Wednesday, October 20, 2010

A Note from the Trenches (Guest Post from Michael Mitzenmacher)

Former blogger Michael Mitzenmacher talks about being chair and not blogging. In a day for guest posts, over at Geomblog, David Johnson wants to know practical applications of approximation algorithms.

It's been about about seven weeks since I've given up blogging, and I have to say, I do miss it. There's certainly been plenty of things I could have written about, from large-scale CS issues (the Simons foundation call for a new institute for the theory of computing, the NRC rankings and the CRA reaction, new people at the NSF, and the movie the Social Network), more Harvard-centric issues (our intro CS class jumping to over 500 students -- about 200 more than last year, Harvard CS is hiring this year, the Harvard endowment performance (+11%), and the movie the Social Network), to more personal issues (my class for this semester, my take on the CoNEXT PC meeting, my very fun trips to Princeton and UCLA, and why I still haven't seen the movie the Social Network). Each of these could easily have been a post, I'm sure. And I've apparently become terribly accustomed to being able to just announce what I'm thinking (as though everyone should care).

On the other hand, I've been busy. Quite busy. Lots of meetings, lots of answering people's e-mail, lots of solving minor problems, lots of pointing people to the right other people to get problems solved. The start of the academic year is always busy anyway, and my graduate class takes up a fair bit of time even if a good chunk of it is material from previous years, because a good chunk of it is also always new. But the new job as "Area Dean" really sucks up a good bit of time. For the first month, I really don't think I got any research done. This month is better, though much of the research time has been going to revising and finishing off old work rather than new work. Next month I hope it will get better still.

I don't want to say the job sucks. (Well, maybe sometimes it does.) But it does take time, and I'm glad there's a planned exit. As I tell my colleagues, not that I'm unhappy with the job, but only 2.7 more years to go.

On the plus side, there's a lot of positive things going on that I feel like I'm pushing forward. To a large extent, that really seems to be the job: just pushing projects (and the corresponding people) forward, so something gets done. I find things like organizing the class schedule for the next X years and getting a slot to hire don't just happen by themselves, but with the right prodding, they do happen. I'm also blessed with incredibly collegial colleagues, many of whom are going extra miles to make my job easier.

My main management technique is to figure out (or get told) what needs to get done, tell people about it, and then see that it gets done, by me or, hopefully, someone else. I'm getting more used to fixing tasks and delegating them to other people. I also use affirmations constantly. I figure if I tell enough people enough times that something is going to happen, they'll all believe it, and so it will happen. For example, for various reasons for several years we haven't hired new junior faculty. One thing I keep telling people is that after my stint, the question every year at Harvard will not be, "Is CS doing a search this year?", but rather, "What areas is CS focusing its search in this year?" We are doing a search this year; I'm already working to get buy-in on next year's planned search, and it looks very promising; and while it's a bit early to start asking the powers-that-be about year three, I'll start laying the foundation there soon. And see, by telling all of you about it, I'm just in my own positive-thinking (or, alternatively, manipulative) way helping ensure it will happen. Once people accept your basic plan, after that it's just (a lot of) paperwork.

So while I miss blogging, it's been good for me to stop. Besides (desperately) needing the time, it's clear that blogging would give me too much opportunity to say my mind about things brewing before the plans are all fully baked (to mix cooking metaphors). Quietly building consensus doesn't go well with writing a provocative blog. The payoff, I hope, is that you'll be hearing lots of good things about Harvard CS over the next couple of years. After all, we have to get ready for the next round of NRC rankings. I'll try to fire off the occasional update from the trenches, and as for returning to blogging, we'll see how things look in about 2.7 years.


  1. Absence makes the heart grow fonder. That was actually pretty newsy, thanks.

  2. Hi Michael,

    Great to have you back blogging! (If only for a guest post).

    Could you clarify -- is Harvard looking to hire a theorist this year? The job announcement was posted on this blog as if it was, but the text itself seemed to specify A.I. as the target area.


  3. The target area is Artificial Intelligence/Machine Learning. However, in the past whenever we've done searches we've tried to do as broad a search as possible. While there are specific areas we need people in, we're also open to hire top-quality people regardless of area when a strong opportunity arises. If the right theory person came up, we'd be making the case for hiring him or her.

    I ran our last search a few years ago; we ended up making 6 offers covering 3 different areas. The economic situation suggests we won't be able to be quite so radical this time around. But (part of) my job is to push hard to hire great people when the opportunity arises. If people apply, I'll do my part.

  4. Michael -- great to see you back on the blogosphere. But just wondering: is it considered cheating if you were to make occasional blog posts (such as this one) on "My Biased Coin". After all, you say you're firing the occasional salvo from the trenches, so it isn't considered regular blogging :-) I, for one, and am sure a lot of others, would be very interested in hearing what you have to say in any one of the many topics that you listed in your post (especially the NRC rankings!)

  5. Anon #4:

    I'll certainly try to comment on news of the day where possible, but probably either through guest posts, or comments on other blogs. For the NRC rankings, for example, Scott Aaronson had an excellent post:
    where I commented. It's no secret the rankings are just goofy. But my main point was:
    "I’d rather worry about how to make CS at Harvard actually concretely better than about what this set of ratings says."
    And that's what I'm doing.