As he was nearing the end of his active research career, Dave Rutledge took a bold step. After decades of work in electrical engineering, Rutledge embarked on a completely new academic initiative: what are the true numbers, models, and data analyses that can provide an accurate and comprehensive overview of the supply and demand trends of energy? It is a topic as complex as it is important for global society. Rutledge demonstrates (by doing more than telling) that this research area sorely needed an engineer's hard-nosed perspective and laser focus on the data, while allowing for others to consider the economic and political implications of his findings. While the topic was new for Rutledge, he was no stranger to book writing. Rutledge's earlier treatise, The Electronics of Radio is a standard text on the engineering behind radio and wireless communications.
In the discussions below, Rutledge recounts his upbringing in Texas and his educational trajectory that took him to Williams College, where his first interest was in biology. In his last year as an undergraduate, Rutledge began to focus his efforts in electrical engineering. Following a formative experience at Cambridge University, Rutledge completed his graduate work at UC Berkeley, where he remembers applied physics, electrical engineering, and computer science all housed together. He reflects on the value of his experience at General Dynamics between college and graduate school, and Rutledge conveys why joining the Caltech faculty was the easiest and best professional decision he made.
At Caltech, Rutledge expanded his research agenda in integrated circuits, semiconductor devices, software design, and cellular communication (to name a few), and he describes the rich opportunities that Caltech offers for new and sometimes unexpected collaborations. He has served as executive officer for Electrical Engineering, and from 2005-2008 he was division chair for the Division of Engineering and Applied Science. His reflections on administrative service provide a master class in the art of academic recruiting, and in fostering an intellectual environment that allows faculty to excel. In retirement, Rutledge quips that his goal is only to look at screens for help in navigating his adventures on boats and camping trips.
DAVID ZIERLER: This is David Zierler, Director of the Caltech Heritage Project. It is Thursday, December 15th, 2022. I am delighted to be here with Professor David Rutledge. David, it's great to be with you. Thank you so much for joining me today.
DAVID RUTLEDGE: You're welcome.
ZIERLER: To start, would you please tell me your title and affiliation here at Caltech?
RUTLEDGE: The current one is Kiyo and Eiko Tomiyasu Professor of Engineering, Emeritus.
ZIERLER: When did you go emeritus?
ZIERLER: COVID aside, of course, and all of the ways that that upended all of our lives, what are some of the things that you've been studying academically since going emeritus?
RUTLEDGE: The reason I went emeritus then is that I had a book writing project that was just giving me fits. There are many stories of senior professors who have a book they're working on, and it never gets finished. I thought, "If I just quit"—which is going emeritus—"then I can finish the book." My wife, Dale, put up with that, because obviously it changes your finances, not having a salary, but it worked. The book was finished at the end of 2019. I retired at the beginning of 2018, so it took two more years, but the book did get finished.
ZIERLER: This is Energy Supply and Demand you're referring to?
ZIERLER: To go back to your title, being named in honor of the Tomiyasus, can you tell me a little bit about them, and if there's any connection to Caltech or to your research?
RUTLEDGE: I know them very well. They have both passed away now. Eiko lived longer than Kiyo. She was at Mount San Antonio Gardens in Pomona, and Dale and I used to go see her before the virus. The reason I know them is that Kiyo was in the area of research I started off in, and he was the editor of the major microwave journal, Transactions on Microwave Theory and Techniques, in the early days, and Eiko helped him. I was the editor many years later. But I had always known him. He stayed active in the Society throughout his entire career, 60 years or so. He lived in Valley Forge, Pennsylvania, but he had no children, and he was an enthusiastic donor to Caltech. There is a little more to the story, if you have the time, because he's a very interesting man, and Eiko is a very interesting woman. He was just a working engineer there, but he saved his money. Then when he retired, very senior, they moved to Mount San Antonio Gardens. That was basically so he could be, I think, closer to Caltech. There are a lot of Caltech/JPL people in that facility. We visited them often there.
The story that they had, and his loyalty to Caltech, was that he was interested in science—engineering, I should say; electrical engineering—from the very beginning. He was from Las Vegas. His family was pretty prominent there. They were very successful farmers, within, I think, the current limits of Las Vegas. He came to Caltech as an undergraduate. He's the class of 1940. From the timing, after he graduated, it was a year later that the Japanese attacked Pearl Harbor. His family was prominent enough locally that the sheriff apparently warned his family that he would be put in one of the camps if he stayed, so he applied and went to graduate school at Harvard. It was safe there. He got his PhD from Harvard. His wife Eiko is from Japantown, in San Francisco. She was in one of the camps, Topaz, but was able to leave during the war to go to Hunter College in New York. There's another member of the family who is significant. Kiyo's sister, Uwami, was a prominent pathologist at the Veterans Administration Hospital in West Los Angeles. When Kiyo gave money for the endowed chair, his sister also contributed. I thought that was a nice family connection that they had. They were wonderful people, very much appreciated in the Microwave Society. He funded many awards within the Institute of Electrical and Electronics Engineers.
Late Career Interest in Energy
ZIERLER: An overall question to gain some understanding about your research interests—how far back have you been thinking about things that might be outside the boundaries of what electrical engineers might be concerned with, namely oil and gas, energy supply, and climate change? How far back do those issues and concerns go for you?
RUTLEDGE: The timing is very specific. I had a stint as division chair, 2005 to 2008. One of the other division chairs, Tom Tombrello, who was in Physics, invited a geology professor from Princeton, Ken Deffeyes, to give a lecture. At the time, November, 2005, there was a great deal of concern about oil supplies. This has been a recurring theme since the beginning of oil production in the 1800s. And it was before the fracking revolution. The first wells had been fracked in 1998, but there was no appreciation at that time at all, by people professionally, or by amateurs like me, that they would affect the long-term supply. I went to Deffeyes' talk. He was one of these people who was very flamboyant. I don't know if you've ever read the John McPhee series on American geology?
RUTLEDGE: McPhee has made a fabulous contribution with those books. Deffeyes is one of his characters. McPhee's approach is that he goes around with people in their jobs. I, in my classes, have used his stories on trucking and on trains. Who works at Caltech and knows a train driver? Not many people. Anyway, Deffeyes was one of the people in McPhee's books, and his expertise was Western geology. He's a very competent geologist, and also an oil supply guy. He had some simple—at least simple for scientists and engineers—mathematical models that he used, logistic fits. I was curious, because I thought the math was kind of fun, so I tried them on some data. The interesting thing to me is that Tom Tombrello, as a sabbatical, had been the research director of Schlumberger, which is a major well services company. While he was the research director, he had sponsored some work like Deffeyes' in trying to estimate long-term trends in oil and gas production. My sense when I looked at the math and tried it out—it really didn't work that well for oil and gas, but just for fun, I looked at some coal records.
At that point, my wife, Dale Yee, was my research assistant. I didn't have any graduate students in this area. She made a complete record of essentially all coal production by all countries, over all time. The thing that I found, to my surprise, was that whereas the logistic fits didn't really work that well for oil and gas—they're only one piece of information—they worked pretty well for coal. I think part of the issue is that with coal, for some of the major producers we have a complete production history—for example, the UK produced something like half the coal in the world in the nineteenth century. But coal in the UK is dead now, and the decline started around the time of the First World War. Then we have our own coalfields. The early production for the US was anthracite in Pennsylvania. The anthracite fields had the same pattern. The mathematical models, while they didn't do that well for oil and gas, did very well for coal. For each of the major areas that had declined—France, Japan, the UK, the Ruhr, and the Pennsylvania anthracite fields—the logistic fit predicted within 20% or so what the eventual total production would be.
Obviously I knew nothing about coal. But as division chair, I had some time on my hands. I view that position as a faculty development job. Caltech is unusual in that the division chair has nothing to do with teaching. They have separate deans for that. So, basically you support faculty, and on the weekends, I could do spreadsheets. That's what my coal studies were, and they were fun. I went around to visit people and I gave some talks. It turned out that the people in the Geological Survey in Denver who were responsible for coal resources picked up on it. I went to visit them and talked, and they turned out to be absolutely wonderful people, in every sense of the word. Fabulous human beings and great scientists. They were aware of a longstanding problem in coal resources. When geologists go out and measure a coal resource, they get a lot of numbers. They take the thicknesses of the seams, they calculate the resources for each of the seams, they add it all up in a spreadsheet, and that's their resource. Well, it turned out that they were grossly overestimating what would actually be produced. It's like a factor of five. It's very significant. The people at the Geological Survey knew this was a problem. But it turned out that my logistic calculations, which anyone with a first-year calculus class in college could do as long as they had the data my wife collected—the long-term production data—gave estimates that were much more accurate than the geologists were getting from digging thousands of wells and taking core samples. It's not to say that you don't do the geologist's measurements—you certainly need them if you're going to design a mine—but it wasn't helping estimate what we might produce in the long run.
Another aspect is that the decline of coal production is as much a social issue as a geological one. Certainly production is affected by the climate change discussion now, but in earlier times, it was affected by the fact that societies were getting richer. Underground coal mining is a pretty tough life, and a lot of people don't want to go down and work in the mines. As the society gets richer, they have alternatives. But if the people in a village do not have family members working in the mines, they won't support the miners when the kitchen rattles from the ground settling as the coal is taken out. There's a social movement away from coal as a country gets richer. The UK certainly showed signs of this, and we have also. The logistic fit seems to pick up this effect even though it's a social factor, not a technical one.
So that's how I got into the problem. It was just to do something on the weekend, but it turned out to be interesting. As a division chair, 99% of the work is talking to people, all day. You have 100 people who think I'm their boss—plus maybe 30 or 40 staff. It's a lot of people, so there's always someone coming in who wants to talk. Then you're trying to recruit new faculty, which is the main job, and what I think a division chair should be judged on. How good are the people that that the chair recruited?
At that time, people were concerned about energy supplies, and they wanted to hear talks from different perspectives. Most of my talks—I gave something like a hundred talks—were to climate groups of various sorts. There was already quite a bit of interest in 2005 in climate. The climate scientists have very little background in fossil-fuel resources. It's just not part of the education of a climate scientist. There's plenty of physics for them to study, but coal supplies isn't something they would pick up—they'd have to go to the mining engineering department, and their university probably doesn't even have a mining engineering department.
ZIERLER: Thinking about the chronology, circa 2005, Hurricane Katrina, Al Gore's An Inconvenient Truth came out. Was there an upswell of interest in climate change, beyond Caltech, beyond being division chair?
RUTLEDGE: I didn't come at it that way. My focus was really historical, in these old mining records. Appalachian people get looked down on, and certainly it's not our richest area. I went to school in Massachusetts in the Appalachian Mountains, and I was fascinated by the idea that clues to our future might be buried in old Appalachian coal records. I was being the amateur historian. You might say it was business history—and social history, because the early history of the labor movement is coal mining. It really is. In that sense, it has another attraction to me, that it is a very significant part of our society that is fading. The fact that their history might be important for understanding our future—that is what intrigued me. The climate audience, those were just my customers. They were the ones who were mainly interested in what the coal supplies might be, but it wasn't that I had much of an interest in climate science, or expertise.
ZIERLER: In terms of your contributions, some of the key conclusions of the book, with your background, with your scholarly expertise more in electrical engineering, how do you think that might have influenced the way the project came about and some of the ways that you made conclusions that might have been different from, say, an economist or a historian, or even a geologist?
RUTLEDGE: Division chairs are perfect outsiders, because the other part of the job is the tenure call. This very much requires you to get into the nuts and bolts of a field that you're not expert in, and you have to make some hard decisions. These affect people's lives, so you're very sensitive to it. After recruiting the professor, the next part of the job is the tenure call. At Caltech, we're relatively easy on tenure. Most faculty get tenure, so it's not as critical as the hiring decision, but it's still life-changing if you don't get tenure. It comes at a horrible time in terms of the candidate's family. So, you spend quite a bit of time on the decision. Tenure is basically a vote of division chairs. You've got the six chairs, and the provost only votes if there's a tie. In practice, there is quite a bit of deference to the person who is the chair of the division that the candidate is from. So, you really have to get into the weeds, and some areas are extremely technical. You just have to face it. It's hard. You spend a lot of time, you call people, you talk to people. You need humility.
I had encouragement by the geologists, who could have just brushed me away, but they didn't. They invited me to contribute a paper to a coal mining journal. They were the leading people in the field. They were just aware that their models weren't working. The starting point for their resource models is only nominally economics, meaning if you have an oil resource that costs a thousand dollars a barrel to produce, it's likely that that barrel will never be produced. But that's a pretty simple call, how much does it cost to produce it? That's not really an economist's job; that's probably more of an accountant and an engineer's job to figure out how much it would cost to produce.
In terms of the overall pattern for the book, it really was just what I was finding out in teaching an energy class. There was a professor, Dave Goodwin, who died of cancer—and he died pretty young. He had taught the class. I was coming back to teaching after finishing as division chair, and I picked up the course. I had help from Joe Shepherd, who is very much an expert in combustion, which again is rare at a university, that you have someone who can answer any question about engines. Most research universities don't have that person. He co-taught the class for a few years, and then he got hooked into being vice president for student affairs. He pulled out gradually, and then I took over, trying to learn the stuff he was teaching by sitting in on his lectures. I found it hard, a lot of new material for me. If you're teaching a Caltech class, you'd like to teach to some depth. The electrical parts, I knew. I had a very traditional electrical engineering undergraduate program, because I went to school in England, and they were probably ten years behind, in a good sense [laughs], the American education, which was really headed towards electronics. Electronics is where I ended up, but I also had the traditional training in electrical machines, and in high-voltage power transmission, which was not part of the modern electrical engineering curriculum in the US, even when I was a student.
I'll get off track just a little bit. My engineering education started after I graduated from college. When I was an undergrad, I was at a liberal arts college, Williams, which didn't have engineering. I majored in math, which I very much enjoyed, but the only thing that I had figured out by the time I had graduated was that I wasn't going to do mathematics for a living. I was going to have to find some other way to make a living. This was before PCs, and the math jobs were teaching jobs. The 1970s was a time after the expansion of universities, so the jobs had all been filled up. Nothing looked very promising. So, I started over in engineering, and then got a bachelor's degree in electrical engineering. I was sick of school at that point and I got a job in my hometown. Fort Worth, when I was there, is not the current Fort Worth that you see on, well COPS, or whatever. It's a big city now. When I was there, it was called Cowtown—200,000, like Pasadena, and you really had two places you could work. One was in what we called the bomber plant, which made military aircraft, and the other was the stockyards. I worked in the bomber plant on the F-16, so that was really my introduction to engineering, and it was an excellent one. The F-16 turned out to be a great program, and I worked on communications for it. It also taught me that I didn't know enough about engineering to make a contribution to the bomber plant, so then I went back to school. So, my training or experience was not nuts and bolts, but it was kind of higher-level stuff, systems engineering. They are not things that are really taught in the university. You have to spend a lot of time reading, and a lot of time visiting people and talking about problems.
ZIERLER: A devil's advocate question about the premise of the book. If we know that fossil fuels are damaging for global warming, that they are problematic in climate change, why is it important to understand how much reserves we have in oil and gas and coal? Isn't it more important just to decarbonize as fast as possible?
RUTLEDGE: The book does not take a position on that. The book really tries to give students from any perspective the analytical tools that would be commonly used. What it does is in the end, is that it comes up with a way to relate the production of fossil fuels to temperature changes and one that could evolve as you get more data over time. Again, it's really meant for the engineering perspective. The engineering perspective is slightly different from the one you're talking about; at least mine is. It's that engineers spend their lives working for someone. We don't necessarily expect to make policy, and we don't expect to determine truth, in the sense that a scientist would. We have clients. At some level, it all comes down to, we work for clients. Now, if we don't approve of the ethics of the clients, obviously it's proper to find a different client, but that's not what we're talking about here. We don't necessarily expect to make the policy, but we're expected to be able to analyze reality. In the end, the client is the judge of what's going on.
There are two words: reserves and resources. Resources is what climate scientists often use. It's a very general word. It's a much larger number. Reserves turns out to be pretty close the production projections. It's not clear we've changed that production trajectory yet. To give an example, if you look at the production of coal—and all these numbers are worldwide, because carbon dioxide is a well-mixed atmospheric gas, so if coal is produced in the United States, whether the UK stops producing coal doesn't really matter; it's the total. More precisely, China, because it's half of world production. Coal production worldwide has gone up in the last ten years, not down. It's a bit of a warning that US coal production has gone down, but the worldwide production has gone up. Oil and gas production also have gone up. So, we're not at the peak yet, and it's not clear how much control we have over that, in the sense that if you look at the countries that are developed—people often use the OECD [Organization for Economic Co-operation and Development] as the definition of developed countries. I would say probably the lowest-income country that is in the OECD, which is maybe 30-something countries, is Mexico. One way to look at it is, Mexico or richer is in the OECD; lower, less rich than Mexico, is not. The OECD now produces 18% of the world's coal. So, if the developed world fades away over some period of time, it's not a significant drop. When I say significant for climate change, I mean like the way you were using it; like a 90% drop, that would be significant. Currently 17% of the world's energy production is not fossil fuels, and 83% is fossil fuels. Now, the share for fossil fuels is dropping, but it's only dropping at seven-tenths of a percent per year. And it's very much the trajectory you might deduce from the logistic models.
As an engineer, I don't make the assumption that tomorrow the trajectory will be greatly changed by policy. There are too many instances where it did not. I give the example in the book of the Kyoto Protocol, which had basically everyone in the world on board except for the United States and maybe, well, North Korea. [laughs] The company is interesting! Production of fossil fuels didn't go down after the Kyoto agreement. But from the perspective of the book, I didn't try to use policy to predict what would happen. I just tried to use the historical trends. That's not to say I'm disagreeing with the policy, or agreeing. It's just simply not something I view as the part of the engineer's toolbox. If I moved into politics, then obviously I would take a position on it. But if I'm teaching engineering—Caltech is unusual in that most of the students I would see in the energy class were not politically interested. I used to tell them, "You should pay attention to the politics, because whether or not you're interested in the politics, politics is interested in you." Meaning, if you work in energy, your job may be affected one way or the other, by politics, in all sorts of ways, and you need to watch out. Obviously if you're running a coal plant in the UK and you have the Ukraine War—to counter what you're saying—all over Europe they are trying to get their coal plants back online again, because they've got a serious lack of energy that the Russians usually supply. But since they are boycotting the Russian production, they're going to have to make some changes. Now, who would have predicted that, 15 months ago? I'm just saying, "Look out, even if you're not interested. Pay attention. Watch the news." That kind of stuff. But I'd guess in a class of 20 or so, I might have one left-wing and one right-wing student, and 18 kids that just want to come find out about energy.
ZIERLER: To that point, even though it's outside of your bailiwick, do you see anything actionable coming from your research, coming from the book, in the political realm?
RUTLEDGE: Not directly. If I have a book customer who writes me and says, "I work for a group that does ESG [environmental, social, and governance] evaluations"—it's basically the social evaluation of publicly traded stocks, includes climate, and it's very much in the political news now. That's one person that would surely come from the left who is interested in the book. The right, I haven't gotten any sense of whether there is interest in it. Where it would be different on the right is that the right tends to view the energy resources as pretty substantial, so that at least in the time frame that's interesting, the energy resources are not a serious limit. In energy, people tend to be quite polite when they discuss different points of view. That's a point of view that I don't have, but it can definitely be argued, based on the history. When there has been an oil crisis in the past, technology has popped up five years later, and then there's a big supply again. Each crisis looks like it's the last one. Well, maybe it is the last one. That's what I think. I think that in coal, the last new technology has already happened. The major change in technology for coal is surface production—people call it strip mining. That has made a huge difference in terms of the cost and safety of coal mining. That's why over half of our coal is produced in Wyoming where it is mainly surface mining. My sense is that fracking may be the last significant technical change in oil and gas production, also. I could be wrong, and the people that are you might call the optimists in terms of production could be right, that there will be something new that comes up that will save them each time, just because that's the way economics works. As far as I can tell, almost all economists have that view, that resources are essentially unlimited. The right tends to focus more on, "Is the climate science good?" That's not something I'm prepared to tackle in the book. That's a much bigger topic. I take for granted that there is a relationship between CO2 and temperature, and then if you make that assumption, you can just plot the relationship between them. It doesn't go deeply into the science; it's just a correlation.
The correlation that is extremely strong is directly connected to fossil fuels, which I do take a position on. It is the correlation between the CO2 levels and the cumulative production, that is over all time, of fossil fuels, which is extremely tight. Depending on how you linearize it, r-squared [the correlation coefficient squared] is 0.98. But for CO2 and temperature, I just make the assumption that there is a linear relationship, and we estimate it, given that assumption.
Assessing Alternatives to Fossil Fuels
ZIERLER: What are some of the big takeaways that you see in the ramping up of both solar and wind, as alternatives to fossil fuels?
RUTLEDGE: I was pretty careful in this part of the book. The share of alternatives, and by that I include everything—all the biological fuels, wood, gas, and liquids, nuclear, hydro, geothermal, solar, and wind; but solar and wind are clearly in the biggest expansion now—is only 17%. It's only increasing at seven-tenths of a percent per year, and the world is spending a trillion dollars a year on it. It's real money. It's one percent of the GDP. It's going up a bit, but it is not clear to me that it could be 10% of the GDP. With solar and wind, I developed what I would call a toy model, of what would happen if you turned California into a place that does all of its electricity without fossil fuels. The reason that's a bit of a dodge is that California may be the ideal place in the world to do this. To start off, it's rich. It doesn't have significant fossil fuel production. It's still got some oil. It also has the political philosophy to try it. And, it has plenty of room for solar. If you go to an agricultural field that is abandoned, which is what is happening all over the country in the US, that's a perfect place to put solar arrays, because you don't have to worry about endangered species. If it's a corn field that people aren't going to plant, then your environmental impact statements are a lot easier to do.
You get into a discussion of how much do the batteries cost, because you have to cover the time that you don't have solar. It turns out that you would be spending 80% of your money on batteries and only 20% on the solar panels. The society hasn't really faced that large ratio yet. Because it is using the storage that fossil fuels provide. Storage is pretty easy for a coal plant. There's a picture in the book. You can see a big pile of coal, and you can store a month's worth of coal, and it costs nothing except for the earth-moving equipment that you need to move it around. Even oil is in big tanks. Gas is stored in large caverns. Each of these is done on a large scale. But batteries are expensive. Now, if we ever bite the bullet and pay for them, batteries definitely have advantages. Getting electricity from a battery is as good as it gets. You're in control of everything. You can measure all the parameters. You don't have anything that's hot. You can do it at any scale—small, medium, large. It's fabulous, except that the batteries cost a lot of money. That's what we have to face.
The other issue is that there is a solar generation shortfall at times in the winter. Maybe 5% of your production over the year, you're going to have to cover some way besides solar and batteries. The reason is, you're going to have a week with clouds, so you're simply not going to be able to cover that with ten hours of battery storage. With that, I kind of wave my hands and say, "What we would do is get a bunch of what they call gas turbines"—there are different words for this, but it's basically a jet engine, like on an airplane, except it is meant for producing power. They're inexpensive, but you'd need a lot of them. You'd set them up, and you would use biogas for the fuel. There are other ways you could do it. You could also produce the winter electricity with wood. I took a trip last spring, driving through British Columbia, up to Prince Rupert, and it seems like almost all of northern British Columbia is big lumber mills, one after another. It would be very reasonable to generate electricity from wood—the lumber mills do this for their own electricity. That's another possible source. So, there are different ways you could cover the winter gap in California, but if you go to, say, Europe, the problem is that their peak electricity demand in Berlin is in January at 6:00 p.m. Well, it's pitch-black then, whether it's cloudy or not, so there's no solar production at all. In fact, their winter solar production is down by a factor of six from the summer. I do not see how they can make it work.
The problem with wind energy is that a weather system can be about the size of continental Europe, and you can have a high-pressure region centered over Europe for a week, with no wind electricity at all, in the winter, when the demand is the highest. I don't know how to solve that problem. Again, I'm the engineer, so I go after what I perceive someone might have clients for, which is something that could actually be done, and who could pay for it, and that's California. But I don't know how Europe goes to full renewables. Certainly there are lots of smart people thinking, Europeans are rich, and they would like to get off Russian and Iranian oil and gas. It's a very productive society with good allocation of capital, so they'll do the best they can. They'll have to do it before we do, and so we will watch what they do. But I don't know how you solve the problem of one week of no wind, because the wind is their major resource, their main alternative source of energy. I don't know what you do with no wind in Europe for a week in the winter.
Also, everyone is busy shifting. Dale and I visited Norway with Finnish friends a few years ago, and in the town of Bergen, they were shifting to heat pumps for heating, which means they are electric. Norway has hydroelectric power all year, so Norway is not going to be in trouble. But the other people are going to heat pumps too, and so the question is, where are you going to get that electricity? There are different ways. I had a sabbatical in Helsinki, and that one is a little different, because the Finns, their history has led them to be extremely practical. Basically they keep their wood capability, and their heat, a lot of it comes from wood, and they keep their ability to generate that. That's what they're going to be doing this winter, burning a lot of wood. But all of Europe can't do that. Europe doesn't have the forests to handle it. You could get wood from Siberia. [laughs] I think we're not doing that one! Europe is getting some wood—a few millions tons a year, which on this scale is not large—wood pellets from North Carolina. But that gets huge criticism politically. I won't give you a political opinion on that one. I tend to like all of the ways people make energy, so you're not going to get criticism from me of the people that make wood pellets, or biogas, or liquid biofuels, or oil, gas, and coal. I didn't try to solve the world's problems in the book. My goal, when someone comes out of the class and goes to work for an energy company or goes to graduate school in energy studies; to know the basics of where the energy comes from, what the trends are, and some of the basic chemistry and physics that is involved.
ZIERLER: Where is nuclear in all of this?
RUTLEDGE: Ah! At almost all of the talks I gave, someone would ask this at the end. I do take a stand on that one in the book. It's not a political one in the sense that it doesn't necessarily represent my opinion. It's more where the trends are. It's a tragedy, in the classical sense of a play where you kind of know how things will work out. The tragedy is that nuclear has always had the connection with weapons. When you look at where the nuclear weapons around the world came from, most of them have come out of electricity programs. If you look at India, North Korea, Iran, Pakistan, the expertise, some of the technology, initially came from people who set up reactors to produce electricity. Obviously some other things are needed—there's refining—but in the end, I think part of the proliferation problem is these electrical power programs. In our country, it's not the problem, since the bomb came first. But in a different sense it's a problem, because in testing, we set off a few thousand bombs in Nevada. Then for some reason—this is the tragedy—we also decided that the best place to store waste was in Nevada. Well, you can be a Republican or Democrat in Nevada and think, "Look, you set off 3,000 bombs, you gave us plenty of radiation exposure in Las Vegas, why don't you put the waste some other state?" During the Obama administration, they had the power to make that stick. Harry Reid was Senate majority leader—from Nevada—and President Obama was sympathetic. So they deep-sixed the waste program. I think most states will not permit a new nuclear plant unless the federal repository is up and running and accepting the waste. So, most states, you can't build one. You're stuck.
Engineering is full of dreamers. There are engineers who say, "Well, we should make a reactor that is like a submarine reactor. Small power. You just bury the thing, kind of wave your hands over where the waste would go, and you produce the reactor in a factory." I just don't see that coming to much. I think the people doing the permits are still going to say, "Where are the Feds? When are the Feds going to take the waste?" The irony of this is that the low-level waste, which by volume is much larger, is stored in salt mines in southern New Mexico, and they think it's great. It's local jobs. In the salt mine, salt mines are dynamic, so if you put things in salt mines, they get buried because the salt moves around. It's liquid in a long-term sense. So, my sense is that's fine, but they don't view that as suitable for high-level waste. The only place they had in mind was Nevada, so we're stuck. They found a place in Nevada which was a thousand feet over the water table, the driest place in the universe that you could bury something. But it has been stopped, and I don't see how we get it started again. We basically are just waiting for each of the nuclear plants to come offline, and they will, in the timeframe of our discussion, which is this century. In that timeframe, it's just difficult to see how we get out of this rut.
The other thing we're seeing, the Finns are bringing on a nuclear plant right now, and it's the new French design, and it's not going that well. I think in a lot of ways we've lost the magic. A lot of things in engineering, in the early days, are due to a group of exceptional people. Eventually, that expertise dissipates unless it is maintained. There is no reason exceptional people would go into nuclear engineering in the last 20 years. You'd have to have a kind of death wish, in the United States, to sign up for a nuclear engineering program. So, you're not going to get the superstars. In fact, people I've known, like Tony Leonard, who is in the National Academy of Engineering, certainly qualifies as a superstar. He got that for work in aeronautics, but he started off as a nuclear engineer and then changed when he saw the way the wind was blowing. I'm skeptical that we have the engineering skills, even if we got over the environmental permits, which I don't think we will, to build them, at a reasonable cost. I've tracked several of Bill Gates' investments. One is in these small nuclear reactors, and one was in a kind of steam storage up in the Bay Area. I haven't been impressed with his judgment. Now, the steam storage one has failed. It was reasonably big money for a startup, because energy startups are big money. The jury is out on the small nuclear reactors. I can't say whether I'm right or not. But I'm skeptical.
ZIERLER: You mentioned the impetus of Europe wanting to become less dependent on Iranian and Russian gas and oil. To what extent is that really going to influence or at least supercharge efforts to decarbonize, the political element of not wanting to trade with unreliable or unsavory countries?
RUTLEDGE: In the book, I do make this point early on, in that in Europe, if the climate concerns didn't exist, they could have justified their renewables policy right down to the last euro, because Russia and Iran are not reliable sources. We've only become a net exporter of oil, gas, and coal during the last year the Trump administration. It is not clear we're going to be a net exporter now, because we haven't encouraged production of fossil fuels since then. So we weren't really a helpful potential factor over the two decades when the Europeans were setting up this policy of aggressively going after alternatives. They could do it just fine on the purpose of trying to get out from underneath the thumb of the Russians. I don't think that's the reason they were doing it, but they could have. I would have. I came, as I told you, from the defense industry, and you can tell from the way I framed the nuclear discussion, that I take security issues seriously. It's the kind of thing that you can ignore until something goes wrong. You're going very much on a gut feeling, because you could go 30 years, and everything is hunky dory, and you say, "Well, maybe we're wasting our time thinking about this stuff," and then something goes wrong. Like Ukraine, which has affected European energy supplies. Chancellor Merkel was always pretty positive about Russia as a partner. Obviously Russia could supply everything. It could supply wood as a fuel. It could supply oil, gas, and coal. And my sense is that the Russians have been pretty reliable suppliers. At the level of "I want to buy a million tons of coal," say, I think the Russians have probably been pretty good trading partners. It's the high level [laughs] which is "the war" that has thrown everything out. Merkel turned out to be mistaken.
We're very close to some Finnish friends from our sabbatical. They were Caltech people in the early 80s. All throughout the Cold War, I would have some interest in security issues, maybe because of my defense background, and they would poo-poo me, saying, "We know the Russians. We work with them. They're good people. There's not going to be any problem." But the Ukraine war has shaken them. It has completely inverted the discussion, in fact, because obviously, there's no sense that the Russians will invade the U.S., but the Finns are very much concerned that the Russians, if they are successful in the Ukraine, might—at least the people I talk to—might attack Finland as well. To an American, it doesn't sound particularly likely, but you have to listen.
ZIERLER: I'm sure you're following the exciting news, the announcement out of Livermore Lab, on the creation of positive net energy gain on fusion. What do you think the long-term implications are of this? Do you really think in 30 or 40 years, this will be the solution to all of our energy challenges?
RUTLEDGE: I'm not expert in this. I followed the fission reactors, because they actually produce a lot—20%—of our electricity, so they're very significant. With fusion, the Tokamak approach is more than 50 years old now. Both Professor Bellan and Professor Gould before him were interested in it. It's not clear fusion is going to lead us to electrical power. I think the challenge is that if you get your energy from fast neutrons, they are going to destroy any container. If you could get the energy from charged particles, like protons, basically then you could capture the energy without destroying the container. Now, at this point—and I'm only going by what I'm reading, so it may not be fair—my understanding is that the net energy calculation is based on the light input, with the output being the neutrons from the reaction. But there is a lot of additional energy consumed in making the light. When I look at how long it has taken them—30 years, something like that—and the absolute precision and high-quality people in engineering that are needed to produce these lasers at such high powers, I don't see it as particularly interesting for energy supplies.
The question I would have is why the program is even there. Since the Clinton administration, we've not been able to test nuclear weapons. I have always been suspicious—and it is not an original idea with me, so I'm repeating something that seems plausible to me—that a lot of the interest in this program is simply to explore fusion physics. When you can't test a bomb, you can explore physics under extreme conditions and timescales that might be relative to your codes that you use to simulate the bombs when you can't test them. I don't think the Tokomaks provide that kind of information. I could be mistaken. But we'll see.
From my background—I was a biology major until the senior year of college. Natural history, camping, hiking. I still like natural history. I've always liked solar panels. The thing that is amazing to me about solar panels is I can go out for 100 bucks and get a solar panel, and it will produce a couple of hundred watts, and that's the same panel that people use when they make one of these enormous solar farms, up in Antelope Valley. They put millions of the panels out there. I'm fascinated with the idea that these scale, that something I might buy and put on my truck camper would be the same or pretty close to the same one that you might put in these farms. They're easy to make. But the warning that I gave you earlier is that the solar panel only solves 20% of the problem. The 80% of the problem is the battery storage that comes with it.
Engineering Focus on Microwaves
ZIERLER: Let's go back to an earlier part of your career—your interest in microwaves, RF, millimeter wave integrated circuits. First as an overall question, to the extent that there is a spectrum between engineering and fundamental research—science—where do you see yourself in terms of both your interests and the kinds of things that you've worked on?
RUTLEDGE: Oh, engineering. I always liked it. I like something that would influence people somehow. Now, as an engineer, if someone wants to use my invention to put on a telescope, I'm happy to have a scientist as a client, but I'm not the scientist. I'm not that interested in the science. But I am interested in a device that a scientist might use. They're just another client. As long as it's legal and ethical. For radio astronomy, I can read their papers. It's fun as a citizen. Because of my background, I can understand a lot of what they're saying. It's easier for me than, say, a smart attorney across town. I can appreciate it more. But no, it's engineering. I like widgets. It was lab stuff. It was not theoretical.
ZIERLER: What is the overall discipline or area of work under which all of your engineering falls? Would it be millimeter wave technology, or is there something even more broad than that?
RUTLEDGE: Microwaves is okay. I told you I worked in an airplane factory, and I thought, "If I want to really do interesting projects, a bachelor's isn't enough." I had a good bachelor's [Cambridge University]. But I started from absolute zero. My father was a physician. To give an example, he would drive his car until it wouldn't go, and then he would call the mechanic who would make it go. There was simply no interest in how things worked in the family. So I started from scratch. The great thing about the program at Cambridge is they had what I would call an open lab, meaning I could go anytime from 8:00 to 5:00 and work in the lab. No matter how slow I was in doing the electronics labs, eventually I would get it. I tried to have something like that when I instructed electronics also, to give people a punch lock on the door to the lab at Caltech. You just go into the lab, 24 hours a day, and work until you are satisfied.
I didn't decide to go back to university until January, and by that point the only two places that I could find that had a very late deadline were Berkeley and the University of Texas at Austin. I was in Texas, and I applied to both of them. For some reason, my application got lost at Texas. It was probably my fault, somehow. You know, you're a kid, you do stupid stuff, and you don't check on things. So, I went out to Berkeley. I met my wife there. She was an undergraduate. And I'm very grateful to the University of California. Anyway, I was in a class, and it was one that used Professor Yariv's book. It was a former Caltech student who taught the class named Steve Schwarz. He's a very fine man. He's retired now and lives in Santa Fe, and we see him from time to time. At the end of one class, he said, "Dave, you ask a lot of questions. I think I've got a problem you'd be interested in." And he gave me a short paper. The style of paper in those days in his field was a letter. It was like four pages. He went to his office and gave me this paper, and I thought this was the most fascinating problem I had ever heard of. It was an antenna problem. You can probably appreciate the wavelengths that are involved. If you take an attack radar, say, on an F-16, the exact frequencies are classified, but the wavelength might be something like three centimeters. What Steve was trying to do was to make an antenna that would work for the CO2 laser, and that's a wavelength of ten microns. It's thousands of times smaller, so you needed a different approach. The question was, how do you make an antenna that would work for it? There are other problems, because you have to make the receiver that goes with it also, but the antenna part was what we were interested in.
I told him I was interested in the problem. He was very gracious. I told him I was just going to do a master's and then go back to work in the airplane factory, but he was very patient, in all sorts of ways. He was interested in some mathematical calculations for it, so I did those and helped the senior student that was working on it. I stayed on with his group.
When I was about to graduate from Berkeley in 1980, it was right about the time the EE Department at Caltech had kind of imploded. Jim Mayer had gone off to Cornell, Floyd Humphrey to Carnegie Mellon, and we had a couple of retirements. The department only had four new graduate students that year. I came down to Caltech to interview, and I thought, gosh, if I succeed, I could really contribute, because the department was clearly at a low point. So then I joined.
There were people at JPL right at that time—and this goes back not to climate but to the ozone problem—they were trying to figure out how to measure molecular concentrations up in the upper atmosphere, the components of the ozone cycle. You needed this kind of technology for that, and so I thought, "Well, there are people I could work with." Then I got a group going, and JPL gave some support. Then after a while, I found a guy in the Army, Jim Mink, who was interested in the work, and who became my main supporter.
Over time, I gradually moved to longer wavelengths, down in frequency, as there were different parts of the problem I got interested in. One of the big problems people were thinking about—if you take the wavelengths that are used for, say, Elon Musk's Starlink, people were complaining they couldn't get enough transmitter power. So, I became interested in increasing transmitter power. We got into a wide range of frequencies. The highest frequencies would have been 30 terahertz, which is 10 microns, where I started. Then—all the way down to 7 megahertz, which is 40 meters. The thing that I found was that if you went to very low frequencies, a student could complete a project in a summer. I had outstanding SURF [Summer Undergraduate Research Fellowships] students, Joyce Wong, Eileen Lau, and Jim Buckwalter, but they had severe time limits. I needed a project where you could work ten weeks and publish a paper or at least be a coauthor on a paper. All of those were, I would say, electrical engineering, not applied physics.
ZIERLER: Some technical questions about the kinds of things that you worked on. Let's just start with integrated circuit antennae. What does that mean? I've heard of integrated circuits. What is an integrated circuit antenna?
RUTLEDGE: What Professor Schwarz introduced me to at Berkeley was a very fine piece of tungsten wire that had been chemically etched, and when it was etched, it makes a point that's atomically small, a few hundred atoms across. It turns out that it will oxidize, and if you put it down on a smooth crystal surface—nickel was the one they used—that junction of the oxide and the tungsten and the nickel has a non-linear electrical characteristic, and that will generate a low-frequency signal if it receives the CO2 light. This detector work was started by a guy named Ali Javan, who was a very prominent professor at MIT. Steve Schwarz was following up on Javan's work. But these whiskers were very fragile. If someone slammed a door in the lab, they would short out. Years later, that turned into a Nobel Prize, but we weren't involved in it. The people at IBM Labs in Zurich used the whiskers in their tunneling microscope. They moved the needle across an atomically smooth crystal surface, and then you can see the individual atoms through the variation in the tunneling currents.
The integrated circuit part of this is that pretty early on we realized that if we were going to make receivers with lots of antennas, we would have to make them by integrated circuit techniques, which is to evaporate thin films of metal on a quartz or silicon wafer, and pattern them through optical lithography, like you do with an IC [integrated circuit]. That's why we called them integrated-circuit antennas. The challenge was that the wafer affected the antenna radiation pattern dramatically, and we needed new antenna designs for this. That was what I was interested in for a few years. The place where this has been applied in a large scale is in radio astronomy. At Caltech, Tom Phillips was one astronomer whom I met when I interviewed here. He was interested in the technology. We later on wrote a paper for Scientific American for these receivers. He recently, as you know, passed away.
ZIERLER: What is quasi about quasi-optical systems?
RUTLEDGE: The analogy is to a laser. If you think of, say, a carbon-dioxide laser emitting infrared radiation, it has atoms in an excited state that are stimulated by an infrared beam, then they produce infrared radiation in phase. Being in phase is the critical part, so that an amplified beam is produced. My thought was that you could take a surface, and through integrated circuit techniques, put a periodic array of transistors, with the transistors being analogous to the excited atoms. With the transistor, you have an input lead that receives the input beam, and then an output lead that radiates the amplified beam. You have to keep the input and output polarizations separate; otherwise you'll get a feedback oscillation like an audio microphone in a convention hall. It's optical in the sense that you have a beam, like the light beam in a gas laser, but it's microwaves rather than light, but quasi in that the circuits that amplify. There's a metal pattern. There are filtering strips and things like that that aren't in any sense optical. There's nothing optical about a transistor or an antenna lead. That was the word "quasi."
ZIERLER: What are power amplifiers?
RUTLEDGE: The power amplifier is that final stage before the radio signal goes to the antenna. The classic one would be in your cell phone. If your cell phone transmits—say a watt—the signal levels that are running around inside the cell phone, when you talk, or when you receive, are much lower than that. On the reception side, it's probably picowatts. Even on the transmit side, you're talking microwatts. To get to a watt, you have to have a final amplifier. A lot of electronics are involved in amplifying, or making a signal bigger. The challenge is always the last stage. The last one is the one where you have the most power, but you may not have much gain. At small signal levels you might be able to get a gain of a thousand, say, meaning the voltage on the output is a thousand times bigger than the voltage on the input. With the power amplifier stage, you might be happy with three to one in voltage. But in power terms, that's nine to one, and all your problems with heat, failures—failures come from heat—are probably going to come from your power amplifier. They're not going to come from the low-temperature, low-power levels. The low-level amplifiers can be much more complicated and complex in terms of the number of transistors. But in terms of where the failures might be, it's where the heat is produced.
Enabling New Research with Software
ZIERLER: When did you start getting involved in thinking about software and applications for computer-aided design?
RUTLEDGE: A lot of things I did didn't necessarily lead to publications; I was interested in classes. There was a guy when I came, named Jim McCaldin, who passed away, a professor of applied physics. He taught a class that Carver Mead had invented. Carver had developed for Gordon Moore a series of demonstration lectures at Intel as a consulting gig there. Floyd Humphrey had taken Carver's experiments and turned them into a set of labs for a freshman class. The students made a light-emitting diode. They made a p-n diode. They made a transistor. I didn't see the class when Floyd taught it, because he left before I came. Jim McCaldin was teaching it when I came. Jim taught it for a few years, but it really wasn't his style of class. Jim was more the kind of guy who liked to work one-on-one, in smaller groups. Floyd's class was a big one. Tom McGill—again, passed away—and Jim asked me to take over the class. I was very interested, and I took it over.
It was a big class. This was the heyday of electronics, and half the students—if you think CS now—half the freshmen in the university were taking the class. It was an organizational challenge. We had 14 lab sections. We had lab sections on Saturday. There was stuff breaking all the time. The students were using glass furnaces. There were acid spills. There were all sorts of interesting things to work out in terms of making it safe for the students, and educational, where everything is accessible. When they do metal evaporation, they can see it, and it's not something that is done in a big complicated system. I was just fascinated by the idea that you could learn something about solid-state device physics by building the different devices. Then following Jim McCaldin, Floyd, and Carver's lead, I gave lectures in how these devices worked. I really liked the idea that we were teaching solid-state electronics to freshmen.
The next class I tried along these lines was for microwave circuits. The students would make different microwave circuits during the term. For microwaves, it is a kind of PC [printed circuit] board called microstrip, where the elements, like inductors and capacitors result from the copper pattern on the board. You etch out different shapes. We set up labs where the students would make filters. They would make amplifiers. Then we'd go out as a field trip to Owens Valley Radio Observatory, because there were some EEs [electrical engineers] there that gave great lectures to the students and gave them a sense of how you could actually use the ideas. One problem, though, was that the commercial software programs for designing these filters and amplifiers were exceedingly complex. I just felt they weren't productive for beginning students. And, they were expensive. You'd have a station—we're talking now 1980s—it probably cost $50,000 a year; something like that. The programs were pretty flaky and slow, and the students would have to share them.
I had a very good student, Rick Compton, unfortunately now deceased, but later a professor at Cornell and then a multi-gazillionaire, chief technical offer at one of the companies Broadcom bought, who had been a scientific programmer as an undergraduate. He was from Australia and he just came by to see me one when he arrived, and he got interested in writing microwave design software for the class. Our idea was that this would allow people to do circuit design without the huge overhead of the design stations, which really were pretty slow and clunky, when the software should be fast and it should be something simple that students could deal with. The way we set it up was that we distributed the software at cost, meaning if someone walked in the door, we counted the cost as zero, and we gave it to them. If they wanted us to mail it to them, we charged them $10. My wife Dale did the distribution; we ended up distributing 30,000 copies of the software. Well, there are only 15,000 people in the Microwave Society, so we covered the market. Now, it did get used for commercial products. I'd get letters from people that said, "Well, watch for a stuffed animal that has a microwave detector inside it"—that used the program for the design. And other universities used it to teach microwaves circuits also.
The program was called Puff. You may have to be my age to remember the Peter, Paul, and Mary song about the magic dragon. He's a companion to a little boy, but eventually the boy grows up, and then the dragon is left by the wayside. So, Puff was really meant to be a starting program. The commercial software available now is much easier to use and much faster than it used to be. The Caltech class still does the same labs, almost forty years later, but now they use the commercial software rather than Puff. But in its time it was a fun project. It coincided with this wonderful, Turbo Pascal compiler, which was vastly faster than the Microsoft Pascal compiler. It was a Frenchman, Philippe Kahn, who came to the United States with this compiler. He sold it for $50, which was much cheaper than the Microsoft compiler, which cost hundreds of dollars. We used Turbo Pascal for the Puff program. Also the source code was open, so we had people who modified it and made different versions. It turned out that over time, the biggest use probably was by amateur radio operators, thousands of them. Amateur radio operators often are technically competent, but they're price sensitive. They used it for their microwave projects, and they still do. They're very conservative, not in the political sense, but often, if old stuff works, they keep using it, because it's obviously the cheapest way to make circuits.
ZIERLER: With the great success of the textbook The Electronics of Radio, as it was going through new editions, did you need to provide updates to keep up with the technology?
RUTLEDGE: I think the publisher does consider it a success. Phil Meyler was my editor, and he is now pretty high up at the Cambridge University Press. I had the idea for the book because of the experience with the Puff software—it's the same kind of idea. This was a class for sophomores, now. It's the first electronics class the students have. The idea is that you could learn electronics effectively by building something. It was a bit of an accident that I came across the design for the book. It was a project for my kids. They built a ham radio transceiver. The output was a couple of watts. The parameters were similar in all respects to a cell phone, except it's a different frequency. My three kids, they each built one and put it on the air. My daughter used the Caltech Radio Club antenna with it to communicate with some kids in Okayama, Japan, so it was not a toy. I looked at the circuit when they built it—all three of the radios worked—it had the basic analog circuits, which is the classic art of electronics. There's a digital side, and an analog side to electronics, and they interact, but properly educated electronic engineers learn both. The classic analog circuits were in this radio, so I thought, "The students, maybe we could talk about a power amplifier, for example, and then they'll go build the one in the radio, and then they'll measure its parameters, and by the time they finish the two terms, they will have built the whole radio, and then they can test the transceiver as a whole."
This took me back to my experience working in a fighter factory. A lot of the work in the industrial world is what you call systems engineering. It's different pieces that go together, and you test them as an assembly. It's hard to teach it in a class, because you don't have enough time for students to put a whole system together. But in this case, we had it. The students could measure sensitivity, non-linearity, and various other parameters for radio transceivers.
I tended to do lectures by typing them up verbatim the night before. I don't know whether you've done a series of lectures, but it's absolutely terrifying. It is the classic—"your mind focuses when you're going to be hanged"—the old Samuel Johnson quote. It absolutely applied to me, and I found all the lectures terrifying. So, I used to type them up. I would get an early dinner around 4pm, and go out in the garage behind our house, and then try to say the words in the lecture, and catch them by typing as I was talking. After a couple years of this, I had a good set of lecture transcripts, and it wasn't the way anyone had ever taught electronics before. There was a guy, Paul Nahin, who had written a book that I very much admired. It was a biography of Oliver Heaviside, who was an early electrical engineer and mathematician, who worked in telegraphy. in the 1800s. Nahin introduced a lot of electrical engineering in the book, so you could read it, say, as a smart attorney, and find out some electrical engineering as you learned about Heaviside. Nahin also followed this approach with a history of radio, and you learned a lot about how radios really worked from his history.
Because I admired Nahin's work, just for fun I sent the collection of my lectures to him. He wrote, "Dave, you should try to get this published. If you go to Cambridge, they won't care whether they sell any." The Cambridge University Press was an unusual publisher. They worked through syndicate that made the publishing decisions. At least they did in those days. Academic publishing is a big business now. There is an inevitable evolution. I'm not criticizing. In fact, I'm very much a fan of the Press. But Nahin was right. I sent the manuscript off, and they had the people in the syndicate look at it. They asked me, "How many do you think will sell?" I said, "I don't think I'll sell any of them, because no one teaches electronics this way." The syndicate approved the book anyway. Then I got this wonderful guy, Phil Meyler, as the editor. There was one stage where I got to one of the chapters, the one on oscillators, and I realized that—and it's the way these things work, right in the middle of a lecture—I realized what I was doing, there was no mathematical error but it simply wasn't relevant to the engineering. It was just a mathematical exercise; it wasn't telling the students anything about oscillators that was practical. I thought, "I've got a problem." I wrote Phil and said, "I've got a problem with the oscillators, because I really don't know what I'm doing." This is way it works, you realize right in the middle of the lecture, that what you're saying is baloney. Phil wrote, "Well, there's a guy Tom Lee up at Stanford that's writing a book now on oscillators, and you might talk to him."
I did, and Tom Lee is a wonderful guy, absolutely one of the great people in electrical engineering. He's completely out there. He wrote a book on kitchen science, which is all the interesting physics and electrical things you can do in the kitchen. He had a huge collection of vacuum tubes, thousands of vacuum tubes. It took him forever to get married, for kind of obvious reasons, but he finally did, because he's just a gem of a human being, but it might take a while for someone to appreciate it! [laughs] Anyway, Tom had a student who was just graduating named Ali Hajimiri. He said, "Ali has cracked this problem." He said, "You're on to something. It's not you that's stupid. It's the fact that the field doesn't describe the way oscillators actually work in electronic circuits. You're right in that you're realizing there's a problem. The whole field is screwed up. But Ali solved this problem." I was running a faculty search right then, and I invited Ali to come down and give a talk. It was the most significant talk I've ever seen by a faculty candidate. I got feedback from the students who came to the lecture, from postdocs, from professors, "You've got to get this guy here if you can." I don't know if you know the advertising code, but if you see a poster that says "Special Lecture," "special" means a faculty candidate. I don't know if the candidates all realize that, because many are students. Well, Ali has turned out to be a great addition the faculty. And he did solve the oscillator problem. Ali's theory was at the level of the really fundamental physics of oscillators for PhD students and industry. My lecture was for sophomores and for the particular oscillator in their radio. I adapted it for that purpose, and it really did make a realistic oscillator section in my book, and no electrical engineering undergraduate textbook that I knew of had one.
Phil was an unusual editor. I said, "Because of my experience with the software, we might sell some copies to amateur radio operators, but they won't buy a hardback." It's 40 bucks for a paperback and 100 for the hardback. I told him, "Look, the way the amateur radio literature is, is most of it is what I would call cookbook. ‘This is how you build something.' Then they just give you the design. But there are some people who are operators who actually have the physics and math background to understand what is in this book, and no one has written a book like this with circuits for the amateur bands. But they won't buy it if you do a hardback." He said, "Okay, we'll do paperback." He did, and I'm sure 95% of the sales were paperback. So, everyone was happy. I was happy. The amateur radio operators presumably were happy, because they got a nice project, because a lot of them built the radio—I heard from a lot of them.
I didn't put the solutions to the homework problems in the book, so a lot of people wrote me to say, "Can you help me with this problem?" I used the problems for a different purpose. The problems were long and hard. I'd say to the students, "All right, this is going to be due in a week. But they're really due at the beginning of class. You've got to turn them in, because I'm going to talk about them." I used the homework problems as a way to launch into a lecture. One of the suspicions you have as a professor is that students, particularly when something is mathematical, can kind of fade out and not get what you're saying. You really have to be engaged to follow a mathematical lecture. But if you can relate the lecture to something they have built in the lab and done a homework exercise for, they're engaged. So, I thought it would help them understand the lecture.
With this tight relationship between the homework problems, the laboratory exercises, and the lectures, it really is not practical to change the homework problems each year. This kind of teaching is very dependent on Caltech's honor code because students could get the answers to the homework problems from the previous year's lecture notes
The other funny thing that we found out—this is sociology—Dale probably mailed out 15,000 copies of the software, and a lot of them were for amateur radio operators, and they would always let you know. They don't sign in a normal way—their language for "best regards" is numerical. The numbers are based on old telegraph codes. The telegraphists had numbers that indicated various things. "73" means "best regards." So the hams would sign "73" instead of "best regards." Then she would know it was a ham. The funny thing with hams is she always mailed the software immediately, because out of thousands of orders, she never had an amateur who didn't follow up and pay. Whatever the discipline is that gets you into amateur radio is a complete honesty filter. Now, I'm not a business man. But I'm sure if I were a business person, I could have figured out a way to make money off that honesty of amateur radio operators. Because if you know something about your customers, something that's special—not true for the general population—you can make money from it. It was a very pleasant experience. A lot of the hams would write back and show us the circuits they had built.
ZIERLER: You mentioned astronomy. Where else have we seen in end use products some of the technology that you have developed?
RUTLEDGE: For quasi-optics, there were three students—Mike DeLisio, and Chad Deckman and Lawrence Cheung, who started a company called Wavestream. They were joined by a professor from Harvey Mudd, Jim Rosenberg, an outstanding lecturer and an outstanding electrical engineer. I think he was getting a bit bored at Harvey Mudd, and he was ready to move on. They made a series of transmitter amplifiers for high-speed satellite uplinks. This is ground to space. And it is high frequency, 30 gigahertz. They also picked up some spinoff technology—I mentioned Rick Compton earlier for the Puff software—from one of Rick Compton's students at Cornell, Bob York, who was a professor at the University of California Santa Barbara. It's a related technology and clearly inspired by ours. Bob's technology is better suited for somewhat lower frequencies, around 18 gigahertz. Wavestream manufactured this, too. The company wasn't what you'd call a home run, but they did just fine. They were sold for, oh, $150 million or so, and the students were able to pay off their homes. Everyone was happy. The University of California was very diligent at keeping the patent rights, much better than Caltech was, so Bob did very well at University of California, from the patent royalties. All good.
Now, for applications—the Wi-Fi transmitters, that connect to satellites from the top of an airliner, would be one. You can see the bubbles on top of many airliners for the transmitters. That would be a commercial application. However, the largest sales, in the thousands—were for soldiers, to send video up to satellites. There's a network of satellites in this frequency range, 30 gigahertz, that the Defense Department has put up. They pick up a huge amount of video. So, it's the communications for the Marines and the Army in Iraq and Afghanistan, and Special Forces. Before, people had used vacuum tubes, which are bigger, heavier things, and the soldiers were always breaking them. They were just too big and clunky. The Wavestream amplifiers were small—a few inches across, and very solid.
After Wavestream was acquired, Mike and Chad left and the two of them formed a company called Mission Microwave, which makes a line of transmitters power amplifiers that compete with Wavestream.
The other spinoff was from a collaboration with Ali Hajimiri. There were two of my students, Ichiro Aoki and Scott Kee, who, with Ali, came up with a transmitter amplifier for cell phones. This was made in silicon. The traditional cell phone transmitter amplifiers, because of some technical requirements about linearity, are made in gallium arsenide, but this was silicon and therefore cheaper. For a while, their company, Axiom, held the market for inexpensive power amplifiers. Something like 10% of the world's cell phones apparently used it, mostly in China, because that was the place that was most price sensitive. Now, that company, Axiom, was acquired by Skyworks, which continues to sell the Axiom power amplifiers.
After Axiom was acquired, Scott and Ichiro went off to form another company, Indie Semiconductor, that is now publicly traded. They have shifted their interests now and they're doing automotive electronics. The company apparently has been doing quite well. I tend to view the product of a university research group more as the student than the research. However, in these cases, it's both the students and the projects.
ZIERLER: For the last part of our talk today, some overall administrative and cultural questions about Caltech. What have you been able to accomplish just by being at Caltech and its unique approach to engineering?
RUTLEDGE: I don't really think that way. I told you I had to start over when I began studying engineering, and I did my second undergraduate degree at the University of Cambridge in England. Of course I arrived there not knowing anyone, and the first guy I met was a man named Stan Whitcomb, who had been an undergraduate at Caltech. I thought he the smartest man I had ever met. He is a wonderfully nice, gentle, kind, thoughtful person. He showed up, later, as assistant professor of physics here, and then he became the chief technical guy at LIGO, not at the Nobel Prize level, but right at the level below that, and he worked on the project all the time since then. I just thought, "Well, if everyone is like Stan Whitcomb, Caltech must be wonderful." I went to a small college in Massachusetts, Williams College, and I loved the small size, and I appreciated the fact that you know your professors, you have them over for dinner, stuff like that. At Caltech I have appreciating knowing professors in other areas socially. When I started looking for a job as a student at Berkeley, Bill Bridges and Roy Gould invited me down to visit, and I just fell in love with the place. I thought I could make a contribution, because of the problems the Electrical Engineering Department was having at the time, but I didn't really concern myself about what the contribution would be.
One of my interests has been in recruiting faculty. As a search committee chair, I recruited Yu-Chong Tai, who is in the National Academy of Engineering. I recruited Axel Scherer, who is in the National Academy of Inventors, and Ali Hajimiri, who may be in one of the academies at some point. When Provost Paul Jennings was considering appointing me as division chair, he asked what I wanted to do, and I said, "I want to do recruiting" and Paul was okay with that. So, I recruited people who are now leaders in the division: Adam Wierman, Azita Emami, and Beverley McKeon, for example. Recruiting was interesting to me. How much of a contribution that is, who knows? I'm just happy to be part of Caltech, and my wife Dale was my assistant in the office next door for 30 years, and she is happy to be part of Caltech, because it's a great institution.
ZIERLER: Last question for today. Between your various administrative service accomplishments to Caltech—being executive officer for EE, serving as division chair for the Division of EAS—what aspects of Caltech might you have come to appreciate that simply wouldn't have been available if you didn't serve in those kinds of roles?
RUTLEDGE: Ooh! [laughs] I think the main one would be if I didn't recruit the people—I told you that when you are search committee chair, the process is like having a baby. With all due respect to my wife, who has had three—it's that tough. The people you recruit, obviously you have an influence on, right then. The faculty hires define Caltech. Caltech is a wonderful environment for a young professor and it very much encourages them.
In Engineering, the department executive officers are quite weak. The power is centralized in the division office. As executive officer, obviously you want to do your job and be good to staff, but it's just not a big deal. But, aside from the fact that you choose division chairs from executive officers, the division chair is the place where you can have an effect, and I would say the best place you have an effect is recruiting. You do the best you can, but you're very dependent on the provost support. If the provost doesn't support you, then you won't have much of an impact.
ZIERLER: This has been a great overview conversation. Next, we'll go all the way back to the beginning, learn about your family background, your childhood, education. We'll take the story from there.
[End of Recording]
ZIERLER: This is David Zierler, Director of the Caltech Heritage Project. It is Monday, January 2nd, 2023. It is great to be back with Professor Dave Rutledge. Dave, once again, thanks so much for joining me.
RUTLEDGE: You're welcome.
ZIERLER: Today I want to go all the way back to the beginning. In our first conversation, we took a great tour of your approach to research, your administrative accomplishments. Let's start with your parents. Tell me a little bit about them and where they are from.
RUTLEDGE: My mother is from Texas, and I grew up in Texas. She's college-educated, an English major. When my parents got married, before I was born, she taught reading—and these don't exist anymore—at a home for unwed mothers. She was a reading teacher, and so reading was important in our family. Reading aloud was important.
ZIERLER: What about your father? Where was your father from?
RUTLEDGE: He's from northern Minnesota. The town is called Detroit Lakes. The closest one that people have heard of, because of the movie, is Fargo. It's just over the border from Fargo. He comes from a family of physicians. His father was a physician. His grandfather was a physician. When you look at the old census records, I think before that, it's all farmers. My father was a physician, my brother was a physician, my daughter is a physician, so there are lots in the family. My wife made a photo montage of 12 family physicians of various sorts. So I'm the odd person out.
ZIERLER: Where did your parents meet?
RUTLEDGE: My father was rushed through during the war—at the University of Michigan in two years, and then he went to Harvard Medical School. The war ended right when he got out of Michigan. My mother was a student at Wellesley, and in those days, people tended—in fact, for my wife and me too—people tended to meet each other, husbands and wives, in college. She was an undergraduate at Wellesley when he was a medical student at Harvard.
ZIERLER: What specialty did your father have? What kind of medicine did he practice?
RUTLEDGE: Most of the family—my brother, father, grandfather, one uncle—are general surgeons, which nowadays, I think it is past its prime because of the laparoscopy. Things have gotten more specialized now.
ZIERLER: How did the family find themselves in Texas?
RUTLEDGE: When the Korean War came, my father was drafted, and he was a flight surgeon in the Air Force. He was very unmilitary. Unlike my mother, he was a slow reader. His name was Robb, and we named our son after him. When he retired, roughly the same time I did, at 67, he spent the rest of his life writing articles on medical history. One was on a surgeon from Vienna named Billroth, who invented several of the early surgical procedures once people started to use anesthesia, which really opened things up for surgeons. He also wrote about the development of anesthesia. He went around to different medical groups giving talks on these things. Anyway, back to the Air Force. He was drafted in the Korean War, but was not sent to Korea. He was a flight surgeon at Hunter Air Force base in the Strategic Air Command in Savannah, Georgia, where I was born. He used to say the problem for a surgeon was that the Air Force in those days did have quite a few accidents with airplanes—not so much now—but he said usually what happened is that the pilot could either walk away from the accident or the pilot was killed, so there really wasn't much work for a surgeon to do. He had to get a certain number of flight hours, so he would hitch rides on bombers and fly around, all bundled up in the cold, to meet his requirements. But when he could get out, he did.
My mother was from Texas, and I think he wanted an environment that was a little more adventurous than the small town in Minnesota that he had grown up in. He practiced surgery in Fort Worth for 40 years. Fort Worth has had an amazing transformation. When my mother was in high school, the county population was around 200,000, somewhat larger than Pasadena. Now it is 2,000,000. Fort Worth was a truly wonderful place to grow up. I've tried to convey to my kids, your generation, what they've missed, because they just don't know how happy this environment and how supportive it was compared to the modern one that I see in Pasadena, where there are only a few kids on the block, and they all go to different schools. In our time, there were 50 children on the block, and so after school, once the mothers were okay with the status of homework, we went outdoors to play, every day. It was just a wonderful place. In high school—again, I think not easy to imagine now—we had an astonishing range of incomes. We had ranch kids coming in, hitching rides from friends, who really were only marginally in the money economy. A lot of the things, they provided themselves or bartered for. At the same time, we had the children of the family who discovered the Libyan oil field. So, we had a range of incomes from essentially zero, before there was much government support, to—infinite. And we kids didn't know. It wasn't really obvious to us at that time.
ZIERLER: What about racially? Was Fort Worth segregated when you were growing up?
RUTLEDGE: I'm one of the last people who went to segregated schools. They desegregated a year or two after I graduated from high school. The kind of integration we had, and it wouldn't be called integration—we had one person who was elected to all of the popular positions in the high school, a Black teenager—let me say teenager; boy is a loaded word now, of course. It wasn't then, but it is now. He was an excellent basketball player, which seemed to overcome [laughs] the other issues. Sports—we had our priorities! Anyway, these issues were simply invisible to us. You really wouldn't know. We weren't taught anything bad; we didn't know anything bad; we didn't know anything good. Just neutral.
ZIERLER: Growing up, did you tinker around? Did you have chemistry sets, race cars, those kinds of things?
RUTLEDGE: Let me back up one more on the race.
RUTLEDGE: Because there's another side. Of course, this is very serious, and again it's a case where the generations see thing differently, and I've had a lot of time to think about it, of course. When we were kids, we were thoroughly integrated with respect to Hispanics. They would be called Mexicans, and they would call themselves Mexicans, then, like Italians, presumably, or Polish, in New England. So, that wasn't something we thought about. Legally, they were white. They were at the same schools. I didn't detect any real friction that I could tell. Maybe it was there. I remember losing in the spelling bee school championship to Yolanda Alvarado, a wonderful person who became a nurse. I saw her at the 50th reunion, and she remembered it too. She came up for a hug. We didn't really think about Hispanic versus Anglo then. There were a huge number of intermarriages. There certainly would be parts of town dictated by, money and migration patterns, but I don't remember tension. Maybe someone else does. The Hispanics in Texas were an important part of the race picture, and my perception is that at that time it was simply ignored. The emphasis on Hispanics as a group that is subject to, say, affirmative action, is modern.
Back to the question about tinkering. Very little. My father was very much the kind of person that would get in his car and go to work, thinking about surgery all of the time. He won the local doctor of the year award, and he gave a speech for that, where he said, "Every waking minute since I've graduated from medical school, I think about surgery." And it was almost true. He was just completely dedicated to his job. If you interrupted him with a serious long-term question, he gave good advice. Surgeons have seen a lot. They have seen people under stress. They have seen long-term and short-term issues. So if you want to ask them advice, they're good people to ask. But he really was focused. It did mean that he would drive the car no matter what the sounds were, and if it didn't drive, he would call the guy at the gas station to come do something about it. That was his attitude toward mechanical things.
The thing that turned out to be helpful for me, because I started engineering after my bachelor's degree, which is certainly not the common way to do it, was that my undergraduate engineering was in England, at Cambridge University, and they had an open electronics laboratory. If you weren't in lectures, from 8:00 to 5:00 you could go into the lab anytime. I found that was a huge help in trying to catch up.
Now, I did have outside interests. I was interested in Boy Scouts. In retrospect, the merit badges in Boy Scouts are I think are a good chance to explore—they had some that were required for the ranks, and they are the classic camping, cooking, and that kind of stuff. Hiking. Those, most people don't make careers out of. But there are a hundred merit badges, and half of them are probably things you could do for a career, like various kinds of farming badges, for example. I did do the radio merit badge, and the electronics one. It was a good thing. I built some electronic gear and learned Morse code. A surprising number of people from my generation in electrical engineering, and certainly the generation before, got some inkling of what's going on in electronics from amateur radio. The thing that is hard for people sometimes to realize—because we're communicating by electronics right now through Zoom—I was born three years before the first silicon transistors was demonstrated. That was the beginning. You cannot do anything like the level of technology we with vacuum tubes or with germanium transistors. The electronics is in our screen. The electronics is in our communications. And in the microphones.
ZIERLER: Another question about a long time ago in Texas. What were the politics like growing up, both in your household and in Fort Worth? Texas was a more Democratic place at that point.
RUTLEDGE: My family were I think mostly Republicans, and in those days—again, you are not asking the right guy, but I think it's fair to say in Texas, there would not be what would correspond to, say—the current left wing of the Democratic Party. In those days, I think you'd probably say the Texas Republicans might have been more to the left than the Democrats. Even in my time, it was the legacy of the Civil War that the Republicans were the party of the Yankees, and the Democrats were the party of the South. That still was in place. You can see this clearly in the 1968 election. By modern standards, Humphrey would have been considered well to the left of Nixon in that election, and Humphrey won Texas.
As a child, I didn't pay much attention to politics. I did run into it when I went off to college. My mother and father, being educated in New England, took me to several campuses there in the summer of 1968. One was Princeton, and one was Amherst, and one was Williams, where I ended up going. At Princeton, the reaction was that my SAT scores weren't high enough. They were more candid in those days, I think, and more specific about scores. Then at Amherst, there was a discussion about the Vietnam War, which I hadn't thought very much about, so the discussion didn't go very well. At Williams, which if you ever get a chance to visit that campus, the town is Williamstown and the setting is beautiful. It's in the Appalachian Mountains. It has all sorts of resonances for me. The person who interviewed me at Williams had been to the same Boy Scout camp I had, in New Mexico [Philmont], so we talked about camping. At Williams, we might even go camping on a date. If someone comes up, say, in those days, from Smith—a women's college—you might well take a couple of sleeping bags and walk up into the mountains and stay overnight. That's I guess a long way from what kids do now. It was kind of non-political.
The other place I was exposed to what would be called race now was my work environment. I worked for three years for a civil engineering company, Freese and Nichols. The first summer—after ninth grade—was as a Xerox operator. In those days, you had an operator for the copy machine, and I was the operator. The second summer, I was a surveyor out in West Texas, on the Robert Lee Dam. Yes, that is the Robert E. Lee, who was stationed in Texas before the Civil War, and much appreciated for his ability to counter Indian attacks. That definitely exposes you, again, not to Blacks, but to Hispanics, both legal and illegal. The third summer, I was an inspector on several highway bridge projects in the Fort Worth area.
Later on, when I was in college, I worked on a ranch in South Texas, the most wonderful job you could imagine. Ranch life has very serious attractions to a lot of people. It's complete isolation from a lot of effects—the people I knew on the ranch, none of them legal, had no concept of time. The only way you could tell how long they had been there was that the owner kept a diary, so you could go look in the diary to see when someone came. It's a timeless place. The way ranch life works is you get woken up by a cook when it's light enough to see. It is tradition to have a cook, and the cook will wake you up, when it's light enough to see, and she will feed you. It was two brothers and a sister from Zacatecas. The sister was the cook. They were wonderful people, and the brothers were superb cowboys. Then, when it was too dark to see, you'd go back to the ranch house, and she would feed you dinner. Then you repeat, every day. Then on Sunday, we'd have off, and I could drive around and see the wildlife. From one of the ranch trucks, you could easily see 50 deer just wandering around the ranch. It was a wonderful life.
So, that's my exposure to illegals. In those days, the Border Patrol flew over the ranch every day, and there were roadblocks every week. They could never catch anyone. No one I knew was every caught by the Border Patrol. But the local sheriff knew everything. One of the brothers got in a fight in a bar and injured someone badly with a knife. The next day, the sheriff was at the ranch. He said, "One of your boys"—remember, this was 50 years ago and "boy" did not have the connotation it does today—" was in a fight, and I don't know if the man he injured is going to live. I talked to people in the bar, and it looks like both are fault." He said, "What you want to do is tell him to go back to Mexico." And that's what the brother did. At the time, I thought that the sheriff's approach was actually quite subtle. It's not an option that is open to citizens, because they cannot go home to a different country. Illegals live a different life. Working with illegals inspired me to spend the summer after college graduation in Spanish language school in Mexico.
One thing that people miss in the discussion about illegals is that ranching culture draws from both the American and the Mexican side of the border. Our ranch was next door to the King Ranch, the most famous of all the American ranches. Captain King—he was a steamboat captain—got his ranch started by first buying land, and then he went down to Mexico and persuaded the people in a village to move to the King Ranch. Descendants of those villagers are still working for the King Ranch 150 years later. Our ranch ran the Santa Gertrudis breed that was developed on the King Ranch. That's a whole side of the US that was always fascinating to me. It was a fork in life that I didn't take. I loved the ranch life, but I don't have the family background or the financial resources to pursue it. But it could have been a wonderful life.
ZIERLER: When you graduated high school, was the draft something you needed to contend with?
RUTLEDGE: I graduated in 1969, and I had a college deferment for one year. Then they had a lottery for people who were born in 1952, based on our birthdays. We sat around in the dorm watching them draw numbers on television. I still remember my number for January 12, 228. That seemed a high enough number to me that I simply let my deferment lapse. You had one year of exposure, and they did not get to 228 that year.
I've worked for the military industry, and I've been interested in military history all my life. I think it is natural when you grow up as a boy after the Second World War has just ended, clearly the greatest conflict in history and with the most impact of any war. But I always felt I didn't have the personality for a military career. When I started in college, we were beginning to pull out of Vietnam so even if I had been interested in participating—because in those days, like now, people volunteered for that way of life, either because they feel a sense of duty, or they just didn't have other things that seemed more interesting to them to do.
I didn't really have strong feelings about the Vietnam War. When you read the popular talk about the war now, I think it has diverged from the views of military historians. I think that the historians' perception is that when the US left in 1973, the North Vietnamese Army was pretty badly beaten up. I think that the reason for that perception is that the aid to the South was cut off when we left, and it still took the North two years to come back and take over the South. This is in sharp contrast to the strategy in South Korea, where President Eisenhower did not pull us out. South Korea today is one of the wealthiest countries in the world, Vietnam one of the poorest. The choice is like Afghanistan. In that the last year we were in Afghanistan, there were essentially no American casualties. There could have been an alternative historical decision to just remain, as in Korea, because the costs were so minor. What would that have done for, say, women in Afghanistan? Who knows? Neither President Trump nor President Biden were interested in staying, but it certainly would have been an option for someone more patient than those two presidents. Clinton and Obama were both more patient than Biden or Trump.
My work in the military industry was in Fort Worth at General Dynamics on the F-16 after I got my engineering degree. I did feel like that I could contribute there. Working on an airplane is a good place for an electrical engineer to be. The F-16 turned out to be an all-time great airplane. They have produced 4,000, and it is still being produced today, 50 years later.
From Williams to Cambridge
ZIERLER: Did the Williams campus feel political to you at all? Did you experience the so-called 1960s when you were an undergraduate?
RUTLEDGE: It was a political environment, but I think I missed a lot of it. Again, you're talking to a guy who worked in the military industry later. We had a strike for a few weeks in the spring of 1970. The Kent State killings were the trigger. I didn't go to Washington during the strikes. I went down to the local Eph Pond, because I had some time off because of the strikes. It was a place I used to go to look for wildlife, and it was walking distance from the campus. I spent a few days clearing stuff out of the pond to clean it up. I remember pulling out some tires. We still had to take finish our classes, so I hung around for a week afterwards to finish up with homework and exams.
ZIERLER: You said it really wasn't electrical engineering as an undergraduate. What studies did you pursue?
RUTLEDGE: That's right. I started off in biology. I was a biology major until the senior year. It was wildlife biology that I was interested in. In addition to the Boy Scouts, in high school there was a friend of my mother's, Margaret Parker, who ran an Audubon group for kids. Once a month she would take us out on Saturday—she was very good—to identify birds. At that time, the ecology concept was really taking hold, but the environmental movement wasn't really going yet. Obviously pulling tires out of ponds is always in fashion. I'm sure it was in fashion 100 years ago. I was interested in trying to understand some of the cycles going on in the natural environment. But I was always taking math classes on the side, and I was beginning to think that the way I approached problems really didn't fit a biologist.
One of the things that I've been impressed with, and I am even more aware now, having been a division chair, is that the cultural environments for different fields of engineering and science are quite different. In order to make progress in a field, you really have to absorb the culture of that field. Even though I was studying biology, I loved my math classes that I was taking on the side. In those days—before PCs—it wasn't clear what you could do with math. I was trying to imagine, what could you do with group theory or number theory or functional analysis. Then I finally decided I was not going to be a good enough biologist. I continued with the math and graduated. At that point, I was thinking, "Well, I'm going to have to do something for a living." In those days it was basically high school teaching, because colleges had stopped hiring after they had staffed up to teach the baby boomers. And there was no equivalent of the quants on Wall Street. I'm not sure that Wall Street would have been very interesting to me, but good mathematicians do work in that area. But my grandfather Simon Freese was a civil engineer. In fact, it was his company, Freese and Nichols, that I worked for, when I talked about working for a civil engineering company in the summers. So, I thought I'd try civil engineering.
At the same time, I was getting a little bit of wanderlust. I thought, "Let me apply to some universities overseas." I applied to different places in England. My grandfather was an MIT engineer. After he started work, he had done a year abroad—in those days, it was a little more free form—and he went to Cambridge University, in England. The English had developed a good way of dealing with the biological waste—black water if you're an RV person, sludge if you're a civil engineer—by encouraging bacteria to consume the waste. He spent some time there, at Trinity College. I applied to Cambridge just because he had been there, and then I chose the colleges that were coed, because I had been to one college that was a men's college, and I was definitely not looking for a men's college again. I was admitted to Churchill College and went there.
Cambridge University turned out to be a fabulous place for me as an engineering undergraduate. There was much more freedom than in the US. There was little homework. You basically just went to lectures, did laboratory exercises, and studied for the annual exams in the spring. That suited me, because at that time, I already had a bachelor's degree in mathematics and I knew how to study. I wasn't looking for the discipline of homework, which is what you start freshmen and sophomores in the US off with. Cambridge got me into electrical engineering.
ZIERLER: Were you specifically looking at a study abroad kind of program? Were you looking specifically at European or British schools?
RUTLEDGE: It wasn't anything like a study abroad. It was a regular undergraduate degree program. I assumed I'd have to start over. However, Cambridge was unique; they had a program where if you already had a bachelor's, and you wanted to study for a second bachelor's—obviously this is a pretty small group of people—then they would let you start in the second year of their three-year bachelor's program. That's what I did. It was very hard, very challenging intellectually, because the European high schools cover much more material than American ones, so by the time you get to the second year of college, you're way in there compared to an American university. They don't really have any ideas of liberal arts and breadth. It's just full-bore, 100%, engineering. So it was hard, but my tutors were very understanding. I told them, "I'm getting swamped here. I am having trouble with mechanical engineering, but I am getting the electrical stuff, maybe because it's more mathematical." They said, "Well, why don't you just come up to the end of the year, and take the electrical exams, and skip the mechanical ones?" By the end of the first year, I was caught up and in the second year I did their electrical sciences program.
ZIERLER: Did you go in with the intention of staying on for the master's? Was the master's a terminal degree? How did that work?
RUTLEDGE: Oh, the master's is a fake. I should say the way the Cambridge—everything in the English system can be way behind the US in all sorts of ways, some good and some bad. When someone says they have an MA from Cambridge, what it means is that five years after their bachelor's, presumably if they're not in jail or something that embarrasses the university, their master's will arrive in the mail. It's easier for me and my resume just to say, "bachelor's at Williams, master's at Cambridge," because otherwise it looks a little strange. But in England, it's the same degree. It just becomes a master's after five years. There's no additional study—the diploma just arrives in the mail.
ZIERLER: Did you think about staying on at Cambridge or in Britain for the PhD?
RUTLEDGE: No. Cambridge has a tutorial system, and most of the tutors are professors, but a couple were graduate students. I remember looking at one of the projects that one of the graduate students was doing and thinking, "They're pretty behind the times here." The other thing is, I had been working pretty hard, I felt, in school. I always found school challenging, even high school, so I was really ready to go work, make some money, and become independent financially from my parents. So, it never really occurred to me to stay. I had a wonderful job in the summer. Cambridge required industrial practice for this bachelors. I worked at Vickers, which is a huge company. They make battleships and tanks. The division I was in made machines to analyze blood. I implemented one of the first bar code readers for vials of blood. These are the labels that you use in the supermarket checkout now. The bar codes allowed the computer to read the patient information from the vial when its instruments were analyzing the blood with chemical and light tests. For that time, it was a very sophisticated piece of equipment. The group of people I worked with was absolutely fabulous to work with.
But at least in those days—this is pre-Thatcher—it's hard, if you visit England, to realize, unless you're my age and lived in the UK before, what a shock Margaret Thatcher was to the UK. The England I knew was poor. The England you see now is not poor by any definition. She really upset the apple cart, in all sorts of ways, and it still reverberates. Every time I visit the Cambridge University Press, my editor and I hang out for hours in the bar and talk about writing and books. She's a wonderful correspondent and loyal friend—but if the word "Thatcher" ever came up in the conversation, there would be this kind of volcano, that this is the worst thing that has ever happened to the UK. In more rational moments, when someone has that reaction, if they're not going to throttle me, I say, "But I was here before Thatcher. You guys were poor then, right?" They'll say, "Right." "And you're rich now, right?" "Right." "Okay, so—" The world is full of tradeoffs. I really liked England, but it definitely was poor then, and part of the reason it was poor was in a place like Vickers, we were probably putting in 20 hours a week of work. The manager was off in a different part of the factory. I rode my bike ten miles to get to work. I'd show up before work like I do in the US and the entire factory would be empty. Nothing would start until 30 minutes after our nominal starting time. We'd take time for lunch, then we would play ping pong. If I waited until the nominal quitting time, I would be, again, the only person in the factory. That whole environment is completely gone now, I think, from England. I don't feel it now. Something radically changed.
So I wouldn't have considered staying, working in an environment like that. I was used to—my civil engineering and ranching experience—well, if you work with illegals, you will work hard. [laughs] If you just work the same hours the illegals do, you will have a long, hot day. Probably you could broaden it and just say if you work with immigrants, you will have a long, hot day. You will earn your pay. I was used to the environment where you're there before the starting time, you work until past the quitting time, and you're tired, you're beat, and you're hot at the end of the day. I just couldn't imagine staying in the English work environment of that time. The people at Vickers were great guys, and they did ask, "Do you want to stay?" I said, "No, I need to go home to Texas."
ZIERLER: Between growing up in Texas and a formative experience in England in the 1970s, do you think that planted a seed in terms of your later interests in energy and economics and society?
RUTLEDGE: Who knows! Texas is the place to consider the relationship between energy and society, but much of this I didn't discover until I started writing lectures. The biggest development in energy in this century has been the development of fracking. Over half of our oil and gas supply comes from fracked wells now. It's a revolution, and it's very much an American story, in that the guy that did it was a man named George Mitchell. He was the son of a Greek immigrant. He had a good long Greek name but his father's paymaster wouldn't learn the name, so his father took the name of his paymaster, which was Mitchell. George Mitchell had a small energy company, Mitchell Energy. He was interested in natural gas. He wasn't really a technical guy—he was a business guy—but he became absolutely obsessed with the idea that he could change the permeability of rock by fracturing it so he didn't have to go to a reservoir for natural gas; he could go to the source rock, where it was actually formed underground. That completely changes the resource level.
The place Mitchell chose to try this was Fort Worth, Texas. The wells are right under my hometown. The attitude there is utterly different from what we would see in California today. By the way, if you want to know why Los Angeles is here, think of the La Brea Tar Pits. In the early days, before there was seismic equipment, tar pits were a good place to drill. Tar pits result from a leak of an oil reservoir. Los Angeles started off as an oil town in 1890. People are not aware of it now. Every once in a while, you get reminded of it, you see oil wells in some west Los Angeles. Attitudes have shifted in California. In Fort Worth, the Trinity River runs a short distance behind my parents' house, and when I was a kid, the area behind the house up to the river was all woods. That was part of our playing around. We would just go back into the woods to play. Since then, civil engineers built a levee to stop the river from flooding, and there are gas wells by the levee. The wells run from there under the city.
Mitchell couldn't himself really make money from the fracking directly, because he only had half of the technology that was needed. The other technology he needed was horizontal wells that gave large contact with the gas source rock. There was a company, Devon, that could drill the horizontal wells, and so they bought Mitchell out. The combination of the fracking plus horizontal drilling got the current oil and gas revolution started. For my book, I went back behind my parents' house—it's my sister's now—to the levee and took pictures of the gas wells. Fort Worth is in the news this year because TCU is the local school, and they're in the national football championship, unusual for a small, private university, but they're there. If you drive down University Avenue, headed to TCU, you'll see a gas well in the parking lot. It's not the same here—you can't just go dig a gas well in Pasadena. But we do burn gas in Pasadena, so it's a bit hypocritical.
Initially, my main influence was Tom Tombrello, who was the division chair for Physics, Math, and Astronomy. Even though Tom was a nuclear physicist, he had had experience in oil. He had a turn managing in the Schlumberger Labs on sabbatical. And while he was in college, he had a job was working for the Shell Labs. I told you last time that I went to see a talk by someone Tom had invited, Ken Deffeyes, who spoke on ways to estimate future oil production. I thought I could solve that puzzle for future coal production. I certainly was going to try to use my hometown culture to try to understand it. Because in Fort Worth, a lot of people own shares in oil wells—the local golf course was known as a place where people would make deals. To dig an oil well, typically you need three investors plus a geologist. Growing up there, I knew oil people. But I don't think it got me into it. I think it was Tom Tombrello.
Career Launch at General Dynamics
ZIERLER: To clarify, you said after Cambridge, you wanted to work. Did you not apply to grad school for the PhD right away?
RUTLEDGE: That's correct. I went to work for a company called General Dynamics. A little bit more about Fort Worth when I was a kid—there really were two places you could work. I think I mentioned this in the last interview. You could work for what we called the bomber plant. That was General Dynamics at that time, and now it's Lockheed. They made the F-16. Now they assemble the F-35. Or, you could work in the meat-packing plants on the north side of town. When I got home from Cambridge, it was 1975, right after the final collapse in Vietnam. It was a terrible time for engineers. They had huge layoffs. I sent out a bunch of resumes, and had little luck. There was one guy in Houston who did seismic analysis, who offered me a job, and that would have been interesting. But a neighbor, Jimmy Vaughan, worked at General Dynamics, and I had known him, because I went to school with his daughter, Carol. He was a wonderful man. I went up to talk to him and see, "Do you guys have any jobs at the bomber plant?" He said, "Well, I will try to get you in to talk to someone, but we're still firing people." A plant like that, when they have a contract, employs 25,000 people; when they don't have a contract they have to get to down to around 7,000, the number that is needed to work on existing planes and to develop a new airplane. But he got me an interview.
The man who interviewed me, Paul Cyr, looked at my resume, and saw that I had a reference to Bobby Brown, a friend of the family. Bobby Brown was a physician at that time, but in his earlier life, he played third base for the New York Yankees. Paul was a Marine veteran, and he loved baseball. He had played against Bobby Brown during the Second World War. Paul wanted to talk about baseball, which I knew nothing about, but I did know about Bobby Brown, and he was interested. And I got an offer. That was my start, there. No one knew anything about a Cambridge degree, or would have cared, really. In that environment, engineers either came from Texas A&M or the University of Texas. Anyway, the way into the factory was knowing a former baseball player.
ZIERLER: Tell me about your time at General Dynamics. What kind of work did you do there?
RUTLEDGE: The work was very much nuts and bolts radio communications systems. When you work on a military product you need to figure out, "Am I okay with this ethically?" I tend to do things incrementally. I met some of the Air Force pilots. They were wonderful people. When the Air Force buys a fighter, they know exactly what they are getting, because they have expert pilots who test everything. It is different from selling a complicated technical product retail, where there's a real issue of, "Does the buyer know what they're getting?" I was very impressed with the pilots, and I thought, "If I do my work right with my radios, then maybe these guys can come home." That's as far as I got with the ethics. I'm not a mechanical engineer. If I was doing the cannons, I might have thought about it longer. I worked on the radios for a year. I began to think, "Well, I'm going to need more electrical engineering if I'm going to advance in this factory." I liked the factory, but I wasn't going to be able to get anywhere, unless I had more courses and learned more. But by this time, it was late in the application season, and I was past all the deadlines for the places I was considering except for two. One was the University of Texas at Austin and the other was the University of California at Berkeley. They're both fine places. Somehow, something got screwed up—it was probably me—but I never got an answer on my application for the University of Texas. Anyway, off I went to Berkeley.
ZIERLER: Did you go to General Dynamics with the intention of this being a short-term experience, that ultimately you were headed to the PhD? Or did you think maybe you'd make a life for yourself in industry?
RUTLEDGE: I never go for short-term experience, meaning I always hope it's the final position. I don't feel like I'm a shotgun guy, trying a lot of things. It's one rifle shot. It's just at that time, the job environment was so awful that I didn't have a lot of choices! [laughs] No, I fully intended to come back and work at General Dynamics. The company was nice. They continued my medical insurance. They didn't pay me money, but they were trying to be helpful. I loved the guys, and I liked the product, the F16. There's nothing quite like walking out of an airplane factory and have the guys testing your airplane taking off, lighting the afterburners, doing maneuvers over the factory. For an engineer, that's some pretty serious positive feedback. If you're not an engineer, it may not make sense, but a fighter with its afterburners low is loud and scary, which of course it is fully intended to be. So no, I intended to be long-term at General Dynamics.
At Berkeley, the first class I had used Amnon Yariv's textbook, and it was a Caltech PhD named Steve Schwarz who was teaching the class. He was one of the early laser people. I was pestering him after class, because from my experience in industry I was used to asking questions all the time. After a few lectures, he said, "Dave, you ask good questions. I think I've got a problem you'd be interested in." I went back to his office, and he gave me a paper to read on antennas for infrared wavelengths. I was interested. By the time I finished the master's, I was thoroughly hooked on his problem, and so I didn't go back to General Dynamics.
ZIERLER: Do you think your time at General Dynamics sharpened your interests for the PhD?
RUTLEDGE: No. In retrospect, the management know their business, and they could see that I was interested in intellectual, bigger-picture things. That is in electronics, not politics. So they gave me a dream project. They were making a proposal for a reconnaissance version of the F-16, the RF-16. They said, "Dave, why don't you design the data link for this project." The design report turned out to be 60 pages long, and it was the best kind of technical writing practice, wonderful preparation for academic life. If you're a math or biology major, you don't usually get enough writing practice for academic life, but I had it because of this assignment. Writing is hard work, and it needs practice. And I even had a paycheck! I loved it. I just wrote all day. The only thing that was awkward in that environment was typing. This was a union shop, and I was not allowed to type. However, my card would get me in the plant at night, so I snuck in at night, late enough that no secretary was there, so I could type and I wouldn't get reported to the union. I'm sure the average professor you have interviewed has written hundreds of papers, a book or two, plus chapters. You need practice before you inflict your writing on reviewers, and I had it.
Tungsten Whiskers at Berkeley
ZIERLER: At Berkeley in the late 1970s, was electrical engineering considered basically the same thing as applied physics, or was there a separate applied physics program?
RUTLEDGE: At Berkeley, at that time, they didn't even separate computer science. It was all one department. So, applied physics was in the EECS department. I always considered myself an electrical engineer. But the group I was in did applied physics, mostly research in gas lasers and dye lasers. But I had an antenna problem, which is electrical engineering. I was an electrical engineer among the laser physicists there.
ZIERLER: How did you go about choosing areas of interest, areas to focus on? EE is such a big field.
RUTLEDGE: The first project was the one that Steve Schwarz presented me with. I was doing the analysis of an antenna based on a tiny whisker of tungsten wire. He was wanting to make antennas for laser frequencies to make accurate measurements of the laser frequencies. He was an interesting advisor, not really typical at Caltech. Normally he didn't schedule meetings with his graduate students. But after a while, he said, "Dave, you're just going to keep working until I ask to meet you. Why don't we set up a weekly meeting?" So, I would work for a week, and then I would come to his office and talk for several hours. Most professors won't do that, but Steve was patient, and he just let me talk. Like the writing assignment at General Dynamics, it was great practice, talking for three hours a week on technical stuff. Both are absolutely great practice for being an academic person. Because I would just tell him what I was doing, and he'd listen. In this way, he let me develop my own projects. Again, great practice for academic life. Steve was a smart guy, and he'd offer comments, but he never really told me what to do. Then he went on sabbatical. He ended up as the contact for Berkeley in Europe, the university year-abroad program. He loved living in Europe. He was gone for one year in the days before email. So, I had much more independence than graduate students usually do.
The other thing that turned out to be important was that I made a valuable contact in the Army Research Office. Steve's contract monitor there was Jim Mink and I got to meet him to talk about my project. Later Jim became my contract monitor for many years. Another good experience was a failed project. In this case the numerical analysis was beyond what was practical with the computers and software available at the time. I spent six months on the project. The key in a failed project is to figure out quickly that you have made a mistake so that you can move on to something else.
ZIERLER: Tell me about the process of choosing a thesis advisor.
RUTLEDGE: In this case, Steve Schwarz chose me. He just said, "You ask good questions." I told him I was just going to be there for a master's, but after a month, he liked the way I was talking about the projects, and he had money for support, so he said, "Why don't I give you an RA [research assistantship] so you're not working for free?" We got along great. He has retired to Santa Fe, he and his wife Janet, and we still go by and see him from time to time.
ZIERLER: Tell me about developing the thesis project. How did that get started?
RUTLEDGE: My strategy—and this may be different from humanities and social sciences—is that I prefer three 1-year projects to a single 3-year project. The most visible professor in our group at Berkeley was a man named John Whinnery. He was Amnon Yariv's and Bill Bridges' advisor. He was a very good man, and I was a TA for him. But he tended to have long, three-year projects that were risky developments in new kinds of lasers. The problem with that, of course, is that you could come up empty-handed in three years, and then what do you do? I always liked things that were settled in a year. I chose those for my own projects. By the time I got near the end of the thesis time, I had papers for each of these, and those papers become chapters. At some universities, they do this formally. I had a sabbatical in Finland, and it's common in Scandinavia that the student literally staples together the papers, adds an introduction and a conclusion, and that's the thesis. I wrote mine like that. My papers had enough depth that I could simply edit each one to make a chapter.
ZIERLER: Did you have in mind applications even as a graduate student? Did you think about where this research could be applied out in the real world?
RUTLEDGE: No, not at that stage. I don't know if this is a normal transition for people—over time, decades, I've gotten more interested in the applications. At the early stage, it was just curiosity, I was just trying to make antennas for the wavelength that comes out of a far-infrared laser, which was 120 microns. I just thought, "That may be crazy, but how do you do it?" I was not thinking of an application.
There was an interesting spin-off from the work on the whisker antennas that I analyzed. The students who were working with them in the lab found that they were very sensitive. If a door slammed, they might short out. The guys at IBM Zurich found that the whisker current was also very sensitive to conditions of the surface of the metal crystal that the whisker contacted. They hooked the whisker up to a fine translator, and they were able to make maps showing the individual atoms of the surface of the crystal. They called this a tunneling microscope, and it won a Nobel Prize.
We were using the whiskers as laser detectors, so we really weren't thinking about that at all. If we thought about it, one of the things that people were trying to do was to directly measure the frequency of lasers. The microwave frequency sources are the standards. They are based on a number of flips of the cesium atom per second. The signal is in the microwave range. If you want to measure the frequency of a laser, which is much higher, you have to have some kind of nonlinear process like the contact between the antenna whisker and the crystal surface it, and it will produce a harmonic of that standard microwave frequency, which will mix with the laser signal, and produce a difference frequency that you can count. Then, after some arithmetic, we can say, "The frequency of this CO2 line is 30.126789 terahertz." That is of interest to the people in a bureau of standards or people interested in the spectroscopy of the atmosphere. I have always felt for applications generally; I am happy if the people applying it are physicists or astronomers. It does not have to be commercial or military. They don't have to be selling something. But at that early time, it was just, how could one make an antenna. It was an intellectual problem.
ZIERLER: Is that to say that the thesis in some ways was largely a proof of concept?
RUTLEDGE: It was an exploration first of, how does this tungsten whisker work as an antenna? Later on, it was developing planar versions, antennas that you could make in an integrated-circuit fab facility. We had a good one at Berkeley. I would say my thesis work provided the foundation for my group's work at Caltech. It was fortunate for me that I was relatively independent as a graduate student, so my group was not competing with my thesis advisor. The first application of my Caltech work was in making antenna imaging systems for tokamaks. You may have talked to physicists who do that—that they make plasmas, ionized hydrogen at very high temperatures, with the eventual goal of producing energy from fusion. We used our antenna arrays to movies of what was going on in the tokamak with a time resolution of a few microseconds. When I was interviewing for faculty positions, I interviewed at UCLA, and there was a professor there, Neville Luhmann. He had the idea that I could you use my antenna arrays to make movies of plasmas. I did not go to UCLA, but my students and I ended up writing 50 papers with Neville and his students.
ZIERLER: What do you think your principal contributions were, with the thesis?
RUTLEDGE: My thesis set up the work for the next twenty years in my group at Caltech. In one of the thesis projects, I worked out a theory for how strip transmission lines on integrated-circuit substrates behave and confirmed the results experimentally. It turns out that the radio signals from the metal strips leak into the substrate. The process is analogous to Cerenkov radiation in physics. This leakage was not appreciated before. The leakage can cause interference between signal lines or the leaking radio waves can be encouraged to radiate to make an antenna. I was quite surprised at how often this work was referenced, much later, by people making pulse measurement systems.
Another application has been in radio astronomy antennas. Tom Phillips and I wrote an article once on this for Scientific American. Jonas Zmuidzinas told me once, "Dave, you really don't know how much your antenna ideas have influenced the design of radio astronomy antennas." Because they have special requirements, radio astronomers tend to build their own equipment, and I was not aware of what was going on.
A final idea from my thesis was that you can increase the sensitivity of a submillimeter-wave detector by making a grid of detectors on a surface and combining the outputs from the detectors. The signal-to-noise ratio improves as the square root of the number of detectors. Radio astronomers have since used this idea in making detectors for measuring the cosmic background radiation. After I started at Caltech, we began to make grids of transistor amplifiers. A major problem in moving to higher frequencies is that transistor power output drops as the square of the frequency. But if you put 100 transistors in a grid, you can get the power back. The students that developed the grid amplifiers started a company called Wavestream that was successful in selling high-power amplifiers. So I would say the thesis really set up the work that I did with my group.
ZIERLER: It's out of the chronology, but you mentioned Tom Phillips, who recently passed. I wonder if you could comment on his legacy, and what he represented in the field.
RUTLEDGE: Tom interviewed me when I came to campus, applying for an assistant professor position. He probably was on my tenure committee, too, because I remember one time I was talking to him, just walking out of the Athenaeum at lunchtime. He said, "Dave, the tenure review is going very well." That's a very nice thing to say, because assistant professors usually do not get feedback during the review. But Tom was English, and the English academic culture is much more direct than ours. I appreciated it, and later on, when I was on tenure committees, I tried to emulate Tom's approach if the review was going well. People approach their own tenure in different ways. I did not really have a tenure strategy. I just did what I wanted to do. I think that probably is the best strategy at a place like Caltech. Just do what you want to do, and then see if other people like it.
Tom's wife, Jocelyn Keene is also an astronomer, and she was the supervisor of a friend of mine, Stan Whitcomb, so Dale and I knew her. Tom had a long period of Parkinson's. Jocelyn has had a tough road. Tom built a major submillimeter-wave telescope on Mauna Loa. There was a conference in Honolulu, and I had a group of 15 of my students and Japanese collaborators there. We rented several Suzuki Samurais and drove up the mountain to get a tour of Tom's telescope!
ZIERLER: To go back to the thesis, who was on your committee?
RUTLEDGE: I remember them all. They were each distinctive in some way. My advisor Steve Schwarz was a member of course. There was a physicist, Paul Richards. Paul was the leading intellectual in infrared and millimeter waves then. He was extremely active, interested in both technology and in the microwave background radiation. I would say just below Nobel Prize level. The next was Ted Van Duzer, who was prominent in superconducting work and he was the co-author of the leading textbook in electromagnetics. The last was Diogenes Angelakos, who was the manager of the electrical engineering department laboratory. Angelakos was from a previous generation, before professors had government contracts. You may remember the Unabomber, Ted Kaczynski, who set off bombs in universities. Well, Diog picked up one of the bombs. The bomb was a Coke can. There was a coffee room and I used to go there, but I wasn't there that morning. Diog was the kind of guy who would clean up things, and he picked up the Coke can to clean up, and it severely injured his hand. Anyway, that was the group. The only thing that was funny about the exam was that Paul Richards was such a dominant personality that he asked virtually all the questions for two hours, and then he said—I can't do it right; I'm not an actor—he said something like, "Humph! You knew more than I thought!" I knew I had passed, then! [laughs]
ZIERLER: You mentioned that the research really set the tone for the next ten years in your research life. That begs the question, what was the jumping-off point? What were the big unanswered questions or projects to do, that built upon the thesis?
RUTLEDGE: There really are three parts. One is this question of how do you make antennas in an integrated-circuit environment, which is challenging. It's planar, you're evaporating the metals. You can do some etching. You can do some polishing, like people who make lenses. That was one problem. The other one was understanding. When you put strips of metal on silicon or quartz, how did the radio waves behave? In this case, the surprise that people had, which I think I was the first one to develop a theory for, and to confirm in measurements, was that it leaked radiation that could interfere with other circuits. You could also use the radiation itself as an antenna, or you can design the structure so that it minimizes the leakage. But in all cases, you do have to take that physics into account. The third was this collective idea that because you're making an integrated circuit, you don't have to make one detector or one transistor; you can make a hundred. Then if you have a hundred transistors, maybe you could make a much more powerful amplifier. If you have a hundred detectors, you can make a more sensitive detector. Those were the three themes.
When I was a graduate student, this would have been strictly an intellectual interest, without a practical application. I just thought, "Oh, there's something that would be fun to do." But as the ideas developed, and you get some students working on different parts of the problem, and then you get the contract monitors interested. The contract monitors tend to view it as their job to poke around in different places for possible applications. So there's an evolution, in intellectual terms, first understanding what is going on, and then figuring out, "Are there any customers for this?"
Falling in Love with Caltech
ZIERLER: Did you look at postdoc opportunities as well, or strictly faculty appointments?
RUTLEDGE: I had one postdoc interview, at Stanford, and I just wasn't interested. [laughs] It's a different generation. I know people tend to do postdocs now. My son Robb is a professor of psychology at Yale, and in that field, you do many years of postdocs. Part of it, I was still a working engineer in my mind. I would have been happy to go back to the fighter factory. I wanted a real lab of my own. So, I just didn't consider it. Stanford is across the bay from Berkeley, and it would have been a good appointment. But after the interview, I just told them, "It's not going to work."
ZIERLER: Besides Caltech, where else were you looking?
RUTLEDGE: It was a different era. I applied to three places and I got an offer from each one. One was Caltech. Then, at the University of Illinois, again I met extremely impressive people. One was Nick Holonyak, who was the inventor of the LED. He did not get a Nobel Prize, but his work was more important than 90% of Nobel Prizes. As we talk, we are looking at a screen lit by LEDs. The guy who makes the first LED, that's a big deal. The other one was Bell Laboratories. The person I met there was Bob Wilson, who was a Caltech PhD, and who won the Nobel Prize for discovering the cosmic background radiation with Arno Penzias. That would have been a great place to work. For a while. Eventually, Bell Labs shut down when the Bell system split up.
My wife Dale is Chinese American, from San Francisco. At that time—1980—when we went to Illinois, she thought, "There are not many Chinese people around here. I don't know if this is going to work." This was not an issue for Bell Labs or for Los Angeles. But I fell in love with Caltech. I think part of it was that it's about the same size as Williams College, where I was an undergraduate. I told you about Stan Whitcomb, the first guy I met from Caltech at Cambridge, who I thought was a wonderful man, one of the smartest people I've ever met, and one of the kindest. I thought Caltech would be a great place to be. One of the things that made a huge impression on me as a graduate student was that I was sitting in my office at Berkeley, probably feet up on the desk daydreaming, and Roy Gould, the division chair, and Bill Bridges, who was the EE executive officer, came to my office looking for me. The Caltech EE Department had lost a lot of people, and they came up to Berkeley and asked around the Department, "Who should we talk to?" I had taken a class from Bill Oldham in how to make integrated circuits, and done a project for his class that he liked, and Bill sent them down to talk to me. Roy Gould and Bill Bridges are very impressive people. Roy just died, unfortunately, and Bill has been retired and living up in Virginia City, on the Nevada border.
When I came down and gave a talk, at dinner I met Tom McGill, Dave Middlebrook, Chuck Wilts, and Hardy Martel, all friends later. Tom McGill told me after dinner, "I think we're going to make you an offer." That again makes you feel good. Nowadays people are more careful because of all of the bureaucratic rules. When I got an offer, I called up Roy Gould immediately. I said, "I'm not going to give you time to realize your mistake." [laughs] Roy said—because he was like this; he was a modest man with absolutely nothing to be modest about—"Well, Dave, we were going to get you down here to try to persuade you to accept, but we'll invite you down anyway and we'll have a party." He also introduced me to Jane Caughey, our realtor, who was married to Tom Caughey in the Civil Engineering Department.
ZIERLER: Joining the faculty at Caltech, what did you feel like your academic area of expertise was relative to the Department or the option? What did you bring to the table as a junior faculty member?
RUTLEDGE: There's two aspects of this. One is falling in love with Caltech, which is the most important. The second is that even given that, you still need to figure out the practical aspects. The practical aspect was that the Electrical Engineering Department—Jim Mayer had gone to Cornell, Floyd Humphries had gone to Carnegie Mellon, Charlie Papas, Chuck Wilts and Joe Langmuir were retired or soon to retire. Hardy Martel was not taking on graduate students. Amnon Yariv was focusing on Applied Physics. Carver Mead was focusing on Computer Science. So we were really down to a small group in terms of Electrical Engineering research efforts. I thought, "From a practical side, I can help." [laughs] I could teach a required class. I liked to teach E&M [electricity and magnetism]. No one else did. JPL was interested in my antenna work for spacecraft receivers. So I had a natural place to look for people that might use my work. Like I said, the chapters of my thesis, when I could find support for them, eventually became student projects.
There was a teaching contribution that was important to me—a wonderful man, Jim McCaldin—he didn't have a large group at Caltech, but he developed some of the very first fast solid-state diodes, at Rockwell. He taught a class for freshmen, Applied Physics 9, that Axel Scherer still teaches now, in how you make diodes and transistors. After a year of teaching E&M, which I enjoyed, Jim came to me to talk about taking over his class. It was a very large class and there were a number of emergencies. Jim was getting near retirement and the class was more than he wanted to handle. He had just had a student sent off to the health center who had squirted hydrofluoric acid on herself. He explained to me about the class. I was fascinated with the idea, and I took it on a year later. The class was a lot of fun to teach, and eventually the enrollment grew until half of the undergraduates in the university were taking it. We had lab sections running seven days a week. But it was a lot of fun, and it greatly influenced my teaching approach when I developed other classes.
I received remarkable support for the class from Roy Gould, the division chair. Roy said, "Jim was complaining it was too much work." He said, "We'll pay for someone to help you with the lab." I hired a man, Bill Rader, to come down from JPL, and when Bill retired several years later Kent Potter took over the job. Kent worked with me for thirty years until he and his family retired to a farm in Idaho. Kent has been a lifelong friend. It is rare to hire a technician to help with a single class, and I was grateful to Roy Gould for it.
Roy also got me involved very early on in recruiting faculty members, which is critical for an assistant professor, because you learn all sorts of things about what is really important in a university, not just your own research project. We were trying to recruit Dave Aspness from the Bell Laboratories. We failed, but one day Roy came in, just poking around my lab. Roy was a lab rat himself. He said, "Dave, I've collected some money to try to get Aspness to come, but he didn't come, so let me give you the money." Again, no dean, no division chair, ever does that! [laughs] The way the world works is if you want money, you have to ask for it. It doesn't just fall out of the sky. But Roy was like that. And there was no way I was going to leave Caltech, ever. [laughs]
ZIERLER: What were some of your early impressions of how Caltech did things, from a faculty perspective?
RUTLEDGE: The main one was how little administrative apparatus there was. I used to tell people that Caltech liked to find out all the things that could go wrong from not having enough administrators. I'm not a historian of Caltech, and my impression is that's not the way it is now, but I'd rather leave the growth of bureaucracy for someone like you to chart. As far as I could tell when I came to Caltech, everything was done by professors, including, say, the registrar, for example. Everything but the vice president for finance. And it turned that he was using Caltech money to pay for an apartment for his mistress. The one person who wasn't a professor turned out to be a crook. So, it was a place run by professors who understood professors, and that made it a great place to work.
ZIERLER: What were your impressions of Caltech undergraduates when you first joined the faculty?
RUTLEDGE: That's a very interesting question. I'll give you impressions. Let me give you a caution—my relationship to the students changes over time. When I'm a young assistant professor, I'm energetic. I'm at the top of my field. When I get older, I am not so quick, not so energetic. I've seen some things float under the bridge, some things that worked, some things that didn't. So your relationship to the students changes. I'll split the undergraduates into two groups, the earlier ones and the later ones. I liked both, but they felt different to me. The earlier group of undergraduates I would say were ruder, less socialized, much more likely to call me "Dave," even in class. But they had a backbone. I could assign a hard problem, and they would really have a go to see if they could do it. I think that experience is good for the students. The students that succeed learn something about the subject. The students that fail learn something about themselves.
I remember as an undergraduate in math—calculus is kind of a baptism for technical people in some ways because it introduces you to a new way of looking at the world. I was so excited in meeting calculus my freshman year that I worked every problem in the book. My roommate, who was a pre-med and thought that you only do exactly what you need to do to get into medical school, and not a minute's worth of anything else, was horrified with the idea that someone would work the problems because, "Wow, this is amazing stuff."
I remember one student who was ornery named Alan Cocconi. He was in my electricity and magnetism class, and he reacted badly to being told "Turn in homework on time." So, I wrote him—in those days, a real, paper letter—that read "Alan, if you don't turn in homework, you're going to flunk the class." I got no homework and he flunked the class. Alan turned out to be an absolutely superb power electronics engineer, world-leading. GM's first electric vehicle, the EV1, and the early Tesla development owe a lot to Alan. But he has his own shop, AC Propulsion. He was and is an independent guy.
There was another student who was also my graduate student, in my microwave class, and he got interested in other things during the class. At the end of the term, I asked the teaching assistants for advice. "What should we do?" They said, "You should give him an F." I ended up giving him a D. His name is Gabriel Rebeiz. Well, he's in the National Academy of Engineering now. Gabriel and I are good friends and we write each other all the time, even though I'm retired. So, you need to give people intellectual space.
Something changed around 2000. The later students were polite—they were angels. My wife was always my assistant and she was in the next office. They'd go sit and chat with her while they were waiting to see me. She ended being long-term friends with some of them. When they talked to me, it was always "Professor Rutledge" or "Prof"; and always respectful of my time. However, if I gave them a hard problem, they were not going to work on it for hours and try ten ways to get the solution. They would cave in.
Of course, some students did not fit these characterizations. There was a woman in my electricity and magnetism class who communicated extremely well, but she was having a terrible time with the material. She was close to failing. But she was thoughtful and disciplined, and she came in to talk to me about it. The two of us worked through my office hours each week on the problems. I told the students in that class at the beginning of the term, "Forty-nine percent is an F. Fifty percent is a pass. She ended up with 51. She needed almost every point she could get to pass. But she did pass. Well, she went off to work for Chrysler, and a couple years later, the recruiter comes back and says, "Do you have any more like this lady? She has saved several programs. She was the liaison between two different factories. She gets everyone to talk to each other, and she is solving problems before they come up." So, you do not know how things will turn out. The Caltech kids, they're all talented in some way. Some of them work harder than others. A lot of them surprise you, and some surprise themselves.
ZIERLER: Caltech is known for supporting its junior faculty members, where the idea is, they want to give them everything that they need to succeed, ultimately to gain tenure. I wonder if you could speak to that for your own experiences.
RUTLEDGE: Oh, more than that, because of course I was division chair, too. It absolutely distinguishes Caltech, and it's a critical retention advantage. It shows up in several ways. In the salary surveys that I used to see, Caltech would be number one for assistant and associate professors but maybe number ten for full professors, replaced by Harvard and Yale, say, at that level. So, we paid the junior people well. I think the senior people know this, and accept it as part of the system. The assistance we gave people to help buy them houses was unique. I may be somewhat out of date on this one. In other universities, the universities co-owned the house, and this was a problem when the faculty member wanted to sell the house. But we just increased the salary to help cover the mortgage. I came to Caltech at the time of the first housing crunch, this mortgage help program did not exist. It was a terrible time. Bank interest rates were something like 18%. We ended up with a 5-year balloon loan for half of the value of the house from the seller, but we still had to come up with a 50% down payment. We had savings, but I was short $30,000 in putting the package together. I talked to Roy Gould, and he told me, "Talk to Greg Van der Werff in property management, and see what he can do." He came up with a $30,000 loan. That support was a big deal.
When I was division chair, if someone came in with a competing offer, when Paul Jennings was provost, I could go talk to him and come back in an hour with a matching offer. I don't think you can do that in other places. When I was chair, I don't remember losing anyone to another university. I'll give another example. Once we were able to give a significant cash boost to a faculty member we were recruiting to allow that person to move a grandmother with them. We can do that because we are private. I think it would be harder for UCLA. And you could help someone at a critical time—divorce, where it may be difficult to make house payments. All of this makes for loyal faculty. This is different from the East Coast universities Our son Robb is going through this now. He is an assistant professor of psychology at Yale. They hire lot of assistant professors. At the end of the tenure period, they may tenure one out of the group, or they may go outside and hire a full professor instead. We just don't treat people that way. We give people a fair shot at tenure.
In the early days, Dale and I went to lot of faculty parties. Our house was not big—1,200 square feet, plus a 150-square foot greenhouse—but we'd have a hundred people at the party. The provost, Robbie Vogt would have weekly parties for ten in his home. I remember once, it was a long evening, and it was near midnight, and I had to go to the bathroom. When I got back, I had broken the ice, and everyone else had gone. So there I was with Robbie, who was definitely the scariest provost we've had [laughs] while I've been here. It did not have to be people in the same department. I remember going over for an evening with Fred Culick, who was in aeronautics, with five or six other people. It was a wonderful way to get into Institute life. But it is gone now. I wonder if it has something to do with the amount of time we spend in front of computer screens. My wife Dale's theory is that the spouses are all working full time now.
Faculty tend to stay at Caltech. There is a loyalty to the place. Much of it is due to the way people like Paul Jennings as provost and Roy Gould as division chair treated people. But I think they felt that there was a loyalty, almost to a level that didn't make any sense, like the things I described from Roy Gould. But it was the way they operated. I don't know if they can now. We have gotten more administration, more rules, and more controls. All of those things reduce flexibility going forward. But that was the way it was, and it was wonderful.
Fabricating Integrated Circuits
ZIERLER: Talking about support of junior faculty, what about in terms of getting your research operation up and running?
RUTLEDGE: It was different then for start-up funds. The amounts of money were much lower. There might be no start-up money if you were a theorist. This was before personal computers. If you were experimental, it might be $50,000. The start-up numbers are enormously bigger now. On the other hand, you could develop funding by building up personal relationships over time. Initially I wrote Jim Mink in the Army Research Office, "Here are some of the ideas. You funded this research when I was a graduate student." He said, "Dave, I don't have any money." But a year later he wrote, "Dave, I've got some money now. Why don't you write me a proposal?" You start small. Now much of the funding is for large groups, and you need to get in one of those groups. When I started, it was individual contracts, mostly.
ZIERLER: What did you need to get started, from an instrumentation perspective, from building up your group? What was necessary for you at the beginning?
RUTLEDGE: Oh, pretty serious. We were making the integrated circuits in my own lab. You needed several lab bays. You needed equipment to cut the circuit patterns for photography. Barrel cameras to expose photographic plates. A mask aligner to position the plates for lithography. Chemical hoods for the acid etches. Furnaces for growing oxides. You needed evaporating systems for making thin metal films. A profiling stylus for measuring the thickness of films. And that's just the fabrication. On the measurements side, the microwave equipment is extremely expensive. You need a network analyzer to characterize the microwave circuits. You need a room to measure antenna patterns. Some of it you build. Some if it you buy. It took many years to put it together. Sometimes the students, when they start, have to build some equipment. I told you about Neville Luhmann at UCLA. He had some vacuum systems that I could use for evaporation that were sitting outdoors in the yard at UCLA. He said, "Dave, you can take them."
My first student, Dean Neikirk, I got through a funny circumstance. I had been sitting in my office, my legs on the desk, looking out the window. It was in late June, 1980, about a week after I arrived. Dean had started with Professor Ahmed Zewail. Dean had a temper, and he had gotten in a fight with Zewail and quit, but he had quit without having a new [laughs] advisor. He was that way. He had his sense of what was ethical and what was not. He went over to talk to Professor Tom McGill. But he had gotten a bad grade in Tom's class, so Tom didn't take him on, but he sent Dean over to me. I talked to him. He came in the office and I said, "Well, what can you do?" He said, "I can put vacuum systems together." I said, "Okay, let's get started." We rented a truck and went over to UCLA and got Neville's vacuum systems. It turned out Dean could put vacuum systems together. He was an etch exceptionally good student. He started a couple of projects while I was out on vacation. He turned out to be a prominent faculty member at the University of Texas, kind of a shop steward for the faculty of the University, representing the faculty to the administration from the point of view of their work conditions.
ZIERLER: Did you take on graduate students right away? Did you want to build up your group right from the get-go?
RUTLEDGE: I absolutely wanted to get things going right away. I started with Dean in the first couple of weeks. The second student, Chung-en Zah, started that fall. It was easier for me than it might have been for people at other times, because the electrical engineering department had gotten so small, and people were taking on few students. I got a chance to talk to several of the students who might have gone to other professors in a different time. Chung-en Zah had an extremely successful career at Bellcore, Corning, and as an entrepreneur. He and his wife gave a SURF Fellowship in the names of Bob McEliece and me. We had a group tradition of going to the Rathskeller in the Athenaeum every Thursday for drinks and chips. People brought their wives, girlfriends, kids. Often colleagues from JPL would come also. We sat around, talked sports, solved the problems of the world, whatever. Any topic was okay in the basement of the Ath. It stayed there, mostly. You had freedom to talk.
ZIERLER: I'm curious if the early years of computer science at Caltech, what people like Carver Mead were doing in the late 1970s, registered with you? Was that relevant at all for what you were working on?
RUTLEDGE: It's not relevant. Carver is interested in a lot of things, as I'm sure you know. One of the things he has been interested in is trying to understand energy issues. I sent him my textbook, and we had a dinner to discuss it. He is a wonderful dinner companion.
ZIERLER: I'll state on the record that Carver is a huge admirer of your book. He just gushed about your book.
ZIERLER: And he's a tough grader!
RUTLEDGE: Oh, okay! Carver and I had some discussions early on about how to judge work in different science and engineering fields. It was very helpful to me. Back to computer science. It was becoming clearer that a lot of people were interested in computer science. My sense is that computer science has a situation that we don't have in electrical engineering and in mechanical engineering. You may need a master's degree to be effective in industry in these fields. This is not true in computer science. You have industry leaders like Steve Jobs, Bill Gates and Mark Zuckerberg with only limited university computer science background or none at all. Apple founder Steve Wozniak went back to Berkeley to get a computer science degree after Apple was well established. It is not clear how much value a university computer science degree is adding.
For a long time, the Caltech computer science department didn't want to have a major. Computer Science professor Mani Chandy felt differently. He and I came up with an idea of, "Well, maybe we could have a hybrid major. Let CS pick half the classes, let EE pick half the classes, and then we'll call it Electrical and Computer Engineering (ECE)." We did that for a few years. The interesting thing about the major was that it was quite difficult because the culture of EE and CS are quite different, and the students have to be able to get into both. It's like learning two languages at once. It's hard. However, the students that came through that were really outstanding, and Dale and I keep track of some of them, partly because my son came through about the same time, and he was friends with some of them. There came a time when CS felt they were really ready to do a major, and they had more students. So we moved on from the ECE major. But I don't think we hurt the ECE students. But there still ongoing questions of what a CS degree actually contributes, when a Harvard dropout can have such success in a computer company. This is different from the early Silicon Valley companies—Gordon Moore had a PhD in chemistry, for example. Something is different about CS, and historians will want to sort this out.
ZIERLER: Last question for today. For your academic area of expertise in the early 1980s, what was the frontier in antenna arrays? What were the big questions in the field, and as a young professor, where did you want to make your contribution?
RUTLEDGE: The first one, I had made some progress on as a graduate student, which is how you could make antenna arrays in an integrated circuit environment for the higher frequencies. The state of that now—I told you about my student, Gabriel Rebeiz at the University of California at San Diego—his work is the state of the art, and he's in the National Academy of Engineering for it. Elon Musk's Starlink system uses his array approach. Starlink is an amazing system. I've been experimenting with it—on my boat. All of a sudden, out in our Channel Islands, 70 miles offshore, I can get an internet speed of 150 megabits per second with Starlink. Since Elon Musk is using low satellites, their position changes quickly, so they have to be tracked, and the tracking is electronic, by the antenna array.
Another area was transmitter power at the higher frequencies. When I started at Caltech, there were entire conferences organized by the government on the topic of "We cannot get enough power at"—pick a frequency—"30 gigahertz, 60 gigahertz, 90 gigahertz." When I heard these people, I thought, "If I replace the detectors in my grids with transistors, I could make a transmitter more powerful." That led to the Wavestream Corporation, founded in 2000 and its spinoff Mission Microwave. These companies have sold tens of thousands of high-power transmitters since then.
Another contribution was in teaching. For my microwave circuits class, we developed a microwave design software program called Puff. We decided to make the program available at our cost, which was 10 bucks. Dale distributed 30,000 of them by mail and through conferences. This is remarkable, given that the Microwave Society has 15,000 members. It ended up in college courses, in amateur radio projects, and in commercial products in countries that do not have a highly developed industrial structure.
ZIERLER: In our next conversation, we'll pick up on the road to the tenure decision for you in 1984, and we'll continue from there.
[End of Recording]
ZIERLER: This is David Zierler, Director of the Caltech Heritage Project. It is January 9th, 2023. I am delighted to be back with Professor Dave Rutledge. Dave, as always, it is great to be with you. Thank you so much for joining again.
RUTLEDGE: You are welcome.
Faculty Support and Promotion
ZIERLER: Today I want to go back to 1984, when you were promoted, when you achieved tenure as associate professor. First, just an administrative question. As you know, of course, today Caltech has done away with the associate designation. When you were division chair, did you have any insight or involvement in that decision, or do you know why Caltech made this decision?
RUTLEDGE: No, that change came after I was chair. I would have preferred to keep the old system. An engineering division chair has narrow views, focused on the interest of the engineering division. In some sense, Caltech is misnamed. Presumably, our name was chosen because of the prestige, well-earned then and now, of MIT, but we probably would be better named as an institute of science. There are four science divisions, and engineering, and HSS [Humanities and Social Science]. Engineering and HSS are the odd ones out. You are very conscious of this when you are engineering division chair. One of the major differences is in the way an academic engineer and an academic scientist are educated. When I got out of graduate school, I went directly to a tenure-track job, with six publications, including several conference papers. Scientists distinguish carefully between conference and journal papers. This distinction is not as important in engineering. Some of our engineering conferences have higher rejection rates than the corresponding journals. So engineers don't really distinguish strongly between conference papers and journal papers. On the other hand, a scientist would go through two postdocs and a stint as a research scientist. In engineering, we'd probably be scratching our head, thinking, "Well, what was that person doing all that time? Why don't I just go work for Lockheed, get a family going, buy a house. That's going to be a lot easier to do with an industrial job than a postdoc salary." But this long gestation period is normal in science.
Caltech approaches tenure by writing prominent people around the world, the further away the better, and asking, "How good is this person?" If someone in Tokyo or Sweden says, "This is a great person, don't let them get away," then we get excited and vote yes on tenure. If the letter reads, "I haven't heard of the person," then the division chair will have a difficult time advocating for the candidate.
As a result, the tenure cases for engineering looked weaker than the ones for scientists. I was always nervous when I was presenting tenure cases because of this. That's not to say they were actually weaker candidates; they were just younger. However, by the time we get to the full professor promotion, the engineers were competitive with the scientists.
ZIERLER: The last thing that we talked about, you were emphasizing that division chairs are parochial in their concerns, their outlook on this kind of thing.
RUTLEDGE: Well, I was. I think it's fair that all the engineering chairs, they will be conscious of the fact that there are more science divisions than there are—there's one Engineering Division, one HSS, and then four science divisions. So, you are conscious of how competitive your candidates for tenure are. Because the engineers often go directly, or at least certainly in my day, went directly to an assistant professor position rather than a couple of postdocs and a scientist position, the resume would be thinner, there would be a smaller number of contacts that would give strong references. But my perception was that by the time we came around to the full professor hearings, the engineers were perfectly competitive, and no one would be able to tell a difference in quality. My sense was that the associate professor rank, from that narrow view, was helpful in maintaining the status of Engineering compared to the other divisions. They can vote your candidates down, so you do have to think about that. If you eliminate the associate professor step, then you are basically presenting Engineering candidates for tenure and full professor at the same time that would not be competitive with the corresponding science ones, because they are younger. So, I thought the associate professor position served a useful purpose.
On the other hand, if I were a candidate, I would certainly appreciate the step forward. Partly because I did have industrial experience, I was used to the idea that people get evaluated every year. When you work as a professor, from that perspective, it takes some getting used to the idea that you may only be seriously evaluated twice in 40 years. The tenure step is a serious evaluation. The full professor step used to be a serious evaluation. When someone gets a chair, which nowadays is over half of the senior faculty, so it's not unusual by any means, that is basically a decision by the division chair or provost or both, but there's no serious review. So, you really only got evaluated twice. If you jump directly from assistant professor to full professor, it is only once. I'm sure you get evaluated seriously every year. And guess what? The rest of the world does, too! [laughs] I thought the associate professor was useful, because otherwise you get evaluated once in your career, when an engineer could be pretty young. I was hired at 28 and had tenure at 32, and the idea that my last serious evaluation would be at 32, and I'd quit at, say, normal time—66, something—that's a long time without anyone saying, "Hey, [laughs] how do things look?" So, that's the perspective from the chair's side.
From the candidate's side, the perspective is different. Caltech really gave a junior professor good access to graduate students. There was not a student culture that says, "Oh, we all have to go to the most famous, senior professor." I just never felt that among the students. If a student comes in your office and is looking for an advisor, and you give them a good pitch and they like the project, then they will come. That's great for a junior professor, obviously, because you can get access to really good students here. The critical time is getting started, and it's wonderful for that. I'm sure you'll know the Shakespeare Julius Caesar line of, "There's a tide in the affairs of men." I view tenure that way. When you have a research career, you go through a variety of projects. There's very much an ebb and flow. There is a time when people are beating down the door to visit your labs and hire your students, and then times that aren't so great. When I felt things were going well I went to the division chair, Roy Gould, and said, "I think it's a good time, if you want to have a go at tenure, because people are coming to visit my labs. They're excited about the work, and they're trying to duplicate it." He went for it, and that is why tenure came after only four years.
ZIERLER: In EAS, when you're coming up for tenure, is there an opportunity to talk about what you've done? Do you make a pitch for yourself, essentially, among your colleagues?
RUTLEDGE: I think that is a change, and I think that is the influence of the scientists, but the engineers, no. We're talking 1984, but at that time, it was really based on the relationship between you and the division chair. You would keep the chair posted, and send him a paper if you've got something interesting. Let them know that Harvard is showing an interest. It doesn't have to be crass, but you can give the chair a sense that people are interested. Chairs want to know. It's not something to hide. They want to know what the competition is. So you keep them informed. I kept a pretty open calendar when I was chair, so a lot of people would simply walk in.
I thought things were going well in my group after three years. Part of the reason was that the guy that I told you about, Jim Mink, from the Army Research Office, had started to fund the work. Neville Luhmann from UCLA also funded the group's work as a subcontract. So I was fine with funding, I had some good students, and we had some results that people were noticing.
ZIERLER: Were there other universities that were expressing interest? Was that a factor at all?
RUTLEDGE: There were several along the way. I did not encourage them, and in one case deflected their interest to one of my students, whom they hired. Unless you encourage them, there is usually not a formal paper offer. I would just pass along information to my chair, and I appreciated when faculty members did that for me when I was chair.
ZIERLER: What was Roy Gould's style like as division chair, and did you take anything from that experience when you became division chair?
RUTLEDGE: I love him. He's a great scientist, a great man, and a good friend. But the way to approach the job—you think about what he could have done better. Paul Jennings asked me, what did I want to do? I said I wanted to do some serious recruiting, because I felt Engineering had not gotten a competitive share of appointments in Roy Gould's time and later. It's not really a criticism of Roy. He was constrained by his own relationship with the provost. Obviously I would to try to behave like Roy did toward candidates. One of the things he did that I copied is that I met all the candidates. We got many search committees going, but I tried to meet all the candidates, and to take them all to dinner, one on one, just to make a strong connection to each candidate. Roy was like that. I told you that he came to my office at Berkeley, which definitely made an impression on a me as PhD student.
ZIERLER: When you achieved tenure, did that affect your research agenda at all?
RUTLEDGE: No. I did what I wanted to do from the very beginning. I even advise people sometimes—if you've got something you want to do, Caltech is a good place to try it, because the resources and the good students that you need to make things happen are here. So, just go for it, and see what happens. The only way I felt that I tweaked it was when I thought things were going well, I went to talk to the chair to let him know. It's still his call, because he's dealing with an audience of the other five division chairs who vote on tenure—and he has to judge his own relationship with them.
Contributions in Nuclear Fusion
ZIERLER: What was your involvement in the mid-1980s in nuclear fusion technology, tokamaks, things like that?
RUTLEDGE: Instrumentation. UCLA had an active tokamak program. It's standard to have a microwave interferometer with one beam through the tokamak. Paul Bellan, for example, will have one on his tokamak, and Roy Gould would have one on his, too. The free electrons in the plasma shift the phase of the beam that goes through the plasma. If you track the phase, you can plot the free electron density over time. It will rise as the plasma is struck, and then it will fall as the plasma decays. The problem is that this measurement doesn't tell you the shape of the plasma. I could fix that problem. We made arrays with 20 antennas that would tell you the electron density at different positions in the plasma. A kind of plasma movie with a time resolution of a microsecond. This would let people map the different modes in the plasma.
ZIERLER: Did you ever talk to Gould about his involvement with the AEC and all of the important work he did for fusion?
RUTLEDGE: It never came up. Roy, probably like me, was focused on the junior faculty development, not the other way around. The job of the division chair is, at least as I viewed it, as faculty development. Old professors telling stories wasn't the focus for Roy, and I hope it wasn't for me. There's nothing wrong with sharing stories over dinner, but when you're in the office, the focus is on the careers of the junior faculty.
ZIERLER: What about the culture of entrepreneurialism? Given the fact that so much of what you were doing was industry-specific or industry-relevant, what was the culture either within EAS or at Caltech about having ideas that might have a business value to them?
RUTLEDGE: That's a really interesting question. I'll tell one story first to show you that I'm not a businessman. Dale also has worked all of her life for non-profits. One of our sons, Alan, is a successful venture capitalist, and I talk to him often about his work, and this gives me a feeling for some of the skills that are needed for success in that work. In addition to Roy Gould, the other person who recruited me was Bill Bridges, the executive officer for Electrical Engineering. Bill was also my office mate for many years, and we are good friends. He was the inventor of the argon ion laser. These are the high-powered green gas lasers in laser shows. He was a board member for a laser company JDS Uniphase, and in his role as a director, he received stock in the company. He often talked to me about the company. It turned out that the company had a 100 to 1 run-up in the stock price, and I missed it completely.
One day, he walked in my office and said, "Dave, I'm quitting. The daily fluctuations of my Uniphase stock are now larger than what Caltech pays me in a year." Just before this, one of Bill's student advisees came into my office and asked if I would sign his registration card. He told me that Professor Bridges said he was quitting and that he should see me for the signature. I signed his card. It was pretty funny and I don't think the student was damaged by it. The students are not looking for a philosophy lesson when they get their registration card signed. They're looking for someone who says, "You've got enough classes, and the courses looks okay for your major." Anyway, I went through that whole time with Bill and never bought any Uniphase stock.
It is the same with my son Alan's companies. When he graduated from Berkeley, it was obvious that he was going to be in business. He had already started a company and sold it as an undergraduate. The funny part—they report company sales in the news—was that when he sold his first company, they met at a bar, but they had to go outside to sign the deal because Alan was underage. [laughs] I thought that was pretty funny. Anyway, that was his first company. I don't own any of his companies, either. Anyway, I would say when I came, patents and startups was pretty much ignored, with one exception. The Institute really wants Caltech to be the primary job. You can be on the board of a company. Bill Bridges was, but you are not supposed to be in a management position. This cost a colleague of mine his job. That policy was fine with me. I find consulting distracting, and I really was only interested in students and their projects. We did make antenna arrays for tokamaks, but that was not a commercial project. We didn't make instruments for radio astronomy, but we influenced their development. Again that was fine with me.
The change came when Caltech hired a new man to run the tech transfer office, Larry Gilbert, from MIT. He got Caltech interested in being an incubator. He pursued many more patents and encouraged startups. Venture capitalists like patents to help a startup company protect its technology. After Larry came Rich Wolf and then Fred Farina, who both continued Larry's approach. Rich Wolf was helpful to my students Mike DeLisio and Chad Deckman with their startup Wavestream. Rich provided management advice at a critical stage and introductions to venture capitalists. Wavestream developed some excellent transmitters for satellite uplinks based on the grid amplifier technology, and everyone, including the Institute, made money from patent royalties when Wavestream was sold to Gilat Satellite Networks.
Two other students, Ichiro Aoki and Scott Key, started a company, Axiom Microdevices, with Ali Hajimiri. Axiom developed a low-cost power amplifier for cell phones that sold extremely well. In this case the Institute traded royalties for stock. Unfortunately, when the company was sold, it was sold for less than what the venture capitalists put in it, so we received nothing. The message I got from this experience is to not give up the royalties. Ichiro and Scott moved on to establish another company, Indie Semiconductor, that focuses on integrated circuits for cars. That company has been very successful, and it is now publicly traded.
ZIERLER: An overall question, staying on the same topic, just the culture of tech transfer, the intellectual property environment at Caltech, what has changed, and what has remained the same, over the course of your tenure?
RUTLEDGE: This is my own perspective, because the kinds of patents I had basically would give a small company protection for its products over time until the product becomes obsolete. I'd take a case like Bill Johnson and his liquid metal patents. His patents likely would be applicable to a range of products that might be sold over a period longer than the patent lifetime of 17 years. Mine were fairly specific; like we can make a cheaper cell phone transmitter with this patent. The company that could take advantage of it is the one started by Caltech students. The patent does not really give enough information to allow, say, Lockheed to come in and make the product. There were some things that really weren't in the computer models that the students had to do by feel, maybe with corrections after a couple of failures. Lockheed might not know how to do these adjustments, because their people would not have enough experience in the area. For this kind of patent, the Caltech Tech Transfer Office is helpful. They were willing to work as an incubator, to encourage the students, to talk to them, and introduce them to VCs [venture capitalists] to get funding. In my part of engineering, you have to talk to a lot of VCs before you find funding, particularly to find a person who will lead the funding round.
Puff the Magic Software
ZIERLER: Lots to talk about with Puff, the design program. First, let's just start with the name. What does "Puff" refer to?
RUTLEDGE: You may have to be my age. In the early 1960s, one popular style for singing groups was a lady with a good folk voice, together with men who play the acoustic guitar and add harmony. Peter, Paul, and Mary was one of these groups and Puff the Magic Dragon was one of their songs. People think of it as a folk ballad, because of the style, but the lyrics are modern. The poem is about a boy who had an imaginary companion, a dragon that he would have adventures with. Eventually the boy grew up, and the dragon faded away. So, the idea was that you would learn on this program, Puff. You could lay out microwave circuits with arrow keys for a homework exercise or lab project. In other universities, the students might design one circuit during the entire semester, because they would be bogged down with the industrial software. Our students designed, built and tested a new circuit each week. Well, most of my students were not going to be professional microwave circuit designers. We set it up so they could etch the board themselves and mount the transistors and diodes on it, all within the framework of a weekly assignment. So in the course of a quarter, they would have built a mixer, an amplifier, a band pass filter, a low pass filter, and an antenna.
ZIERLER: Tell me about the collaborators on this project.
RUTLEDGE: They were my students. When I did a software project, I would typically write a short version and then see if I could find a student who was interested in doing something more formal. The person who first took on Puff was an excellent student named Rick Compton, from Australia. He was on a Fulbright exchange, and he came in to see me. I asked him to give a presentation to my group. He had looked at a metal structure in the shape of one of the coils that you would burn to discourage mosquitos. He had calculated somehow the radio wave performance for this structure. He was obviously a very good student and very determined. My wife and I visited Australia for a month and I worked at their standards bureau, CSIRO [Commonwealth Scientific and Industrial Research Organisation]. Every industrialized country needs someone who can tell you how to measure a meter or a volt precisely. It is interesting scientific work. But if they call it a standards bureau, its sounds boring, so they probably won't get enough funding to support it [laughs].
Anyway, I visited there, and it turned out to be commuting distance from his father Lyn and mother Eireen. Dale and I stayed in a small house in their backyard and Lyn drove me to work. He ran a concrete pump company, but that wasn't the interesting part. He was a teenager when the Second World War started, and he was one of the Australians who went and flew Spitfires over the UK to stop the German air attacks. He stayed in the Royal Australian Air Force until he reached the American equivalent of a three-star general, Air Vice Marshal. He ran all of engineering for the Australian Air Force. There was another connection, because he bought the F-111 that was made in the airplane factory I worked at in Fort Worth. Some F-111s were still on the assembly line when I worked there. He knew the test pilots. So we had a lot to talk about. After that, he managed the Sydney Opera House. He was clearly a broad guy like you rarely meet in life. He said two things that stuck with me. He said the hard part about being a fighter pilot was that you'd get the new guys, and you'd have dinner with them, and you could tell whether they would be back for dinner the next day, just by the talk. It's a tough deal. It's very dangerous to be a new fighter pilot. He said, "But you could just tell. You just knew if they weren't going to come back the next day." The other thig he said—he had a great sense of humor—in the middle of the war, they transferred him to the Pacific. The boat delivered him to Sydney but the war was up in the north, in Darwin. He said he figured out a way to drive to Darwin, which in Australia, in those days, took a long time. He said he had been in one war and he really wasn't in a hurry to go to the next war.
His son Rick went off to Cornell, and got tenure there. I think he could have done a software company, starting with his program Puff. He was so early that I think he could have produced competitive industrial software. But he thought he wanted to be a professor. He realized after he got tenure that he really wanted to be an entrepreneur, but by then, the software had moved on. However, he had an extremely successful career in communications startups. A very good guy, and a good friend all through life. He died young, unfortunately, at age 59.
ZIERLER: To broaden the context a little bit, in the mid-1980s, what was happening in the world of personal computers? Were they being adopted? Was there something about personal computers that you were specifically looking to address?
RUTLEDGE: Like the previous, you've hit a really excellent question. For the people who were interested in instrumenting labs, personal computers were a major step forward. There was an engineer who worked in my lab for a long time, Sandy Weinreb. He was the chief electronic designer for the Very Large Array. I don't know if you've ever seen the movie Contact with Jodie Foster. The Very Large Array is the star of the movie, along with Jodie Foster. I had met him when I was a student at Berkeley. Sandy was a visitor in Professor Charles Townes' group, Townes was a Caltech alumnus, and a Nobel laureate for the invention of the laser. Sandy wasn't a professor, but he taught a microwave class while he was at Berkeley, and he taught it in a novel way using forward and reverse waves rather than voltages and currents. I was a student in his class, and I was inspired by the way he taught it. It heavily influenced me when I began teaching microwaves. The funny part about the course was I had gotten married before the class started. My wife Dale and I went on a honeymoon, so we missed the first several weeks of class. He didn't take it personally and he allowed me to join the class late.
Anyway, Sandy, very early on at the Radio Astronomy Observatory Labs in Charlottesville, Virginia, had adopted the Apple III computer for making measurements. It was quite innovative. I didn't have any measurements that needed that, so I did not copy him. But we did have antenna patterns and impedances that we wanted to calculate. We had two choices, and they were both terrible. I did my calculations on a programmable hand calculator, and just let it run all night. The other approach was to share time on a big computer. The big computers had all sorts of quirks, and they really weren't that fast, because they were a shared resource. But little handheld calculators, you could definitely optimize. The programming language was a kind of machine code, and you could really pretty easily get to the limit of what they could do. But the limit of what they could do wasn't very impressive. They could give you a series of numbers that you could write down and plot. However, it was certainly better than ten years earlier, working a slide rule.
Then in 1983, the IBM PC [personal computer] came out. It cost 3000 bucks. We had several antenna problems that we needed calculations for. We wanted to put metal strips on top of insulators. In your cell phone, for example, that metal strip might be plated on a fiber-glass PC [printed circuit] board. That needed more serious calculations than people were able to do on a programmable calculator or a shared computer. When the IBM PC came out, we realized that it—not Apple—had a slot for a $50 numerical coprocessor integrated circuit that you could plug in and it sped up the calculations up by a factor of 50. It completely left Apple in the dust. That's why a lot of technical people, to this day, use the descendants of the IBM PC, rather than Apple, because of that original coprocessor.
Rick had some experience with the time-shared computers in Australia, and their problems, so we decided to have a go with PCs. He said, "Dave, I can run one frequency point with an overnight run with one of these." I went and got him ten Compaq computers. That was the start of the PC age for us. These were IBM compatible and they were the first portable ones. You could stack them. You couldn't really stack the IBM PCs. So we made a couple of stacks of these computers in the lab, and Rick would set them running, 24 hours a day. He could get 10 frequency points a day. We had the kind of problem where we wanted to make plots of the performance of an antenna shape versus frequency. If we interpolated carefully, we needed about 20 points to make a good curve. The speed was fine for us to make plots for our papers, and Rick said, "It's better than the big time-share computers." Then we'd go build the structures in the lab and measure to compare to the numerical calculations. I don't know anyone else who went and bought a stack of 10 PCs, but it was enabling for the kinds of calculations we wanted to do.
The other place you could see the impact, is that all of a sudden, the instruments would incorporate fast numerical coprocessors. For example, if you wanted an instrument to characterize an antenna, the instrument couldn't do the kind of interpolation a human would do, but it could take a lot of measurements and do the calculations that are need to make a plot on a TV screen. So a set of measurements that would have taken several hours to do by hand could now be done in a fraction of second, compatible with a TV display. You could see the plot and adjust the antenna to see the improvement in real time. That definitely helped investigators in microwave circuits.
ZIERLER: Who was the end user of the personal computer that you had in mind in creating the Puff program?
RUTLEDGE: Puff was written for my microwave circuits class, but the audience turned out to be the entire IEEE [Institute of Electrical and Electronic Engineers] Microwave Theory and Techniques Society. In addition, Europe had an active amateur radio community that focused on microwave electronics and Puff was popular with them. Later, with another student Scott Wedge, we did a 2.0 version. Then the European amateur radio community stepped up and Andreas Gerstlauer made further improvements. Andreas is now a professor at the University of Texas at Austin.
We did not do any marketing. I warned you that I'm not a business person. We just went to the microwave symposium. We got a poster session presentation accepted and then just brought a stack of floppy disks and gave them out at the session. There are 15,000 people in the Society and maybe 8,000 show up for the symposium, so it's good coverage. After that, we sold the program for $10 to cover the postage. We got letters from many people. It was used a lot in classes, and for some industrial development for simpler projects in developing countries, things like toys.
ZIERLER: Did you have a sense of just how well received the Puff program would be, how successful it would be in the long term?
RUTLEDGE: I guess I really didn't think that way. It gets back to the teaching. I wanted it to be useful in the class, and I liked the idea of students doing a large number of small projects. After each project was turned in, I would go over the project in the lecture and talk about how the simulation fell short. In an industrial program, there are lots of corrections that the user doesn't know about, so the user may not understand what is going on. I was interested in showing how a simple calculation doesn't quite work, and there are corrections you need to make it right, or things you need to think about to tune the circuit to get the best performance. So, my interest was almost always teaching. Puff lasted as long as DOS did. People ran it in a DOS window for many years after Windows came out. Probably 20 years.
ZIERLER: A year later, a paper that you coauthored in 1988—"An Alternative Approach for Designing Microwave Circuits Using a Personal Computer"—in thinking about Puff, what was it an alternative approach to? What were the other options?
RUTLEDGE: That's actually a broader question. This goes back to the class Sandy Weinreb taught at Berkeley that I told you about. If you just buy a book on electronics, or electricity, let's say, they will talk to you about volts and amps, voltage and currents. By the time you get to the microwave frequencies, the circuit component lengths are a significant fraction of a wavelength. This means that you can interpret what is happening in the circuit in terms of electromagnetic waves bouncing around in the circuit. Now, from the point of view of the computer, you can do the calculations either way. You can do it in terms of voltages or currents, or you can do it in terms of these forward and reverse waves. It was a little more natural to use waves when Sandy was a student, because microwave circuits then were actually like plumbing. There were pipes and the microwaves went around in the pipes, rather than in strips of copper on a circuit board. So it was natural to think about waves going back and forth, bouncing back and forth in the pipe. There was some supporting mathematics for signal flow graphs that was developed at MIT by Samuel Mason, that could be adapted to analyze the behavior of the waves in microwave circuits. In the paper, we developed the formulas that were need to do the wave analysis for any microwave circuit and we used that approach in the Puff computations.
Scott Wedge, the student who wrote Puff 2.0, later wrote a paper that showed that you could also use the wave approach for analyzing noise in circuits. In communications circuits there is often a competition between the signal that you want and the underlying noise, which is made up of unpredictable fluctuations in the voltage and current. If you listen to an AM station, you can hear the noise as a hiss. Scott developed formulas that predicted the statistical properties of noise in circuits in terms of waves and incorporated these formulas in circuit design software.
MESFETS and MOSFETS
ZIERLER: I've come to appreciate how historically important the Schottky diode is. I wonder if you can talk about it and how you've interfaced with it in your research.
RUTLEDGE: As I'm sure you know, there's very much a Caltech part of the story. There are two parts. I'm sure Carver Mead has told you his part of the story, but I can give you some perspective as a microwave engineer. The first thing that Carver did was to describe the physics of the interface between a metal and a semiconductor, that is a Schottky diode, in an intuitive way. This is characteristic of Carver's approach—he can do serious mathematics, but he only uses it when it helps. Schottky diodes have an energy barrier that only allows current to flow freely in one direction, and blocks the current in the other. Carver predicted correctly how large the energy barrier would be for different semiconductors. This one-way current flow is not perfect; you need a small turn-on voltage to make the current flow. Schottky diodes have a smaller turn-on voltage than ordinary diodes made entirely of semiconductors. This makes them more efficient in high-power circuits.
Schottky diodes are important for microwave engineers because they only use electrons to carry the current. The ordinary semiconductor p-n diodes, use both electrons and another conduction mechanism called a hole, which can be thought of as a shifting electron vacancy in the crystal lattice. Generally speaking, holes slow down the response of the diode. At microwave frequencies, you want fast diodes. The Schottkies are fast. With a Schottky diode, you can construct a microwave device called a mixer, which allows you to shift the received signal frequency down to a lower intermediated frequency [IF] where it is easier to manage the signal. The mixer forms the front end of the classic super-heterodyne receiver that dominated radio design for a hundred years.
The other thing you can do with a Schottky diode is to make a power source at extremely high frequencies through a process called multiplication that produces harmonics of the original frequency. For example, if you want power output at 300 gigahertz, that is a wavelength of 1mm, you can introduce a 100 gigahertz signal, say from a vacuum tube, to a Schottky diode. The result will be power at the second harmonic, 200 gigahertz, and power at the third harmonic, 300 gigahertz. So, it allows you to make a high-frequency transmitter just with a Schottky-diode multiplier, without a transistor or a vacuum tube that works at that high frequency. And you can bring a signal down with a Schottky-diode mixer. That's the importance of the Schottky diode.
Now, the Schottky diode can also be used as the controlling terminal, the gate, of a transistor. As you know, Carver is the inventor of the Schottky-diode gate transistor. The modern acronym for the Schottky-gate transistor is MESFET [metal-semiconductor field-effect transistor]. I got him to give me a photograph of his first MESFET, patterned with a razor blade. I had the photograph framed and mounted in the lobby of the Moore Laboratory [the electrical-engineering building].
The MESFET is an enabling device for microwave engineers for two reasons. First, it is fast because the operation does not depend on holes. The other thing—and Sandy Weinreb was the person who uncovered all of this—is that if you cool a MESFET, you can make an a very sensitive receiver, good enough even for radio astronomy. MESFETs formed the dominant radio-astronomy receivers for decades.
So Carver made two critical contributions, first the theoretical understanding of the Schottky barrier, and then the invention of the MESFET that uses the Schottky diode as its gate. Generally speaking, in radio engineering, until you have a working transistor, your technology progress is slow. Transistors allow an input and output to be separated in a stable fashion, and they produce gain, where the output is larger than the input.
ZIERLER: You mentioned MESFETs. I wonder if you can explain, from a semiconductor perspective, what the difference is between MESFET and MOSFET, and why you have been more on the MESFET side.
RUTLEDGE: In a MOSFET [metal-oxide semiconductor field effect transistor] the control terminal, the gate, is a capacitor rather than a Schottky diode, and the substrate is silicon rather than gallium arsenide. Generally speaking, MOSFETs are not as fast as MESFETs, and they used at lower frequencies. The MOSFET technology allows an enormous number of transistors. Computer circuits use MOSFETs. In the freshman Applied Physics device class we talked about, the students built MOSFETs. On the other hand, when we developed high-power 30-gigahertz transmitters, we used MESFETs that were fabricated at the Rockwell Science Center in Thousand Oaks. Our collaborator was a Caltech alumnus, Emilio Sovero. Emilio was in Fleming House, and he is an immigrant from Peru. He was wonderful with my students.
ZIERLER: Moving into the early 1990s, what were some of the advances in gallium arsenide and the technologies that were possible as a result?
RUTLEDGE: Well, there were two that were important for our work. The first was Rockwell Science Center's gallium-arsenide MESFET process. At the time, I had a student, Jeff Liu, who had superb lab skills. In recent years, Jeff was first mate on my boat. The Rockwell MESFET allowed Jeff to demonstrate the success of the grid-amplifier approach for making high-power transmitters based on combining the outputs from hundreds of transistors by radiating the power into space. However, the problem with the Rockwell process was that the surface of the gallium arsenide wafer was not stable, and the gain of the transistors dropped rather quickly, in weeks. The measurement was quite difficult, but Jeff succeeded in measuring the gain before the transistors faded. His measurements allowed us to write a paper on the device.
The second was the development of a stable high-power gallium-arsenide MESFET process by a company called WIN Semiconductor in Taiwan. A student who came after Jeff Liu, Chad Deckman wanted to sell a transmitter based on the grid-amplifier that Jeff had demonstrated. Chad persuaded another student Mike DeLisio, at the time a professor at the University of Hawaii, to join him, and around 2000, they started a company, Wavestream. One of the investors in Wavestream was an Oregon company, TriQuint. TriQuint had solved the stability problem in their gallium-arsenide MESFETs and they had a commercial foundry to sell circuits based on that process. They did not invest money in Wavestream, but they said they would make several wafers with Wavestream's designs on them. This was worth several hundred thousand dollars per wafer run. This turned out to be critical because Chad and Mike's theory was not good enough to get the design correct the first time. They needed the TriQuint runs to get the design right. However, by the time they had used up the TriQuint runs, they also figured out that the gain in TriQuint's process was not good enough for them to make a transmitter that they could sell. TriQuint was a good company, but their process was optimized for digital signal processing rather than transmitter power. However, WIN Semiconductor in Taiwan had developed a process specialized for power, and the gain was high enough for Wavestream to make a commercial success with their transmitter. The Taiwanese chips ended up costing about $800 apiece, but they were the critical components in the transmitters that Wavestream sold for $15,000.
Wavestream has a large facility in San Dimas now, but the process of starting up was interesting. Mike and Chad rented a few rooms in a bank building in Covina. They bought computers and rented design software to make the design files. The design files go to Taiwan, where the wafers are made in a facility costing a billion dollars. For the mechanical parts of the transmitters, different design software is used, and these files go to American foundries. Initially the assembly at Wavestream was done in a small room. To test the devices, Wavestream bought specialized measuring equipment which ordinarily would have been expensive, except that many companies had failed the dot.com crash in 2000, and the equipment showed up on the used market at low prices. The measuring equipment also fit in a single room. These much bigger companies, the wafer fabrication company, the mechanical parts foundry, and the instrument company, allowed Wavestream to develop their product. Without them Wavestream could not have succeeded.
Particularly note the dependence on Taiwan for the electronic circuit fabrication. Taiwan and South Korea have specialized in this, and they are good at it. Recently, I bought a Corolla. The saleslady was quite a character, and when she found out I was an electrical engineer, she said, "Why can't you guys make more circuits? I can't get cars to sell. Toyota can't get the circuits. If I could get the cars, I could sell them." And she could.
ZIERLER: A technical question—what are grid amplifiers? What is being amplified?
RUTLEDGE: I can try a couple of ways to describe it, but first, the goal. Within a month of the time I joined Caltech, I went to a NASA meeting, where many people said that they could not get enough power for communications and radar transmitters at high frequencies, 30 gigahertz and up. My sense of what was going wrong was that to make a transistor fast, you have to make it small, and then you can't get the heat out, so it can't be very powerful. There are other limits. When you make a transistor small, the voltage you can use gets smaller, and that shrinks the power, too. The transistor gets narrower, and that shrinks the current. Everything goes in a bad direction. I had the idea that in an integrated circuit, you could make a lot of transistors, but for them to generate a powerful radio wave, they had to synchronize somehow. What I did was to put the transistors in a periodic structure on a surface, and then hit them with an input radio wave that was coming not along the surface but rather normal to the surface. That way, each transistor would see the radio wave with the same phase, and then the transistors would radiate out the other side. Again, all with the phase. That is the grid amplifier.
In those days, when people made amplifiers that combined transistor outputs, a major issue was that they would use strips of copper on the surface to join them up, and when the signals were combined at the end, the transistors didn't agree on what phase, and the signal would get all jumbled up, and this reduced the power. The grid amplifier was our idea, and that was the basis of Wavestream, and they sold lots and lots of powerful transmitters for about 20 years. Then they moved on to a new technology now, gallium nitride, that can handle the power we were talking about with a few transistors, so they do not need the grid.
Optimizing Cellular Communications
ZIERLER: When did you get involved with the cell phone antenna work?
RUTLEDGE: It would be cell phone transmitter amplifiers, not antennas. The lead on this project was Ali Hajimiri. I had a couple of students, very good—Scott Kee and Ichiro Aoki. They were very energetic, close buddies, and still are. Now, they've founded a publicly traded company, Indie Semiconductor, that I don't own any stock in, so [laughs] again, I missed my business opportunities. But I set Scott and Ichiro to making powerful lower-frequency transmitter amplifiers that were extremely efficient. These were amplifiers where they mounted several transistors on circuit boards. They were taking Ali's class on integrated-circuit design, and thinking "We could make this kind of transmitter as an integrated circuit." So it was a collaboration between the groups. Scott and Ichiro were very independent and excellent students, and Ali is a very smart man and good advisor. They came up with a nice transmitter for cell phones in silicon, that probably cost a dime, quite inexpensive. It was used for the final power amplifier in the early cell phone technology that used the GSM [Global System for Mobile communications] standard. In the early stages, most of the world was using GSM, and many people wanted the absolute cheapest phone, so their company Axiom Microdevices sold a lot of power amplifiers for these phones.
ZIERLER: Was this really big business as cell phones were becoming ubiquitous?
RUTLEDGE: It's big in the number of power amplifiers. They were inexpensive individually. If you ever do an interview of Ali, you could get the whole story about Axiom. Ali and Ichiro were on the board, and I was not. There was a former Caltech student, Arati Prabhakar, now the Presidential Science Advisor, who represented the lead investor. The company went through a really unfortunate sequence. They had a product, they had huge sales, and they had an offer to buy, which—let me try to be tactful—the board did not accept [laughs]. Ichiro said he voted for it. Ichiro had already built and sold a company, in Brazil. Anyway, the unfortunate part after that was that there was an entangling lawsuit that took a while to get out from, and by the time they came around to selling the company, the value of the company had dropped. When that happens the investors get paid, but not the founders. There was a different company, Skyworks, that marketed the amplifier, and it went on selling it for quite a while. Then Ichiro and Scott went on to form their current company that they're with, which is focused on electronics for cars, and has been very successful.
ZIERLER: When did you first come up with the idea of the project that ultimately would become the book The Electronics of Radio?
RUTLEDGE: I told you in an earlier session about teaching the freshman class that Axel Scherer still teaches, APh [Applied Physics] 9, where the students make and characterize a number of devices—an LED [light-emitting diode], a Schottky diode, a pn diode, an MOS [metal-oxide-semiconductor] capacitor, a MOSFET [metal-oxide-semiconductor field-effect-transistor]—over two quarters. The labs are structured progressively, so that the later labs repeat the fabrication of the earlier devices as part of the construction. For example, the transistor structure includes an MOS capacitor as its gate terminal and pn diodes as the source and drain terminals. I was really impressed with the idea that you could engage students in a strong way if there is coherence to the laboratories, and not just random exercises. My experience of taking the Physics 1 kind of lab is, one week you roll stuff down an inclined plane, and the next week you have a laser measuring the speed of light, but they've got no relationship to each other. The cool part about APh 9 is that each lab builds on the other, because in the end, when you build a MOSFET, it includes a pn diode, it includes metal evaporated on a surface, it includes an MOS capacitor. It's the climax of the two quarters, and by the time the students get there, they are quite expert. They are comfortable with the evaporator. They are comfortable with the acid etches. They are comfortable with the furnaces. They've used them several times, and they have probably failed and succeeded at each step.
There was a transceiver project that my daughter built as an amateur radio operator when she was 11. Hams have had a tradition of building their own gear. There's a whole community that will do a design, make printed circuit boards—because that's often the limiting step, a board with the metal pattern—and a parts list. The transceiver was the NorCal 40a, designed by Wayne Burdick. Kate built it over the weekend. I sat at the computer in my office and she soldered away. She had one oopsie moment when she powered it up and dropped a metal screwdriver across the power amplifier. That shorted out the power transistor and it burnt up. But it turned out there was a transistor in the stock room in the sub-basement that was a good enough match. She soldered it in. Then she had this complete transmitter/receiver, functionally like a cell phone, except it was for a frequency of seven megahertz, not gigahertz, and it was used Morse Code, not voice. And it worked! In those days, there was a huge antenna on top of Winnett, for this wavelength, 40 meters. Kate had some contacts in Okayama, Japan, kids her age that she corresponded with. She could communicate with them in Japan, with this two-watt radio, with the antenna on Winnett. It was cool!
I looked over the radio, and it had the classic analog circuits in it. We were just before classes started in the fall. There was a professor, Dave Middlebrook—a guy who had won all the teaching awards, Feynman Prize, and ASCIT [Associated Students of Caltech] for his graduate student class. My students cherished their notes from his class. They revered him. He was a wonderful colleague, also. He has passed away now. But he really hadn't faced teaching beginning students, and he decided the class just wasn't a fit for him. I didn't hear a complaint from students; he was a very good lecturer. I think it was more he wasn't getting the results that he wanted to get. So, we needed a new instructor [laughs] for the introductory class.
I was thinking I could make this class like APh 9, where the students built a transistor, progressively, over two quarters. In the class I was thinking about, they could build the transceiver over two terms. By the end, they would know a lot about radio design. For example, they would build an oscillator, a circuit that produces a sine wave. Transceivers have several oscillators. To make one of these oscillators, you've got to take a magnetic donut and coil a wire around it, and hook up some parts. And I would give them a theory, and the theory won't quite work, so then in the class we can discuss, why doesn't the theory quite work? What's going on? That engages them, because when they come to the end of the class, they're really expert.
They also have some frustrations. They probably spoiled a part soldering, and spent more time in the lab than they wanted. Then when the circuit finally works, it's pretty cool. They're engaged, and they're ready for the lecture on the circuit. The other thing is, by the end of the second term, they really know how the radio works as a complete system. You can do the kind of tests that a company that buys radios—I used to work for a fighter airplane factory, and we bought radios, and then we did tests on the radios—how well does this radio hold up to jamming, for example. That is a standard test. The companies that make cell phones do all these tests for jamming and interference, because that's what limits how many conversations a cell phone company can carry. And the students would understand what the tests mean, because at that stage they were experts on that radio. No one had taught a class before where students could do the tests that industry actually wants engineers to understand.
This idea was partly based on my own experience as an undergraduate, I think I told you—I graduated in mathematics knowing nothing, zero, about electrical engineering, except for a Boy Scout merit badge in electronics. Then I went off to try an undergraduate program in electrical engineering. But Cambridge had an open lab. Even though I was definitely not up to the standard of the other students in terms of lab ability, I could stay all day in the lab, any time there wasn't a lecture, and work on the projects. So when I started the electronics class, EE [Electrical Engineering] 20, I got a lab room, and I put a punch clock on the door. This is convenient for Caltech students, who like to work at night. It can be a social time, because other students would be there. They could ask each other questions, show each other their circuit problems, stuff like that.
The social dynamics of the class was great. We wrote up an article for the Microwave Magazine about how we were teaching the class, and my wife Dale went in at 10:00 p.m. at night to take photographs for the article. The lab was packed, and the students were chatting away. It was not scheduled, but people were there, building stuff and talking, doing stuff engineers like to do, having fun, sitting at the bench and getting circuits to work.
Once I gave a talk at MIT on energy or a later topic, and I met the people in climate science, and one of them wanted to talk about the radio book, because he built radios. There was a former student, Sarah Warren, who wrote me last weekend. Her married name is Rose. She took the class and graduated in 2003. She wrote to say she really liked the class. She was a mechanical engineer, but she just took the class for fun, and she went off to UCLA and got a PhD in mechanical engineering. But then her first job was in radio engineering, and she made it her career. She goes around the world testing military communications. She sent me a picture of herself in a plane, with a laptop. She said, "It's great. People have no idea, unless I tell them, that I'm not educated as a radio engineer. I'm educated as a mechanical engineer, properly, with a PhD from UCLA." Then she tells them, "But I took a radio class once."
ZIERLER: To clarify, did you write the book with the intention of reaching out both to students and to amateurs, to radio enthusiasts?
RUTLEDGE: No, it was more just when you try to persuade a publisher to do a book, the editor will want to know what you think the potential markets are. I have seen biology textbooks where there are a corps of writers and several artists, and the book sells for several hundred dollars. People buy that book only if it is required in a course they are taking. On the other hand, the editor can come out with a paperback for $40 and see if he can pick up sales outside the academic community. I was telling the editor, Phil Meyler, "I think you could pick up some hams, if you sold it as a paperback. Students will appreciate the lower price also." He did come out with a lower-priced paperback, and I believe it turned out to be the correct decision. Phil was unusual; a lot of editors will not even consider a paperback. In the publishing business, people like to start off with a big price tag and hope that professors will require it. [laughs] When you go to Cambridge University Press, it's like visiting a steel mill in Pittsburgh. I mean, it's a huge operation, the largest academic publisher in the world. It's not a couple of guys sitting in an office in a bank building somewhere. It is a big, serious business. Someone has to think in business terms. Obviously that person is not me. Since then, Phil has risen up through the ranks at the Press. I think I told you in a previous interview that he was very unusual for an editor. I told him when I was having trouble with one chapter, and he pointed me to the right person, Ali Hajimiri—and—to help me formulate a response to the chapter. Ali's theory is too serious for sophomores, but I could use a simplified version of the theory that was good enough for the class and the book.
ZIERLER: Has the book needed to be updated? Has the technology changed, or is it still pretty solid almost 25 years out?
RUTLEDGE: The next editor tried to get me to do new versions, but I told her, "I don't have anything new to add to it." Where the technology has changed—and there's a couple of classes that still use it, but they have had to do some modifications—is that the original electronic parts aren't available now. People tend to buy components that include several of the parts we would have used. So, you need new circuit boards. A former student, Zoya Popovic, teaches it at University of Colorado, and she designed new boards. She still uses the book, but with updated circuit diagrams and homework problems. She has taught it for quite a few years. Most of the people who buy the book buy it to read about electronics. The difference with APh 9, the class on making a transistor, is that the equipment for EE20 is basic and available. The equipment is pretty easy to set up and maintain
ZIERLER: The so-called tech bust, the dot-com burst of the late 1990s, early 2000s, did that register with you at all? Did that affect the kinds of things you were doing?
RUTLEDGE: Surprisingly, no. It registered in the sense that a lot of my colleagues in industry—again, you're talking to someone who doesn't see business opportunities— left their companies to join startups to make amplifiers, so I probably could have gotten involved with one of the startups.
There was a way that the dot-com boom affected us. David Lee, who has been head of the Board of Trustees, was a longtime personal friend. We met through our wives. Ellen Lee and Dale carpooled when our children were in nursery school. David Lee was a student of Kip Thorne's, but after he received his PhD in physics, he switched to business. He was an executive in a microwave circuit company, and then a vice president at TRW. Then he became a venture capitalist, working at Pacific Capital. He had gone around visiting people looking for ideas, and he talked to some people at AT&T who were having trouble financing an undersea fiber optic cable. I think AT&T tended to finance their undersea cables themselves, but apparently they were short on this one. They told David, "If you guys want to raise your own money for this, you can lay the cable." David said he went around and quickly raised a billion dollars, from all sorts of places. The company that was formed to lay the cable was called Global Crossing. It was spectacular financial success, and it made several billionaires, including David Lee.
We had a group that used to get together for Chinese New Year. We were sitting around talking, and Kwang-I Yu, a Caltech computer science PhD, said to David Lee, "You had a good year. You could do something good for Caltech." David turned to me and said, "Get me a meeting with the president [David Baltimore]." I went to the president's office, and they checked with development, which at the time was run by Jerry Nunnally. Development reported back that he wasn't a good enough prospect. There I was, without an appointment with the president. I went to the engineering office to talk to the chair at the time, John Seinfeld, who like Roy Gould is one of my heroes. John would meet with him. I arranged the dinner, and I got Mani Chandy and Shuki Bruck to join us. We went to the Athenaeum, and we were about ten minutes in when David said, "I'd like to give ten million dollars." I don't know how many development people you've talked to, and whether you realize how hard they feel have to work for $10 million, and we hadn't even gotten our food yet. The donation funded the Lee Networking Center. Because John Seinfeld, the engineering chair, came to the dinner, rather the president, the Lee Center was an engineering project, rather than a university project. Paul Jennings was the acting vice-president for finance at the time, and he let us invest the money in the endowment without charging overhead, so end it turned out to be worth quite a bit more than ten million dollars. David let us do a lot of small projects as opposed to one big focused project the way donors usually want, so we could use the money to fund new faculty. This was David Lee's reconnection with Caltech. He ended up an important person in Caltech history, as trustee, chair of the trustee Finance Committee, which is the power center of the board, and then as head of the Board of Trustees.
One interesting sidebar is that David's donation was in Global Crossing stock, but initially we couldn't sell it. He was the president of the company, and because of this there were restrictions. There was a good gain in the stock over the next few months. Then we were allowed to sell it, and we did. Shortly after that, Global Crossing cratered. Companies had laid too much undersea fiber. It was one of the most spectacular failures of the dot.com era. At that time, they couldn't fill the fiber with data. The companies that were laying the cables went broke, and Global Crossing was one of them. The value of the stock crashed, but Caltech had sold it all by that time.
There was another way, not so obvious, that we were affected by the dot.com crash, At the time, my students were looking for funding to start Wavestream. Of course, everything looked terrible, and that made it difficult to get funding. But once they got it, there was no other startups in the area. They could develop a market undisturbed for many years.
ZIERLER: Another technical question—flux weakening schemes, what is that?
RUTLEDGE: I don't know. [laughs] No clue.
ZIERLER: It's a term from a few papers you were a part of.
RUTLEDGE: Oh! Well, can you help me? [laughs] I was 70 years old when you started this. If you can give me a title of a paper, maybe I'll remember.
ZIERLER: It's the late 1990s. You have "Two Flux Weakening Schemes for Surface-Mounted Permanent Magnet Synchronous Drives."
RUTLEDGE: Oh, boy! You'd better give me some coauthors on that one! [laughs] Oh! I know! And I may not even know I'm an author! Maybe Dragan Maric.
ZIERLER: That's it. Yes.
RUTLEDGE: Okay, good. He probably included me on the paper. Let's put it this way—I can tell you I didn't understand the paper. I may have been included as an author in a weak moment on my part. I'll try to give the circumstances. He's a wonderful person, and his wife is wonderful, too. Dragan was recruited as a graduate student from Serbia by Professor Slobo Ćuk. Slobo is a pioneer in power electronics and a circuits genius. He and Dragan had a disagreement. I'll repeat what Dragan told me, although it doesn't make sense in American terms, but maybe it does in Eastern European terms. Apparently Dragan's politics were Royalist, whatever that means, because it's anachronistic, and Slobo's were Communist. As a result of the disagreement, Dragan was initially forced to raise his own money, which you don't usually ask a first-year student [laughs] to do. But Dragan's a persuasive guy, and he went down to the GM research lab in Orange County, where they were making the EV1. I don't know if you've ever seen the movie Who Killed the Electric Car?
ZIERLER: Oh, sure!
RUTLEDGE: This is part of the history. The EV1 is the star of the movie. GM was making the EV1, and their research engineers had serious power electronics questions to investigate. Dragan persuaded them that they should send him money so he could study some of these problems at Caltech. The relations between Slobo and Dragan did not improve, so eventually Dragan was pushed out of Slobo's group, but he had his own support! [laughs] He came to me. I was the graduate option representative at the time. That person is the advisor of record for anyone who needs a research home. I thought, "If this guy has raised his own money, I had better go talk to these people at GM." I went there, and they were absolutely first-rate, and they thought Dragan was first-rate. I said, "Okay, Dragan, I know less than nothing about power electronics [laughs]"—which I just demonstrated to you, David!—I can assure you it is not for the uninitiated. People sometimes don't realize how complex technology can get for the stuff that makes your life go, like power electronics for your EV. The GM people came up for his thesis defense, which I clearly needed help with. They loved the thesis, so he got his PhD. He went off to work for International Rectifier initially, and he has had a nice career. He married a Serbian American who still speaks Serbian, and she has been a diplomat. He spent some time as a house husband in different countries where she is the star diplomat, and then they come back here, and he's the guy making the money as a star power engineer. Anyway, that's the story about the flux weakening, but I haven't answered your question! [laughs]
ZIERLER: [laughs] But it is good insight.
RUTLEDGE: Presumably he persuaded me to put me on it. Who knows!
ZIERLER: It is a good story of there's often a back story about when you get included in a paper and when not.
RUTLEDGE: Could be. I tended to be pretty inclusive, but I don't remember what happened on that particular paper. I couldn't have made a technical contribution, but I tended to view that if someone has an advisor and the advisor is meeting with them weekly, as I was, that the advisor should be on the paper. But the student's name came first. We had student first, advisor last, and then the other helpers in between.
Named in Honor of the Tomiyasus
ZIERLER: Finally, last question for today, as we move into the twenty-first century, tell me about the process where you were named the Tomiyasu Professor in 2001.
RUTLEDGE: [laughs] I knew Kiyo well. He was a leading person in microwaves. He did not have children, and he was very much interested in supporting Caltech. It was a tough time to grow up as a Japanese American, obviously, because he graduated in 1940, just before the war. I think I told you his wife Eiko, who just passed away, was in the Topaz concentration camp. Eiko was very patient with him, a true love match, 60 or 70 years. Kiyo absolutely loved Caltech, loved to donate to Caltech, loved scholarships, all of it. Most of it I think was his own money that he saved as a working engineer. He did have enough to donate an endowed chair. He recommended me for the chair, but he wrote, "I recognize that it's not my call." It's Caltech's call who gets the chair. That's the background. But we were good friends before. We had a nice dinner at the Athenaeum to celebrate the chair with a couple of my students—because Kiyo already knew them. Also Bill Bridges, who Kiyo knew well. And Dale and Eiko already knew each other.
ZIERLER: As an addendum to that, I wonder if you can explain the mechanics of the endowment. What does that mean? How does it change things for you?
RUTLEDGE: This may have evolved over time. When I first came, only a very few people had chairs, so someone like Dave Middlebrook who as I mentioned before, was truly stunning in terms of his contributions—he basically founded power electronics—that's a huge area now, because you can't make electric vehicles or wind turbines without serious power electronics. He didn't get into the National Academy and he did not get a chair. So, there weren't many. I was in a middle time, where some people were getting chairs, but not as many as now. Kiyo's I would call a standard chair—there's a standard amount that Development will say, "You endow a chair for this amount." I'm sure you could give more, and there would be more benefits. What it meant was when you get a contract or grant from, say, the Army, you could put in a certain amount of professor time. In most universities, what that does is pay their summer salary, because most places pay a nine-month salary. In Caltech's situation, when you put in professor time, but it ordinarily simply goes to the Institute, because you've already got a 12-month salary. But if you had a chair, for any salary that a grant paid, the Institute would put an equivalent amount, funded from the endowment that Kiyo established, into a discretionary account. It allowed you to build up discretionary money to start a new project without having to have a grant for that project. For senior engineers, that's very important, because when you're beginning, you're hot. That's why they hired you. But later when you want to fund something new, it is not hot yet, and you need to figure out, "How do I fund this?" Having a pot of discretionary money to do that is just extremely valuable.
ZIERLER: Next time, we'll pick up 2001. We'll bring the story right up to the present.
[End of Recording]
ZIERLER: This is David Zierler, Director of the Caltech Heritage Project. It is Thursday, January 12th, 2023. It is great to be back once again with Professor Dave Rutledge. Dave, as always, great to be with you. Thanks for joining me again.
RUTLEDGE: Thank you.
Perspective on Administrative Responsibilities
ZIERLER: Today I want to start on an administrative term. In 1999, before of course you are division chair, you are named executive officer for EE. I know that more recently, the Division has adopted the department model that is sort of more common at other academic institutes. I wonder if you might have a unique perspective on that decision, given that the executive officer is in some ways a department chair, like it would be at other colleges and universities.
RUTLEDGE: Yes, I can. At the time I was division chair, we had 13 options. These would be called departments at other universities. It was a very centralized administrative model, and virtually all decisions that were of interest to faculty were made in the division chair's office. An executive officer for an option was kind of like a building supervisor. Over that period—this was before my time as division chair—Elliott Andrews, who was the division administrator, had built up a system of lower-level staff positions called building administrators throughout the division, and that took away a lot of the work of being an executive officer. The executive officer is also concerned—for the options that have serious teaching loads—Electrical Engineering did then, Computer Science does now—with instruction. A division chair will know only a couple of the 13 options well, and the executive officers will serve as points of contact for advice on that option. Finally, the executive officers form a good pool for the provost when there is a search for a division chair.
The move to a department model was after my time as division chair. I think a major motivation for it was that people thought it could improve development within the division. One thing that the new department chairs are supposed to do is to establish a department committee made up of prominent alumni and others who might help with development. However, my sense is that all this simply added another layer of professor administrators without a real benefit to engineering. I think that the major step forward that the engineering division made in development in this time frame was to identify an untapped source of donations, the local Chinese community, first with David Lee, for the networking center, and then with Peggy and Andrew Cherng, for the medical engineering option.
The tendency in virtually all institutions is to add administrators and to not subtract them. It's not something I sympathize with, but it seems to be essentially universal. When I came into the division chair office, I did a reorganization of the office, and the division administrator was moved to a different department on campus. That person's salary was quite significant. A peculiarity of Caltech—possibly only at Caltech—was that I could convert that salary to three graduate fellowships. I viewed that it would cut out one administrative layer, and it would allow me to get to know the building administrators, who were truly outstanding employees. No one got to that job unless they absolutely knew how to work inside Caltech. No one was hired for the job from the outside. They could do magic.
I'll give an example. Each organization has to figure out what happens when we have a budget crunch. Budget crunches at Caltech come from the top. If there's a perception, say, that the endowment return is bad for a while, the question is, how do you handle this? You can do it at the top. The provost can look around and say, "Ah, there's some unit that to the accountants looks like it's losing money, like catering. Maybe we could buy those services on the outside." So he shuts the unit down and fires the employees. This hurts the employees and their families. Caltech loses good people and the excellent service. There is a better way to approach this problem when you have a relationship directly with the lower level building administrators. First the provost has to agree to flat funding the divisions until the budget crunch recovers. Then you can tell the building administrators, "We're going to have flat funding for a couple of years. Why don't you look around? Maybe there is someone doesn't really want to work 40 hours, or is at least comfortable working 30." They will still get their benefits, and the budget can be brought into line. What you find under those circumstances, when you have good, smart people as lower-level administrators, you can handle flat funding for quite a while, without the trauma that comes from shutting down a complete campus unit.
ZIERLER: Do you see something so unique in EAS that it is the only division that has adopted this model, and the others have stuck to the options model?
RUTLEDGE: I'd be shocked if any other division has anywhere near 13 options. The options are really quite different intellectually. The education of a computer scientist, mechanical engineer, and electrical engineer have little overlap, aside from things that are required in the core, like physics and math. You have to have some way to deal with each of these fields. It isn't just that the intellectual aspects are different. The culture of the working engineers—60% of Caltech graduates go out to work without a PhD, and their working environment is quite different from the alumni who get PhDs. We do get feedback from these students on whether we are on the right track or not.
ZIERLER: In light of your comment that a provost, in considering the next division chair, will strongly look at people who have served in the executive officer role, did you see, circa 1999, 2000, that that might be the next step for you? Did you consider that at all?
RUTLEDGE: I did consider it, and the reason is that on a previous division chair search committee—the one that brought John Seinfeld—Paul Jennings had chosen a couple of junior people, Demetri Psaltis and me, presumably because he thought we might be future leaders in the Division. He also asked us to run faculty search committees earlier. Except for the division chair itself, a faculty search committee chair is the most important job in the division. The Seinfeld search made me aware that my name was popping up as a candidate in faculty polling. There wasn't any particular reason to hide it from me, because I wasn't a serious candidate at that time. But I had fair warning that I might be considered as a candidate in the future.
ZIERLER: When you did get that opportunity and were considering it, what was your research at the time, around 2004, 2005? What did you think the impact might be, given all the administrative burdens of taking on the division chair job?
RUTLEDGE: I was at a stage where intellectually, I felt okay taking on the job as one hundred percent administrator. My graduate students were pretty senior. They were quite self-driven, so all I had to do was meet with them for an hour a week. I had enough funding for them. The truth is, if a division chair runs out of money and his student's salary can't be paid—I'm sure it has come up with other division chairs in various kinds of circumstances—a division office can certainly come up with enough money to make sure that the student is paid. It's pretty embarrassing if a division chair's student does not have a stipend. The division budget is roughly $100 million. So, someone in the office can make sure that the student has a salary. In that sense, I just let things run out. The students graduated, and went off to become professors and work in industry. And I did not take on more students.
ZIERLER: When you came into the role, the duality of, on the one hand, "don't break anything, it's working well and just keep the ship running," and on the other, you now have an opportunity to improve things as you see it. How did you balance those priorities when you first started?
RUTLEDGE: I don't think I had that concept. I did certainly have the concept that I was not looking to make changes, administratively. The one that I talked about, which is trying—uniquely, I think, among Caltech divisions—to run without a division administrator, which I thought worked fine, and was worth the three graduate fellowships—that was accidental. It had to do with personality issues, and it wasn't something that I planned to do. What I planned to do was recruit, and Paul Jennings did give me the slots for that, so I did it. I didn't plan to change anything else. The previous division chair, Richard Murray, the current Biology division chair, is very much the kind of person who plans, in the sense that would make industrial people comfortable. For industrial people, planning is a serious part of the job. Universities are somewhat different, because professors have their own sources of money. I had Army support. If I had moved to a different university, it is likely that the contracts would have gone with me. It's not like a company that is producing one product, where you truly have to get everyone on board and engaged. But Richard did thorough planning, and that included planning for recruiting. So, when I started, I was perfectly happy to propose to Paul that we use Richard's recruiting plan. It had been discussed in the Division. I didn't invent the plan; I just tried to execute what Richard had done.
ZIERLER: What did you see as the major challenges facing EAS at that point, circa 2005?
RUTLEDGE: I would say, hiring new professors. In those days, there was not much discussion about minority candidates. At that time, there were only a few minority candidates who would have passed a search committee's review. I did recruit quite a few women when I was chair, some who are now leaders in the division. I had a strategy for this that was not ever legally tested. Paul Jennings did not micromanage which options I hired in. I told the search committee chairs that if their first candidate was a woman or a minority that they could keep searching for an additional candidate. That meant that often a committee ended up hiring both a woman and a man. It also meant that if a man applied, he wouldn't necessarily feel that the position was closed off to him. I think that's the reason this approach worked. I did not pass down the pressure that was applied to me to search for women. I simply said, "If your preferred candidate is a woman or a minority, you'll get another slot," and I left it at that.
I did meet with the candidates from all the search committees across the division. I did not see other division chairs do this. I took each candidate out to dinner, one-on-one. There were a lot of candidates and a lot of nights out. I was not trying to impress the candidate with a fancy restaurant. In fact, the people doing my reimbursements gave me a hard time, because I used to take candidates to Fu-Shing, an excellent but inexpensive Chinese restaurant. My wife is Chinese American. The Chinese know the good places to eat, and it was a great place to talk. The people doing the reimbursements would ask, "You really took a candidate to a restaurant that charged only $15?"
RUTLEDGE: None of the candidates complained. But it's a lot of stress to meet the dean, even if the dean is trying to be friendly and just sell Caltech. I'm sure there were some that the meal didn't agree with them, just because interviewing for a job is a tough business on the candidate side.
ZIERLER: The impetus for recruitment, what aspects were about replacing retired faculty, and what aspects were expanding the Division into new areas?
RUTLEDGE: I told you John Seinfeld was my hero in a lot of ways, but he came from a different division, Chemical Engineering and Chemistry. When that happens, there are lots of good things. Certainly he can take a fresh look. I don't think it's good for recruiting, because a lot of the recruiting really depends on someone who has been in the weeds of the division, and I had. The same thing with Richard Murray. I don't think Richard had run a search committee before. He gave his focus to reorganization, and—Steve Koonin was the one who chose him—maybe that's what Steve wanted. I was never privy to the discussion of what Richard and Steve agreed on. Steve Koonin is a smart guy, he has been honest and straightforward, and supportive of Engineering, so I don't have trouble with any of those things. But it did mean that I tried to make the case to Paul Jennings that the Division of Engineering, while we had a lot of interest from students and good research projects, was not keeping up with the other divisions in recruiting. It wasn't specifically replacing, say, someone in liquid metals if Bill Johnson retires.
I encouraged each option to write the advertisements as broadly as they could, recognizing that you could get pretty disparate candidates that may be tough to compare. Nevertheless, writing the ads broadly might allow us to pick up a field that we don't have at Caltech but that we might want to have at Caltech. For example, I was the search committee chair for Professor Tai, who is now in the National Academy of Engineering, and who started the Medical Engineering Department. At the time we hired him, he did micromachining, which at Berkeley was an enormous effort that covered several different professionals. We didn't have anyone in the area at all, so we were recruiting in a new area. We did consider a competitor to Tai. It's clear in retrospect that Professor Tai was the right person, although the other person was very good, too. For that committee, we were trying to consider new areas, and so that may answer your question. We did it by writing a broad ad, where a lot of people could feel like, "Oh, I should apply for that position."
ZIERLER: Beyond the informality of a good dinner with a prospective candidate, what role informally and formally does the division chair play in the recruitment and successful landing of a new professor?
RUTLEDGE: That, I can certainly comment on, and every division chair will probably tell you almost verbatim the same words. The candidates in the modern world, not necessarily when I came, but by the time when I became chair, will be offered substantial startup packages at other universities, particularly for the experimental people. These packages have grown over time, and the growth has outpaced any kind of inflation index. It is the way things have gone. When I came to Caltech, part of the reason the budgets were not large, even for experimental people, is that you built most of your own equipment. Over time, the scientific and engineering instrument market has developed, and there is now equipment to buy that my students and I would have built ourselves. The commercial equipment will be better in many cases than something you could build, and it will be nearly stress-free to get going.
One of the things that the chair does is to ask the candidate, "What do you need to start up?" Caltech tended to take the view that there wasn't a set amount of money in mind. It was, "Listen to the candidate. What does the candidate say she or he needs to start up?" There's a negotiation on that, because, the amounts of money were usually larger than the resources available in the Division. That meant that you had to go to the provost to get the money. You had to persuade the provost that this is what we should do to get this person. So there is a back and forth.
From the perspective of the candidate, there is also a significant waiting period. The candidate is going to be one of several people interviewed, typically at weekly intervals. If the candidate is at the beginning of the interview list, a candidate could be waiting a long time just for us to get through all the interviews. After that, the committee meets to decide which candidate to put forward. Caltech tends to take the view that we're really just going after one person. I gave the one exception to that, when I told them that if their first choice was a woman or a minority, that they could go one down the list, and put that person forward also. But that could take a while for a committee to sort out. We're not professional HR people. We're dealing with a great deal of uncertainty. We're responding to different things in the candidates. There's nothing standardized, equivalent to the SAT, where you could say, "He had an 800 on the math. That might be a Caltech person." It's just not that way. The disagreements can be intense. A lot is at stake—a person might work here for 40 years. It really does affect the Institute. It affects who your colleagues are. It affects the students. The whole nine yards. So, the process can take a while.
Next the search committee writes up a case for the candidate, and then they need to persuade the division that they should vote for the candidate. One of the problems is that you have 13 options, so an electrical engineer has to persuade a mechanical engineer that this is a great person to hire. The other options will tend to view, partly correctly, that there are some elements of zero-sum to this, meaning if electrical engineering gets a position, that may be that we don't get a position in mechanical engineering. The number of positions in the Division is watched carefully by the provost, so to some extent, it's hard to argue with that view. It means that people tend to be careful with the appointments. They want their colleagues to really be good. You get the vote on that issue. Then the division chair needs to go persuade the other five division chairs that this is a good person. To a somewhat lesser extent, the division chairs will also view this as zero-sum. If Engineering gets a lot of appointments, they'll wonder, "Hmm, will Biology fewer appointments?" But usually the issue is more that you're trying to persuade someone about a candidate when the intellectual areas and culture are quite different from the fields they are familiar with. To give an example, the Physics Division chair must try to persuade the engineering chair on a math appointment, although there is almost nothing in the research that an engineer can understand. That has to be done in a thoughtful way. Once the chairs have voted, then it goes to the trustees, and they've got their schedule.
All of this takes time. The wait can be longer than a candidate might feel comfortable with, so you have to definitely keep them interested. Once they get the offer, then you need to persuade them to accept. You may also have to persuade the spouse. The spouse may want a job. The spouse may be technical. Hmm. Does that mean we need to look at JPL, or USC? You can contrast with, say, a large state university. I'll use Illinois, a superb place for electrical engineering. I visited there as a candidate myself. In that case, the department chair could simply tell the candidate at dinner, "There's going to be an offer," and then the paperwork may follow in a week. Our structure simply doesn't allow this. If the candidates have to wait a couple of months, they may be gone. The candidate may think, "Hmm, someone doesn't like me." It's tough being a candidate. Even for the ones who win, it's tough.
ZIERLER: On the other side of the life career cycle, what role did you see, if any, having in more senior faculty discussions with them about when they might consider retiring, going emeritus?
RUTLEDGE: I don't remember ever talking to a professor about retiring when I was chair [laughs]. In fact, I can tell my retirement story, on the other side. Ravichandran was the division chair. I was in the middle of the two-year leave before retirement. However, the two-year leave can be commuted to a lump sum payment. I had a financial plan, but I could not get a financial advisor or tax person to understand what I was trying to do. I don't know whether [laughs] that makes it a good idea or a bad idea. But the plan was to convert all of the money at TIAA that was not locked up in the TIAA Guaranteed funds and convert it to a Roth IRA. You have to do a bunch of things all in one day. That means you need to pay a great deal of taxes. But once it is done, if you are patient enough to wait five years, any capital gains are not taxable I had become concerned that the market might be starting to rise, which turned out to be correct. I thought, "Hmm, I'd better go ahead and pull the rip cord." It was funny; I had always told my wife Dale that "I'm just going to wake up one day, and say ‘I quit.' You're not going to get any warning." I don't think she really believed me, because it's not the way she thinks. She thinks in terms of relationships. In her mind, you don't just wake up and say, "I quit." But as an academic researcher, you're very used to the possibility, "This line of work is a dead end," and you decide that quickly. The decision may be based on events that have happened over time, but you may make the decision quickly. Then, you've got to move on.
Anyway, I made that call. Caltech was simply marvelous. Ravi had to sign paperwork, obviously, to commute the two-year leave to a lump sum cash payment, and then the people over in Accounting had a lot of figuring to do, because there are some funny factors in the contract, and of course the tax withholding aspects are awful. But I started in the morning, Ravi had letters to HR by noon, and the people in payroll—amazing—stayed late to cut the check for me that night. That freed up the money at TIAA and the next morning Dale sent TIAA a fax to do the Roth conversion. So it was all done in 24 hours. Caltech was magical. The division chair was magical. I didn't ask Ravi ahead of time; it was just something I decided to do on the spot.
In my experience, a lot of people retire if a spouse gets sick and then they take care of their spouse. In a way, this happened to me. One factor for me in deciding to retire was that I wanted to finish a book on energy, the one that Carver Mead talked about. I was having trouble writing the book, but I figured that if I quit Caltech, I could finish it, and I did. But then three months after the book came out my wife was diagnosed with a horrendous cancer. But there was a new treatment, CAR T—just approved by the Trump administration—that saved her. She would have been dead in six months. By then I had quit Caltech, and the book was out. She needed someone in the home who was available 24/7 to respond. That can be a reason to retire, too.
So, no, I don't remember ever advising anyone on retiring. The only type of advising—a faculty member says, "I've got an offer from ETH [Swiss Federal Institute of Technology, Zurich]." The ETH offers were the toughest ones I dealt with. I would be in the business of trying to match the offers. I had to persuade them, "Please show me your offer letter, and let me see what I can do with the provost." When I was chair, that always worked.
ZIERLER: Perhaps it's a function of your gentle personality that you never talked to anybody about retiring, or did you truly never see the need to help ease someone out the door?
RUTLEDGE: I'll make a strong statement on that one. Again, 13 options is a serious intellectual head stretch, trying to understand the work that different people are doing. Particularly you need it for promotions. We had quite a bit of talk about promotions and associate professor and full professor, last time. You start with some preconceptions which didn't last very long. After a while I got the sense that there were two types of Caltech professors. Let's use engineering. I really can't judge biology, but I do know how to judge engineers. There were two types of engineering professors at Caltech. There were people who were absolutely outstanding in their field, and there were people who I hadn't figured out yet that they were absolutely outstanding in their field. And that was it. The fraction of what I would call "dead wood" was very low when I joined Caltech. By the time I got to be division chair, basically it was zero. So encouraging people to retire was not on the agenda. It was more my learning about the research in the division. We didn't really have this idea that we needed big programs, so a lot of faculty members simply go to their Engineering Society to present great work there. They are appreciated in that Society, but there may not be another Caltech person there. They will get tenured if they are appreciated in their society, but the division chair may be clueless. But you try to educate yourself, talk to people, socialize, and go to all the Christmas parties. [laughs] Anytime you're invited to something, go, talk to people, and get a sense of what they are doing. Caltech is a small place, and the professors here are all used to the idea, that even if they are personally modest individuals, they have to do their own marketing for their research. And my experience is that even modest people can market their own research. You just have to listen, poke them, and they'll talk about tit. And it can be a shock for a new division chair to realize how good people are that he only knew socially before.
ZIERLER: As you well know, Caltech boasts of several professors who are active and they are well into their nineties. What did you learn about professors who wanted to do that, who wanted to stay active at that stage of their life?
RUTLEDGE: I did not impose a judgment on that. We are legally required, if they can do their job, not to try to push them out. I did successfully push out a building administrator who I felt was no longer not able to do her job. She was quite old and she was nowhere near as good as when she was hired. She did resent me, but I view it as part of the paycheck. Nevertheless, she did get the message. She retired with a nice party happily and is very much still alive and kicking at near 100, a very vigorous lady.
Personally, I feel that faculty members have to make this call themselves. For me, I felt that when I watched the young people, I thought they were doing better work than I was, so I was ready to move on. As I told you, in the end, I was grateful that I had retired because of my wife's sickness. I don't know how I would have handled it as well if I was still drawing a Caltech salary, because she needed me 24/7. Now, it's not 24/7 like working in a salt mine, where you come home and your body is broken up, but it's still 24/7, and you're taking someone to the hospital, 50 times.
People have different life cycles. You can have a cycle that is perfectly ethical and legal that goes like this. You start as assistant professor, when most of the work is based on your own intellectual capacity, even though you have some research assistants. Near the end you may have a postdoc running the lab with outstanding independent graduate students, and you may not be adding much intellectually. You would still be the connection to the funding agencies, which is necessary, so I'm not disparaging this. But would be difficult for me to argue in a field where mathematical skill is important that a person at 65 can do the mathematics that they could do at 25. It's just not the way the brain works. I've never seen an exception to this.
ZIERLER: As you well know, one of the big educational trend lines over the course of the time when you were division chair was the growth of computer science and the interest in undergraduates in pursuing this option. What was your perspective on that and how it changed EAS?
RUTLEDGE: I'm not very helpful on that one. The reason is that I viewed the division office primarily as a faculty development office, not in the sense of raising money, but in the sense of, what do new faculty need to get going. Students really are covered more at the option level for the academic curriculum, and with the undergraduate and graduate deans. We definitely tried to recruit computer science faculty. It was always extremely difficult. My sense of what was happening was that, if we split the CS [computer science] Department into an engineering side and a mathematics side—that it's extremely difficult to hire someone who is near the top on the engineering side, say in robotics. There are a couple of reasons for this. Computer science has a very strong pecking order among universities, and it is very difficult to unseat. For example, Carnegie Mellon, would have been near the top in robotics—and there's a whole local industry there in Pittsburgh. I had a sabbatical at Carnegie Mellon. I knew some of the people, and it helped me appreciate what you're up against when you're trying to compete against Carnegie Mellon in computer science.
There is a large infrastructure that is important to computer science with massive support from Microsoft, Google, and other companies, that we wouldn't have. But on the mathematical side on computer science, we can recruit top people. The question is whether this gives us the computer science department we want, in terms of our standing among other computer science departments and in serving the needs of the students. There is another problem with the computer science curriculum—an intellectual one. My sense of electrical engineering and mechanical engineering as they are taught now, is that they are simply marvelous intellectual activities. You could teach electrical and mechanical engineering in a liberal arts college, and you would get all of the things you would want liberal arts colleges to do, in terms of really stretching students mentally. You change them. You don't have dropouts leading the electronics industry. They may even have PhDs. I'm thinking of Gordon Moore. He has a chemistry PhD from Caltech. You can't say this about the computer industry. Think of Bill Gates and Steve Jobs. Neither finished college. So, what is the role of computer science at a university?
ZIERLER: Just to push on that a little more, from an administrative perspective, what burdens did that place on the Division, given all of the demand, all of the requirements for teaching these students?
RUTLEDGE: [laughs] Great! You're assuming—I took teaching seriously. In my workshop in the back garage, I keep one plaque. Any professor my age will have a whole stack of plaques, of various kinds of things, and I'm sure at the time it was great; you could give a nice speech and thank everybody. There is only one plaque in my workshop, and that is the teaching award from the Associated Students at Caltech. When the student called to tell me, I told him, "You're joking." Nevertheless, that was the way it was. Yes, teaching is important to me. However, I can say that at the provost level—and I have known many—the way they allocate faculty positions has little relation to student interest. For example, GPS is a fabulous division from the point of view of its contribution to Caltech over the years, with a top national ranking in its departments. But, at the time I came to Caltech, you might have three or four undergraduates a year in the entire division. In electrical engineering alone, one of many options in the engineering division, close to half of the undergraduates would have been electrical engineers. The teaching at Caltech is pretty much [laughs] disconnected from the faculty recruiting.
ZIERLER: Were there any discussions during your tenure as division chair about splitting the division into two component parts, one that focused on information science and the other that was the remainder?
RUTLEDGE: Oh, yes, there was! In fact, because I was part of the—you can use a Star Wars metaphor— the Rebel Alliance.
RUTLEDGE: Even the least political division chair will think politically about the position of his own division. They won't choose anyone to be division chair who can't at least sort out the political aspects. The politics are simple compared to a state university, but you do have to think about it. The six division chairs essentially run the Institute in that they make the critical hiring and promotion decisions. I never liked the idea that when I vote as engineering chair, there are one engineer and four scientists. I liked the idea of a split in the engineering division to get two engineers and four scientists. The division voted on it when Richard Murray was chair, and I voted for the split. A majority voted for the split, but Richard wanted a super-majority vote in favor of the split to bring the issue forward to the trustees. I think my mechanical engineering colleagues are wonderful. That's why I want them to have their own chair! [laughs] I assume it's reciprocal. All types of engineers can make stuff, magic stuff, so you appreciate the other engineering specialties. In any event, a possible split was discussed before I was chair and it wasn't discussed when I was chair. I viewed the issue as moot by the time I became division chair, I had lots of faculty hiring slots, and I worked on filling them.
ZIERLER: What perspective did you appreciate about Caltech from sitting on the IACC, just working with the other division chairs, the president, the provost? What kind of issues and things did you learn about Caltech that simply wouldn't have happened otherwise?
RUTLEDGE: I don't have a strong sense of that. We had six people who were all—well, you have a dual job. You represent the division and you represent Caltech. There was a negative issue—from my perspective—that arose when Jean-Lou Chameau was president. We had a feeler from the major Saudi technical university that they were starting up.
ZIERLER: KAUST [King Abdullah University of Science and Technology]? King Abdullah.
RUTLEDGE: Good work, thank you for filling that in! They wanted to set up a relationship with Caltech. I said to the president, provost and the other chairs, "Wait until we approve of the way they treat women, Jews, and Christians. Wait until I can take a Bible into Saudi Arabia. Then let's talk." This is not a criticism of the Saudis trying to establish the relationship. It's their country. It's a question of our relationship. And it is not making any statement about what US foreign policy should be toward Saudi Arabia. This was strictly Caltech. I was really the only one who thought that this was a problem. I felt that with President Chameau, money was the priority. Of course, in some sense, that was tested, because he left Caltech to become president of KAUST at a large multiple of his Caltech salary. I think money also influenced other people's minds. Now, I knew colleagues that I very much respect who have worked up individual relationships there and I've got absolutely no objection to those. There was one professor, Yaser Abu-Mostafa, who is Egyptian, and Arabic-speaking. He had arranged to be their dean of engineering. I talked to Yaser about it after it fell through. What he felt happened, I think, was that the place was so centralized that all the money was going to the top, and that by the time it got down to the level of the dean of engineering, they weren't able to keep their commitments to him, not just in terms of research support and faculty slots, but even in terms of his salary, so he backed out.
ZIERLER: As you alluded, tell me about the timing of stepping down as division chair in 2008, which is of course a relatively short time to serve as division chair.
RUTLEDGE: It is, yes. It gets back to the recruiting. By coincidence, I saw Ed Stolper today. It's my birthday today, and my wife and I went to—
ZIERLER: Oh, happy birthday!
RUTLEDGE: Thank you!
ZIERLER: —Noda's to celebrate. Her birthday is the tenth of January and mine is the 12th, so sometime during the week of the 10th, 11th, and 12th, we go out. Anyway, I ran into Ed, and we had a nice catch-up conversation, because he was meeting with the president there. There is a letter I wrote to Ed at the time I was chair expressing my wish that he find a different chair. The archives asked if I wanted it to be private, and I said no, I wanted it to be public. I don't have anything to add to that letter here. The letter goes into details, and it is contemporary, so I will just give the outline of the problem as I see it now. I felt I wasn't able to recruit faculty, even with outstanding offers. Ed restricted the startup funds so severely that I wasn't able offer startup packages that were competitive. I thought maybe a different chair would be more successful with the startup offers. Richard Murray filled in for a while, until Ares Rosakis took the job. Ares told me that because I had been public with my complaints, that Ed established an income stream for the division based on the executive education classes that greatly improved the financial situation of the division office. However, the new faculty recruiting was essentially shut down.
ZIERLER: To clarify, your thoughts about your capacity or not to recruit, this was a self-assessment? This was not coming from anyone else?
RUTLEDGE: I copied a lot of people on the letter that is in the archives. The feedback I got from John Seinfeld is, "What a surprise. I thought you were doing a great job." My sense was that people outside the division office could not see any problems yet, but I thought that they would see the problem in the future. My staff and the electrical engineering faculty were extremely supportive.
ZIERLER: Being constrained in your capacity to make offers, looking at the year—2008—did the financial crisis play a role in Caltech's overall ability to disperse funds?
RUTLEDGE: No. It was before that. The funny part—I talked to Ed—you run into him from time to time—and Ed said, "Dave, you got out of here before things really went haywire." He meant financially, because the provost is not the VP for Finance, but the provost is definitely on the hook when it is difficult to balance the books. We talked earlier about the approach I liked to take when there are difficult financial times, and it was the opposite of Ed's, which was to pick out campus units to shut down. I probably would have had a conflict with him about that too. We definitely had some rough times when Paul Jennings was provost and I remember collecting Ken Farley, the geology division chair, and the two of us going to Paul and telling him, "Look, if you just give us flat funding for a while, we think we can handle this at our level." Paul went along with that. But it wasn't Ed's style. Ed's style was much more, "Let me identify some unit at Caltech, and we'll shut that down, and we'll balance the books that way." I view a university differently from a company. People understand that if a company makes airplanes, and they're not selling airplanes, they're going to have to have layoffs. That just the way it is. But I have always felt that one of the advantages of a university, with an endowment, is that you could hold on to good people and even recruit people in bad times, because of the endowment.
ZIERLER: Just so I understand, was the strategy that in stepping down as division chair, this would create a shock to the system so that ultimately EAS would get those recruitment funds?
RUTLEDGE: You're thinking probably more than I did at the time. I think at the time, I felt that I just wasn't making the progress that I should, that I just couldn't get the startup funds that I needed, and maybe someone else could figure it out. I think that's probably as much as I can say.
ZIERLER: Because you went to one hundred percent administrative responsibility, stepping down earlier than you might otherwise have, what did that mean for your research, just returning to so-called civilian life?
RUTLEDGE: That was interesting. The EE Department executive officer, or rather the whole EE Department, they were absolutely marvelous. They had read the letter, and they were extremely supportive. My students had graduated by that time. One thing that was happening in parallel was that I had gotten hooked on this problem of energy supplies. Babak Hassibi, the electrical engineering executive officer, was wonderful. He let me teach an energy class instead of an electrical engineering class. Eventually I relabeled myself as a Professor of Engineering, not electrical engineering, without any criticism of electrical engineering or criticism from electrical engineering. We always stayed on good relations. They let me take a sabbatical to Pittsburgh [Carnegie Mellon] where they know something about energy. [laughs] Pennsylvania, it's oil, it's coal. It's all in western Pennsylvania. There's a long history there, so it was a good place to pick up the topic. At the time when there was a concern about oil supplies—it was called "peak oil"—and Congress enacted bipartisan legislation to encourage ethanol from corn. It was thought at the time that future oil supplies were quite constrained and that we needed a substitute that you could put in the gas tank of your car. It was a fun time to enter the field.
Funding tended to come from people who had political angles, so I was nervous about taking on PhD students. I had $300,000 in discretionary money saved up, various donations from microwave companies from before the time that I was chair. I was able to self-fund the research that I was doing myself, with my wife Dale as the research assistant. She used to be the Sierra Club librarian, so she was actually professionally qualified to do research in, say, locating things like old mining records. And I developed energy classes that I taught at the freshman and senior/graduate level.
ZIERLER: I'm not sure where it fits in the chronology, but when you stepped down as division chair, what was the state of the Wavestream Corporation at that point?
RUTLEDGE: Wavestream was founded in 2000. It was successful and acquired by Gilat Satellite Networks in 2010. I was considered a founder, but I probably only did two days of consulting for chip design reviews. The students built it themselves. With the acquisition, the students could pay off their houses and the Institute and I made a modest amount of money.
ZIERLER: What did you see as the main niche for Wavestream? What role did it fill in the industry?
RUTLEDGE: With satellite communications, the tendency is to go to higher frequencies. When you do that, the transmitting antennas become more directive, as the square of the frequency. This is a big effect, and it allows you to send data faster. The down sides of going to a higher frequency is that you become more susceptible to rain, and you need to know where the receiving station is. If you don't know where the receiver is, you do better going to a low frequency. I think I told you my daughter built the radio that I ended up using in my class. The transmitter power was two watts. That's like a cell phone. She could talk to Japan with that radio. But the frequency is only 7 megahertz. Elon Musk's Starlink system uses frequencies as high as 50 gigahertz, more than a thousand times higher.
The technical challenge at the higher frequencies has historically been, for decades, getting enough power. If you have twice as much power, you can send twice data twice as fast. That is simple, but it's not simple to get the extra power. There are two ways you can go about it. One is that you can somehow combine the outputs from a large number of devices. That's what Wavestream was doing. The other is that you can develop a new material technology that can handle higher voltages. That is happening now. Gallium nitride is replacing gallium arsenide. Gallium nitride was not available when Wavestream started more than 20 years ago. So now, 20 years later, an individual gallium-nitride transistor has significantly more power than a single gallium-arsenide transistor. Wavestream now sells gallium-nitride amplifiers.
Coal Production Estimates
ZIERLER: Your paper in 2011, "Estimating Long-Term World Coal Production with Logit and Probit Transforms," looking back, do you see that as the intellectual seed for making a book?
RUTLEDGE: Oh, very much. That had a whole history with it. Around the time I was division chair, I was giving talks on future coal production. From a research perspective, the 2011 coal production paper was unusual in that, with one paper, I achieved my goals from this line of research. I told you in an earlier interview that the seed for my interest in coal supplies was a talk by Professor Ken Deffeyes, who was a buddy of Professor Tom Tombrello's, who was Physics Division chair. Deffeyes used a logistic analysis to estimate future oil production. I liked the mathematics, but it wasn't clear to me that the technique really worked very well with oil, and it really hadn't worked with natural gas. But as I was poking around in the fossil-fuel resources literature, and I came across this wonderful book on coal by Stanley Jevons, who was an early economist from the golden era of British intellectual life, the reign of Queen Victoria. This was the time of Faraday and Darwin and Maxwell, amazing people for the ages. Jevons talked about the future of coal production in the UK as seen in 1865. I was just fascinated. He did a marvelous job with the data that was available at the time. Tragically, Jevons died young, drowning on holiday. I was interested in whether I could adapt Deffeyes' techniques to coal production. Deffeyes was a superb geologist, but his math was shaky.
I tweaked the math a bit, and I found that, although the math did not work that well with oil and gas, it worked extremely well with coal. Then I got in touch with people who know things about coal, particularly Romeo Flores—Romey to everyone who knows him—in the Denver office of the USGS [United States Geological Survey]. The coal geologists were extremely welcoming to me. I think they were conscious that the geologists had calculated the coal reserves exactly the way they thought it should be done, mapping out at enormous expense the coal deposits of the entire United States. This knowledge is useful to coal companies, of course. But, the major intent of the work—going back to the 1800s—was to tell national leaders what the total future coal production of the United States would be, based on the reserves. But this turned out to be a spectacular failure, and the geologists knew it. It turns out that these reserves, as estimates of future production, were five times too high. That provoked me, because I found that the logistics fits predicted the correct future production within 20%. I'm not talking about overall US production, because the US isn't done, but I'm talking about one particular field in the US, the anthracite field, as well as other countries around the world that are finished with their coal production. The logistic fits worked in every case with typical errors of around 20%. Compared with a five-to-one, this was a huge improvement.
Romey Flores picked up on this work, and he persuaded the editor of the major coal journal to invite me to write the paper. This was important, because when you have an electrical engineer writing a paper on coal, there are all sorts of things that can go haywire. A warning—when you want to make a contribution in another field, you have to master the vocabulary, obviously, but less obviously, you have to understand the culture. But the paper was, I think, appreciated. Every once in a while I hear from people about the paper. I think I told you about a conference a couple months ago where because of the paper more than ten years earlier, they invited me to be the engineer representing future coal production in the discussion. I like that as a contribution.
Richard Heinberg, at Post Carbon Institute, wrote a book where he used many of my talk slides, so in some sense, he did a book on the work. But I did write up a manuscript for Cambridge and I sent it to the same group that had published The Electronics of Radio. The first editor had moved on, so he wasn't involved in the evaluation. They rejected it. I said, "Okay." That's the way writing is. It's their call, obviously. I did not try to shop it around, trying different publishers. I suspect in retrospect that if I really wanted to, I could have found a publisher for it. That's also the way it is with a scientific paper. If you really want to get it published, keep trying. There have been Nobel Prizes won by papers that were published in obscure journals. [laughs] The reviewers don't always get it right.
I went back to work, teaching the material in my class. A couple of years later, the editor got back to me. Julie Lancashire, from Cambridge. She said, "Dave, we've been rethinking here. Would you like to have another go with that proposal?" I said, "Well, the original plan was really just the logistic analysis, one paper expanded to a book." But after teaching the material for several years, I wanted to do something broader. I was picking up historical themes in my classes. I taught topics in traditional energy sources, like whale oil for lighting, horses for transportation, wood for heat. I was seeing some things might repeat in the future, like biofuels. In some quarters biofuels are controversial, but they are a significant part of our energy production, and we know how to make them, which [laughs] is a big deal for an engineer.
Some things are speculative, like fusion, and an engineer like me tends to discount it until people show us the approach really works. People often say, "Ah, we've got the new Livermore fusion results, break-even." But the Livermore setup is not remotely practical for large-scale electricity generation. On the other hand, we do know how to make biogas, bioethanol, and wood pellets. I would always get a couple of people who really like horses in the class, and they appreciated the connections, where hay and oats are the transportation biofuels for the 1800s. The transition from horses to cars took a lot longer than you might think. I know some of us men, maybe some women too, really liked their hot cars when they were teenagers, but a lot of people liked their horses—men, women, children—a lot more than cars, frankly. There's a lot of things that horses can do—a horse goes off road.
RUTLEDGE: If you want to go across that field or down a forest trail, you could do it on a horse. With ATVs, you have to be more cautious. Horses have a spectacular spatial memory, and before GPS, that was a big deal. In the wild, a horse may have to go 30 miles for water. There is serious selection pressure for a good spatial memory. The idea of a horse pulling the milk cart, where the horse knows all the stops and can get the milkman home if he falls asleep, is pretty cool. Maybe cars will be able to do it in the future. These were the kind of things that were kicking up in my mind.
Each time I visited Cambridge, the editor Julie Lancashire and I would go out to dinner and we would talk for several hours about the project. Anyway, I wrote the book proposal again, with a broader approach. But the writing was painful, and I told Julie, "I feel that for every paragraph I write, I have to read a book." It took much longer than I thought it would. But she put up with it.
Multiple Meanings of Environmentalism
ZIERLER: As you alluded, from geologists to economists, there's all kinds of scholars who specialize in the question of energy supply and demand. What unique perspective, as an electrical engineer, do you think you brought to the field that ultimately convinced you that there was a book in the works?
RUTLEDGE: You've asked the right question. The central intellectual question is whether the production history can be used to predict the total production in the long run, with the emphasis on coal, which has the largest resources of the fossil fuels. The logistic function is a kind of s-curve. It starts off with an exponential rise, and then approaches an ultimate value with a matching exponential decay. The logit transform that you mentioned from the title of the paper linearizes the logistic function, and this allows you to use the standard regression mathematics to calculate a slope, intercept, and correlation coefficient. These enable you to make the prediction for the total production in the long run in a relatively stable way.
Now, there are times when this obviously doesn't work. Oil has had several technical changes that are very large, like seismic technology and fracking technology, and these disrupt a logistic curve. You might say, "I'm going to do a logistic fit for the western United States for oil." But then you find some new wells, big new fields. Or you go to fracking. Something disrupts the logistic fit. On the other hand, coal mining is something people are good at, and they have been good at it for a long time. The only significant new development for coal is surface mining, using explosive to remove the overburden. Even surface mining for coal is 100 years old now. I was intrigued by the idea that the past would somehow be important in telling about the future.
Another theme in my class that I was fascinated with was that the students in my class were almost all urban kids. It's rare to have a farm kid. But when you talk about energy, you are talking about people in the country. My home town, Fort Worth, is the exception. People there seem to think it is a great idea to have an oil well under their houses, as long as there are royalties. I wanted the students in my class to try to understand the rural communities that produce our energy and our food and our lumber. The people in the cities make the rules, and they can destroy the rural communities.
The classic example that I covered in the book, was the Spotted Owls taking land that was used by the lumber industry in Oregon. City people made the rule that because of the owls, they were not going to be able to cut down trees. The country people did have a plan in place—which did not involve owls—with goal of having a stable lumber industry. They cut a certain number of the trees each year on the private lands, and then they would be planting new trees to replace the ones they cut down.
I drove through Oregon once, with my wife Dale, towing our RV, and we stopped off in Douglas County, which used to be one of the centers for the Oregon lumber industry. We went into the quilting shop. My wife wanted to meet people there and talk. I listened. They were all married to foresters. They said, "There used to be 100 lumber mills here, and now it's less than ten." That really hit me, because the truth is, the owl they were trying to save, the Spotted Owl, has a very similar subspecies in California that is not threatened. The other thing is, after they shut down the lumber industry, the number of Spotted Owls in Oregon kept going down. The federal biologists' model obviously didn't work, because they destroyed the lumber industry and it didn't fix the problem with the Spotted Owls. Now their idea is that an eastern cousin, the Barred Owl, which unless you're a birdwatcher, looks pretty similar to the Spotted Owl, is moving in, and outcompeting the Spotted Owl. So their solution now is to shoot the Barred Owls.
I just don't have that mental framework. I view the interactions of the Spotted Owl and the Barred Owl as part of a natural process, and a reasonable approach is to do nothing, to let it play out. Most species on Earth become extinct. This is a natural process. But it clearly isn't the way people think in Fish and Game, or they wouldn't be out shooting Barred Owls. But it's also not clear shutting down the lumber industry helped the Spotted Owl. The locals had severe disagreements with the biologists' models in terms of where they had seen the Spotted Owls nesting. The biologists' theory was that the Spotted Owl has to nest in old trees, so they made a rule you can't cut old trees. The locals said, "We see them in our backyards, so your model is wrong." But the biologists made the rule anyway. They did not save the birds, but they destroyed the lives of country people.
For more than 40 years, my wife Dale and I have lived in New Mexico in the summer, up in the mountains. Our county seat is Taos, which is not a big town. We're an hour's drive into the mountains from Taos. So we're definitely in the sticks. Our neighbors are different from the people we know in Pasadena. We know people who can run a power plant and an oil refinery. We know farmers; we know foresters. I think it has made us sympathetic to the challenges that they have. I tried to get the students to think, well, "I know you're a city person, but this person who lives in West Virginia, why do they do this?"
Several of those things that we have run through in the last 20 minutes of talking were things that weren't showing up in other books. Often other books would have a particular focus like, "I think we can bury carbon dioxide," for example. That might take over a third of the book. This idea seems to infect people who live within 50 miles of Boston—MIT and Lincoln Labs. Anyway, that one didn't seem very interesting to me, because the only people that have been burying carbon dioxide are people who want to get oil out. They put carbon dioxide down the oil wells. The problem is, that for every carbon atom they put down, they get two carbon atoms up.
ZIERLER: In the way that you organized the book, did you think about the public policy implications? Whether you wanted to have that impact or not, did you see that to some extent as inescapable?
RUTLEDGE: I definitely tried to escape it. I'm not criticizing people who don't take this philosophy, but I always had the philosophy—I didn't want anyone to know my politics during the class. Students come by, after. Caltech is a small place. They come by and hang out. They want to talk to Dale. So, I wouldn't hide my politics then, and certainly not after they graduated. I feel I have done my job if they tell me they could not figure out my politics during the class. I wanted to present tools to analyze energy production. I didn't really want to talk about policy. I would have preferred at the beginning to say almost nothing about climate. When I started thinking about energy, the issue was the perceived scarcity of oil, and how the production of fossil fuels might evolve. But in the time that I was thinking about it and teaching, the emphasis was shifting to climate policy.
Even in climate policy, what I did was to present at the beginning—the best data connection, that I mentioned before, is that, if you plot cumulative production of fossil fuels in terms of their carbon content versus the atmospheric CO2 concentration over time, half of it stays up in the atmosphere. It's a spectacularly linear relation. The correlation coefficient squared [r squared] is 0.98. So there's absolutely no doubt about what is going on. The data for fossil fuel production is reliable, as is the data for the carbon-dioxide concentration. In addition, during the time that accurate carbon-dioxide measurements are available, fossil fuels production has risen by a factor of six, and the relationship between the cumulative production and the atmospheric CO2 level tracks perfectly. It is an excellent test.
As an analysis tool, I give a relationship, based on historical correlations, between temperature and the changes in the production of fossil fuels. For example, this might be the temperature change that we might expect from a policy that reduces our coal production by a billion tons. One of the themes of the book, from a modeling perspective, is that if you want to predict something, you need to use a small number of parameters. This is in contrast to fitting something that has already happened. Here are you are just interpolating, and you can use many parameters. You can have one fitted parameter for every data point. But if you want to extrapolate into the future, you don't do that. The reason is that adding the extra parameters will often make the prediction unstable. Once that happens, it is easy to fall into the trap of predicting the outcome that fits your own ideas about how the system might evolve. So, I work with a small number of parameters.
In contrast, climate models have hundreds of parameters. The problem is that climate has a large range of space and time scales. The role of many of the parameters is to simplify the calculation. Otherwise the calculations would overwhelm by a large factor our data collection and computing ability. The result is that the climate models can't be validated to give us assurance of the accuracy of the predictions. This why I use a simple historical correlation.
It is ironic that, in contrast, weather forecasts have become quite good. I am a boater, and I have had times where I am planning a return from the outer islands a few days out, and think, "Well, if I come back before noon, I'll beat the swells that are coming from the Gulf of Alaska." Sure enough, I get back, and then the big swells come in an hour later. They're incredibly accurate.
In my talks and in my books, I try not to prescribe a policy. Possibly the one weakness that I have, and I mentioned it, is that I do want people to understand the perspective of people who live in rural areas. I think they often can get the short end of the stick in many ways, like applying to a selective university. I don't think we do as well for them sometimes. Some of these rural kids, they won't have the city programs for science that would tell us, "This is a Caltech student."
I've told colleagues that every once in a while, I could feel like by the end of a talk, I could lose 95% of the audience—some people on the right feel that fossil fuel production is relatively unconstrained, it's simply a matter of economics. And then some people on the left believe that I should emphasize negative consequences and tell people what should be done about it. I can feel it sometimes in the questions, "Hey, you're not talking about how bad things are going to be," or "You're kidding me that we're going to run out." It comes with the territory. I had a paycheck the whole time, so I can't really complain. But the discussion has gotten more ideological than I would like.
The example that I would give for how rural people are treated is Sri Lanka. It is heavily dependent on agricultural exports. The people at the top in Sri Lanka are very much associated with people at the top in other countries, and they are exposed to their discussions. They got the idea that Sri Lanka could do something for climate. Of course, Sri Lanka is not like, say, Canada—with massive consumption of oil for cars and trucks. Sri Lanka can't help on fossil fuels. But, they thought, "Well, if we go to organic farming, we'll reduce the emissions that are associated with agriculture, like the production of nitrogen fertilizers that use a lot of natural gas." So, they went organic. Well, the estimate I've seen—and I've checked the calculation—is that half of the nitrogen atoms in our body—all amino acids contain nitrogen—are synthetic, meaning they came into the food that we eat through the fertilizer that contained nitrogen from the factory that synthesized it from molecular nitrogen in the atmosphere. If you take the synthetic nitrogen fertilizer out, you cut food production in half. Well, maybe that's okay if it's 1800s US and you can clear more forest for agriculture, but Sri Lanka is not that place. You can't double the area under the plow. So, their food production crashed. That meant instead of food being their main export paying for almost everything the country imports, they ended up importing food without the hard currency to buy it. It was an absolute catastrophe. I think that the people at the top, presumably for what they thought were good reasons, talked to their colleagues in other countries and wanted to impress them with their leadership in climate policy.
ZIERLER: In the course of researching and then working through the ideas and writing the book, what were some of the biggest surprises you encountered?
RUTLEDGE: The biggest surprise was how long it took to write the book. It was important to have a really patient editor, who would have to explain to her bosses why I was so slow. Energy systems are highly constrained by history and by the enormous capital investments that are required, and you really need to understand that it's not a mathematical exercise. You can't just sit and scratch out the plan on a napkin when you're eating lunch at the deli. You've really got to go read the history, and understand what smart people have said about energy over not just the last two years, but the last 150 years, going back to Stanley Jevons. The transition to alternatives—it's not if; it's when—is going to be expensive and slow, costing something like the annual GDP of the world, and lasting the lifetime of my grandchildren.
Maybe that's what's different from electronics, where I started in engineering. Electronics, possibly because it has a strong mathematical foundation, developed quickly—the transistor itself was invented only four years before I was born—and history is not particularly important or relevant in electronics. It's interesting to historians, I'm sure, but it's not relevant to working engineers. When I was a kid, electronics was a tiny industry. Cars and chemicals were big industries. Electronics is a huge industry now. Apple, Google, Facebook, are some of the biggest companies on the stock exchange.
So, I wasn't prepared for how much background reading I had to do in energy. For example, to understand geothermal energy—it's important for California—I really had to read its history. I really had to understand, why does geothermal work for us in California, but not in Japan. Japan is like giant volcano, but the Japanese get little electricity from it. Maybe if I had done energy work all my life and it was just second nature, then maybe I could just sit and reel five pages off every night, and produce the book in six months, and get on with life.
ZIERLER: What was the timing of the book's status when your wife fell ill?
RUTLEDGE: We were extremely fortunate. I told you that I retired so that I could finish the book. Dale was patient and she said she was okay with the reduced income, and I really wanted to finish the book. I just didn't want to be a person who always was going to write a book and not finish it. The book came out in January 2020. Dale's cancer was diagnosed in April. But for the miracle of the CAR T therapy, it would have killed her within six months. But by then, I really was retired. I also really wanted to be retired. I really wasn't that pleased with the politics entering into the energy area. I told people, "Whatever I do in retirement, it's not going to involve looking at a screen. It's not going to involve checking and seeing what the policy issues are, that are coming up, day to day." So I was available for Dale when she got sick.
ZIERLER: Have you paid attention to how the book has been received, the debates that it has ignited?
RUTLEDGE: I tend to avoid that. I do try to keep track of things that I might have gotten wrong. The bitcoin page has bothered me. Mining bitcoin consumes 1% or 2% of the world's electricity, enough to include in the book. To my engineering mind, bitcoin is pretty sound, in the sense that you need a certain amount of energy to mine bitcoin, which tends to put a floor on the price. I thought bitcoin could eventually be a store of value, like gold is. But the warning, of course, is that governments, historically, have not liked gold as a store of value. I don't think any major country has a silver or gold standard for their currency today. We got off gold in the Nixon administration. If you look at economic growth during the time when we were on the gold standard, it is pretty hard to argue that the growth was worse than it has been since. What you can say is that when the US had a gold standard, there was almost no inflation, aside from the times the government turned on the printing presses during wars. On the other hand, since Nixon, we've averaged 4% inflation. You may not feel it, because you have a salary that may be keeping up with inflation. But retired people get hit twice. First the prices they pay go up. Second their assets lose value. The inflationary 70s were a terrible time for the stock market. And last year has been characterized as the worst year for bonds since the American Revolution.
I thought, with bitcoin, countries may at some point simply outlaw it—China has done it—because they don't want miners to determine what has value. They want the government to determine through its fiat currency what has value. I was right with China, but China is more paranoid than Western countries about these things. But what I missed is the group that is interested in bitcoin—and I own none—for speculation. This group also, for whatever reason, tends not to look at it like an engineer, and say, "Because it takes many kilowatt hours of electricity to find a bitcoin, when the price of bitcoin drops below the electricity cost, mining bitcoin will stop, and that will support the price." What the speculators bought as an alternative to bitcoin was the currency created by Samuel Bankman-Fried, simply based on personal trust. Well, that's a lot like government fiat, except it looks like he's a fraud. [laughs] And he damages the entire industry.
I think of the book as primarily providing tools for analyzing energy. If I hear from a guy who does ESG [environmental, social, and governance] analysis, evaluating stocks for whether they're environmentally sound, and he thinks it's a useful book, that's fine with me. He is probably on the left politically. If I have someone who thinks we're running out of oil, and he likes the book, that's okay, too.
ZIERLER: Moving the conversation right up to the present, since the book's publication date, is the energy industry so dynamic that you're already seeing parts of the book that might need to be updated or expanded upon?
RUTLEDGE: I haven't seen that. I told you about the one thing that worried me, as I was writing it, which was how to deal with crypto currency. I have not seen anything that I would want to change. It would be tricky to publish the book right now. I use the most recent data, but there were energy consumption changes related to the virus, like more people working from home. This was already happening before the virus, and it may be that the virus accelerated the trend.
In the book, I did discuss how self-driving cars might affect energy consumption without predicting when they will work or even whether they will work. I showed a picture of driving in India, with something like 30 cars all in the same intersection. How are you going to manage that with self-driving cars? I think it's hard enough for self-driving cars just to find the lanes. So, no, I haven't found anything I really want to change. The proper time to reevaluate the energy production predictions is probably two decades away.
I am seeing some interesting changes in energy supplies for RVs and boats. I like the new inverters and LFP [lithium iron phosphate] batteries. They are fabulous. You can now have essentially full AC power, on a boat or in a camper, and it is quiet. Before we only had 12V, and there are a lot of things that do not run well on 12V, like air conditioners, air driers, and Elon Musk's extraordinary Starlink internet antenna. To me, it is a revolution. The LFP batteries don't catch on fire. That is a big deal on a boat. Recently a dive boat out of Santa Barbara caught fire, possibly from lithium ion batteries, and it killed more than 30 people. The LFP batteries also do not need cobalt. That means that you don't have to worry about the conflict minerals issue of cobalt in Africa. But, probably connected to the fire safety, the LFP batteries do not have the energy density of the NMC [lithium nickel manganese cobalt] batteries that dominate in the high-end EVs.
EVs might work for a vehicle that is just going around town, but the way I use a vehicle, say if I want to tow a boat to Canada, that's 2,000 miles. The total weight for everything now—and an EV would be heavier—is 18,000 pounds. When you do the math, it may not be possible to do this job with the EV technologies we can visualize. You're talking about a truck for a single person, not a commercial person, that might cost hundreds of thousands of dollars, and might weigh five tons. You might have to charge up every 100 miles. Would there be charging stations for a towing rig that is 50 feet long every 100 miles? How would you do that? I just don't see how that is going to happen. It helps to be a California resident here, because this is probably version five, of the edicts related to electric vehicles. I think you said you had seen the movie Who Killed the Electric Car? That was edict version one. X percent of cars sold will be electric. So now, it's 100% in 2035. I don't think we have met any of the previous goals. I think I just want to be an observer, go out on my boat, and leave it to your generation.
RUTLEDGE: They are challenges, and engineers like challenges. In that sense, I'd be happy to try, if I were younger. It's not in a timeframe that I have, but for a young person, it might be very attractive. If it were me, the question would be, "Is this an interesting engineering problem?" Not saving the world.
ZIERLER: On that note, now that we've worked right up to the present, for the last part of our talk, to wrap up this excellent series of discussions, I'd like to ask a few retrospective questions about your career and work, and then we'll end looking to the future. First, in all of your work in electrical engineering and communications—antenna, microwave, radio—what are you most proud of? Either because it was so significant, or because it really influenced the kinds of things that your students went on to do?
RUTLEDGE: You asked a less sweeping version of this question earlier, and I wasn't really comfortable answering. I don't really think in that way. If I have to say something, I like the idea that a group produces students. The product is students, and the students do great things. It's not just that. They're colleagues, they're family, they're friends. Their children and their spouses are family and friends. On the engineering side, as long as it is legal and ethical and interesting, I just think it's a great life. Now, the technology changes enough that you don't say, "Ah, Professor X did it this way, starting four years ago now, so I'm going to try to mimic that." You have to make the contributions within your own culture and intellectual environment, with the students you get, and the kind of problems they can tackle.
ZIERLER: I don't have to tell you what a unique place Caltech is. What aspects of your research career do you really associate or were made possible just by virtue of being at Caltech? As a counterfactual question, what are the areas of your research that are just so close to you, so close to your graduate training, that you would have pursued them no matter where you had a professor's career?
RUTLEDGE: I was a student at Berkeley, and they tried to get me to apply for their faculty position, but it was pretty clear that Berkeley had a different faculty development model from Caltech, which is, Berkeley tells you, "We have this research hole that you will fill." The hole that they had was from Tom Everhart's leaving. Tom later became president here. Tom did electron microscope work, but he was going up the administrative ladder. They wanted someone to do electron microscope work. It didn't matter to them if the new hire didn't do electron microscope work before. Now, they did find a person that did it, Ilesanmi Adesida, a very good guy. He has done fine at Berkeley.
But that wasn't me. I was interested in chasing my own fox. In those days, Caltech was, "Here's an empty lab bay. Here's an office. Go to work." That fit me. It wouldn't be a fit for everyone. A lot of people want mentoring. A lot of people want to work in a larger group when they get started. A lot of people may want to be told what to do. But I wanted the empty lab bay.
A Legacy in Recruiting
ZIERLER: What about on the administrative side? I tried in our first conversation to get you to reflect on your accomplishments on the Engineering side. You gave me a little bit of a more expansive answer this time. What about administratively? What are you most proud of in your accomplishments in EAS?
RUTLEDGE: That is easy. Recruiting. Within the Electrical Engineering Department, I was the search chair for the three committees that brought Axel Scherer, Ali Hajimiri, and Yu-Chong Tai. All outstanding researchers. Then at the division level, in addition to the faculty who focused on research, the faculty who became administrative leaders in the division—Azita Emami, Beverley McKeon, and Adam Wierman. And we just had a power satellite launch from Sergio Pellegrino and Ali Hajimiri.
ZIERLER: I know you don't want to talk politics in the classroom. Have your politics changed at all as a result of writing the energy book?
RUTLEDGE: Well, on the climate side, I am more of a skeptic than I was when I started energy studies. If you interview 20 people who study energy, you would find some skeptics in that group. They might differ a lot in their thoughts about fossil fuel resources. For fossil fuel resources, I'm on the more pessimistic side. For me, politics just is not part of my job. In contrast, Steve Koonin, former provost, and, I thought, a superb provost, has waded into climate politics. Steve has had a great career, chief scientist at BP, and then head of science within the Department of Energy in the Obama Administration. Now he runs the Future of the City program at NYU. He has just published a book called Unsettled, about climate modeling. Steve has a background in large-scale computing, and he is qualified to write the book. He also did an experimental study of earth shine, the light reflecting off the earth's surface that reflects again off the Moon surface. Reflected light is one of the components in the energy balance of the Earth. It was a very interesting project. Unfortunately, I think his accuracy was about the same as the other ways of doing it, so it would not be considered a breakthrough. Anyway, Steve has attracted a huge amount of publicity for his book. He has debated some people. I have seen the video of the debates and I think he has done well in them. He does need some body armor, because climate science is a tough neighborhood and skeptics are subject to vicious personal attacks.
In contrast, microwave engineering, where I started, is a gentle intellectual neighborhood. Colleagues are supportive of students and their research. And we have delivered. Cell phones changed our way of life. Satellite communications has changed our way of life. And we did it without beating up on people. Back in 2009, there was a group of emails from climate scientists at East Anglia University that revealed how temperature series were manipulated and how nasty the climate scientists were to those who were skeptical of their claims. I never saw any of this in microwave engineering. Steve takes a lot of arrows for his book. I've read his book. I've also dealt with him in some situations that related not to science, but dealt with personal integrity, and I would say he's a person, in those situations, of the absolute highest integrity. I read what he says in the book; I can't punch holes in it. There might be some things I wouldn't state the same way and I might not agree with him on. But on the other hand, he has presented something that is coherent, and honest, as far as he sees it. He's certainly in the limelight these days about it. But it's not me.
ZIERLER: Your own skepticism, in questioning the orthodoxy of climate change alarmism or whatever you might call it, does that at all change the equation from the science side, from the energy supply and demand side, in terms of what should be done with fossil fuels?
RUTLEDGE: Here is how I would frame my skepticism. The agreement between the carbon dioxide concentrations and cumulative fossil fuel emissions is so spectacular, it simply admits no other theory. The other data point that is pretty good—I would call it second best—is what is called global greening. This gets into the other side of the carbon dioxide emissions—if half of the carbon dioxide from fossil fuels is still up there, what happened to the other half? We're producing—burning—a lot more fossil fuels, like five or six times, as a world population, than we did 70 years ago when good atmospheric carbon dioxide measurements became available. What happened to the other half? Initially people thought, "Well, it's going to end up in the ocean." And presumably if you wait 10,000 or 100,000 years, that's what will happen. But in the short term, a lot of plants really like carbon dioxide. Gross photosynthesis is up by one third in the last 100 years. That's a spectacular rise. And three quarters of it is from the increased carbon dioxide concentration.
This came up in my class—if you want to find a carbon dioxide supply to use for the purpose of growing your plants bigger, who do you go to? You go to the guys who are selling supplies for growing marijuana in homes. They certainly figured out that if you have a higher concentration of carbon dioxide in your home, you're going to get a lot more marijuana to sell. My sense is that as a result of the increased photosynthesis world agricultural yields are up15% to 20%. Also I suspect some areas are more interesting places to live in now than they were before, because there are more green plants. My sense is that the positive effects of increased photosynthesis outweigh any negative effects of climate change.
Now, in my book, I'm writing it as an engineer, speaking to engineers. Most engineers, once we have an employer that sets a legal, ethical goal, we'll try to make it work. We think we can make it work, because we're good engineers. So, in the book, I take the view that, "Yes, we can do the transition to alternatives, because my audience, my students, are good engineers, and they'll make it work. But it looks like it will be expensive.
ZIERLER: Finally, last question for you, a more personal one. As you look to the future, are you enjoying a pure personal retirement? Is it about the boat? Do you want to keep engaged in the literature? What do things look like for you?
RUTLEDGE: As I told you, I would like problems that don't put me in front of a computer screen, and that don't involve politics. Boats involve all sorts of puzzles. To the extent that I look at a screen, I am looking at a weather map. I like truck camping too. I don't know if you've seen the statistics, but old people aren't necessarily worse drivers, for some reason. They're not as quick as young drivers, but they've seen a lot more things that have gone wrong, and maybe they plan a little bit better to fit in with the other drivers. I think Northern America, that is, the US and Canada, are spectacular places to visit, both on the water and on the land, compared with any place in the world, and they are completely accessible without a visa. There is a common culture. You can go anyplace, and talk to people, and they will make you feel at home.
ZIERLER: I want to thank you for spending this time with me. This has been an excellent series of discussions, great insight on engineering, academics, energy. I want to thank you so much for doing this. I really appreciate it.
RUTLEDGE: Thank you.
- Late Career Interest in Energy
- Assessing Alternatives to Fossil Fuels
- Engineering Focus on Microwaves
- Enabling New Research with Software
- Texas Roots
- From Williams to Cambridge
- Career Launch at General Dynamics
- Tungsten Whiskers at Berkeley
- Falling in Love with Caltech
- Fabricating Integrated Circuits
- Faculty Support and Promotion
- Contributions in Nuclear Fusion
- Puff the Magic Software
- MESFETS and MOSFETS
- Optimizing Cellular Communications
- Named in Honor of the Tomiyasus
- Perspective on Administrative Responsibilities
- Coal Production Estimates
- Multiple Meanings of Environmentalism
- A Legacy in Recruiting