The Research Career Constraint Satisfaction Problem
14 Jul 2021The subject here is : suppose you are trying to have a research career, and your life has more constraints than the average person’s. These constraints can range from your health condition, your partner’s health condition, a special needs child, single or co-parenting, a disability, mental health, etc. In any event, the things you were able to do, either because of your own abilities or own time, you are not able to do because of the new constraints on your abilities or time. How do you continue this research career under the constraints?
I have had the conversation now contained in this post, privately, three to four times. Usually when that happens, I decide that it may be useful to also have the conversation more generally in public. In this case, moving the conversation I have been having privately to a public form is a little tricky because I am not going to reveal some aspects of my constraints here in the same way I do in private communications, but I hope that it is useful to people all the same. I’ve been dealing with my constraints for about ten years.
I will not use an analogy with the Constraint Satisfaction Problem any further in this post – that was just to get you reading, computational person – and I do not mean constraint in a negative way necessarily. Only that this item, whatever it is, cannot be changed.
Realization 1: There will be disappointment.
My experience is that I had disappointment over time, as I set research goals, and then failed to meet them not because of lack of will or hard work, but because of my constraints or other factors. I revise the goals, break them up smaller, and still not get there. When the disappointments are because I couldn’t work for months, because I was dealing with my constraints, well that hurts and is upsetting after being able to muscle through most anything in my earlier life. I see the holes in my C.V., and … what would it have been like otherwise?
I found that dealing with these disappointments gets easier over time.
Realization 2: There will be local minima.
My husband and I have learned that while things can generally be “good-ish”, there will always be local minima, or time periods when dealing with the constraints goes badly. Sometimes these local minima are blips. Sometimes they are severe. We can’t predict when the local minima will happen or their severity.
Realization 3: Planning without the constraints doesn’t work.
At one point I thought I could plan my life without incorporating much thought about the constraints. Perhaps I had this idea that at some point the constraints would become much more manageable. This has not happened, and if anything, the covid-19 pandemic has reinforced to us how much we have to factor the constraints into most everything we do as a family. So change – of job, location, spouse’s job, activities for anyone in the family, vacations, requires a lot of discussion within the family about how to make it work within the constraints.
Coping Strategy 1: Workplaces.
It is my view that if you have special circumstances you need more than a particularly good manager at your organization. In this respect, policies and procedures that provide benefits for whatever you have going on are very important to ensure you have an income and job protection for the inevitable local minima. Universities with no sick leave for faculty? No, I think not.
This part was very USA-specific, but well, I am in the USA.
Coping Strategy 2: Collaborators.
Great collaborators can be a source or great joy and energy. They can also help to get the work done; one of my academic bros and I called it, “divide and conquer.” Choosing collaborators with slightly different expertise can make for great projects.
I found that with limited time and energy, I needed collaborators with whom I could be very direct and expect the same in return. This directness includes tricky things including the budget, unhappiness with me, and research setbacks. I can deal with negative news, as long as I know what it is.
I have no tolerance for bad working relationships now. I do not join grant proposals unless I have worked with the PI before. So everyone I don’t know, sending me proposals the week before they are due, asking me to join – that decision was made way before the invitation with a one-sentence description of the project landed in my email!
Coping Strategy 3: Saying no.
There are many required activities in a research career. On the other hand, the activities that are not mandatory, but expected, can take much more time than the mandatory items. Get used to saying “no,” frequently!
I will discuss three cases where you can decline: peer review, invited talks, and service.
Peer Review.
The rule of thumb from the IEEE’s Robotics and Automation Society – I thought I read this from the conferences but right now found it at IEEE’s Transactions on Robotics journal, was for every paper you submit, the set of authors of the submitting paper are responsible for reviewing three papers. Here:
Reviewing Responsibilities of T-RO Authors
For each paper that proceeds to the full review process, the editorial board of the IEEE Transactions on Robotics (T-RO) attempts to obtain three high-quality reviews. By submitting to T-RO, you and your co-authors agree to provide up to three high-quality reviews of other T-RO submissions, if called upon. If you are unwilling to provide reviews, please submit to another journal.
Using this rule, if you review three papers for every paper you submit, you are golden. If your co-authors review none and you review three, you are still holding up your side of the reviewing bargain.
Of course, some conferences / journals may have different rules concerning their ‘exchange rate’ of submissions to corresponding reviews. Some things to consider: 1) peer review is a system that relies on a lot of people, and 2) you are one person. So if you find yourself buried in review, and not in your research, this is not your solitary problem to solve and you can say no. Or you can ask for a lighter load.
Invited talks.
Invited talks are another time sink. They are lovely! And an honor! And take me forever to prepare. I also frequently get invited, I suspect, because I am a woman in a field with few of them (agricultural robotics). After quite a few of these during the pre-covid-19 times, I realized one invited talk, just the talk, took five workdays.
- Day 1: deal with constraints at home, work maybe 2 hours, mainly administrative things so I can leave.
- Day 2: fly to the talk location, usually the opposite coast of the USA.
- Day 3: the talk. I learned to request a less packed schedule because I was exhausted at this point.
- Day 4: fly back across the USA in usually a not very efficient manner.
- Day 5: deal with constraints at home, work maybe 2 hours, mainly administrative things so I can get reimbursed.
You can see that no personal research gets done during this five-day period, which did not count the talk preparation. So I learned to choose talks carefully given the various costs to me involved. I know that some people just zip out and back for a talk. I am not one of those people.
In any event, you can decline invited talks, physical and virtual, as long as you give the host their answer quickly they will understand and go on down the list. I also take this opportunity to suggest other people working in the field as speakers to the host, if asked. I love suggesting early career researchers (ECRs)!
Service.
Committees also take a lot of time. I do not volunteer unless I really, really, really feel passionate about the subject.
Since most people do not like serving on committees, if you have done a period of time on them it is appropriate to ask for a break, a reassignment, or other modifications so that you can work. Especially if you have constraints, but also, in general, it is my view that suffering in silence does you no favors. This doesn’t mean you have to tell your supervisors every detail, only that the current setup isn’t working.
Editing is seen as a positive thing on your C.V. but also takes a lot of time. I have learned a lot being an associate editor, but doing so during the covid-19 pandemic was crushing on many levels – others, not just me, are exhausted.
Coping Strategy 4: Scheduling.
Some aspects of scheduling seem to help with the management of my constraints, such that I can work efficiently with the time that I have and also rest.
I try to keep a regular daily schedule. I work no more than 40 hours a week. I do not stay up all night, ever, except with a sick kid, which has not happened for a long time. Even then, my husband and I took shifts so that we could rest.
Given all of that, if I plan to submit for a conference, I’ll have the paper ready about a month before the deadline. Then I can send the paper around to collaborators for feedback. With my constraints, if I hit a local minimum, I can’t plan on the paper being ready otherwise.
Another aspect of scheduling is within-day scheduling. Paul Graham has written about this, that meetings can chop up your day and break concentration for programming and writing. If I have a lot of meetings in a day, I will use the breaks in between for email and administration and not even try to program. I try to keep Fridays meeting-free. Sometimes this works.
You need to determine your best working hours for specific activities, and guard them. I work best in the morning, so I tend to program and write then, and use breaks and the end of the workday for going through email and administration. I have separate work and personal email accounts, and don’t check the work account on nights and weekends so I can rest without thinking about work.
Coping Strategy 5: Communication.
I do tell my collaborators and co-workers if I am going through one of those local minima. I don’t include all of the details, but I do give them an indication of why they have not heard from me and how we should plan for the research in the meantime.
Conclusion: low tolerance for standard academic lines, “Just do one more thing,” “Wait 3 months,” etc.
Because I have been trying to have a research career with constraints since mid-PhD grad school, to now, somehow seven years later as a PI, I have low tolerance for bureaucratic academic rules. Some people do not have extra time to jump through more hoops. I am one of these people, and I can imagine many more people in not the same, but similar circumstances where research time is precious.
Thanks very much to an early readers of this post, including Dmytro Mishkin.