Oral-History:Arthur Stern (1993)

About Arthur P. Stern

Arthur P. Stern

Born in 1925 in Budapest, Hungary, Arthur Stern attended rabbinical seminaries and the University of Budapest. In 1944, he was deported to concentration camp Bergen Belsen. Upon liberation, he attended the Universities of Lausanne and Geneva and earned a Dipl. Ing. degree in Electrical Engineering in 1948, at the Swiss Federal Institute of Technology, Zurich, Switzerland, where he later became an instructor. In 1955, he obtained an M.E.E. from Syracuse University.

He came to America and joined General Electric in 1951, where he pioneered in color television. He published the first technical paper on transistor radios, did seminal work in transistor circuits and led GE’s initial efforts in integrated circuits. He left GE to be Engineering Director of Martin Marietta’s Electronics Division, and later Operations Director of its subsidiary Bunker-Ramo Corp. He became Vice President of The Magnavox Company in 1966, where he guided development of spread spectrum systems and led the introduction of satellite navigation to commercial ships and to navies worldwide. He was a leader in developing key elements of the Global Positioning System. He retired in 1991 as Vice Chairman of Magnavox and President of Magnavox Advanced Products and Systems Company.

Stern was also a long time IEEE volunteer leader, and served as IEEE President in 1975. For a discussion of Stern's IEEE activities see Arthur P. Stern Oral History (2009.)

A Fellow of the IEEE, he was a recipient the IEEE Centennial and Millennium Medals.

About the Interview

Arthur P. Stern: An Interview Conducted by William Aspray, IEEE History Center, July 13, 1993

Interview # 175 for the IEEE History Center, The Institute of Electrical and Electronics Engineers, Inc.

Copyright Statement

This manuscript is being made available for research purposes only. All literary rights in the manuscript, including the right to publish, are reserved to the IEEE History Center. No part of the manuscript may be quoted for publication without the written permission of the Director of IEEE History Center.

Request for permission to quote for publication should be addressed to the IEEE History Center Oral History Program, 39 Union Street, New Brunswick, NJ 08901-8538 USA. It should include identification of the specific passages to be quoted, anticipated use of the passages, and identification of the user.

It is recommended that this oral history be cited as follows:

Arthur P. Stern, an oral history conducted in 1993 by William Aspray, IEEE History Center, New Brunswick, NJ, USA.

Interview

Interviewee: Arthur P. Stern

Interviewer: William Aspray

Place: Edison, New Jersey

Date: July 13, 1993

Background and Education

Aspray:

Why don't we start by having you tell me something about your background: where you were born and what your early education was like?

Stern:

I was born in Budapest, Hungary, in 1925, to a religious orthodox Jewish family. I went to elementary school, and then to middle school, both orthodox Jewish schools. Then I went to "Yeshivas," i.e. pre-rabbinic schools, up to the age of 17. While I was doing this, I was also pursuing my secular studies. I finally joined the Gymnasium of the Budapest Jewish Community, where I graduated in 1943 with what was called a Maturity Certificate. I've just come back from Hungary from the 50th anniversary of that. I studied law in Budapest from 1943 to 1944. In 1944 our Holocaust started and we were deported to the concentration camp Bergen Belsen in Germany. From Germany, toward the end of the war, I got to Switzerland. It was there that I decided to switch to engineering. I studied at the University of Lausanne and then at the Swiss Federal Institute of Technology in Zurich, where I obtained my engineering degree in 1948. A Swiss "professional" engineering degree (Dipl. Ing.) is somewhat similar to a master's degree in this country.

Aspray:

What was the field?

Stern:

Electronics. I think you can call it electronics, but electronics was very different at that time. It was electrical engineering with some specialization in high-frequency technology. I then worked for an industrial firm in Switzerland.

General Electric and Martin-Marietta

Stern:

I taught at the Institute of Technology in Zurich for about a year and a half before I came to the United States in 1951 and joined General Electric's Electronics Laboratory in Syracuse, New York. I had the exhilarating experience of participating in the development of the country's color TV system from 1951 to 1952. It was a wonderful opportunity to make some contributions.

In late 1952 I felt that transistor technology was far enough advanced to develop a transistor radio receiver. I was appointed project engineer on that job by the management of GE's Electronics Laboratory. In early 1954 we were the first ones to demonstrate and publish a comprehensive paper on transistor radio receivers. Shortly thereafter I became manager of the GE Electronics Laboratory's Advanced Circuits group. It was my first foray into management. I was responsible for the development of all kinds of solid-state techniques, systems and subsystems in that capacity.

In 1957 I became manager of the Electronic Devices and Applications Laboratory, which gave me responsibility over things like lasers, masers and various solid-state devices, semiconductors, magnetics, dielectrics, etc. I stayed in that position until 1961, when I left GE to serve as director of engineering of Martin-Marietta's Electronics Division in Baltimore.

Aspray:

Why did you make that move?

Stern:

I got fairly high up in the research and development hierarchy of General Electric. I was sort of boxed in. My salary level was too high to switch over into engineering without taking a loss in grade. Then Martin-Marietta came along and offered me the management and development of a large engineering organization, just on their faith that I would be able to do it even though my entire background was in R&D. So I took it, and that was an interesting thing. I stayed with them from 1961 to 1966. It was a very good change. The Martin-Marietta Electronics Division merged in 1964 with TRW's Computer Division. These two together formed a jointly owned subsidiary called Bunker-Ramo Corporation. I became director of engineering and then director of operations of Bunker-Ramo. The latter position included responsibility for everything except human resources and marketing. I was number two there like I had been at GE since 1957.

Then Magnavox came along in late 1965. They had the Magnavox Research Laboratories on the West Coast, and that was an opportunity for becoming number one. I switched in early 1966 from the Martin-Marietta empire to Magnavox, and I stayed with them until my retirement in January of 1991. I spent about 25 years there.

During my association with GE, after I switched from color television to solid state, I had an opportunity to be a member of the group of people, centered mostly on IEEE that promoted and developed solid-state technology throughout the country. There were maybe thirty or forty people throughout the United States who were leading in that area, and that was a tremendous experience. Later at Martin-Marietta I was involved more in the development of substantial military systems, anti-submarine warfare, missile guidance, and command-control systems. At Magnavox I had an opportunity to start new things and see some of them through. In particular, we established spread spectrum communications in the world. It was quite difficult because you needed a lot of components and integrated electronics wasn't around yet at a sufficiently high level of integration. It was tricky to keep these very complicated systems and subsystems working for more than a few hours at a time without failing. But then, of course, as solid state came along, these things became very practical. Today everybody uses spread spectrum technology.

Magnavox

Satellite Navigation

Stern:

The next interesting challenge was satellite navigation, where we at Magnavox became world leaders. We brought satellite navigation to commercial shipping and a variety of other commercial applications such as high-precision position location for surveying. Then came the next generation of satellite navigation, which was the GPS system. GPS stands for "global positioning system." We started it conceptually under a subcontract with Hughes Aircraft. There we built the first practical demonstration of system feasibility. We won several competitive contracts starting in 1968, only to lose the big production contract to Rockwell in 1985, 17 years after we started. But it was an exciting experience. We did a lot of new things. I think I was very lucky to be associated with people who enabled me to make a small impact on the lives of many people. I retired in early 1991, and since that time I've been consulting and participating in a variety of volunteer and committee activities.

Corporate History

Aspray:

Can you tell me about the nature of Magnavox's business? How was it structured? In what areas did it start?

Stern:

Magnavox was established, if I'm not mistaken, in 1908. Within the past decade we had our 75th anniversary. It was named Magnavox, which comes from Latin and literally means "magna vox" or "big voice." They were in the loudspeaker business. Then from the loudspeaker business they went into a variety of other areas. During the Depression years it was difficult to survive. They almost declared bankruptcy in the late 1930s. Then World War II requirements for electronics rescued them. They became a substantial military producer. After the war they went into television and later into color television, while maintaining their defense activities.

By the time I joined in early 1966, the company was in home electronics, including color TV, hi-fi consoles, etc., a fair amount of military and industrial electronics, and some non-electric activities such as furniture. Furniture was very important for electronic consumer products at that time. In those days the hi-fi set would be a large nice-looking furniture console with some electronics in it.

Aspray:

Yes. My parents had such a Magnavox product in their home.

Stern:

Then things were difficult in the early 1970s because of Japanese competition. The Japanese had a better understanding of what the consumer market needed and made many technological innovations. They implemented solid state in TV at a time when Magnavox still used tubes. They came up with modular consumer electronic systems, while Magnavox still tried to sell big pieces of furniture. Things got tough. The company was very profitable at the time I joined it and for a few years thereafter, but then it started losing money.

Philips Acquisition of Magnavox

Stern:

Philips, the Dutch electronics giant, via its North American Philips arm, made a tender offer for Magnavox and in 1974 acquired Magnavox in toto. Philips did this in order to acquire Magnavox's consumer electronics business, and at the time they were hardly aware of the Magnavox defense and industrial electronics businesses. We told them after they acquired us. But they were very good to Magnavox for many years. They gave us a tremendous amount of freedom. They chopped up Magnavox into a variety of companies. One of them, the Magnavox Government and Industrial Electronics Company (MAGIEC), dealt with the defense and industrial business. I was first a vice president and then a senior vice president of that company. My organization grew from a laboratory to a pretty sizeable product organization between 1966 (when I joined) and 1980 (when it was established as a separate company). I then became president of what became known as Magnavox Advanced Products and Systems Company (MAPSC) and stayed there until my retirement.

Aspray:

Why did Philips decide to break it up into separate companies?

Stern:

Philips management principles at the time called for very limited interference into the businesses that they owned. They were delegating as much of the responsibility as possible to the local management of the business. Philips management at the central level recognized that they were not experts in many industries in which they were engaged. In the U.S., Philips was a real conglomerate. (Later, in 1989-1990 when Philips got into financial trouble, they sold most of their non-core, peripheral businesses.) So it was in line with the company's thinking to delegate a lot of responsibility, and in turn to encourage businesses that could stand on their own. They would give the management of a business authority to do whatever they could do, and then hold them accountable for the results. After acquiring Magnavox, itself somewhat of a conglomerate, Philips divided Magnavox into its component business elements and gave each its autonomy.

Aspray:

What sorts of things were centralized within the Magnavox group or within Philips itself? Research? Capital? Marketing?

Stern:

If you consider the period starting with 1974, at the time of Philips' acquisition of Magnavox, and ending in the late 1980s, Philips really did not involve itself in any significant business decision, except when we needed substantial capital. We went to them, and they either made the capital available or not. They generally did. As I recall, they hardly ever didn't. If we wanted to do something big, like building a new facility, they exercised the authority to decide whether to own or lease the facility. In other words, they assumed a more active role whenever substantial capital amounts were involved. Otherwise, Philips did not interfere. As a matter of fact they couldn't because, as a predominantly defense business, we were under the special supervision of the U.S. Department of Defense (DOD). Philips was not even supposed to know (and they didn't know) about the classified things we were doing. By agreement with the DOD, they were not allowed to interfere in any way. They behaved like a bank. We would simply give them a major presentation at the end of each year on what we intended to do next year, mostly in financial terms. We would give them a little information about technology, whatever was not classified. We also gave a mid-year presentation, which was shorter. That was about it. We were governed by our management and by our Board of Trustees, which was appointed jointly by Philips and the DOD.

The central Magnavox management was more active in managing things. There were several companies and organizations that reported to the central Magnavox management, including mine. The central management allocated part of the research and development funds for us. They also approved or disapproved major bids — those in excess of $10 million. Otherwise, during the last ten years I ran my company pretty independently.

Nature of the Defense Business

Aspray:

I don't know much about the defense business. What are the kinds of things that make a defense company successful?

Stern:

There was considerable difference between the two major components of Magnavox: the so-called Magnavox Electronic Systems Company (MESC) headquartered at Fort Wayne, Indiana, and my company, MAPSC, headquartered in Torrance, California. MESC was 95 percent defense while my company was about 65 percent defense and 35 percent industrial. I really had very little to do with what was going on elsewhere. In my company the backbone of the defense business was spread spectrum systems and subsystems. We were responsible for applied research, technique development, product development, and ultimately the production of some major military communications systems. In particular we developed, designed, and produced the core of the country's successive strategic communications systems: the AN/URC-55, AN/URC-61, and the AN/USC-28. The systems we designed had all kinds of ramifications. For example, when the President of the United States — don't forget that those were Cold War days — left the country to meet with Gorbachev in Iceland, he required instant access to the country's strategic assets, including the nuclear capabilities. So Air Force One carried one of our systems, which made it possible for him to have access at all times to the levels of military management with which he had to be in contact.

The strategic communications network was rather extensive. Around the world there were many terminals. It was expensive, very important stuff. This system used protected communications. Protected communications means that you can communicate with your friends without the enemy being able to interfere or listen in on what you are talking about. The enemy or potential enemy would not even know that you were communicating. This is the kind of system that we developed and manufactured.

Aspray:

What did the business of that division amount to?

Stern:

That division was doing somewhere on the order of $80 million a year. Not very large but nevertheless a sizeable business. It was quite profitable and always on the edge of the technology because improvements had to be generated all the time. Of course the military budget in the 1970s and 1980s was sufficiently large that these almost continuous improvements could be sustained financially. Thus multiple generations of these systems came about, and the equipments that we built in the late 1980s were not comparable to the equipments we built in the mid-1960s — although the fundamental principles remained the same.

Aspray:

How does a division like that plan for future developments in a product area?

Stern:

Obviously, any organization engaged in that type of activity must constantly be aware of technological trends. It is impossible to be successful in a business of this type without knowing exactly what's going on and without being able to do something about it. You are engaged in continuous technology monitoring and technology forecasting in a very practical fashion. You know that the customer expects to use our equipment for somewhere between five and ten, maybe even more, years. You know that the customer would not view your company favorably if you pursue major modifications every six months. But you also know that the customer will consider you unreasonable and not really on the ball if you didn't come up with potential improvements within the next year or two that will be ready in three to four years.

So you have activity all the time. You are constantly evaluating new technology. You are forecasting which parts of the new technology will last, which make sense and which don't make sense. Sometimes you make the wrong bet. Sometimes you do not perceive that a new technology really impinges upon you, and then you are late. In other cases you bet on some new technology that doesn't make it. We had our share of both. It's a continuous process; it's a way of life. The function of top management is to allocate resources in such a way that your access to new technology is assured, and that you have the right people, who in the majority of cases make the right decisions and bet on valid kinds of technology. This is not necessarily a very efficient process. It takes people of different abilities and you have to be careful of the type of abilities that you are accumulating. You also have to be very careful of the type of people because some people are tremendously productive and others are not. They may have gotten the same degrees in the same year at the same institution, and the one who is not productive may have had a much higher grade point average than the other who's productive.

Managing & Monitoring Technical Business

Aspray:

Is there a special set of techniques that are used for doing your forecasts?

Stern:

I don't know how special these techniques are. People who are in top management in a highly technological organization, while not necessarily technically creative, must be on the ball and must know what is going on. I always considered it one of my responsibilities to know what's going on technologically.

Aspray:

Did most of your senior managers have engineering backgrounds at some point in their career?

Stern:

No, I wouldn't say that. My chief financial officer did not. But even in that kind of a position it is good to have some technical understanding, and the bright person is going to pick it up very fast. You have an important combination of people who have the technical savvy, the manufacturing savvy, and the marketing savvy. In order to succeed, they'd better know everything. There will be a redundancy of knowledge and a redundancy of objective, which they can work out among themselves, obviously with my participation. Then when the chips are down, the person responsible for the whole thing — in our case, it was me — will make some good decisions and some bad ones. I made both. I made some outstanding decisions and I made some outstandingly lousy decisions.

Aspray:

What about technical monitoring? How do you do that?

Stern:

We had a number of excellent technical people on our staff. We conducted design reviews for every activity on a regular basis. That must be part of life. You cannot let technical groups of people go off on their own. Not because they are not bright. The reason actually is that they may become too bright. You must have a number of crucial things in your organization if you want to be successful. One of them is standardization, and that means that you ought to tell the bright engineer, "I know you want to use this component, or this approach, or this tool. And I know that it appears to you as superior. Who knows? Maybe it actually is superior. But we can have only so many approaches inside any organization. Otherwise, we are fractionated. So, let me tell you something: We are not going to use that." That's part of life. Incidentally, that results in resignations and in upset and angry people. That's part of dealing with technical people. It's exciting, it's productive in a practical sense, and it's very rewarding. Sometimes it's absolutely frustrating. The young engineer says, "But I know that this is better than what we have used." And you have to be firm, "Leave it that way. You may very well be right. As a matter of fact, I know it. But we can only have so many approaches. And this is one that I'm not going to use. I'm going to wait for the next generation and skip that one. I'd rather skip two stairs than one." Many young people will not accept this.

Aspray:

What about monitoring what's going on in the government laboratories, the academic sector, and your competitors? How do you do that?

Stern:

If you are set up reasonably well, that's not a problem. Are we still talking about defense electronics?

Aspray:

Yes.

Stern:

As I mentioned, that was two-thirds, and in fact, the less interesting two-thirds, of my organization. I was spending most of my time on the one-third that was commercial. But monitoring is not much of a problem. If you are set up properly, your marketing people will be dealing with a variety of government customers and will learn from them what's going on and what's being planned on the system and subsystem level. On the techniques and technology level, your engineering people will follow up on their own or because you insist on it, and they will participate in relevant meetings of technical societies such as IEEE, NSIA, etc. But I should say that the IEEE is no longer the home for engineers that it was in my time. The members will follow what's going on technologically. But a lot of them just read what the magazines have, and it's often outdated by the time it's published. The leading people, however, will go to these meetings, and they will meet people informally. They will know their peers in other organizations. And in no time you will know what's really going on, well before the publication of it.

I didn't find that with an organization of reasonable size and reasonable emphasis on gathering technical intelligence that it is terribly difficult to know what is going on. You do this legally. It's all clean and there is nothing improper about it. To me the problem never was knowing what was going on. You knew it practically as soon as it was thought of — with some exceptions, of course. I don't want to exaggerate here. There were very few things that were successfully kept quiet by some companies. Usually they don't succeed because people are migrating, and this is a free society. More important, people who are plugged in talk with each other. The problem is how to have a modicum of control and what to do about it. Marx said — he is not frequently quoted today — well over a century ago that German philosophers have evaluated and talked about the world for years. The issue is not what we talk about. The issue is what we do about it. That's the situation with technical management: What do you do? Do you sit at these meetings and pick up information or have your people come in and tell you, "I picked up this, I picked up that." There comes a point when you have to say, "Now we do." And that's difficult.

Satellite Navigation Sales to Shipping Companies

Stern:

I recall one of the more important success and failure stories of my career. In 1969, when our Fort Wayne sister division failed to exploit satellite navigation profitability and when the company's top management was about to terminate the satellite navigation activity, they decided to offer it to my division if we wanted it. This was before the acquisition by Philips. It meant moving it from Indiana to California. We decided at that time that this was very exciting. So starting in July 1969 we were engaged in satellite navigation with the objective to make it a commercial product. Just to give you a little background, satellite navigation was really established by the U.S. Government, with the intent to equip the country's Polaris submarines and their ballistic missiles with precise position-location and navigation systems. When you shoot something off and you want it to go exactly where you want it to go, you've got to know where you are with high precision. This is not only good advice for life in general but it is absolutely mandatory for missiles. That which you deliver will arrive with lower precision than the precision of your location. So if you've misjudged your own location by two miles, the missile will not fall on the target but two miles off — at least. Then there are additional errors, which accumulate placing the missile five miles or ten miles away. So precise position location from the point of view of moving ballistic missiles was essential. Johns Hopkins University's Applied Physics Laboratory developed a very fine position-location and navigation system using satellites. They were called Transit satellites or the NNSS — Navy Navigation Satellite System.

This was a tremendous opportunity. Vice President Humphrey announced in 1968 that we will make a large part of this technology developed for Polaris submarines available to the world at large for commercial applications so the world can benefit from these improvements in satellite navigation. So we said, fine. We went for that. ITT made the same decision at the time. This became — between 1969 and 1973 — an area of intense competition between ITT and us. We finally beat them out, and then there was just us. We made a lot of mistakes in doing this. We did the job, but we did not become a big commercial success. The market wasn't ready. It took a lot of learning.

I remember my first trip to Norway in 1970. At that time Norway was a major power in commercial shipping. We met with one shipping company after another to learn what they really wanted in precision navigation. We learned after a few fascinating meetings that in each shipping company there were two powers: the director or vice president of operations and the director or vice president of engineering. The guy in engineering always wanted the best. He wanted complete control. The vice president of operations said, "No. The 83 captains that I have around the world on 83 ships don't want us to know where they are and what they are doing all the time." The director of engineering would say, "I want a permanent record. I want the printout of where this ship was all the time." The director of operations would say, "Look, you're crazy! You don't understand! The captains will sabotage the system. We don't want this thing." We learned that our systems had to be satisfactory not only from a shipping company's technical viewpoint, but also from its internal politics perspective.

We also learned that military reliability was not good enough for the commercial shipping people. If you're going to use our equipment on a ship, and it fails within a year, they'll send it back and never order any more. Formal acceptance doesn't really mean a lot. If they accepted it and it fails, they won't reorder. Whereas the military at that time, if they accepted it and then it failed, they would pay you well to repair it. So, there is a big difference.

For two or three years we concentrated on the competition with ITT and learning about the shipping industry. After we won, we still didn't have products that were sufficiently popular. We had an engineer in our organization then who was really bright. He said, "you know, we've got to come up with a system using microprocessors." Microprocessors were quite new then. We got together, debated it and agreed to do that. So in mid-1976, we came out with an instant success! It was the sort of thing that is almost a once-in-a-lifetime event: You develop the thing, you sell a few of them, and you get wonderful feedback. Suddenly you don't know how you're going to fill the orders. That's what happened to us. We were in a dream in the second half of 1976. We tried to increase our capacity. We were somewhat lucky because I decided that in the original run we would produce 250 instead of the usual fifty.

By the end of 1976, we knew the product was a tremendous success. I got the guys together, and I said, "You know, this is great! And it's going to last for a while, but everybody knows about it. Somebody is going to come up with something better within the next two years. Therefore we're going to design something new and do something much better." We established the priorities on what this "something better" needed to be and development started.

Let me make a long story very short. I became involved in many other activities. I did not check on the progress. I assumed that the written directive which defined in fair detail what should be done would be followed. I found a couple of years later, when we were supposed to have the product, that we were far from being through, and furthermore the directive was ignored. A number of expensive technical "innovations" were introduced that shouldn't have been there. These were bright things — new, daring, super high-tech — but too expensive. We had to regroup. Instead of two years, it took us four years, including a major redirection after two years. It was a catastrophe. We did finally recover from this, but we had several down years in the meantime, and we had a loss of market share which should have never been.

Why do I relate this? It was one of the really bad things in my career. You cannot walk away from a project. You cannot say, "well, I've given instructions, and everybody agreed, and it's going on." Unless you constantly monitor, unless you constantly emphasize, unless you issue minor redirections every two or three months as they may become necessary, you are bound to have problems. If you have looked the other way for a year the redirections will become major and you will have lost time. That's the way life is.

Doing Business with the Government

Aspray:

In regard to the defense business, I have this impression of business lock-in. Once you get in with a system, you continue to modify it and improve it. You have a great advantage over your competitors in continuing business with the government. Can you talk about that kind of market and how you protected it?

Stern:

Well, to a large extent it is true. At least it was true during most of my career. Toward the end of my career it changed somewhat.

<flashmp3>175_-_stern_-_clip_1.mp3</flashmp3>

Each business has its traditions and its ethics, which are not necessarily the ethics of others and of other businesses, and which sometimes are not even rational or in conformance with general ethics. But a business has its practices, which then become its ethics unless something happens — which actually did occur in the 1980s when the Department of Defense suddenly declared certain practices of the defense industry illegal by enforcing laws which were previously ignored. Some people went to jail or were fined. But that's also a question of monitoring. When you have laws, you've got to monitor your organization for compliance with the laws. The fact is, for a long time during the first 35 years of my career in the defense industry, from 1951 when I joined GE until the mid 1980s, the key to success was to win the R&D phase of a new system. You did that by underbidding it. You could overrun it if it was a cost-plus situation, but if you overdid it, that made the customer very angry. So you frequently decided to swallow much of the cost overrun even though the contract was cost-plus because you didn't want to make the customer angry. Typically, you lost a lot because either it was a fixed price and you had to lose, or the customer relations were such that it was better for you to lose than to compel the customer to come up with the last penny he had, thereby earning his undying enmity. But once you were through with that, nobody had a chance of taking the production away from you. And on the production you did well. From the initial production you recovered that which you lost in R&D, and then as things went on for years, you made a lot of money.

This approach was supported by a number of things. It was supported by the procurement methods that were used. What do I mean by that? First of all, a low R&D bid generally got the award, even though everybody, including the evaluating government officer or team knew or should have known that you would be losing money on it. Or, there would be an overrun and they would have to come up with more money. They had, in principle, the right to say, "Oh, you are underbidding. You are overly optimistic. Therefore, we are going to give the job to the other guy." It happened on rare occasions. It was a great exception. Suppose one was coming in with a fixed-price bid of a million dollars; they would suspect it's going to be two or three million, but they aren't going to do anything about it. So that was one procurement practice that contributed to this kind of thing.

The second one was tricky from a legal viewpoint. When you bid the production phase, which was for a fixed price in most cases, you had to face price negotiations. Price negotiations are a very sensitive thing because you must comply with "truth in negotiations" requirements. That means if in the system you are going to use six widgets and you have the vendor bids for these at, say one dollar each, you put them into my bid at a dollar each. If this is then agreed on in the negotiations, there is a limited profit on that production element, say a profit of 10 percent. So you get one dollar and 10 cents for every dollar cost. But you may suspect that, when the chips are down, and particularly in the second and third phases of production, you're going to buy these things for 60 cents each. If you have evidence that it will be 60 cents and not one dollar, you must reveal it to the government during the negotiations. But if you have no evidence, you don't have to reveal hunches. So you can make a tremendous amount of money because you have this production for years. You are buying these things at the beginning for a dollar a piece. Then you buy them for 60 cents. Sometimes you may buy them for 20 cents. The government would still pay one dollar and ten cents. And with that, you made up the losses that you had in development — and much more. It was a very good business. This is of course a highly simplified example. Reality was more complicated but, in essence, the same and very profitable.

Aspray:

But that means that your parent company has to have fairly deep pockets for a while because the payoffs are pretty long term.

Stern:

That's right. That's the way it worked. Much money had to be put into certain things until the money came back. But then it came back in large amounts. Obviously a small company couldn't afford to do that, and obviously this all tilted things toward the big companies. These were the practices, and there were frequent arguments with the government about truth in negotiations. A very difficult situation, which we had, was the following: Out went solicitations for this cup, which we needed. We got an offer from one vendor for a dollar each. This was incorporated into our bid. The bid went out on July 1. The government found out that one of our engineers, without reporting it to management, had already discussed on June 25 with another vendor the possibility of changing this cup slightly so that it would cost 50 cents. These were often quite honest mistakes. But communications inside the company were not that good. So there were a lot of misunderstandings. All this then changed in the mid-1980s when the government became very touchy about things, particularly as available defense funds decreased. Today things are much more complicated because the government is more precise in its evaluations. Life has become more difficult and more bureaucratic.

Expanding a Defense Business

Aspray:

In the defense business, how would you expand your business? Did you try to build up your market share? Did you just assume that you'd be making more money on these products that you already had?

Stern:

It depended on the kind of company that you were. There were always companies who were primarily production companies who were built on taking the production away from the original developer. For example, Magnavox developed the VRC-12 army field communication system and produced it for I don't know how many years. Then somebody came on the seventh or eighth production run and underbid us. They took the business away from us. These were usually comparatively low technology and therefore low overhead companies who were good in manufacturing techniques and were able to manufacture inexpensively. Full-range companies started with applied research and stayed in the business through initial production and services or even longer. Once something was somebody else's, we generally forgot about it. We didn't go after that because it was a bad deal. I would usually not go after something that Hughes Aircraft had already had. But that's why you have marketing people, and you have technical people. You could get into it in two ways: One is that you internally initiate new functions based on the needs of the government, as you perceive them. If you proposed things that were satisfying a need and ultimately satisfying a formal requirement, and if your proposal was a reasonable, near optimum balance type proposal, then while others will have an opportunity to bid, you have a good chance of getting that contract. From then on you're in. In other cases, the need was identified by the government rather than by you. You learned about it relatively early because you have marketing people with contacts in the most important government agencies. Then you started working toward satisfying the need. You worked with the government in developing technical methods by which the need can be satisfied. But in both cases, whether you originated the lead or the government initiated the lead, you were knowledgeable of the need and proposed something new. So you expanded your business by getting into something new.

One example of the exciting things that we got into was just at the time when we lost the big GPS competition to Rockwell. We bid on a project for which the government had a need, which we perceived very early. It was for a field facsimile system. Many Japanese companies make facsimile systems, but I'm talking about a military field facsimile system. That's a facsimile system that you can drop from a helicopter, and it will still continue to work. If you work in a physical environment, which is very hard, you can't buy it for $500. It will cost many thousands of dollars. We proposed this, and it happened to be in 1985, just at the time we lost the GPS. The government decided in our favor in the competition. That became a major multi-hundred-million-dollar business. We had developed the first version military fax in 1972 or so. Then we modified it, and modified it again. We sold a few to some odd government agencies. By odd I mean not the ones that buy a lot. Then we sold it overseas. I remember the Israeli Army bought some. Then came the serious interest of the U.S. Government, so we redesigned the whole thing. That which we produced had nothing to do with the various models that we had already sold. We finally sold to the U.S. Army, and created a lot of new technology at the right time. Timing is very important.

Industrial Sector Business

Aspray:

You said earlier that two-thirds of your responsibility was in the defense business and one-third of it was in an industrial business. Do you want to talk more about the industrial side?

Stern:

Important parts of the industrial group were the satellite navigation business and communications for ships, primarily. But then we went after a variety of other types of vehicles and also for precision survey instrumentation. It was very inspiring because it was really new. At some point I got sick of the military marketing process that I described earlier. The commercial business was pure competition and more exciting. We did the right thing, and we knew what the results were. In some cases we lost. In many other interesting instances, we had tremendous success. For over a decade starting about 1976 we had the majority of the world market in both satellite navigation and in precision position location.

Aspray:

Was the product life expected to be shorter in the industrial sector than in defense?

Stern:

It was hoped to be longer, but it didn't work out that way. In commercial electronics, starting around 1970, there was no product life of any length. The technology moved so fast that you had to get used to recovering your investment over three or four years, or you were out. So in our field, satellite navigation, precision position location, and satellite communications for ships, we just had to get used to the idea that we would have three or four years, and that's it. That had to do with the rapid changes in technology, which then had an impact on the amount of time that companies took to develop. When you knew that three or four years from now entirely new technologies would be available, you could not take three or four years to develop a product. So you had to learn how to develop in a year or a year and a half.

I gave you an example before where instead of two years it took us four years, and that was useless because by the time we came out with this, a new technology had arrived. You had to shorten your development cycle, and you had to develop special techniques to do so. You don't shorten your development by simply saying, "Okay, in the past it took you six months to develop a circuit. Well, now you take only three months to develop a circuit." You had to come up with simultaneous engineering and manufacturing developments that previously were sequential. You had to come up with methods that were unheard of in the 'fifties or the 'sixties. You had to pull the whole team together and tell them, "between such time and such time you're going to do the development work, and your manufacturing and quality people will be involved from the beginning and you will productize as part of the development effort. Manufacturing will be in the picture from day one, and they will provide instant feedback, and then you are going to be able to produce it." You have to have all of them — design engineering, software design, production engineering, manufacturing, and quality assurance — in there together right from the beginning to make sure that when the engineer uses nuts and bolts, that the manufacturing people say, "yeah, we can use those nuts and bolts, and they are in the right spot."

Aspray:

So they don't have to completely re-engineer the product.

Stern:

That's right. You had to do the electrical and mechanical things in parallel that were historically done in sequence. That really changed our entire development process and our decision process. The decision process was very strongly affected by it because many management decisions, which were usually done a year later, had to be done right away.

Aspray:

Did you have to retrain management staff because of that?

Stern:

Yes. We had quite a few people who were able to adapt.

Aspray:

But, for example, did you decrease the number of layers of management?

Stern:

Oh, yes. First of all, we went over to what at that time were somewhat less conventional methods of development. The defense industry had these for a long time because defense systems were so complex that it became clear relatively early that rather than organizing into engineering, manufacturing, and so forth, you had to have product teams to put them together. The concept of project managers was developed in the 1950s, where the basic unit was the project rather than functional organizations: engineering, manufacturing, and so forth. This became mandatory in the industrial area also, to put together a project team that was self-sufficient and competent to do everything that was necessary. They had some consultants from the outside because they couldn't get the best people for every one of the teams. People like Charlie Cahn, who was our chief scientist, worked on many project teams. But fundamentally there was a project manager, and he generally had his project instructions. The progress was measured against those instructions. Of course, the instructions were not immutable, but they were usually followed meticulously. I told you before what happened when I didn't look at something; it was a catastrophe. Incidentally, the distance between success and catastrophe is very small. One or two minor mistakes can cause the difference between success and failure.

In the particular case that I mentioned of our third-generation satellite navigation system that didn't succeed, what really caused the failure was an improper interpretation of a written instruction. The instruction clearly pointed out that the major objective of the new development would be reduced recurring costs. Specifically, reduced manufacturing cost. The second objective was reduced development cost. Now, when you give instructions of this type, they must be interpreted intelligently and loyally. They are clear. That means you may spend a little bit more for development if it saves you on the recurring manufacturing costs. The third objective was the smallest possible size and objective number four was a special feature compatible with the first three objectives. That priority list was neglected. For the sake of new product features, expensive specialty components were developed. Huge costs. It therefore contradicted priority number one.

Aspray:

I see.

Stern:

I recall, on the other hand, another case where a similar thing happened, but it was caught. That must have been around 1978, the development of our first high-precision position survey instrument. I caught it in the early design phase. I just said, "You start all over." But it was within a few months from start, and there was time to start it all over. Engineers all have the tendency to go for all kinds of product features and embellishments, so-called "bells and whistles." Particularly in the defense industry. I feel it is going to be very difficult to change the defense industry to commercial products, as people have been discussing, because there are major cultural differences. You can't change a culture easily. If the culture of your engineering organization puts features and size onto a higher emotional priority than cost, then they are going to fail in a commercial market. I'm not trying to generalize in some nasty way but engineers will have a tendency to do as much as possible for technical characteristics which are not essential. That is their pursuit of beauty.

Aspray:

In this defense environment does that also apply to your manufacturing engineers, who might be more concerned about things such as reliability than about cost in manufacturing?

Stern:

In defense everything is very costly. In the commercial business, things are conceptually pure. That doesn't mean that you are not making mistakes. But they are relatively pure. You know that in a certain business, the customer will pay only so much for a feature, and that fundamentally this customer wants to do something specific with this equipment. The thing must be delivered for that specific purpose at the lowest possible price with the highest possible quality. The commercial customer almost never falls in love with fancy technical features. There is no such thing. Again, I'm saying there isn't, but of course there is. In the military business it was a multi-variable decision. First of all, the manufacturing engineers generally were not accustomed to the idea of lowest possible cost. But even if they were, the individual customers — and the individual customer frequently was an engineer — decided many things. If the individual customer wanted to have a certain feature, even though it upset the apple cart and cost more, he got it. They knew that it was going to result in substantial cost to the government that was unnecessary. They knew that it would reduce our profit in cases of fixed price contracts. They knew that in some cases it was crazy. But we would do it anyway because that customer controlled much of our income. Being friendly with him was important. Nothing pure. In the commercial business, everything is very pure. Again, there are exceptions.

Industrial Products & Military R&D

Aspray:

The examples you've given from your industrial products business were ones that were based upon technologies developed for the military. Was that true in every case in your business?

Stern:

That was our predominant experience; because that's what I chose at the time I accepted that failing product development in satellite navigation. I felt that taking military technology and applying it with appropriate changes to the commercial market would give us a special position and a special opportunity. So that's what I did.

The interplay between military and commercial products has really been the substance of my career, and it's inevitably somewhat narrower than it could have been. But to go back to earlier years when I was very active in solid state, we applied solid state in military programs at a time when there would have been absolutely no economical way of doing that in commercial products. So I think that there was a great contribution made by the U.S. military to the advancement of technology, both in my organization and other organizations — obviously I was only a small cog in this huge system.

This military-commercial arrangement was both good and bad. It's good in the sense that the U.S. military became the sponsor of technology at the time when others probably could not have sponsored it for a variety of reasons. When the U.S. military sponsored semiconductor technology and integrated circuits technology, other technologically advanced countries were recovering from World War II and they didn't have the means to do any of that. Look at the publications. In the 1950s and early 1960s, we were competing for being first in this and first in that on an individual basis in terms of publications. All of us worked in the U.S. There were no significant publications originating in France or Germany or Japan. There were a few in the U.K. So my only concern at GE in the 'fifties, when I was in a high publishing mode, was what does that person at Stanford do, and what does the other person at MIT do, and what do the people at BTL or at the RCA Research Laboratory in Princeton do? I wasn't really concerned what the people did in Germany or France or Japan.

We published the first paper on transistor AM broadcast receivers in the January 1954 issue of Electrical Engineering. There was a race to get it published because we at GE knew that RCA in Princeton was working on the same thing, and it was a question of where to publish: in the Proceedings of the IRE or Electrical Engineering, which was published by AIEE. I made the decision to submit it to Electrical Engineering, where we would have a better chance of an early publication. It was a good bet. But within weeks RCA had the same thing. Building an AM broadcast receiver, even though transistors were not very good, was not that much of a miracle. The interesting thing was, and at that time we didn't know that this would soon become a common pattern, neither GE nor RCA built the first commercial transistor radios. Instead, it was a Japanese company we had never heard of. Their name was Sony. They are quite well known today.

Moving from Defense to Commercial Industry

Aspray:

Do you want to make a few additional comments about this great hope today about moving from a defense industry to a commercialized industry? I know you have made the comment about corporate culture before. Do you want to say more about this?

Stern:

I think that most of that is politicians' talk. Politicians from the President of the United States down can talk about so-called "conversion" from military to commercial, but to do it is really a different story. If you look at the term "conversion" itself, it sounds and is simplistic. Maybe I'm too literal, largely because I have a fairly strong Latin background. I was asked to go to the Soviet Union a year and a half ago to help them in military to commercial conversion. They wouldn't even be able to convert from inefficient military to efficient military. But to speak of converting from military to commercial operations...that's dreaming. It is understandable that large military factories which have a lot of floor space and a lot of employees, for the sake of maintaining reasonable social conditions, should want to use that space and whatever machinery that can be saved in some useful manner. It is the essence of our public responsibility, to make people reasonably happy, and to keep them usefully active. But to say we are going to take this factory and convert it from military to one that produces competitive commercial products just cannot be done.

Aspray:

Do you see a solution to this political problem, or the problem that the politicians are facing?

Stern:

We are getting into the area of social conditions and political predictions. I believe that we are looking at difficult times to come. I think that the strong reduction of the defense industry in this country is going to have a permanent effect on us, and it's not going to be easy to replace. We have thousands and thousands of highly skilled workers and highly skilled engineers — technical personnel with magnificent technical backgrounds who have become permanent surplus. We have a situation in my state, California, where people are moving out. For the first time in the history of California, people are moving out rather than moving in. Some people will say that's not true, that others are coming in. Most of the people who move in come from Mexico, Guatemala, etc. So maybe the total population of California is going up, but the standard of living is decreasing. A lot of valuable people are leaving because there's no work for them.

A similar situation exists in Europe, although the causes are a bit different. Today we are paying for the leadership that we have given to the world in resisting the Soviet Union. We have led the so-called free world in providing a counterweight to the Soviet Union, and we have spent a lot of our budget, a lot of our resources, on defense. Meanwhile the Japanese, for example, have hardly spent any and said, "We will let you defend us. And you are going to pay for it." Today, here we stand. We have a technically brilliant but politically antiquated and largely useless defense industry while they have a magnificent commercial industry. I went into retirement at the right time.

Aspray:

I see. New challenges for new people.

Stern:

Yes. It's partly challenging, partly tragic. The situation requires a lot of patience and thinking that is really new. Many people in this country — certainly the ruling set of the country — are not prepared for it. It may require conclusions like this: That in the past you saw that everybody who demands a reasonable living must work for it, and that somehow the economy will provide the work. So that when the economy is in high gear, everybody who wants to have a decent living will find decent work. Occasionally there will be low points, and somehow we will get over them. We never really said, "Why should people not be entitled to a reasonable living at those low points." We always got out of the valleys and scaled the next peak. We may have to make up our minds that people are entitled to a decent living; regardless of whether they work or don't work because there never will be enough work for everybody in this society again. But try to explain that to a righteous Republican, or even the many Democrats. I'm convinced that we have reached that point. I think that if we don't do that, we are going to have an increasing number of homeless, an increasing number of beggars, an increasing number of college graduate destitutes, seething inner cities and more and more crime. I don't expect to be around to see what the world is going to do about it. But you will.

Value of IEEE Membership

Aspray:

That's right. You said earlier that the IEEE is not as valuable an organization for the young engineer today as it was before. Can you explain that?

Stern:

I'm not sure I can explain it.

Aspray:

Could you elaborate then?

Stern:

I will say this. At the time when I was a young engineer — I came to this country in 1951 — one of my first professional actions in this country was to join the IEEE. It was the organization to belong to. It was the organization which had the right publications. For an electrical engineer, it was the organization which from a social point of view was the organization you wanted to belong to and you wanted to be part of. In turn, it was an organization that satisfied most needs of the young engineer at that point. Let's not forget that at that time a very large fraction of young graduates, somewhere between a third and a half, worked in the defense industry. R&D was very important for a large fraction of IEEE (actually its predecessors: IRE and AIEE) and for a large fraction of engineering throughout the country. In the engineering groups to which I belonged in the 1950s, the vast majority were members of IEEE.

Toward the end of my career, due to my personal loyalty to IEEE and because of my experience with IEEE, I tried to push my people mildly into being members. I didn't succeed. By the end of the 1980s, too many of our people IEEE had become irrelevant. Management pressure was of no use. I don't know what the story is numerically, but my conviction is that only a small minority of engineering graduates — electrical engineering, electronics engineering, software engineering, computer engineering graduates — are joining the IEEE.

Aspray:

What is it that's lacking in what the IEEE has to offer now?

Stern:

I don't think I can answer that in any satisfactory fashion. I did not do a thorough analysis. I would say that, from what I have picked up, IEEE is an elite organization for the older people, for the R&D people, for the people who have international interests. But for the typical engineer, who is doing everyday work in software, engineering design, marketing, or particularly manufacturing, IEEE is irrelevant. Even though there are IEEE societies that are marginally involved in practical areas, it's still considered an R&D type organization. That plus the deteriorating economy has had a very negative impact. Having put so much effort into IEEE as president, officer, board and committee member, from a personal point of view, I deplore the situation greatly. I remember the first engineering group I belonged to in GE. Everybody was a member of IEEE (at that time, IRE). That's why as a young man I became an associate member of IEEE even before knowing why. But I also know that in the 1980s when as company president I put a lot of emphasis on IEEE membership, people talked about it. "Will I be penalized by the old man if I don't join IEEE?" As the Romans said, "Tempora mutantur"... times change. And they will continue to change. I hope that on balance the changes will work out for the good of most people. That will take time, a great deal of time.