4 ms·

I never met Mihai in my life and will never achieve anything close to what he has but I do remember trying to read his PhD thesis to while away my time and I ch

I never met Mihai in my life and will never achieve anything close to what he has but I do remember trying to read his PhD thesis to while away my time and I chanced upon the section where he said that he solved a problem he was working on after pulling an all-nighter while he was sick and lying on a couch to recover. The guy was brilliant and truly enjoyed his work. Few people are so capable of consistent excellence as Mihai was.

dfc 14y ago
"In the spring of 2008, I was trying to submit 4 papers to FOCS, during my job interview season. Needless to say, this was a challenging experience. I had an interview at Google scheduled two days before the deadline, and I made an entirely unrealistic plan to go to New York with the 5am train, after working through the night. By 5am, I had a serious headache and the beginning of a cold, and I had to postpone the interview at the last minute. However, this proved to be a very auspicious incident. As I was lying on an MIT couch the next day trying to recover, I had an entirely unexpected revelation: the 4-dimensional lower bound could be proved by a series of reductions from lopsided set disjointness! This got developed into the set of reductions in [82], and submitted to the conference. I owe my sincere apologies to the Google researchers for messing their schedules with my unrealistic planning. But in the end, I’m glad things hapenned this way, and the incident was clearly good for science." Lower Bound Techniques for Data Structures - http://people.csail.mit.edu/mip/docs/thesis-phd/phd.pdf http://people.csail.mit.edu/mip/docs/thesis-phd/phd.pdf