Hamming, Intro to The Art of Doing Science and Engineering: Learning to Learn (March 28, 1995)

Hamming, Intro to The Art of Doing Science and Engineering: Learning to Learn (March 28, 1995) - https://www.youtube.com/watch?v=AD4b-52jtos

Well, welcome to EC4000 and I believe it hits on Tuesday, Thursday and Friday. Is it 3 o'clock? 3 o'clock on Friday. Right, 3. Right, 3 o'clock. There's been some confusion locally. I'm in Spanigo 518-2655. I'm typically in in the mornings because I'm only teaching half time, so I go home and work in the afternoons. There are some notes here and for those not here who don't come. I can't give you out sets of notes by mail because the book is being published by a publisher and he has a right to profit by it. It will be out within a year. But meanwhile, students in this class can get a set of notes and if there aren't enough at the end of the day, I'll bring some more tomorrow. The first lecture is on orientation. What am I trying to do? The purpose of this course is to prepare you for your technical. There really is in this course no technical content. Although I'm going to talk about digital filters and all kinds of things. They are things you presumably know. I am concerned about style. I have studied great scientists ever since I was at Los Alamos during the war. What is the difference between those who do and those who do not do significant things? Mainly it's a matter of style. Many a person I've known worked just as hard as others but didn't have much to show for it. So my problem is to instill in you something called style, so you'll amount to something. After all, the Navy is paying a large sum of money to have you here and it wants this money back by your later performance. Now I will examine, criticize and talk about various people's style, mainly my own, but other people's, where I can use it. Now, there are many things I'm going to tell you. I wish somebody had told me. I had to find out for myself. This course is not a normal technical course. It's all about the topics they never told you in class that they should have because each course is taught this way and a large amount falls in between. That's what I'm trying to pick up. Now. Style cannot be put into words. I can only approach it by particular examples and let you infer what it is. Now, there is a belief that you probably have that anything can be talked about. This goes back to Socrates, Plato, Aristotle and the early Greek times. They thought they could talk about the gods. Truth, beauty, justice, love, all those things. At the time they were saying these things, there were the mystery, cults, in Greece who said, you must experience, you cannot talk. And if you remember the Middle Ages, various saints said, you can't talk about God, you've got to experience him in the same way the Mohammedans about Allah, you can't portray him, you can't put pictures. You must sense. So there has long been a school which says you cannot put everything into words, and one of them is style. I really cannot say what I mean. I can only give you these examples and struggle hoping that you will get the idea. Now, to be effective in a course like this, I have found that I have to talk about myself. If I make abstract remarks, it just sounds like so many pious words. If I talk about me and what I've done, maybe it will penetrate you. Now it gives, of course, an attitude of bragging. I'm always talking about myself. But I will tell you several mistakes that I made. Lulu's so you won't do the same sort of a thing. Similarly, I have to get you to quit your modesty. I have to get you individually to respond to my challenge that you're going to be great. You have to say to yourself, yes, if that guy Hamming can go out and become a great scientist, I can, or I can become a great person. I have to get you to say to yourself that you want to, that it's worth the effort and you're going to try to be something more than just the average person. Now, while we speak of teachers, we are really coaches. I cannot run a mile, the four minute mile for you. I can comment upon your style, but you know you must do the work the same way. I cannot make you a great scientist. I can criticize style and other things, but I cannot by mere words make you a great scientist. You, just as in running a four minute mile, must do the work. Which means you have to take what you hear and read, think it over carefully, discuss it with your friends and see what you can adapt to yourself. There is no one style which is successful. Painters paint many different styles. You have to find a style that fits you, which means you have to take what fragments you can from other people, use them and adapt them, and become yours. You can't copy me directly. You won't get away with it. And I will use the analogy of painting as an example. In painting, once you've learned color mixing and form and sketching and so on, you study under a master who you temporarily accept as knowing what he's talking about. Well, there are limits to what can be done. You know that if you Copy the master style exactly. You will not be a great painter. You know also that if you paint in the style he did or she did, it's too late. The future wants a different style. Thus I can tell you about the style I used in the past. But that won't be the style you'll have to have. To cope with the future, you must manufacture the style which will make you a significant person in the future. So it's not easy. While I only talk about past ones and make references to possible future ones, it's the problem you face. What I did would not make me successful if I were starting now. Just as my predecessors got successful on other things that I couldn't do and get successful on now. It's another part that's very difficult for you. When I went to Bell Tel Avo Laboratories in 1946, I looked around since I was already interested in what made great scientists. And I looked at what they did, and when I looked at what they did to become famous, it didn't look that difficult. They tend to do the easy problems I found in the course of my time there. A couple of holes they left, but fundamentally they did the easy problems. My generation did the somewhat harder ones and we left to the others the harder still. Every generation has more difficulty, but you stand on our shoulders to some extent. Yet the task is harder. Having gotten man to the moon, the next real good feat in space is going to be a lot harder. Therefore you have difficulty. It's very definite now. When I came to Bell Labs, there were four of us at the same time. About. We came in about the same time and we were about the same age. Within a year we privately called ourselves the four young Turks. And many, many years later, I discovered top management called us the same. We were troublemakers. We didn't do things the way the previous generation did. We did new things. The previous generation didn't like it. We didn't do things right. For example, my boss, Hendrik Boda in network theory had made a reputation doing network theory with complex variable and knew that's how you do things. After all, that is what made him famous. This guy Hammond comes along and keeps using computing machines, which is not the way to do it in his eyes, but it was the thing that needed to be done. This is a lesson which I want to get across to you regularly. Supposing I am successful and you do rise to the top. Will you please remember that what made you great is not appropriate for the next generation? You know how to get great, because after all you were great, but the things that you did may not be appropriate for the next generation. All too often we have the troubled bosses. They know by God, this is the way I did it and I got to the top and it must be right. They're very often wrong. And I want you to think seriously when you rise to the top, that your method of success is not appropriate. Now that the world has changed, I want to talk to education. Education is what, when and why to do things. Training is how to do it. Most of your courses have been training. I'm trying to talk about the education part. It's not easy, but the school has allowed me a great deal of latitude in putting this course together, which is concentrating on education. Now, if you have one without the other, it's not much good. I've had very able technical people reporting to me who applied their technology and methods to the wrong problem and it had to be undone. And I have other people who had all kinds of theory but couldn't do anything. They're not much use either. You need both theory to guide you and skill and technique. To do one without the other isn't too good. Now, in a certain sense, I'm engaged in meta education. I'm talking about education constantly because that's what you're going to have to do. You're going to have to educate yourself constantly. That's what the future says. Now I'm going to constantly try and project forward what the world's going to be like. Let's look back first to history. The modern era in science and engineering began with Sir Isaac Newton, roughly 1642. He was born Christmas Day, the same year that Gallio died, and he lived to be about 85. So we can say it's around 1700. From Newton's time to ours, we have about double the knowledge every 17 years. The doubling period of science from then to now is roughly 17. When I came to Bell Laboratories in 46, they were trying to shrink down from war size down to 5,500 people. I watched through 30 years of management putting hiring freeze and doing everything else like that. Double every 17 years with small wiggles. They had to hire the people to keep up with the expanding knowledge, publication of books, journals and so on. For example, I think I have the numbers here. No, I guess I don't. Now I'm going to make a digression. Oh. Another thing about the situation is that 90% of the scientists who ever lived are now alive. It's a common statement. I'm going to now Turn to a back of the envelope calculation which I learned by watching Fermi and other people and shock the other people. I used to lunch with them. I'm going to suppose first we have an exponential growth of the number of scientists that comes from a differential equation. The rate of change is proportional to how much you have. And the solution is, as you know, the exponential growth. Now if I assume that the amount of knowledge being generated is proportional to the number of scientists, this is the amount of rate. And in the up to 17 years ago, this is how much we generated. This is the amount up to now. Now I put minus infinity on because it doesn't matter what lower limit I put. It's so small it doesn't matter. The exponential is very, very small out there. So who cares? Well, I simply work it out. I do the integration, I come up with that. And the statement was half the stuff has been done. The doubling every 17 years from 17 years ago, now we've doubled. That says the ratio of half. I've got a formula for B. Now take the other statement. 90% of the scientists who ever lived are now alive from now back 55 years. That's what I'm going to take for lifetime of a scientist. You probably don't mean a living scientist when he's 2 years old. You probably mean a scientist alive when he's become or beginning to be a scientist. And until he decay somewhere in the 80s, you consider them a sinus. So 55 years is a reasonable number. If I put that in over the whole of all sizes who ever lived. I come up with this using that. Whoops, this is B. I will come up using a substitution here. I come up with 0.89 which is close enough to 90%. Now let's see what happened. I got a clearer idea of what I was talking about and I had to answer the question which I hadn't thought about. What did I mean by a lifetime of a scientist? But you see, those two statements are compatible. We double every 17 years and 90% of the scientists who ever lived are now alive. You have seen enormous growth of science from Newton's time to now. Well, let me project. Well, let me say now a good estimate of the number of various branches of science which we have developed. In Newton's time we had only one thing called natural philosophy. Now we have lots of specialties. There are something like 10,000 specialties. There certainly is more than 1,000 and almost certainly is less than 100,000. So 10,000 is a good number. Now if I project forward doubling every 17 years for 340 years. That's a million fold to the 20th. That would make 10 billion fields of specialty. Well, you don't believe it? You don't believe in 340 years there'll be 10 billion fields of specialty? Consequently, science cannot go the way it has been for the next 320, 40 years. The doubling and the growth cannot go on. One of the things we have done is we've got an exponential number of people in the field. We can't go on that either. Well, everyone would have to be a scientist. So you know that the past is not too good a guide to the future. Now, the reason why I want to put those back the envelope book in is it's widely used. I observed that Fermi and Shockley and those, I used to eat lunch with them, they did back the envelope book. And you saw what I had to do. Not only that, but it also does two things. It puts the thing firmer in your mind. Having shown you the calculation, you may retain it a little longer. Plus it gives you practice in quick modeling. Nobody pretends this is really accurate. I don't pretend 17 is the exact number. It's somewheres around there. But back of the envelope calculations are very useful. I have found it very, very useful when I hear things over TV or something else, radio, read newspapers and so on, do a quick modeling and ask myself, are these numbers possible? And very frequently two things emerge. Either they're not possible, or B, you didn't even know what they were talking about to make a model. You found that they failed to tell you what they were talking about, just gave you a spectacular answer. So doing back the envelope modeling is a very, very big help. Now, this doubling business is a very serious one. I've had to live through my life with that fact. So I put over here at a table, double the 17 years, triple that. 4, 5, 6, 7, 8, 9, 10 times about 56 years, something like that. Tell you how you read that One way is ask the time from now to retirement. Look at this column. That's how much knowledge will be that much times what you now have. If we go on the same way, you face a rather horrendous future. Another way to look at it is this. Suppose you were 34 when your child was born. Now your child goes to college, there's four times as much knowledge, not just mathematical theorems. Recordings of Beethoven's 9th, where to go skiing, what channels to read, listened to on tv. There's going to be four times as Much knowledge for your poor child to face. Now, you remember when you hit college how much there seemed to be. Don't be surprised if your children are somewhat more disoriented than you were. And God knows you were sometimes disoriented. This is what that means. Furthermore, the doubling, all the doubling occurs worst in the last period. Almost half to half the episodes occur in the last doubling period. And that's what causes saturation. Saturation comes down quite rapidly. So another way of looking at doubling is simply this table here, which is disconcerting if you think you'll be chief of staff in, say, 44 years from now, say 39 years, there'll be five times as much knowledge needed to run the Navy as is needed now. That is what you face. Well, what's my answer? My answer to that is learning to learn was the only thing I could do. Things become obsolete. Something like half of what we have taught you lovingly in other courses will be obsolete in 15 years. Either we're no longer doing it, or it's been replaced by something else. Consider what I had to live through. I came to Bell Laboratories in 46, and they were running vacuum tubes and so on. It was a very important part. So I started having a mathematical background, studying electrical engineering and what vacuum tubes were and so on. But in some years, I began eating with the physics department and I ate with the guys while they were perfecting, not when they started, but when they were developing the engineering side of transistors. I did a great deal of calculation for them on transistors. I obsolete. All the knowledge I knew. I haven't seen a vacuum tube for a long while, except in my friend's office where he keeps going around to show students what a vacuum tube is. You don't see them very often. Now you can say, well, the original transistors were little tin cans and three legs. Yeah. Now there's a million on a chip that size. I've had to endure that. At Los Alamos, we calculate atomic bomb designs on relay calculators, which probably averaged maybe an operation or a second or maybe a second and a half around the clock, six and a half days a week for a month, sometimes three months, but typically about a month to get one solution. Now you can punch in a modern machine, go boop, boop. And there's the answer. I've had to live through a tremendous change. Furthermore, I was educated as a mathematician. I certainly had no course in numerical analysis. I never knew about a computer. I knew a little physics, but Los Alamos Taught me some more. But fundamentally, when I went to Bell Labs, because I believed that the computing I did, I should understand the nature of the problem. I had to learn something of the breadth of physical sciences, some chemistry as well as a lot of physics, some social science and a little bit of biological science because the laboratories had such departments and some social science. I spent a lifetime getting background knowledge on something. You have to have background knowledge enough to penetrate jargon, which I'll talk about extensively later date. Now, one thing you could do is to try and cling to fundamentals, which is very glib until you ask what do I mean by fundamentals? Well, I have two criteria which are not adequate. One is from the fundamentals you can derive the rest of the field. Secondly, they've been around for some time, but the fundamentals of amplification, which were vacuum tubes, doesn't count now. True. Hartley. The formula for gain. I have trouble with names frequently. I'll come to it pretty soon. Nyquist. Nyquist formulas are still good. The gain formulas used out of vacuum tubes are still useful, although we have to apply it to other things. Feedback is still the same, but a lot of things are not. Now I need to discuss science versus engineering science. If you are doing it, you shouldn't know what you're doing. If you know what you're doing, you shouldn't be doing it. Not in science, because science is supposed to be the exploration of what you don't know. Engineering, you shouldn't be doing it unless you do know what you're doing. Well, nothing is pure. Science involves a great deal of engineering and engineering involves a great deal of new material. So it's a great blend. But what is painful to you, and it's going to be worse, is that the two fields are growing together because of a simple fact. Again, going back to when I first came to Bell Labs, when something was discovered in physics. The telephone company was not in that great a hurry to get it developed into the field. After all, it had pretty much monopoly. Why hurry now? As you know, we are not willing to wait for scientific principles to develop. We want it in the field tomorrow. So the two fields are coming together like that. And the leisure which we used long ago and which we are still using to some extent, develop the ideas first and then apply it is going to be less and less acceptable. When an idea is first around, you want to apply it. I just read last night that one of the presidents who were was at a museum. One of these world's fairs was shot and the bullet Was in his back. Right in his back bullet. The doctors refused to operate because they didn't know where the bullet was. But there at the Muslim were X rays being demonstrated. They didn't use the new technique was readily available. They could have wheeled them in there, got in the picture. No, they were conservative. We don't allow that much anymore. We're pushing very hard and you're going to be pushed very much to go from idea to developed item and get it on the market rapidly. Now, I once read there were some 76 different methods of predicting the future, which is what I'm engaged in doing to some extent. One is to predict tomorrow will be like today. Whatever temperature it is today, predict tomorrow is the same. It's a pretty good prediction. A somewhat better one is to note the linear trend and predict a linear trend. And that's good for a while, but not too long. And furthermore, it depends on which variable you pick to be linear. If you pick the coefficient in front to be linear, it's one thing, if you pick the exponent, it's something else. It doesn't work too well. I made many predictions on how much computing I'll do pretty soon because I need to know how much computer capacity we would need and so on. I was regularly wrong on the low side. So one time I got miffed and said, I will predict high. So I got out some formulas and predicted real high. A couple years later, the piece of paper turned up my desk. I looked at it, I was low again. The growth of computing has been unbelievable. On the other hand, on the other side, take artificial intelligence. The predictions made by almost all the experts 10, 20, 30 years ago have not been realized. So you can't always go on things. Nevertheless, there's a saying. Short term predictions are optimistic. Long term predictions are pessimistic. And the reason is very simple. The long term are pessimistic because nobody can believe a geometric regression. I say again, when we got transistors going, nobody in his right mind would have predicted a million transistors on a chip that big. Nobody. It's beyond belief, but that's what we did and you know it. So predicting the future is a very, very hard business, but you have to do it. History is important. Now, some people believe that history repeats itself, and some people believe exactly the opposite. But one thing you can be sure of, what we now regard as the past was to some people the future. And what you think is the future will be the past. There will be a time when some of you will be in the history books? Yes. You live long enough and do enough and you end up in the history books. So what you think is the future will become the past. Now another thing against history is Henry Ford Sr. S remark, History is bunk. And I think he said it for two reasons. One is history is rarely reported correctly. There are a great many descriptions of what happened at Los Alamos during the war. No two of them agree and they don't agree with what I think happened. Indeed, one time a mathematician, Ulam, wrote his experience about the matter and published it. And I came into Los Alamos on my regular summer visit and said to my friend, I just read Ulam's book. That isn't how I remembered it. He said, that isn't how I remembered it either. I was just going to say, how did you remember it? And I suddenly realized no two people remember the same. Now you're familiar with this, an accident. Several witnesses see it, they report different things. There is no reliable report of what happened in the past. It's what has come down to us and accept it. Secondly, I think in Ford's mind was the fact that the past is being more rapidly disconnected from the future. The invention of a computer tells you how much the world is different than what it was before computers appeared. It's a change in the way we do things. Engineering now is to a great extent getting a computer to do job, writing a program and putting some terminal equipment around to affect the real world. But the heart of much of engineering now is a computer. Now some historians, when you read them, they will give you the impression that it was inevitable this was going to happen. It was inevitable that Rome would fall or this or that. And on the other hand they will tell you the future is very open ended. Many things are possible. Can this be true that the past was very determined, the future is very open? It seems unlikely. So you're left with saying maybe the past was not so determinate. For example, consider the individual lives of Alexander the Great, Napoleon and Hitler. If they had died in their childhood, would not the world be very different? On the intellectual side, Pythagoras, Aristotle, Newton, Maxwell, Einstein are examples of people who had they died in their youth, the world would be rather different. So individuals do matter. I suggest that the past was less determinant than historians like to make and the future is less open ended than you would like to believe. But there's a great many possibilities for you. The future has got great possibilities. Now one other thing against history is unforeseen. Technological inventions can ruin anything. Like I told you, the transistors, the development of vacuum tubes was practically cut off. A technological invention can change completely the history of something, and one can hardly foresee technological inventions. But there are also social inventions which are important. You people have been trained mainly in the physical side, and I've got to make you more sensitive to the fact that all of your life takes place in a social society which has restraints. Thus, I will claim that the future of technology will be less determined by what technology can do than social, legal and other restraints on what we can do. Thus, if you stop to think about highway controlled, computer controlled highway traffic, it sounds good until you ask yourself, who do I sue in an accident? And you begin to decide, you know, it's going to be a very, very difficult thing to get going. Very difficult. Social conventions are going to stop a great many things from happening. Now I want to talk another thing, a story which I'll use several times. The story of the drunken sailor. He staggers a couple steps this way, and he staggers this way, and he staggers this way, and he staggers this way. In N steps, typically, he'll get the square root of n distance. In 100 steps, he'll get about 10. In 10,000 steps, he'll be about 100 times. He may be right where he started, he may be further away, but that's typical. On the other hand, if there's a pretty girl over there, he starts like this, back like this, over like this. He's going to get a distance proportional to N. If I can create in you a vision of where you are headed, you will make a progress proportional to end. If you do not have a vision, you will wander like a drunken sailor and get very little. So one of my major purposes is to get you to form a reasonable vision of what you are going to do in your future, what kind of a person you're going to be now. You're going to say to me, hammock, how do I know the future? I'm going to say, it doesn't matter much from what I've examined in life, what goal you set, whether you want to march that way, that way or that way. If you have a goal, you'll get somewhere near it. And if you don't have a goal, you're a drunken sailor. My problem is to make you form your goals and to some extent try to achieve them, to make you something important rather than just drifting. Now it's comfortable to drift through life, and a great many people, when questioned closely will assert that they're perfectly content to drift through life. I don't think too good an idea of the whole thing. Now, it's none of my business what goal you take. It is my business to force you one way or another to set up some reasonably decent goals to try and achieve something in your life. Again, the society is paying a great deal of money for your education. It's entitled to something. Those who do something generally have somewhat kind of goals and see where they're headed and their lives add up. Those who don't are just a bunch of isolated events. They did this, they did that, they did the other thing, but nothing added upon. So my problem is to get you to choose your goals, even if you want to merely be a great guitar player. I don't mind so long as you set a goal and is struggling. That is the essential part that I'm really after. And that's what this course is about to some extent, forcing you somehow or other to do more than you would have done otherwise. Now, the standard method of teaching is to have departments, departments break things up into subject matter like calculus, linear programming and so on. Too much falls between. And this course is an attempt in one way to plug all those holes in the engineering courses you had. You had a lot of engineering courses, they taught you this, that and the other thing. There are vast holes between them. The optimizing of the components, the individual courses, is not optimizing a total education, as I will come to in system engineering. Now, another goal I have is to show you that in spite of different departments, there's essential unity of all knowledge. When you face a difficult problem of unknown type, it doesn't matter whether it comes from chemistry, physics or anything else, you have to find the answer. And knowledge is pretty homogeneous. Then it's no longer divided up into courses, no longer divided up into departments. Although at Bell Labs I was in the math department almost all the time. In fact, I was doing a great many other things. I was doing statistics, I was doing computing, I was doing physics, I did a lot of other things. Chemistry, we did not observe too tight a division. But for purpose of organization you do have to have some structure. But I want to get in your minds. Knowledge is sort of a homogeneous body which we have specialized with certain names, but it's all reconnected together. Now the course will center around computing. Not, I like to think, because I'm prejudiced my life in computing. But rather, in fact, they are going to dominate science and engineering. And there are reasons for this very Powerful reasons. Economics. For example, computers are far, far cheaper than human beings. Far cheaper. And they're getting cheaper by the year. Humans are getting more expensive by the year. Speed. Far, far faster. Your nervous system. If you drop something on your toe signals up your head about 100 meters per second. Light is 300,000 kilometers per second. You aren't in a league. You can't even touch electronic speeds. There's no way you come near it. So speed is overwhelmingly on the side of the machine. Accuracy, namely number of digits of arithmetic. A carry. Yes, they can be quite precise. They can do double precision if necessary. You would have trouble doing double precision arithmetic. Probably if you tried doing it, you could work it out, but you'd have trouble. Reliability. They're far, far ahead of you. God or nature, however you want to do it didn't make you to be a reliable thing. You've been walking for years and still every now and then you trip and stumble. You can't do anything really reliable. That's why man ended up on top of the heap. He has the flexibility built in. But don't ever try to get humans to do something reliable. Take for example, bowling. All you can do is throw the ball down the alley exactly the same way every time. Have a perfect game. Perfect games are rare, even among the most skilled experts. Precision flying and other things are very hard to do. We recognize that. Being very precise. Drill teams and so on. Or something remarkable. The human animal wasn't really designed to do that. He was designed for something else. Rapidity of control. Because the machine's got rapid control. We are now building airplanes which are basically unstable. And we have a computer. Every millisecond is correcting the instability so we get better performance out of it. But the pilot couldn't do it. If that computer goes out, the pilot's through. The pilot is left with a large scale. The broad planning. But the millisecond. A millisecond is left under a computer because a human just can't act that fast. Another one, which I dwell on very much freedom from boredom. It sounds trivial. You cannot put a human being on a job to look for something for three years and when it happens, respond promptly. You can put a computer on the job. You can put the computer on the job to watch for the rare event. If such and such happens in the atomic pile. Do this. Well, it hasn't happened for four years now. The dial goes over like that. The human being isn't going to do very well. He hasn't been looking at the thing for the last two and a half years. Even you can't get humans to be freed from boredom. Machines don't know what the word is. Bandwidth in and out. In any rapidly changing situation, the person in charge can only get so much information in and out. And there's a general belief that really you can process only about 50 bits per second, maybe 60, something like that. But you can't process 10,000 bits per second. A machine's got enormous more bandwidth. Not only visual, auditory, or put all your inputs together, they won't match a modern machine for bandwidth not only coming in but giving orders out for central control. The human simply cannot in a complicated situation compete with a machine if it is merely bandwidth in and bandwidth out. If it's making judgments, that's another story. But the machines simply cannot cope with it. Thus we no longer have a crew aiming a gun at an airplane. We have a self contained. The human is too slow, he just isn't much good. We need much more rapid things than humans can cope with. The bandwidth in and out, which is really speed of getting information is fundamental. Computers have gotten all over you. Ease of retraining. Training to great extent is you learn to do something and now I change the equipment. You've got to unlearn the old habits and learn some new ones. And you've got to repeat them many, many times to learn them. With the computer I change the program and it's done. No elaborate training, no endless hours of constant practice. Bing. You just put a new program and the machine behaves a new way. Very easy. Hostile environments, outer space, underwater, high radiation fields, warfare, manufacturing situations that are unhealthy and so on. I can put machines in those situations where humans are very, very difficult in space. I got to keep this human being in an atmosphere somewhat. He's used to oxygen, so on has to be employed. High radiation will kill them and so on. How we're going to manage to get people to Mars and back in the radiation field that's coming from the sun. I don't know whether we'll sort of radiate them thoroughly or maybe decide not to send human beings that far. It's a problem now. Personal problems is another one, and it's one I'm much sensitive to. Personal problems dominate management. There are all kinds of troubles with people with machines. There are no pensions, there are no personal squabbles. Two machines don't get squabbling with another. But I've had two girls squabble and wouldn't even share the same room together. Unions no personal leave, no egos, no death of relatives. Your mother died. Machines don't have that recreation. If I turn the machine off, that's the end of it. If I have a human being, I have to provide reasonable recreation. Machines have gotten all over humans. Now. All of you probably already been saying, oh, yeah, but what about the advantages humans have? I won't have to list those. You're trying to do it already, but I gave you a bunch of details which you could find very hard to get around. The machine has got great advantage in many places. And because it's economically sound, you are going to see more and more machines running organizations. Some computer, well, let's say computers. The design of chips is under computer control to a great extent. Some computers are actually being assembled heavily by machines. I was on the board of directors of a computer company for a while, and at one point, more than half the computers coming down the production line, we were grabbing to mechanize the production line. We're mechanizing the building of computers. More than half the computers we sold less than half of them because we're mechanizing the line and getting production much cheaper. That shows you how rapidly a company in computing business was really mechanizing itself. And one of my friends said he ordered a bunch of machines. The message came in overnight. A bunch of machines assembled those particular computers they wanted. And the next day those computers were on the loading dock, designed just what they wanted with the parts they wanted. Now, lastly, this is, in a certain sense a religious course. I am preaching a message that with one life to lead, you ought to do more than just get by. Now there are a great many religions, and I don't want to get involved in one or the other too much. It is, however, an emotional matter I'm really appealing to now. It is very frequently said that a happy life is one who has some goals. They achieve well. Studying the matter over and reading about it and talking to people, everybody pretty much agrees that it's not the achievement of the goal that really is the best part. It's the struggle. The struggle to success is what makes you what you will be. Remember, in your old age, you're going to have to live with yourself. There's no escaping living with yourself. In your old age, you're stuck with yourself. And in old age you can't change much as you can when you're younger. Consider the kind of person you wish to be in your old age and start now being that kind of a person. This is what the course is all about. Really, in one sense. Now, it's an opinion. It's not a fact. It's an opinion that most people believe that the struggle to achieve excellence is worth the struggle. Also, when you look at people's lives, I can tell you a story which I may repeat a couple of times. As a child, I went to a movie. They were called Nickelodeons in my day, but we actually spent a dime to go to the movie. One Saturday, I went with a friend of mine, and it was one of these. You laughed and laughed and laughed. All ridiculous situations. We walked out and he said to me, you know, that wasn't a very funny movie. I thought for a while and said, you're right. All the laughter did not make the movie funny at all. In the same way with life, the pleasant life is not the one. The sum total of the pleasant moments, somehow or others added up very, very differently. The good life is not the life of pleasure from moment to moment, and you know it. In fact, you are well aware that you cannot get up in the morning and say, I shall be happy today and make it work. The good life has to be snuck up upon. And I'm saying an opinion of myself and many other books. The way to do that is to take yourself in hand and manage yourself, to be the person you wish to be, to achieve the goals you wish and be more our ticket than just idle drifting like a drunken sailor. Now, in ancient Greece, our boy Socrates said, the unexamined life is not worth living. So what I'm saying goes back that far. I was crossing the Yale campus one time as a consultant, for the present on a job a whole committee put together. I walked across and I heard a professor walking across the campus right ahead of me, saying to a student, the unexamined life is not worth living. And in the course of crossing one quadrangle, he managed to say it three times. So I'll repeat it the third time. The unexamined life is not worth living. See you Thursday. Right. And there are notes here on the course. And if there aren't enough, I'll bring some more tomorrow. Thursday.