BIO: Don Shafer
TITLE: Cofounder and chief technology officer, Athens Group; adjunct professor of engineering at the University of Texas at Austin
ACADEMIC DEGREES: BS from the United States Air Force Academy, MBA from the University of Denver
CAREER HIGHLIGHTS: Shafer has contributed to three books, written more than 20 published articles, and is coauthor of "Quality Software Project Management," published by Prentice-Hall. He is a contributor to the 2010 edition of the multivolume "Encyclopedia of Software Engineering" and is a Certified Software Development Professional.
CS ACTIVITIES: Twice treasurer of the IEEE Computer Society Board of Governors, past editor in chief and chair of the IEEE Computer Society Press, IEEE Senior Member, Golden Core Member.
Complex Control Systems
An interview with Don Shafer
Don Shafer built Athens Group's oil and gas practice and leads engineers in delivering software services for exploration, production, and pipeline monitoring systems for clients such as BP, Chevron, ExxonMobil, ConocoPhillips, and Shell. He led groups developing and marketing hardware and software products for Motorola, AMD, and Crystal Semiconductor. Shafer managed a large PC product group producing award-winning audio components for Apple. From the development of low-level software drivers to the selection and monitoring of semiconductor facilities, he has led key product and process efforts. In December, he spoke with Dick Price, the IEEE Computer Society’s Director of Sales and Marketing.
DP:Today I’m talking with Don Shafer, a longtime Computer Society volunteer and friend of mine who is completing a tour as the Society's treasurer. Don, how you doing?
DS: Very good, Dick. Thank you very much for allowing me to do this interview today.
DP: We're conducting these interviews to give people a sense of how they can progress through a field in computer science and software development. Could tell us where you are in your career right now and a little bit about how you got there.
DS: Well, I'd be glad to. Right now—and it seems like only yesterday—but in six months, I will have spent 40 years in engineering and the computer industry. I got my undergraduate degree in 1970, and today, I'm the chief technology officer of a software engineering firm here in Austin, Texas. We've been around 11 years. I'm one of the founders, and we've got international operations where we're doing both systems engineering and software engineering verification and validation of fairly complex control systems for the offshore oil and gas industry.
We've got teams all over the world, from the top of the North Sea to Australia, off Africa, to India and China, doing the commissioning, the analysis, and the verification and validation of control systems on these very large, deep-water oil rigs. Each one of these rigs costs close to a billion dollars to get out of the shipyard, get it configured, and get it onsite so we can start drilling for oil. It's pretty exciting.
Throughout my career is I've always tried to find out what was the exciting part of software engineering and work in those areas.
DP: I believe you travel out to these rigs at times?
DS: Absolutely. I've gone through the all the training. In fact, we got into this business about seven years ago, and at that time, we were doing analysis work for one of our other groups in Houston. I said, “gee, you know, we've got some of these interesting issues with these drilling control systems that don't work, could we go out and take a look at it?”
I had to go back through not only safety training, buying some steel-toed boots, but also had to go through the helicopter training, where they strap you into the helicopter fuselage, turn you upside down and drop you into the water, and you have to go through the escape training.
It's a lot of very interesting stuff for software engineers to do. You don't see a lot of other software engineers out on these rigs because most of them don't want to go through either firefighting school or getting dropped upside down in a swimming pool.
DP: Yeah, I bet. What kinds of things need to be controlled on one of these rigs?
DS: Well, think about where these rigs are, especially in the Gulf of Mexico. One of the first systems we did was a drilling rig that has dynamic positioning. I look over my career and think of the work we did in sensor systems in the 1970s, where we had uplinks to orbiting aircraft—we weren't using satellites that much.
Today, 40 years later, we've got fully dynamically positioned ships. These ships and drilling platforms, their size is pretty hard to grasp unless you've seen one. Some of the largest ones are about 140,000 tons displacement. The largest aircraft carrier in US Navy is only 100,000 tons displacement, so this is larger than the largest aircraft carriers we've ever built.
We're looking at controlling the systems that keep these enormous rigs within 10 meters of a spot on the ocean floor—which is anywhere from 10,000 to 15,000 feet below the rig. They're drilling in 15,000 feet of water, and then once they get to the ocean bottom, they put a whole set of equipment down there to handle different types of protection and the wellhead itself, and then they drill up to five miles beyond that.
So if you can think of the enormity of this and how much software control—wow! This could never have been done without software and without good software controls. Humans just can't keep enough of the data points in their brains to make this happen.
DP: I know you're one of the few fellow Vietnam veterans I'm going to talk to on this series, and I believe you attended one of the military academies. I wondered what factors pulled you from that direction into the career you ultimately pursued.
DS: My father gave me my first soldering iron and breadboard when I was about 7 years old. He was an engineer with Bell Labs. During World War II, he worked on radar systems and then on the original ballistic missile defense system that was put in place in the '50s around Washington, DC. So my bent was always to go into electrical engineering, and I spent a lot of time building radios, doing things like that, building stereos—all that stuff.
I realized that when I was about two years from graduating from high school that I wanted to go to one of the services academies. I had been to West Point; I had been to Annapolis. I had grown up on the East Coast and decided, well, you know, that's not what I really want to do. What I really want to do is go fly. So I went to the Air Force Academy.
My first two years, I thought I was going to be a EE, an electrical engineer. But some time in the second semester of my sophomore year, I went up and talked to the dean of the department, and I said, “you know, all this stuff's really interesting, but I'm not learning anything. I've done all this stuff before.” I explained some of my background and what I'd been doing. "So, when am I going to learn something new?"
He said, "Well, probably when you get to graduate school, but you're not going to learn much new as an undergraduate."
So I decided, okay, I want to change my career. I didn't finish in EE. Instead, I decided to go into astrophysics. When I finished my degree, I had had two years of Double E and two years of astronautics. And then I got my commission in the Air Force.
At that time, we were just using large-scale mainframe computers and everything was on punch cards. When I went into the Air Force and I had done a lot of parachuting at the academy. I was on the freefall parachute team and had several hundred freefalls, and in one of those I had had an accident with a parachute, which knocked me out of going to pilot training.
Instead I went to intelligence school, which many of us feel is a contradiction in terms. But I have papers that say I've been to intelligence school, which is more than some of my colleagues have.
After that, since I had had a background in parachuting and had also gone through Army Airborne School, HALO (High Altitude-Low Opening) School, Ranger School, and I had also gone through underwater demolition training.
So with that, when I went into the Air Force, after I finished my intelligence training, I immediately went into a detachment that was part of Defense Intelligence Agency. I ended up trading that to be a basically an Intel combat controller.
In the mid-1960s McNamara had decided that technology was going to help win the Vietnam war. One of the things that he put in place was a project called Igloo White, which was a set of remote sensors on the Ho Chi Minh Trail. The sensors had been designed by companies like Bell Labs—a lot of folks from the different national laboratories, both Los Alamos and Livermore, worked on the design.
These sensors were either earth penetrators or they hung from trees on parachutes. They had sniffers in them, seismic gear, and acoustic gear. All this data was uplinked through aircraft orbiting over South Vietnam—well, actually, mostly over South Vietnam but other spots as well. We had a large computer center in Thailand where we did all the data reduction.
My team did everything from doing the data-reduction sampling to being the Maytag repairman. I had a team that went into the Ho Chi Minh Trail and checked on the sensors, which is something that causes a very high increase in one's pucker factor.
DP: You said your dad was a big influence as a very young boy. But through your college years, and perhaps graduate school, is there somebody else who played a big role in moving you along in your career?
DS: I was lucky. When I got out of the Air Force, I went right to work for Boeing, where I worked on the AWACS—the Airborne Warning and Control System. It was an interesting project. One person who had a real strong affect on me there was a gentleman by the name of Ben Prince. Ben was one of the senior program managers and then became a vice president of Boeing Aerospace.
I had briefed Ben a number of times on some of the things we found. This was 1972-73, the beginnings of really some of the initial software engineering work that was going on. We were putting together teams with folks like Barry Boehm. As I briefed Ben, he helped me understand how to really stay focused on technology—not just from the research point of view, but to take technology and implement it and build products with it.
DP: As you moved through your career later on, was there one decision that you think was particularly smart on your behalf?
DS: Starting out as an engineer and being very technical, I realized when I was at Boeing and interacting a lot with other aerospace companies, that as an engineer, even as an engineering manager, if I didn't understand how the companies worked and how the organizations worked, I was always going be at a loss. So, in the late 1970s, I decided to get an MBA. I tell everybody that I'm just an engineer with an MBA, and the only reason I got an MBA was purely out of defense, so I knew how badly the finance guys were sticking it to my projects.
For all of us who are technical, whether we're in engineering, whether we're in the private sector, the public sector, or even at a university—if you're the dean of a university department or heading up some projects or getting research funds—you better understand the finances and the business aspects, because if not, you're going to become a victim.
DP: Good. So was there a particularly dumb decision, some place you left too quickly or stayed too long?
DS: Probably, I left a job at Advanced Micro Devices too soon and went to another semiconductor company. I left AMD because I had just finished building up an about 160-person team and through no fault of ours—we were hitting all of our targets, but one of the major processors tanked—I got a call from the folks in Sunnyvale saying, “We're sorry, we're going to cancel your project, you've got to lay everybody off.”
So I went to another company where got a very nice promotion, but about two weeks afterwards I got a call from the president saying, “We need to talk to you.” And again, back in Sunnyvale—this was another semiconductor company—he said, "I've got some good news for you and some bad news. The good news is you're now going to be in charge of three divisions, and the bad news is you've got to fire half the people."
So I should have stayed where I was, because I just jumped from the frying pan into the fire on that one.
DP: At the top, I said that I know you through your long involvement with the Computer Society. I wonder if you could describe one of your more rewarding experiences as a volunteer.
DS: Right now, I’m ending my three-year tour on the Board of Governors and two years as a Treasurer. Being on the Board of Governors these past three years has been a real challenge. We've made a lot of changes to the Computer Society. One of the more rewarding areas on the Board of Governors was pushing for resolution of the Massachusetts Avenue headquarters, getting that sold and really helping fix our financial position and also get something that could have been a fairly long-term liability off our hands.
When I first started out as a volunteer, I worked as the CS Press editor-in-chief. Working with authors, working with the publishers, working with the staff—it was the first time I really had a chance to work closely with the staff in both Los Alamitos and Washington, DC—that was very rewarding because you saw the tangible results by getting books published.
I couldn't recommend becoming an editor in chief more highly to anybody who wants to really see something get done as a volunteer. It's the hardest job but it's the most rewarding job we've got in the Society.
DP: Right. So either in your day job or in your volunteer work, is there one particularly interesting project you're undertaking right now?
DS: We're putting together a fairly extensive system, a rules-based system, where we're using a set of sensor data that's never been used before, to predict the optimized drilling parameters for some of these long-distance drilling wells.
When you start looking at energy exploration and production, software and software process is about where US manufacturing was in the 1970s. There are no standards, there's very little reuse of anything, and they are not taking advantage of what they can do with software and technology. That's one of the areas where we really focus as a company.
One of the more fun projects right now is being able to take the data that they've had for years and years and years and never used, go through the process of data mining and generating rules from it and saying, “look, if you can go out and change how you drill these parameters, you're going to be able to improve productivity.”
Not only have we changed the science behind it, we've written a couple patents on it, and we've actually improved productivity by about 30 percent. Well, 30 percent on a 10-day well, if you can save three days on that well, you're going to be saving at least a million dollars, if not more. And this is very simple, straightforward software approach to fixing problems.
It really excites me to find new uses in these industries for some of the software that we worked on in research labs 20 or 30 years ago, where we can make a difference, making things safer and more productive.
DP: Other than learning how to jump out of airplanes, what advice could you give to a recent graduate who might want to pursue a career something like yours?
DS: The first thing I'd want to tell them is to get your bachelor's degree in engineering and go get a job. Don't go to graduate school until you've had three or four years, maybe five years of experience. Go out and understand what it takes to get some of this work done. And don't shy away from taking the difficult courses.
You really need to start sometime in the middle of your junior year and make sure you've done some internships—go do internships with companies you're interested in.
DP: In your field, what do you think the most startling advance or change will be in the next five to 10 years?
DS: Oh, the next real big changes are going to be in nanotechnology. We're going to see the smart, clustering nano objects, where you pump a whole bunch of microscopic glass bubbles, which have a certain amount of intelligence in them, into a system and they decide how things are going in the system. They report back and they may even be able to go in and fix things. That's where we're headed.
We think, “Oh heavens, that's absolute science fiction.” Well, if you go back and read science fiction from the 1920s and 30s, they were talking about stuff we're doing today on a regular basis.
DP: Here in the US, what scientific or technological policies would you like to see the Obama Administration pursue?
DS: Two things: One, they've got to get healthcare for everybody. And you say, “Well, gee, you know, you're an engineer, why do you care about that?” Well, I care that the children who are going into school are healthy enough to be in school and their parents are healthy enough to take care of them. Because I want those kids to take advantage of the educational policies that the Administration needs to put in place, so that we get back to where we were in the late 1950s.
Second, we need to have another space program. There's got to be another Sputnik that pushes us out there. The children today who are in elementary school have got to get excited about going into science and engineering. We're going to see an enormous gap if we don't do that.
One of the strengths of our country is the ability to innovate. We've got the freedom to go out and try things, and we've also got the freedom to fail. But the only way we learn and are able to push through technology and really understand how to get new products out the door is by taking chances. Before we can take a chance, we've got to be smart enough to know what we need to take a chance on. So education's got to be one of the top priorities.
- Sumi Helal: Mobile and Pervasive Computing
- Nur A Touba: Design and Test Research
- Deborah Cooper: Reaching the Under-Represented
- George Cybenko: Dorothy and Walter Gramm Professor of Engineering at Dartmouth College
- Sorel Reisman: Technology and Teaching
- Susan K. (Kathy) Land: Software Process Improvement
- Sajal Das: An interview with Sajal Das
- Don Shafer: Complex Control Systems
- Natalia Juristo: Mastering Experimental Software
- Nigel Shadbolt: Huge Amounts of Connectivity
- Shmuel Shottan: A Passion for What You Do
- Elena Ferrari: Improving Security & Privacy in Social Networks
- Harold Javid: Developing Global Understanding
- Dawn Song: MacArthur Award for Computer Security Specialist Dawn Song