Joan Feigenbaum – not the Best, but Good Enough

The next installment of our sharing-advice project. For the ten colleges who only hire “the best in five years” and for all of us as well.

Just a few weeks ago, I heard a fascinating, partly autobiographical talk by Richard Tapia, who is an applied mathematician at Rice, a National Medal of Science winner, and a (successful!) champion of under-represented minorities in STEM.  He has given me permission to quote the following great line from his talk (one of many):

“I’m not the best, but I’m good enough.”
During my more than thirty years (and counting) of scouting, evaluating, interviewing, and discussing job candidates, I’ve never heard anyone admit that he or she was not “the best.” Nor have I ever heard my colleagues who have championed various candidates admit that one of them was not “the best.”  Of course, the official policy of Yale and all other selective colleges and universities is that we only hire “the best” faculty and that we only admit “the best” students.
Think about it, folks.  None of you is innumerate.  That means that, deep down, you know that you cannot all be “the best.”  Even the top departments in the country have hired (and tenured!) and admitted people who are not “the best.”
Following Tapia’s lead, I’d like to point out the completely obvious fact that the last time I personally was “the best” in my peer group was in high school.  I wasn’t the best math undergrad at Harvard or the best CS grad student at Stanford or the best research staff member at AT&T, and I’m not the best CS faculty member at Yale.  But I’m good enough to have made a go of it at every stage.
Next time you feel tempted to make an offer to the same (“best”) faculty candidate who’s likely to get offers from every CS department in the country and … surprise, surprise … to accept an offer from MIT, pause for minute.  Look more seriously at some of your other candidates, particularly those who are members of under-represented minority groups.  Perhaps some of them are not the best but are plenty good enough.

9 Comments on Joan Feigenbaum – not the Best, but Good Enough

  1. While I fully sympathize with the idea of “good enough”, I think the larger problem is with assuming (even implicitly) that there is some single total ordering of people (candidates, colleagues, etc.). Which is better, the lion or the cheetah, the giraffe or the eagle or the snake? etc. Given what I feel is a partial order at best, I suggest replacing “best” with “second to none”, enlarging the pool significantly.


  2. I agree with L: a key point here is that there is no total ordering on computer scientists. We ought all aspire to be (and hire) people who are on the Pareto frontier. There are plenty of such people to go around.


  3. I have thought about this before because similarly to you I have not be “the best” since highschool. Over time I have learned to accept and even embrace this. I now enjoy the fact that our jobs give us the possibility to interact with such extremely smart people.

    In fact, I like to think that if you are the best in whatever you are doing, you are likely not surrounding yourself with the right people. After all, working with stronger people gives you the chance to learn from them and thus become better yourself.

    So no, I have not problem admitting that I am not the best. After all, there is enough to be done also for the “good enough”. (Of course it help that I have a stable position…)


    • I could not agree more. I’ve never looked back to high school and wished that I could still be “the best.” It’s been a great privilege to be “good enough” to be in environments in which I could look up to and learn from my coworkers.


  4. Not even good enough // February 19, 2018 at 3:44 pm // Reply

    Whom is this “sharing-advice project” targeted at? This post is certainly not encouraging or helpful advice for those candidates that didn’t even get interviews at Yale! (And maybe also not for those who did.)


    • I don’t know about the whole project, but my own “tip for life” was targeted at computer-science faculty members (and others who make admissions and hiring decisions) who are unjustifiably narrow-minded in their consideration of candidates. Instead of looking at the full range of candidates and potential candidates and asking themselves “would this person’s presence significantly improve my organization,” they ask themselves “who are ‘the best’ candidates?” Wrong way to pose the question!!! — JF


      • The project in its entirety is aimed at a wide population of theoreticians (though most advice is not strongly tied to the area). Each individual contribution covers a more narrow aspect but put together I believe there will be something relevant for anybody. I’m happy to receive other advice and to post anonymously (or you can add advice in comments).


  5. This is a good advice for the hiring committee members of any school. I will add one more bit to it. When deciding the adjectives “good enough”, “second to none”, keep in mind the amount of opportunity a candidate has received so far. Doing undergraduate from a prestigious Ivy league school –> Ph.D. from MIT/Stanford/Berkeley –> writing papers with heavy-weight advisors and a cohort of famous collaborators —> faculty position in top school –> great students …..the ripple effect keeps on adding. Sometimes people are not so fortunate–they had to start their career at a wrong place for many reasons, and then it becomes an uphill task to catch up with those privileged peers without any support. Pay some attention to their cv before brushing them off.


  6. Other have already pointed out what seems to be a conceptual mistake in this post. Let me make it clearer, if you will: there is no such thing as a “best” researcher. There may be “best results” though, though this in itself is unclear. So a candidate can be defined as “best” if his/her results up to this point are “best”. This has some correlation with having “best” results in the future, but not strong enough to justify hiring based on such a metric (for instance, if A has “best” results up to the present moment and B has only slightly “less good” results, there is no significant higher chance that B will not overcome A in the near future).

    Oded Goldreich had a good essay about a related misconception caused mainly by awards (they tend to make people focus on personal greatness instead of greatness of results).


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: