Saturday, February 12, 2005

Just Say No

I have one word of advice that applies especially to junior faculty: No.

You will be asked to referee papers, look at graduate applications, look a faculty applications, write reports and recommendation letters. You will be asked to sit on committees: curriculum committees, space committees, program committee, web page committees, budget committees, planning committees and other committees you would never have imagined. You'll have committee meetings at the departmental, divisional and university levels as well as committees to serve the broader theory community. You will be asked to organize workshops, conferences and edit special issues. You'll also be teaching, writing grants and going to faculty meetings.

All of the above are good things to do. But do all of the above and you'll never get tenure. You need time for research. So learn to limit yourself, learn to say "no". I'm not saying not to do any of the above. Most of these tasks have to get done; you should do your fair share and be a "good citizen". But you don't need to agree to every request, feel free to turn down requests when you feel yourself getting overloaded. If you gain a reputation as someone who can't say "no" people will take advantage of you. And don't fall for the "You are the only one who can do a good job" ploy.

If you try to do too much you won't do anything particularly well. I would much rather have someone say "no" to me than saying "yes" and doing a mediocre job.


  1. Excellent advice! I am a faculty member (not exactly junior) and have been grappling with this for a while. Any thoughts on appropriate ways to say "No" - e.g., by clearly explaining the sort of things one has to finish soon?

  2. I can see why you suggest this to junior faculty, who do have tenure to worry about, and I understand the sentiment. I also agree that when you get overloaded, you need to say no. But as someone who just "can't say no" -- as unforutnately I think my record shows -- let me take the other side of this issue.

    I think the problem in computer science as a whole is that we say no far too often.

    Let me go over the specific types of work you bring up in the original post.

    I consider teaching, looking at graduate applications/faculty applications, and writing letters of recommendation just part of the job description. If you can't generally say yes to these, I suggest you look at research labs (what few there are left) rather than academia when planning a career path.

    If you're unwilling to serve on committees at the divisional or university level, or beyond (say NSF panels), you shouldn't be surprised when things don't go the way you think they should for your CS department (or for the CS community in general). Perhaps if we had more "good people" serving on university committees, basic computer science would be part of the "core curriculum" at most universities. Perhaps if we had more "good people" in our community willing to say yes when the NSF comes looking for volunteers, theory would have a higher standing with government funding agencies, and we'd have less to complain about funding.

    Moving on to refereeing and similar work -- my rule of thumb is that for every paper you submit, you should expect to review at least one paper in return. It's pretty clear to me that a lot of people aren't living up to this standard, which may explain why we complain so much about journals being so slow in getting things published in our community. For every 6-10 submissions or so, you should serve on a PC. And a few times in your life, you should organize a major conference. This may mean local arrangements, not just being a PC chair.

    In science in general, and I beleive in our community in particular, we glorify the research. We don't glorify the "good citizens", or the outstanding citizens, as much as we should. And this particularly is an issue come tenure-time. But the whole enterprise depends on good citizens.

    So maybe everyone should think twice before saying no. Have you been putting into the system as much as you have been taking out? Maybe you can trade the time for one more paper for "community service". Maybe the whle community would be much better off if everyone did.

  3. This seems to be an avatar of the "tragedy of the commons". As Lance says, it is in each (junior) faculty member's interest to avoid non-research work as far as possible. And as Mike points out, the community would suffer if everybody followed this approach. Much like its in an individual's best interest to drive a humvee, though it hurts the society as a whole by guzzling gas and making the roads more dangerous for others.

    What Lance may be suggesting is that a junior faculty member is better off sticking to research until she gets tenure and turn to other stuff once she gets tenure. Provided people don't get too used to saying 'no' in their junior years, that approach might succeed.

    On the other hand, some cities do have more hybrids than humvees because the people there consider it "cool" to have one and "uncool" to have the other. The question is how our community can encourage community-beneficial behaviour.

  4. I appreciate the larger points being discussed in the two previous comments -- but I really fail to comprehend the analogy of humvees. Why is it "in an individual's best interest to drive a humvee"? Are they fun to park? Is it fun purchasing gas for them?

  5. I strongly disagree with the third post.

    Humvees are so ugly and unwieldly that their only use is to flaunt that their astronomical expense.

  6. Admittedly this is completely besides the point and I agree with all the criticisms of Hummers. But the sales of these vehicles seem to indicate that may ppl think its in their best interest to buy them:
    (maybe the tax break helps)

    H2 sales in its first six months (2002): 19000
    Annual sales of prius (2002): 20000
    Annual sales of insight(2002): 2716

    In any case, I think the hummer in the analogy can be replaced by SUV.

  7. I disagree somewhat with Lance's message: As a junior faculty member you will be judged mainly on your research and your teaching and not your service. However, a very important aspect in judging your research is your reputation. Building that reputation can often only be done by saying 'yes' to some things.

    For example, a junior faculty member should almost always say 'yes' to serving on the PC of a major conference. It can be a lot of work, and for most theory conferences it means you have to forego submitting to the conference, but being invited to serve is a key measure of your reputation.

    Visibility is also a key. If people know you because you have said 'yes' that can be very good by the time it comes for them to be writing letters.

    Tha main take-away message should be to avoid being swamped by tasks that won't lead to increased visibility rather than just to say 'no' indiscriminately.