By Don Dugdale
Name: Rob Gingell
Position: Vice President and Sun Fellow, Sun Microsystems
Time in Current Position: 2 years
Years in the Industry: 19
Family: Married, no children
Cars He Drives: 1966 Jaguar XKE and 1984 Nissan "with about another year's worth of paint left on it"
Airplane He Flies: Twin-engine Cessna
Favorite Nonwork Activity: Aviation (takes at least one coast-to-coast trip a year)
On UniForum: "I heard people say UniForum is to some extent 'Usenix with ties.' Its focus is on people dealing with the stuff for business rather than creating technologies. And it's more of an applications group than a systems-creator group."
Rob Gingell had his moment of epiphany as a freshman chemistry major at Case Western Reserve University in Cleveland. A few months after starting classes, he realized that the computers he had been using in his science experiments were more interesting than the experiments themselves. "I'd figured out that computing was not just a hobby," he says. "So I ended up mentally switching that year. I think I'd actually switched sooner than that but I just wouldn't admit it to myself."
That switch eventually led Gingell to Sun Microsystems in Mountain View, CA, where he's a vice president and Sun Fellow. He was also recently elected to a two-year term as a director of UniForum.
Gingell (pronounced "jingle") had plenty of chances to cut his technology teeth in high school. Born and raised in the Washington, DC, area, he was lucky enough in the early 1970s to attend a school with computer timesharing. "In my junior year, the school bought a small minicomputer, which we got to be very hands-on with. At that time I was interested in how I could use the hardware with experiments."
In college, Gingell had a chance to work on a number of projects for the Advanced Research Projects Agency (ARPA) associated with the U.S. Department of Defense, an early sponsor of Unix system development. As an undergraduate, he worked on a computer-aided design system for programming, using graphs. "I got to play with computers that most people, even then, didn't get to play with," he says. "I got a lot of empirical knowledge about how stuff worked, and that was a great opportunity to play in the golden age of computers. Computers are now so 'submerged' that we're graduating a lot of people that know how to use them, but not so many people who know how to build them. I'm not sure whether that's a bug or a feature, but it's a difference."
After getting his bachelor's degree in computer engineering from Case, Gingell decided to stay in the academic environment, working in a variety of research projects at the university for the next eight years, including development of systems analysis tools to verify the properties of high-end avionics systems for the U.S. Air Force. He also helped put together a campus network before networking became popular. "Around the middle 1980s that work was coming to a plateau. We had succeeded with what we started out to build and learned a lot from it. It was time to think about the next stage."
The next stage turned out to be a move to the West Coast, where Gingell has been with Sun since 1985. "At the time, Sun didn't have positions and titles," he says. "They told me I was a 'project leader,' a fabricated name designed to put something on the paper. Everyone was sort of a member of the technical staff. When I left the university, I had a lot of interest and energy about working on the next-generation operating system technology. Sun looked like the most likely place that would be doing stuff like that."
First, Sun had Gingell work on applying operating system ideas to products in the field. That led to the creation of a new memory management system using ideas pulled from MULTICS, the AT&T Bell Laboratories/Massachusetts Institute of Technology time-sharing program that preceded Unix. "Now, almost everyone runs a virtual memory system like the one we put together, and even people who don't run that VM system do dynamic linking the way we put it together," Gingell says. "So they've all become prevalent technologies."
After three or four years, Gingell began working on a joint project with Bell Laboratories to produce Unix System V Release 4 (SVR4). Sun also launched a number of research projects for its next-generation operating system. "I was so absorbed with the day-to-day aspects of the products that I actually didn't get to play with the research that much," he says. "Sort of tragically, what I came here to do happened without me."
In addition to the SVR4 work, Gingell dealt with the practices and procedures of an organization in rapid change. "How does a rapidly growing organization change itself to cope with the fact that it's a much larger place, and how do we continue to create innovative things without becoming hidebound the way some other organizations always seem to when they get very large very quickly? I worked on a variety of problems like that. So today, I'm sort of stuck with owning the practices and procedures stuff because I made the mistake of volunteering for it."
Gingell became a Sun Fellow in 1994, which brought more prestige but no real change in responsibilities, he says, as part of Sun's dual career track program that lets engineers advance without necessarily becoming managers. "This is supposed to be the same job I had, only more so. They gave me a more pretentious-sounding title to go with it. I'm supposed to do whatever I think is interesting and important. It sounds as though it ought to be a great deal of fun and for some reason, I don't seem to get around to having that. I keep having to deal with day-to-day problems."
"It would be really nice if you could treat software the way we treat a lot of hardware pieces"
He is now working on technologies to help componentize software so that every piece of a configuration doesn't have to be in place in order to test it. "People are building their systems today with networks of computers from different sources, running different applications and perhaps different versions of software. The idea that you're going to test a system by putting together an entire configuration and hold that static while you evaluate everything--that doesn't reflect the world where people are always dealing with change of some sort. It would be really nice if you could treat software the way we treat a lot of hardware pieces, as a replaceable unit. That's probably the most interesting and important thing I've been working on."
Gingell has charge of another group that's responsible for overall engineering and software standards and procedures for the engineering community at Sun--things that transcend any one product. He deals with attributes like quality, architectural issues and the way software works together. He also serves as chief of technology for software, advising SunSoft and the company at large about how to integrate all Sun's software. "I'm sort of a technology generalist," he says. "I get to learn a lot of things all the time. It has the somewhat unpleasant property that as a generalist, you know a little about a lot."
Role with UniForum
As a new member of UniForum's board of directors, Gingell sees himself as more interested in the development of technical professionals than in representing his company. "I certainly can't leave behind the biases I have as an employee of a vendor all the time, but my motivations are more about trying to have the industry be more effective," he says. "That, in the long term, benefits the vendors. An educated and empowered consumer base can't do anything but help us. It's nice to be able to go into something without having to worry about whether I'm representing the company correctly. I just focus on what I think is the best thing to do, and I don't usually find much conflict there."
Gingell hopes to help UniForum educate computer buyers and relieve confusion about Unix and open systems. "What I wanted to do was to try to meet these problems as I saw them. I'll be spending some time working on various aspects of the open systems industry in terms of how technology gets created and how we go about agreeing on what constitutes the body of things that users want to buy from. It's only natural to want to talk about how we educate people to succeed with this stuff."
Back | Table of Contents | Next