In the late 1960s, you found yourself involved in what was called the "fighter mafia." Where did that name come from?
What made the fighter mafia threatening?
Does technology discourage simple approaches in aircraft design?
What are the advantages of simple and small?
How were these relationships used in the development of the lightweight fighter?
Had anyone looked at the problem this way before?
We were trying to determine the trends. We didn't spend a lot of time looking for exact values. It is one thing to agree that something is better. But how much better is another question. The answer involves finding a trend and asking more questions. Is the design being improved by these actions? How fast is it improving for a given amount of change? The person most responsible for this approach was John Boyd.
What were some of the conventions the fighter mafia challenged?
Why were they accepted as truth?
Did those involved in the early days of the lightweight fighter program take such a historical perspective?
How was this design approach different from the norm?
What was the riskiest portion of your lightweight fighter design?
Were other companies looking at fly-by-wire control systems?
Why didn't they incorporate these technologies?
How was it that you could take these risks?
Why was Northrop unwilling to take the risks involved with the new technologies in their prototype for a lightweight fighter?
Are you comfortable with the title "Father of the F-16"?
In the first part of the interview, you talked about the fighter mafia’s unconventional approach to fighter design. What were some of the conventions you challenged?
Why were they accepted as truth?
We had to reduce weight because we had to use a given engine, the F100, which had been developed for the F-15. John Boyd [a fighter mafia member] had played a part in defining that engine, and he felt comfortable with it. So the engine was fixed. That meant that the thrust was fixed. If we wanted a high thrust-to-weight ratio, we had no choice but to reduce weight.
The range equation can be treated like the thrust-to-weight ratio. The typical approach to increase range is to simply increase fuel capacity. But increasing fuel capacity increases volume, which means more weight and more drag.
People think that big is better. It’s not. With the lightweight fighter, we wanted to achieve our ends through different means. We increased range by reducing size.
Innovation requires breaking rules. But organizations are based on rules. How can this fundamental conflict be overcome to encourage innovation?
The easiest way to change something, unfortunately, is to have a disaster. You take innovative approaches when you have to, when you’re forced to.
Can companies encourage innovation by tolerating eccentricity in the right places, in places like advanced design departments?
Who most influenced your career and how did they influence it?
Then there’s Ed Heineman. He was the equivalent of vice president of engineering for the corporation. Earlier in his career, he was responsible for the A-4, the Heineman Hotrod. He did for the Navy with the A-4 what we did for the Air Force with the F -16. He brought a lot of his enthusiasm to the F-16 program.
There were others who influenced my career, people I debated with frequently, John Boyd and Pierre Sprey, the other core members of the fighter mafia. With them, I was hearing music that I liked, so I danced to the beat.
To sum up, design approaches - Widmer and Heineman. Working relationships - Bill Dietz. Fundamental concepts and approaches for aircraft and their use - John Boyd and Pierre Sprey.
Boyd and Sprey would later admonish you for not sticking to the fighter mafia’s original intent summed up by the group’s motto “make it simple.” They fault the aircraft for getting heavy and overloaded with gadgetry. What is your response?
The F -16 has far exceeded my expectations. However, if I had realized at the time that the airplane would have been used as a multimission, primarily an air-to-surface airplane as it is used now, I would have designed it differently.
Is this difference represented by the F-16XL?
With the F-16XL, we reduced the drag of the weapon carriage by sixty-three percent. The drag of the XL with the same fuel and twice as many bombs is a little over thirty percent less than today’s F-16 when you load it up. This points up a fallacy that has existed for thirty years, and I’m concerned that it may still exist. Our designs assume clean airplanes. Bombs and all the other crap are added on as an afterthought. These add-ons not only increase drag but they also ruin the handling qualities. They should be considered from the beginning.
We ought to start with the weapon. That’s really the final product. We ought to determine what the weapon is and what it will take to deliver it and then do the airplane. Now, we design the airplane and smash the weapon on it.
Is anything being done to address your concerns?
What is your conception of concurrent engineering, and why isn’t it being used to address these concerns?
I can’t fault that objective. But I don’t think it can be achieved by simply glamorizing a process. We wouldn’t need to do these things if we had the right attitude and dedication to begin with. The relationship between the Air Force and defense contractors is also important. The Air Force would be well-advised to concentrate on what to and leave the how to to the contractors. I think you would end up with a better, lower-cost product.
Furthermore, this system - concurrent engineering - does not let you select a seed that you can cultivate. You need a skeleton from which to start. We now have people working on a project before we know if its skeleton has seventeen ribs or four ribs, three or two arms.
Is concurrent engineering made necessary because people are specializing?
Is this approach affected by the way engineers are educated?
So, specialization is a problem?
Most people assume that lengthy development times increase cost. That’s a false assumption. You don’t necessarily save money by shortening development time. It’s what you do during the development time that matters. That gets to the heart of my concerns about concurrent engineering and integrated product development. We want to bring everyone on board right away to avoid problems later on. But you can’t have thirty-three engineers run through one door at the same time. You have to queue up in some fashion.
I will end up with a better and cheaper product if you let me add six or nine months onto a schedule and give me some time to sit back and contemplate what I’ve done. Our programs, with their tight schedules and payments conditioned on meeting those schedules, just don’t allow us to do that.
I know of no airplane flying that didn’t go through an additional phase of development. For example, after building the first eleven F-l11s, we found out that it was not right. The design needed some changes. So the next airplanes were changed.
Those first eleven airplanes were really prototypes, though they weren’t done under a formal prototype contract. The F -15 went through the same sort of thing. McDonnell built so many, then they changed them. People don’t go back and examine the history of these programs. You always have a certain number of production planes that never make it to the operational units.
Did those involved in the early days of the lightweight fighter program take a historical perspective?
They started with the P-51 Mustang. The minimum increases in cost in jumping from one airplane to another was a factor of 1.9. They were as high as 3.1 in same-year dollars. You can see the increment involved in going to jet engines, to swept wings, to supersonic, missiles, and big radars. You could see what was different between this airplane and that airplane and its effect on cost.
The cost per pound of succeeding airplanes went up at the same rate as the overall cost. This is true even for the F-16. That is, if I plot a curve of cost per pound for succeeding aircraft, the F-16 is right on the curve. Its increment of cost per pound has gone up the same as any other airplane. However, if I plot a curve of unit flyaway cost, the F-16 falls off that curve. It reversed the upward trend in unit flyaway cost. It was the only aircraft to do this. So the way we got the cost down was by getting the size down. That was another motivation for reducing size.
Has this approach been taken for aircraft following the F-16?
As much as I think things are a matter of attitude, here’s a case where attitude alone is not going to get you where you want to go. You must incorporate certain features to have low cost.
I’m concerned that we’re only saluting the flag when it comes to low cost. I don’t see anything specific being done. There are a few exceptions to this. But these are relatively small.
How can affordability be addressed in aircraft design?
You said earlier that specialization is more of an effect of tight schedules than a root problem. How has an overemphasis on schedules resulted in specialization?
So we create reliability groups, parade them over here, and give them their own vice presidents . Now the designer has even less reason to address reliability because there is this other group that does this. Pretty soon, you have reliability engineers talking to reliability engineers.
Like the tight schedules, our increased size is also getting us in trouble. For the lightweight fighter, we had a small group. We had one maintainability engineer and one reliability engineer, not a huge organization or organizations devoted to reliability and maintainability. Again, this gets back to schedules.
All our efforts on integration within the organization seem to have the opposite effect: they have pulled us apart into little factions.
You seem to be dissatisfied with integration attempted from within an organization. Can external factors lead to integration?
Technology itself may have an effect. With today’s computer systems, we have a better chance of integration because they’re changing the way we communicate
One of the problems we’ve had over the years has been our principal vehicle of communication - a drawing. With a drawing, coordination can only come about by two means. Either people gather around the drawing to see what was going on or they take the drawing and copy it and pass it around. Both of these are intrusive and interfere with the designer. A computer drawing, on the other hand, can be immediately transferred to other computers, to other places, without interrupting the person doing the work. So, technology may bring about new ways of organizing engineering groups.
With the lightweight fighter, you seemed to have disregarded the organization. You worked outside of it, or at least out of its sight. Two questions: Is this an accurate observation? Could it be repeated today?
In answer to your second question, I refer to the prototype phase of the lightweight fighter as a Camelot- a bright, shining era that will never return. Nobody told me to start working on the airplane. I just did. I could because of the environment that existed at the time. Nobody told me to do it. Nobody told me to stop.
The system was such that it would let me do it. Today, we’re more concerned with style instead of substance. You couldn’t do it today because it would look wrong. We have too many controls. So-called controls. I don’t think they really control that much. If they did, I don’t think GD would have run into what it did with the A-12.
On the lightweight fighter prototype, the entire SPO [System Program Office] consisted of five people. Five people were just enough where you could have hands-on, face-to- face control, not paper-to-paper control. We had several advantages. The Air Force got exactly the same data, in the same format, as we prepared it, in virgin form. The information didn’t have to go through various cycles to meet some military format. And it didn’t go through a bunch of approval cycles. The way our engineer prepared it was the way the SPO engineer saw it. Because we didn’t have to go through all this formalization, they also got to see it a lot sooner.
What brought about all this formalization?
The result is a formal system that insulates engineers from their counterparts on the other side. We don’t have that face-to-face contact that I mentioned. We are rarely asked to communicate with people outside our own group, let alone our own company.
As your structure becomes more complex, you lose that personal contact. Back in the B-58 days, the people working for Bob Widmer were all located in the same place. We could see each other. Our offices were next to each other. We’d find ourselves in big shouting matches. We weren’t concerned about using direct language because we were close. We knew who we were talking to. When we were through shouting, we knew where we were going and why we were going in that direction. Today, you don’t have shouting matches. Because of this, you don’t know the background to the decisions being made. This is a result of our size and complexity.
The formalization also has a lot to do with the attitude of management. Management involves a lot of give and take. While engineers tend to think more in terms of black-and-white decisions, management does not usually involve problems with black-and-white answers. That’s one reason good engineers don’t necessarily make good managers. Some managers just want to dictate. They never get out of their offices and talk to their employees. That’s what we miss most, that face-to-face contact. We’re in a culture or an era in our culture in which we don’t value contention. We place a greater value on conformity.
Instead of asking what’s good for the company, which reflects on what is good for the customer, people are more often asking what’s good for their own personal pursuits. Some people base their decisions entirely on what they think their boss wants to hear.
In the early days of the F-16, our energy was focused on the airplane. We had to promote our design. We had to promote not only the airplane, but also the concept behind it. We were breeding a thoroughbred, but that thoroughbred wouldn’t do any good if there wasn’t a race for it to enter. That’s something GD was very good at, that is, promoting the concept behind a plane, seeing that it fulfilled an Air Force need.
Today, I’m not so sure companies do that. I don’t think they have the people who look that far into the future. They’re more concerned with near-term problems and personal pursuits. We have too many people thinking about how they are going to come out in pursuit of something as opposed to how the company will come out. Now I don’t really fault them for the pursuit of their personal interests. For me, the company’s interests and my interests have always coincided. As a company grows, however, that basic relationship can be lost.
What part does communication, or lack of communication, play?
More importantly, you have to believe in something in order to articulate it. Today, it is harder for people to assess their roles in an organization because organizations are so large and because there are so many specialties. This uncertainty makes it more difficult to have convictions. It makes it more difficult to be confident.
I always knew where I stood with the people I worked for. It’s extremely difficult to work for people who never tell you what they’re thinking. I’d much rather work for someone who criticizes me than someone who says nothing. When you hear nothing, you tend to assume the worst.
Our education and training play an important role, too. What little report writing we get in school, for example, usually discourages the use of active voice. This practice is often reinforced at work. Active voice requires a subject. When you associate yourself with your subject, you are more likely to feel more responsible for what you write. It’s ‘easier to express confidence.
The proposal for the lightweight fighter was written in active voice. We used action titles on all our figures, and we threw out a full third of the figures that various people submitted for the proposal because they could not come up with an action title. Our rationale was that if you can’t think of an action title for a figure, that figure doesn’t say anything. It doesn’t convey a message.
How was the proposal received by the Air Force?
We were pretty let down by this. We figured our proposal wasn’t good enough for them to ask us any questions. We thought they just weren’t interested. Like I mentioned earlier, when you hear nothing, you assume the worst. We assumed the worst.
It turns out we were wrong. We won the competition far and away. We learned later that they didn’t have any questions because the proposal was so clear and concise. It turned out to be a classic. The ASD said that it was the best proposal it had ever seen. General Dynamics and other companies subsequently used the proposal as a model.
What made your proposal so good?
How has the way proposals are evaluated changed since that time?
Back then, you could offset a higher cost if you could show technical superiority over your competition. But now you’re judged to be technically acceptable or unacceptable. So everybody is normalized. Today, cost and politics playa much larger role.
This approach diminishes the advantages of technically superior companies like General Dynamics. In my mind, GD is far superior to any other company in fighter aircraft design, with the possible exception of McDonnell Douglas. I’d say GD and McDonnell Douglas are about equal, but not equal in the same sense. McDonnell Douglas does some things better than GD, and GD does some things better than McDonnell Douglas. But the system for evaluation no longer accounts for these differences. It wipes them out.
That’s not good. The system inhibits companies from taking technological risks.
Has the system changed, become normalized, because there are fewer people who are capable of making these highly technical distinctions?
The reason behind the changes gets back to distrust. We seem to be consumed with all the standards and ethics.
Your current activities seem to concentrate on aerospace systems of the future. Do they include space?
He talked about a system of space-based sensors that would provide continuous data on the origins of acid rain, flooding, weather patterns, and a host of other things that impact on our well being. I can now see the potential for some direct benefit to the inhabitants of this increasingly ravished planet. Before I get into trouble, I should quickly add that my attitude on space does not carry over into hypersonic vehicles, such as the National Aerospace Plane, which, by my definition, is a transatmospheric vehicle, not a space vehicle. I am a strong supporter of hypersonics.
What is the biggest challenge facing today’s aerospace engineers?
Lockheed’s F-117 Stealth is an extreme example. The plane represents everything you would not do for both performance and flying qualities. But the design works because it has a flight control system that is a direct descendant of the F-16’s flight control system.
The answer to your question goes back to communication. Our biggest challenge relates to integration and interfaces. There is no one person today that can, within his own discipline, come up with a decent system. We may have seven engineers working on a system. They might talk to each other. And even when they do communicate well, they don’t really understand what impact their function has on another function. You will always have the problem of getting people to work together.