Monday, September 20, 2010

Joseph B Kruskal passed away (Guest post by Clyde P Kruskal)

(Guest post from Clyde P Kruskal.)

Yesterday (September 19, 2010) my uncle, Joseph B. Kruskal passed away. I just wanted to say a few personal words. Maybe, later, someone will discuss his work. Although he is famous in computer science for Kruskal's Algorithm and Kruskal's Tree Theorem he was not primarily a computer scientist. He was also a statistician and psychometrician.

I recall as a child how happy I was when he came to visit, usually after giving a talk somewhere. I also remember that once, when our extended family had a get together on the Long Island Sound, he spent all day taking the children one-by-one out sailing.

I always knew the three Kruskal brothers were well-known mathematicians. But it still surprised me the first time I actually saw my uncle's work referenced in a book. I was in college working at a summer job, learning some graph theory, which at the time I knew nothing about. As I turned the page of the book, there was Kruskal's algorithm! I had never heard of it until then.

Now I have the pleasure of teaching the algorithm in my classes. I used to think that when I got old enough I would be able to fool students into thinking that it was actually my own, but it never quite worked out that way.

I did use his theorem on Totally Unimodular Matrices that he co-authored with Alan Hoffman, in a paper I wrote with Marc Snir. When I told Joe we were referencing the theorem, he told me that he rarely collaborated on research, but this was an exception, which he very much enjoyed. When looking up the reference just now, I found an interesting description of their collaboration, which confirmed how I recalled it.

There was a period of time when I used to visit Bell Labs, where Joe worked. I would stay with him and my Aunt Rachel, who were wonderful hosts. I used to enjoy our wide ranging dinner conversations, and I learned so much about words, politics, statistics, my family, etc.

I apologize if this is a bit disjointed. It makes me feel better having written it on this very sad day.

1. So sad. We are at the end of an era. In the introductory class of ever field, we learn the basic theorems of that field -- theorems that were proven way back in the days "when giants walked the earth". In computer science, I remember always being shocked to learn that those giants still walked among us.

2. A while back I heard it said that in
Mathematics you learn theorems proven
by people who are dead, while in computer science you learn theorems proven by people who are alive. This is getting less true every day. Dijistra, Floyd,
and now Kruskal...

3. My older brother (and best friend) passed away just a little more than a month ago, and have an idea of how you feel.

There are so many Theoretical Computer Scientists I wish I had met. Luckily, I'm still young. I hope to meet the ones who inspire me the most - and to inspire my peers into being the ones some even younger me will someday want to meet.

4. Kruskal's MST algorithm was one of the things that attracted me to the algorithmic side of graph theory, before that I had only taken a course on "pure graph theory" with S.P.Mohanty and greatly enjoyed it.

I wish I had a chance to meet him when he visited his nephew Clyde. Sorry that it will not happen now. I wish his family all the best and the strength to cope with the loss of their loved one.

5. Soda accepted papers posted on website.

With all this time, you would think they could get the list of authors correct. Many titles have incomplete authors and/or wrong ordering.

6. I learned about you a year and a half back. You are part of one of my favorite subjects, and proving your algorithm was one of my first assignments in graphs. I did not know you still walked this earth. It is a pity I never got to meet you.

Farewell, mr. Kruskal. May you find your brothers that I see have passed as well.

My condolences to your family. Be proud that you are related to men of such stature!

7. About the incomplete authors, the table only claims to provide the authors who labeled themselves contact authors, and not the full author list.

8. I am so sorry for the loss of your uncle Joe; we wish you and your family peace and comfort at this sad time.