Goodbye, Uncle Steve

©Copyright 2007,2008,2009 by Howard S. Modell

Stephen A Underwood Christmas 2003.

During an otherwise very pleasant break from work, I got some sad news. I found out today that my Thesis Advisor from Graduate School, Dr. Stephen A. Underwood, died a few months ago.

Uncle Steve is gone.

This is Steve's obituary. (originally published here. I'm not going to repeat anything said in it. Instead, I just want to set down some of my memories of a man I was very pleased to count among my friends.

I did my initial graduate study at the University of Texas at Arlington. in the new Computer Science department of the College of Engineering. I started there in the Fall of 1975, and after taking time to "interview" each of the 5 profs, I came to an agreement to do some research for a Master's thesis under Dr. Underwood. The obit says that Steve was new to UTA that year, too. I knew the department was relatively new, only a few years old at the time. I think I even knew at the time that Steve hadn't been there long. I was probably his first Graduate Student (at least at UTA).

I don't really know why I went with Steve. Dr. Kurt Schember (also deceased, long since) who's specialty was compilers and compiler languages might have been a better choice. I was conscientous and took time the week before classes started to stop in to introduce myself to each of the profs, sounding them out about doing research under them. Steve's area of interest was more hardware than software, and his Ph.D. work had been (according to my 30 year old memories) in the area of Infra-red detection and analysis. However, when I asked him whether he had any topics he was interested in having researched, he mentioned computer image detection. He admitted that there'd already been considerable work done in the area but one idea that he was interested in (but hadn't had the time to poke into) was investigating whether parallel computing (designing programs such that the work could be automatically split up among multiple CPU's) could be used to improve the speed of detecting the outlines of images encoded into digital scan data. This ended up being the subject of my Master's thesis: Computer Image Feature Extraction Using Parallel Processing: A Simulation Study (University of Texas at Arlington Press, University of Texas at Arlington, July 1976.).

It's strange. For all my affection for the man (enough that I occasionally called him "Uncle Steve"; he never contradicted me, just smiled that sheepish grin of his. I think it amused him.), we didn't actually do much together. I mean, we didn't get together outside of school. No particular reason, I suspect; it just didn't occur to either of us to do anything socially.

I remember being his grader for one of those auditorium classes. I remember it particularly because while grading one particular test, I found not one, not two, but three distinct pairs of cheaters. (I have enough of a "photographic memory" that when I see something a second time, I tend to recognize it. In this case, they were answer papers that were not just full of incorrect answers, but full of wildly incorrect answers. Things like that stick in your mind, so that when you see a second instance, identical right down to the crossed-out corrections, you have to go burrowing through the stack of "already-graded" just to be able to convince yourself that you're not crazy.)

When it came time for me to actually write my thesis, I made a decision to let my committee members see the first chapter so I could gauge how well i was doing in (what was to me) a new style of writing. Steve was quite content with it. Dr. Schember less so, but he wasn't that upset, either. Dr. Ward, on the third hand, returned his copy to me so thick with blue pencilling that I could barely read what I wrote. Upon seeing this, Uncle Steve felt honor-bound to go back and re-read it and be a little more critical. In any event, when it came time for my "defense" oral exam, Steve was very definitely on my side, helping me defend at times.

I know that Steve liked what I'd done. He indicated at one point (after I had decided to leave the school to pursue a doctorate elsewhere) that he might continue/extend what I'd done himself. As far as I know, he never actually did so.

I tried my best to keep in touch with him over the decades after I got my MS. He got cards from me every so often, and on the one or two occasions I found myself back in Arlington, I made it a point to stop by and visit with him. One of those times was after I was married and I brought my wife and infant daughter to visit him. After he got married, the "staying in touch" graduated to exchange of Christmas/New Year's cards & letters, which let's face it is a direct consequence of having a woman to "civilize" you. I tried to get in touch with him when we were in Texas this past summer, but we missed each other. My fault for not letting him know ahead of time that I'd be out there. I regret that I didn't get to talk with him one last time.

Steve Underwood was a very warm human being, friendly, out-going, who enjoyed his work, enjoyed teaching, and did his best to help his grad student "survive the process". I don't think I could have done better with any of the other profs. I'd like to think that some of what I've learned about teaching and coaching students I learned from him during the more than two years I worked with him.

I will miss Steve. He was a very important person in my life, and I know I'm a better person for having known him.