Friday, July 14, 2006

We don't need no stinkin' publication count.

Presented for contemplation: Richard Feynman wrote only 37 research papers.


Categories:

5 comments:

  1. Yeah, but some of those weren't even Nobel prize material ;) 

    Posted by Dave Bacon

    ReplyDelete
  2. For truly bright people out there reading this, there's a lesson to be learned: if you want to be a superstar today write 250 medium-to-good papers, butif you want to be a superstar for generations to come write 50 good-to-excellent papers instead.

    N.N.

    p.s. and then there's the rest of us struggling to write 50 medium-to-good papers.
     

    Posted by anonymous

    ReplyDelete
  3. ...if you want to be a superstar today write 250 medium-to-good papers, butif you want to be a superstar for generations to come write 50 good-to-excellent papers instead. 

    No, that's not really the way to become a superstar for future generations. You need a few (or just one) result which is very high profile, makes it into textbooks etc. Take Peter Shor for example. Despite an excellent publication record and some major contributions to CG, his superstar status is related to exactly one paper (sad...)

    On the Complexity Blog, Patrascu suggested aiming for a high percentage of good papers, rather than a high publication count. That seems like good advice: you have stuff to show current generations (instead of spending all your life hunting for one big result), while maximizing your chances of making a lasting contribution. 

    Posted by Anonymous

    ReplyDelete
  4. That seems like good advice: you have stuff to show current generations (instead of spending all your life hunting for one big result), while maximizing your chances of making a lasting contribution.  

    So in the end we sort of agree. Here's my revised dictum: publish 50 good-to-excellent papers in the hope that while doing so one of them will turn into a memorable result, then you'll be a superstar for years to come.

    About the lots of FOCS/STOC papers argument, I've noticed that quite a few of the people who at some point published many papers there while young, go on to rather less memorable careers later on. I'm not sure why is this though.





     

    Posted by anonymous 50-250

    ReplyDelete
  5. Regarding the last point on FOCS/STOC, I have noticed this as well. It's actually quite hard to maintain the level of focus that you can achieve as a grad student; responsibilities of either being a faculty or a member of a lab kick in, and you end up being more selective, and in many cases much broader than when you started out.  

    Posted by Suresh

    ReplyDelete

Disqus for The Geomblog