§1. Materials hype and the Sprague/Williams corollaries [00:00]
That's my tokenism to structural materials. Now this is a quote from Daniel Kahneman, and next week I will probably give everyone a copy of this book. Has anyone ever heard of this book, Thinking Fast and Slow? Kahneman won the Nobel Prize in economics back in 2002 or 2001. He's a psychologist, and he's also a statistician. An MIT grad recommended it to me, and it's one of the few books I've read that I put a lot of highlights in. One of the highlights is: I've yet to meet a successful scientist who lacks the ability to exaggerate the importance of what he or she is doing, okay.
Materials and new materials, or whatever you read in the MIT news and see on the screens at MIT — most of this is oversold. Ninety-eight percent of it is a sales job, and it's not really quite that great. There's a quote that I like. There was a guy at General Electric in Cincinnati where they make aircraft engines, head of materials, name was Robert Sprague. And Robert Sprague had a quote: whenever you first hear about a new material, write it down, because those are the best properties that material will ever have. And then there's Jim Williams' corollary. Jim Williams was a titanium metallurgist who became department head at Carnegie Mellon and Dean at Ohio State, and then he replaced Bob Sprague at General Electric. Jim Williams says his corollary to Bob Sprague's quote is: whenever you first hear about the cost of a new material, write it down, because that's the lowest price the material will ever have.
So not only am I unconventional in my teaching, I'm a little bit unconventional in my approach to: is material science as wonderful as everywhere it says? Actually it is pretty wonderful in many ways, but most people are exaggerating how wonderful it is.
§2. Syllabus and externalities [02:35]
Okay, so today we're supposed to be starting What Is Engineering. It's sort of a catch-all, in part because students have asked me to talk about different things at different times, and I finally decided to put some of these things together in terms of what is engineering. This is the syllabus, it should be on Stellar. We're going to start today with definitions of engineering. Then we'll talk about externalities. Do you know what an externality is? It's an economics word, and it's all the things that don't have to do with what you think the main technical subject is — like politics.
Student: Economics, or cost, public perception, environmental.
Yes. And externalities are often the biggest hurdle to introducing a new technology. The technology is great, but there are other factors.
We're going to talk about the history of engineering, just because I think you'll be surprised at where engineering came from and its various diversions. We're going to talk about the scientific method, and I'm going to talk about Kahneman — that's why I bought you a copy. Amazon sells these in paperback for nine dollars nineteen cents. I figure you're all worth nine dollars nineteen cents; if it'd been more than ten, maybe not. They should be in tomorrow or Monday, and I'll bring them to class. Kahneman talks about thinking fast — your intuitive judgment — and thinking slow, your analytical judgment.
We get into what's an expert. If you're going to be an engineer you're going to be engineering experts, and what does it mean to be an expert. We're going to talk about design. To a certain extent engineering is design, and how design has evolved. We're going to talk about forensics and how do you get to the root cause of a problem. We're going to talk about errors and logic and learning to be honest with yourself and forcing other people to be honest with themselves, which is not always a way to win friends and influence people. And we're going to talk about language translation and communications, which I said are important. Then we've got some miscellaneous topics, and if you want some other topic let me know, I'm happy to try to incorporate it.
§3. Pre-assessment exercise [05:30]
Right now I want you to take a couple minutes to try to answer some of these questions. You got the paper, you're going to keep it, I'm not going to read it. Write down what you think engineering is, and go down as far as you can in the next three minutes and try to give me your idea.
Okay, you keep writing. What I'd like you to do is keep this — if you want to, bring it to class every day. But maybe once a week you can pull it out and write down any other thoughts you might have, fill in the blanks. At the end of the term I might give you an assignment to tell me how your opinion has formed or changed.
There's a couple other things with this exercise. I just pulled a teaching technique on you that you should always be aware of. Anybody know what it is? It's called pre-assessment — to find out where your students are. You're all at different levels of understanding of what is engineering, and whenever you're teaching someone, the first thing you should do, no matter what the topic is. If you're teaching a cooking class you might want to know if you've got one of the world's top chefs as one of the students.
Professor Sadoway and I were asked back in the late 1970s to teach a course to 40 IBM students one summer, and we were supposed to teach them introductory materials science. Fortunately we did ask them what some of the background was. A couple of students had PhDs in materials, and we were supposed to be teaching them introductory materials science. We also had electrical engineers who didn't know what a material was. So we came back and said, well, how are we going to teach this when we've got people who know nothing all the way up to people who have a doctoral degree in the area? But we found out the old MIT technique sort of works: you go back to the basics. If you go back to the basics, people usually can find some insight, because the basics are actually simple.
So pre-assessment is one thing. The other thing is, I sort of sandbagged you — there is no good definition of engineering. As Kahneman will say, I've essentially pre-programmed you. Actually I think there is a good definition. About 25 years ago, during a Columbus Day holiday, about 24 people from the School of Engineering — department heads and a few others; each department head was allowed to pick one younger faculty member, and I was the one Professor Flemings chose — went down to a retreat on the Cape to figure out the strategy we should have for the next 25 years for the School of Engineering. We spent a lot of time trying to define what is engineering. You would think that 24 engineers from the School of Engineering at MIT could define what engineering is. We spent about a day in little breakout groups, and we finally came up with what I still think is a very good definition, and I'll give it to you later. But about 10 years ago I asked the assistant dean, Donna Savicki, for a copy of it because I didn't have that little report. She barely even remembered it, okay.
So we had this little meeting. This is another lesson. The meeting had to have cost twenty-five thousand dollars — 24 people to go spend a three-day weekend at the Cape in a hotel, probably cost more than that. We had some output which I thought was valuable, and it gets filed away in a file cabinet, and it gets thrown out in the trash someday. So I don't really have the exact definition, but I remember enough of it.
§4. Student definitions and von Kármán [12:35]
So anybody have something they'd like to submit, or shall I start telling you what some other people said?
Student: What I'm hearing is the application of logic to generate products, services, or actions that meet some given need.
Did you write that down or did you look that up? Okay, fine — that's very good. You could Google, and I'll show you some Google definitions, that aren't all that different from that. Logic, and people's needs. That in fact is one of the definitions we're going to go through, that other people come up with. I actually gave all of you something a few minutes before that: I said engineering is design.
Student: Making new things or making things in any way. Originally I just wrote "make anything."
Well it is. There's a creative process to engineering that is not involved the same way in science.
So those are all actually pretty good. I pre-assessed it: you know more about it than I do. So maybe we're ahead of the game and we can skip time. But it turns out, here's a quote: if it stinks it's biology, if it stinks chemistry, doesn't work it's physics, if it works but no one knows why it's engineering. So that's the cynical view. Cynical is funny. One of the reasons humor is often funny, particularly things like Dilbert, is because it's true. There's a certain truth to it.
Let me give you the best succinct definition that I've ever heard, and it's from Theodore von Kármán. Who was Theodore von Kármán?
Student: Partly legendary aerodynamicist and founded a laboratory in Pasadena, California.
Did he found Jet Propulsion?
Student: Jet Propulsion Lab, he founded it.
He's the founder, over in the 1940s. Okay, you guys are pretty good. Von Kármán said: the scientist explains that which exists; the engineer creates that which never was. He was both a scientist and an engineer. And there's more depth to that definition than you might think. Joel Moses, who used to be Dean of Engineering here — he was head of electrical engineering and then became Dean of Engineering, became Provost — he used to say, and the Media Lab creates that which never will be.
There are a lot of wannabe scientists at MIT, most of them in the School of Engineering. I often say eighty percent of the faculty in School of Engineering are wannabe scientists. There's only about twenty percent engineers on the faculty in the School of Engineering, in terms of people who can actually go out and solve a real problem. A scientist, if you present a problem to him, will throw out all the hard parts so he can solve it in closed form with a fourth-order differential equation or whatever sophisticated technique. And in fact that's what you've learned in many of your classes to do.
Whereas Thomas Edison was an engineer. Edison sort of was intuitive — he just kind of did things, and he made them work, and he created things that never were. He didn't always understand why they worked. He said it was ninety-nine percent perspiration, one percent inspiration, whereas a scientist would much rather be on the ninety-nine percent inspiration and one percent perspiration. That's why they do calculations — not a lot of heavy lifting to do, but it takes deep intellect in a very narrow field.
Phil Morse and Herman Feshbach — Feshbach was head of physics 20 years ago, and Morse was his mentor — wrote a book called Theoretical Modern Physics or something. It's basically a bunch of mathematical techniques to solve very complex equations. It's a very famous book. But most physicists will throw out the difficult terms so they can solve it explicitly. Then they pound their chests: "I solved this problem." Well, they solved a mathematical problem, they didn't necessarily solve a real problem.
§5. Brian Josephson, Margaret MacVicar, and the Josephson effect [17:33]
Does anybody know the story of Brian Josephson, the Josephson effect? Brian Josephson was a graduate student about 1962 at, I think, Cambridge — might have been Oxford, somewhere in England. He went to see his thesis advisor in physics, wanted to know what should I do my thesis on. His professor was busy, so he said, well, go solve this problem, and it had to do with electrons. Brian Josephson goes out and writes down the differential equation. Most people who had solved this problem in the past — it was sort of a standard problem in physics — knew that you threw out this one term because it was inconsequential. Brian Josephson, being young, not knowing much about the world, kept the term in, and went through the harder problem of solving it with that term, and he discovered the Josephson effect. The Josephson effect is basically tunneling of electrons through insulating layers. At age 24 or so, he won the Nobel Prize.
I met him when he was about 30, because of my thesis advisor by 1965, Margaret MacVicar. Ever heard of Margaret MacVicar? The MacVicar Faculty Fellows at MIT. Margaret was a physics student who couldn't get into the physics department at MIT, because they won't admit their own. That's why Feynman went to Princeton, because MIT wouldn't admit him into the physics department. Margaret came to the materials science department, wanted to do a doctoral thesis, went to Bob Rose and said what should I do my thesis on. He says, go up and see Pfeffer, who was the technician up in the lab. Pfeffer trained me in the lab, and Pfeffer was a genius — should have gotten an MIT PhD, but he grew up in the depression, his family couldn't afford to send him to school. Pfeffer showed her how to grow single crystals of niobium, lay down an insulating layer on it, put some electrodes on it, sit it in liquid helium, and do tunneling experiments and show the Josephson effect.
So Brian Josephson did the calculation saying, electrons can go through insulating layers, and Bob Rose and Margaret did the experiment. Margaret graduated in two and a half years, went off as a Kennedy Fellow in England, came back here as an assistant professor of physics, and rose to Dean of Undergraduate Education, but died of brain cancer at about age 41 or 42. A scientist explains that which exists, and Brian Josephson made the mistake of solving the harder problem and not throwing out the insignificant term, which turned out not to be insignificant. On the other hand, engineers create things that never were, and they don't always know why they work. Edison — his first useful filament for a light bulb was what?
Student: Carbonized horsehair.
So he just burned up horsehair and turned it into a carbon filament, had a life like 11 hours if you're lucky.
Student: Did Margaret get any kind of recognition for actually showing that it's true?
Margaret got a lot of honors when she came back here. Bob Rose gave her a hundred-thousand-dollar DARPA contract that he had gotten, to help her get started, and she had a budget. She created the UROP program — that's what got her tenure. Margaret actually turned out to be a fantastic administrator as Dean of Undergraduate Education, but the physics department would not tenure her, because she could not teach freshman physics. She was incompetent as a teacher. But the MIT teaching award is named after her, because Paul Gray thought she was great. She was on the Board of Exxon when she was 37 years old, because Paul Gray was president of MIT, and he thought she was great. She got tenure at large, not in the physics department, because Paul Gray was president. Margaret turned out to be a fantastic administrator. The reason we have a biology requirement now at MIT is because Margaret blasted through all the naysayers. Margaret was an interesting individual, but that's not the subject of this course.
§6. Webster's, the first engineering schools, and Harvard's missing engineering school [22:27]
Here's a picture of good old Theodore, if you want to see what he looked like. He was a Hungarian-American mathematician, aerospace engineer, and physicist, active primarily in the fields of aeronautics and astronautics. So now you can go to Webster's Dictionary, which is always the sterile way to do it. Engineer, noun, is the builder of military engines. That's the number one definition. Why? We're going to find in the history of engineering — this is posed by Encyclopedia Britannica, so it might be about 1985.
The first engineering school in the United States was West Point — I've got the date in here later. Until 1823 or 1824, when they were building the Erie Canal in New York State, when Rensselaer Polytechnic Institute started the second engineering school in the country. And they had one curriculum, it was called civil engineering, to distinguish it from the only type of engineering that was known before that, which was military engineering. So now you know where we get the term civil engineering.
But the word goes back to the 14th century in the French, which means a person who designs and invents or contrives. It actually comes from the French word meaning contrived — a designer, a builder of engines. These could be catapult engines or breastworks and dams. An engineer would build military facilities. In fact, the commandant of West Point until 1845 was required to come from the Army Corps of Engineers. They changed that, but that was part of the history.
Engineer, verb, means to act as an engineer in laying out construction or management of something like some great dams. To guide the course of, manage or supervise during production and development. What's the interesting word in here that you don't usually associate with engineering? Management. I'm going to tell you the history of Harvard's first engineering school. They tried to purchase MIT three times, 1873 to 1917, were unsuccessful because the Supreme Court of Massachusetts turned it down. So they built, on the land that Andrew Carnegie had bought for them in Allston, their engineering school, and today it's known as the Harvard Business School. I'll tell you that story.
Student: Allston is across the river?
Allston is also across the river, in Allston, Massachusetts. It's not in Cambridge.
Engineering, the science by which the properties of matter and the sources of energy in nature are made useful to man, in structures, machines, and products. Both of your definitions had some of these types of things. So that's a common definition.
§7. Roads, the Romans, and Maslow's hierarchy [26:09]
What I'm going to do — I've never taught this course before — is, I have a handy-dandy little book here called The Builders: Marvels of Engineering. This book was written by none other than National Geographic. Here's the table of contents. Of course National Geographic's budget is a picture book, not too many words, mostly pictures. I've been reading National Geographic since I was eight years old. I used to have a couple of tons of National Geographics holding down the floor in my basement, but I finally got rid of them.
They have a bunch of different examples of major engineering marvels. Since I can't afford to give you this book — it's out of print, but it's a nice book — what I'm going to do, probably every day, is spend a couple of minutes on each one of these things. Today is roads, to give you some examples of major engineering achievements over the centuries. I'll have Jerry put that chapter on roads on Stellar, and the next time I'll do canals, and Jerry can put that on, and by the time the course is over we will probably have everything on Stellar. I will have violated the copyright laws, but Stellar won't know, because fair use says I can take pieces at a time — as long as I haven't taken a hundred percent of the pieces.
So, roads. Overcoming distance is the main chapter — roads, canals, bridges, railroads, pipelines. This is something you haven't seen before probably: this is the National Road, started in 1811 in Pennsylvania — Cumberland, Maryland, to Vandalia, Illinois, 600 miles. It's an interesting story. This is one of the first major projects in 1811. The federal government paid for it; it was the first interstate highway, if you will.
If you go on you'll learn about — this is a Roman road in Syria, still in use. The Romans built a fantastic series of roads all over — 50,000-mile network of roads. Here's the Incas. The Incas did it without wheels. The Incas didn't have a wheel in their technology toolbox, but they built these huge roads. This is in Sichuan, China — how the Chinese had 200,000 people working on a road and completed it in a couple of days.
It goes on, it talks about macadam. Anybody ever heard of macadam? That's the macadam on the pavement at the airport. Tells you about John McAdam, who he was. There's the first toll road, 1940, Pennsylvania. Here's the Glenwood Canyon — there's actually several pages on Glenwood. Obviously the National Geographic had to run a big article on that. So it's some reading. The evaluations — some students said they wished I could give more reading so you could follow up — so I'll give you more reading. You can read about roads.
Going back, the Romans' roads are still working after 2,000 years. If you read about how these different people came along, they improved on the Roman system by making it cheaper, faster to build, but it doesn't have the same longevity as some of the Roman roads. Those are some of the trade-offs that you have in engineering. You also learn, why did the Romans build these roads? It was for their military — to be able to get the supplies and the soldiers where they wanted.
Anyone ever heard of Maslow's hierarchy of needs? This is a simplified Maslow; this is a more complex one, coming right off of Wikipedia. The Romans built roads as part of their military security program. Why did we build the interstate highway system in the 1950s, called the Eisenhower Interstate Highway System? He liked the autobahn, but there was another military reason that the generals wanted.
Student: Had missiles on trucks.
The only way you could keep the Soviets from targeting our launch sites was to move the launch sites. The Soviets did the same thing — they built mobile launch sites, and we built mobile launch sites. We built a huge interstate highway system to be able to move the missiles around. It's harder in the Soviet Union to do that — it's a little larger.
The first level on Maslow is physiological needs — food, shelter, clothing. The next one is safety, security — that's your military security. The Romans built roads for the second level on Maslow. We built the interstate highway system for the same reason. It was a military endeavor.
§8. The space shuttle, Wernher von Braun, and Feynman's failure statistics [32:44]
How many of you think the space shuttle was built for civilian purposes? It was built to hold the lasers they were going to put in space to shoot down the missiles coming at us. The cargo bay just happened to be the exact size of the chemical laser weapon — amazing. But you couldn't have gotten it through Congress if you were just proposing it as a 20-billion-dollar military project. But if you said we're going to make space travel cheaper so it only costs a thousand dollars a pound to get into space, rather than twenty thousand dollars a pound of payload — now, did they achieve it? The space shuttle over the 25 years or so, they increased the price to about fifty thousand dollars a pound in space. They didn't get it down to a thousand; they still talked about trying to get it down to a thousand dollars a pound.
Student: Which means the space shuttle is safer?
Have you ever heard the story of how Feynman analyzed the space shuttle? NASA had done this big safety analysis, and they proved by all this probabilistic — probably paid some professor at some well-known Eastern University to help them with this — there was one chance in 10,000 of the space shuttle failing on launch. And then when it failed on launch, Feynman was part of the blue-ribbon panel that was supposed to figure out what was going on, and they learned it was the O-rings, in the cold weather — there's a whole book written on this stuff. But Feynman basically pointed out that, since Wernher von Braun — we stole him from Germany — we took all the great scientists out of Germany before World War Two. The center of science in the world was Germany, not the United States. In 1900 you were required at MIT to take several semesters of German so you could read the scientific literature. All the great scientists — think of the quantum mechanics folks, start naming them off — mostly Germans. Americans would go over to Germany to learn the science and bring it back. After World War Two, the United States became sort of the center of science. Why? Because we picked and chose — we brought them back. There was starvation going on in Europe after the war, and so we brought the people back that we wanted to bring back, and we essentially made the United States great in science by bringing all these other people. Wernher von Braun — they set him up in Huntsville, Alabama.
Feynman showed that four percent of the missiles they had shot since 1947 had failed on launch. He also pointed out, this was the 25th shuttle flight — four percent. Now, that's kind of coincidence, since we all know statistics in small numbers — it shouldn't have been just the 25th, it just should have been within the first 25. But the media doesn't understand those types of statistics. In any case, the shuttle has not been safer.
Student: [comment about prevention].
Yeah, but that's because we've improved safety of aerospace flight generally since 1947 — that's the general trend. But there's not some big blip because of that program.
Student: I was just saying it got more expensive because after they start blowing up, they're like, oh, we better check every part after every mission, do failure analysis, replace every part all the time.
Right. NASA actually learned what Admiral Rickover had found out earlier.
Kahneman makes the point in Thinking Fast and Slow that all the pundits that are predicting politics and elections — he's a statistician, and he uses the term: a blindfolded monkey with a dart could do better than what these people predict. There's an interesting chapter where he and another guy are kind of at odds — the other guy studied firefighters and basically said those guys are experts, they make intuitive snap judgments that are right. And Kahneman says, well most experts make intuitive judgments and you can do better with a blindfolded monkey with a dartboard. And he proves it. That's why he's a nonconformist, is what I like.
§9. Webster's expanded, ABET, and what engineering really involves [38:18]
Any questions on this? That sort of answers part of the second question — we talked about what is engineering. I'm going to talk some more about what's the difference between science and engineering. Von Kármán sort of gave us that to a certain extent, but I'm hoping you might refine some of these things. If one of you wanted to do your presentation on answering some of these questions or how it's evolved, you could do that, but you probably have more interesting things to do.
How old is engineering as a profession? Anyone have an idea? It's actually both very old and very new, and I'll go through that when we go through the history of engineering. What types of disciplines — we're going to go through that. What's a PE license. PE license is something totally ignored at MIT, and almost essential for civil engineers practicing. So this is just what Webster's gave us — I got it out of some dictionaries, not Webster's, but it says the same thing. The dictionaries all plagiarize each other, did you ever know that? Or maybe it's because the words have similar meanings.
Contriver, laying out usually with more or less subtle skill or craft — which means it's sort of an art, and some people might say it's not a science. To guide the course of, manage or supervise during development — management is an important part of engineering, and that's going to be one of my themes. Engineer, from an old French word, engineer or war machine — all these words derive from the Latin meaning genus, genius, divine spirit presiding at birth, or a talent or natural gift. So we now have an etymology of engineer: a war machine or a divine spirit.
Engineer, noun, from French, to contrive — the builder of military engines, a military engineer, contriver, someone who designs, inventor, contrives, could be called a schemer. That's not exactly a polite term for engineer, but, someone engineering something. The Mafia engineers the internet or something. A person who runs complex machinery — a railroad engineer. A person engaged in occupation requiring special skill — like a sanitary engineer, fix up garbage. A person who carries through an enterprise or brings about a result, especially by skillful or artful contrivance. That's a quote from a number of dictionary definitions.
Something I've said before in this course: engineering is design. Anybody know what ABET is? Accreditation Board of Engineering and Technology. If you get an engineering degree from MIT you will all be ABET-accredited. MIT participates in ABET, and ABET has a design requirement. When they come to evaluate us once every four or five years, we have to prove in each department that we have a certain number of units that involve a design component, because ABET feels design is essential to engineering. 3.042 is part of that. Actually this course could qualify, but it's not required.
We're going to talk about design later, but engineering does design. We talked about creating, making something new or different. There's a creative aspect of engineering. Remember the stinky quote at the beginning — if it works but no one knows why, it's engineering. You create it, but you don't — like Edison didn't know why horsehair worked so well. He later found out tungsten was better, and then General Electric spent billions of dollars over the next hundred years perfecting tungsten light bulbs, only to have LEDs come along.
Engineering is problem solving. One of the two of you said problem solving in your definition, and that's something I kind of came up with on my own when I was thinking about this 10 years ago. Engineering is the application of science and math to solve problems. I got that off the internet — by googling "what is engineering," I can go through a bunch of sites, and one of them says engineering is the application of science and math to solve problems. And it is.
§10. Complexity, ambiguity, uncertainty, safety — and everything else [43:09]
Now let me tell you what the School of Engineering came up with. I can't tell you what they came up with because they've lost it. But engineering involves complexity, ambiguity, and uncertainty. That's why it's not what a scientist likes. A scientist doesn't mind something's complex as long as he can write down a mathematical formula for it, to describe it. That's what chaos theory is — you can write down the formula for chaos. But ambiguity — your boss doesn't give you the right amount of information, no more, no less. Sometimes you don't have all the information, so that's ambiguous; sometimes you have conflicting information, so that's ambiguous. And uncertainty — if you have conflicting information, you don't know what's right.
But it also should involve safety. I thought of this earlier this week, up with this last weekend — what's Labor Day for? Being an engineer involves complexity, ambiguity, uncertainty — that's the School of Engineering. Safety is critical. If you read the engineering code of ethics it will say that safety is paramount. As a professional engineer — and we'll talk about what that means later — you have a duty to hold the safety of the public paramount above your job, above anything else. So I put safety in there. Actually I came up with CAU, and then I came up with safety, so I came up with CAUS, and I figured I could make it "because."
I've always said, if your boss gives you a job to solve, and the solution is to the accounting problem — I don't care if you're a chemical engineer or materials engineer, you've got to solve the accounting problem. Accounting is not always the problem, but I can tell you what ninety percent of the problem is in most cases. Anybody know what the number one problem is? Yourself, exactly. People. Managing people is the biggest problem in getting anything done. Anyone who's actually been doing something will tell you that — trying to get other people to work together as a team is the hardest job.
So, what is engineering? This is my summary: complexity, ambiguity, uncertainty, safety, and everything else. Engineering is everything that's good. I like that. I'll probably be here tomorrow — Dr. Belmar will be lecturing, but I don't know who it'll be, me or him.