say, luck favors the prepared mind. If you go with your mind both open and prepared, you’re more lkely to get lucky. Part of the point of going somewhere is to meet new people and learn new things. Attend seminars. Chat with faculty. Visit local companies. Go to local gatherings. Do the things that will expose you to new ideas.

I went to Edinburgh because of the fantastic programming languages and verification researchers. Purely by accident (an interesting story that I’ll tell you over a drink) I ran into Keith Stenning, who was just wrapping up a monograph. He was kind enough to summarize its core argument.

In ten minutes, Keith had completely changed my perspective on computing. I realized why I had to learn about cognitive and human factors. Soon after I hired a post-doc in usable security, and my career has taken a completely different trajectory. All from ten minutes, that wouldn’t have happened if I had gone with a single-minded focus. In my sabbatical report from that year, written a few months later, I wrote:

I view it as an achievement that my work now requires communication with the IRB: this is a sign that I am following through on my desire to move my research in this direction.

There is no way I would have predicted such a change before I left.

Many of us have a choice between a semester and a year. Yours should be as long as your partner, your finances, your child, and your research need it to be. But if you have flexibility on all these matters, let me just say…

A year was too long for me.

Here’s what happened.

The summer was a summer, as usual. That fall, we focused on releasing Flapjax, and it was wonderful to have no other constraints. In particular, it was a relief to not be teaching my usual fall course.

By mid-spring, however, I was getting restive. I’d been away from a classroom for too long, and really wanted to be back in one. (I rather love teaching.) I almost considered teaching during my sabbatical, but decided that would be silly, and resisted the temptation.

By the time the next fall rolled around, the truth is I was rusty. At best, I was sluggish. Worse, it had also been so long since I’d done any kind of teaching (adding in summers and breaks, about 16 months since a seminar ended, nearly 21 since in-class lecturing) that I felt slightly resentful about having to do it again. It took me about a month to get my groove back. I don’t think my students noticed (at least, it didn’t reflect in my teaching evaluations), but I decided I didn’t want to go through that again.

Since then, I’ve only taken semester sabbaticals. Combined with a summer on one side or the other, inter-semester breaks, and so on, it adds up to about 9 months, which is plenty. By the end, I’m twitching to be back in a classroom.

Your mileage will vary.

Universities have turned sabbaticals into a precious commodity. On the calendar where you can get one every seven years,Brown, like many peer schools, currently lets you take one every seventh semester, which I think mitigates this phenomenon. for the typical academic career of about 35 years, that’s only five sabbaticals. That makes them a really big event, and one we’re loath to put off even a bit.

If you need a break, or have other personal or professional needs, by all means grab it when you’re eligible for it. But if you can afford to: consider resisting that temptation.

I think it’s best to take a sabbatical when you can justify why you want one. I don’t mean “justify” in the sense of memos to Deans—academics are experts at writing those. I mean when you can truly explain to someone you can’t fool what you want to do with it, and they believe you.

I took my first sabbatical when I was ready to reflect on what I wanted to do with tenure. I took my second when our research program on scripting languages was at a peak, but I was starting to get disillusioned by it, and also starting to think about MOOCs, which were nascent. My third was when I began to rethink my career direction and involvement in education, which I set down in a Manifesto.

I haven’t taken a fourth sabbatical, despite being well overdue by the clock. I’m waiting for something big and new that requires months of dedicated thought.To be entirely honest, in the meanwhile, I’ve found a way to rearrange my teaching schedule so I get a quasi-sabbatical most years.

Ah, you have difficult questions!

There are no easy answers for dealing with graduate students. You can try to take them along, but it can be expensive, and they may not be portable; if some are and others aren’t, you’ll end up fragmenting your group.

Late-stage graduate students are perhaps the easiest to deal with, since they should be relatively on auto-pilot; your main input might be into their job strategy, job talk, etc. A lot of that can be done remotely. For younger students, if they can’t join you, you’ll need to make other arrangements: regular call-ins, a secondary advisor, pairing up on a project with another student, and so on.

The case where things get particularly tricky is if you go off to explore truly new directions. Then you have tough work ahead.

One way to address this is to time your sabbatical (remember, you don’t have to take it as soon as it’s available) to be suitably between students. The problem is that many of us always have a student or three, who’re at different points in the pipeline, so there’s never a good time. I think viewing it that way is counter-productive. Many people switch areas but support the students from their “past life” through to graduation. You have an obligation to remain supportive of them—you got them into this!—even as your excitement may have shifted.

For the younger graduate student, your sabbatical could be a great opportunity as you come back immersed in some new area that they hadn’t imagined you might work on. What you do need to do is give them fair warning, and help them transition to another advisor (or even institution) if your plans don’t follow theirs. I was in this situation, but fortunately my plans meshed with my advisor’s. I’ve actually seen this happen in the other direction, too: a student goes off on a summer internship (the closest they get to a “sabbatical”, in terms of a break from their routine) and come back with different ideas. In one case, a PhD student took such a strong turn that we helped him move to another institution (and he’s gone on to great success).

In some fields, there’s one other very interesting possibility: have your graduate student do an internship. While I was in London, Joe Gibbs Politz spent the spring with the security team at Google. This way they really are not your responsibility, since they’re answering to someone else. A pleasant side-effect for me was that Joe was back for the summer (when students usually go on internships), which is my most productive research time. We not only began new research, we plotted out the MOOC that we ran the next semester.

In short, think of students not as a “problem” that blocks your path but as a “situation” to be managed, fairly. Be flexible and be creative. But—as long as your students fully understand the consequences—it’s actually okay to be a wee bit selfish. It’s your sabbatical, and you earned it. Remind them that there’s a great opportunity awaiting them here: to join you as you start on a new quest. People are often excited to join a neat project on the ground floor! That can be so much more fun than just doing more incremental work. You just need to be open about what you’re doing and supportive of them satisfying their needs, even if that means helping them leave you and work with someone else.

Yeah: live your life.

What I mean is this. Academia has many pleasures; it can be a great life. But it also has its own relentless clock.

For instance, one of the most remarkable trips I’ve taken was to visit the Trinity test site in New Mexico. But you can only visit on two days of the year,Pro tip: Go early, and visit through the Stallion Gate. and my university would not appreciate me swanning off for a week or two of vacation to New Mexico in the middle of the semester. (Which rules out the balloon fiesta too.)

Make sure you have time to catch up on the things you want to do: see balloons or nuclear sights or hike trails or bike across a country or travel around the world. Personally, when I went to London in 2012, I left this on my calendar scheduling page:

Until September, I am on

sabbatical.

For March-May I will be in

London,

that ancient center of

civilization and

banking.

I will revel in the

British Museum, while

casting figurative stones

on behalf of

wronged nations

everywhere.

I will re-visit the

Tudors and Victorians

at the Portrait Gallery.

I will marvel at

Uccello's linearity in

The Battle of San Romano.

I will thrill to the

Hogarth's in

John Soane's house while

squeezing through its

interstices.

I will watch cricket

at Lord's.

I will watch footy

on the telly

while eating curry.

I will color the city

Monopoly hues, and

distinguish the

Jubilee, Bakerloo, and

Metropolitan lines.

I will refine

old thoughts

and think

new ones.

I will even, after

tea and cakes and ices, if

the mood strikes,

the time permits, and

the spirit is expansive,

engage in some of the

Trades and Things

that I have promised my

Dean.

Fortunately, I don’t think my Dean saw that.