Victor Klee

The Seattle Times printed an obituary for Victor Klee today. Klee was a professor of mathematics at the University of Washington from 1953 to 2000. He was mainly known for work on convex polytopes (generalizations to higher dimensions of convex polyhedra), optimization theory, combinatorics, and discrete geometry.

I got to know him when I was an undergraduate at the UW majoring in Math. I signed up for a course that was called Honors Problem Seminar, not really knowing what to expect. By that time Victor Klee was very well known. But I didn't know that because I lived in a world without famous people in those days. I didn't know who Mick Jagger was either.

Klee found my lack of interest in the personalities of math odd but unthreatening. He was singularly unthreatened by everybody. That was what made him so great. He had fun doing what he did and wanted everyone to join in.

I think that had a lot to do with his choice of research topics. It may not be evident to a layman reading the list of topics but the problems Klee chose to devote his energies to were problems that tended to have elementary statements. That doesn't mean the problems were easy, but it does mean they are easily shared.

In Mathematics people try to understand each others work. There will be visitors to the department to give lectures and everyone will go and sit in and try to get a sense of what the visitor is working on.

But usually that's all most people get, except maybe the one guy in the room who's up on that particular area. In an hour the lecturer wouldn't be able to even state the problem or problems he/she was working on. Instead you'd get a description of the nature of the problem usually using graphic metaphors that give the illusion of understanding, because we all think we understand what we can visualize. So you might hear a lecture on 93 and higher dimensional manifolds, and get a lot of talk about "kinks" and "measuring kinks" and "kinkiness", to find out that the actual definition starts on page 23, and goes 6 pages, in a research paper you can't read because the first 22 pages assume you've read ten others you haven't, because among other things you never heard of them. But you'll feel like you get it, because you have a feel for kinkiness.

Victor Klee didn't like that kind of mathematics. He liked to work on problems that were relatively easily stated, that didn't require a lot of theoretical machinery to set up.

As a result his work was appreciated by everyone and influential. He had students galore and no end of people willing and eager to collaborate with him.

I'm guessing that was the real reason he preferred those sorts of problems. He enjoyed the sharing. A lot of other people enjoyed that sharing, and will miss it very much.