WEBVTT 00:00:09.757 --> 00:00:15.364 In problem solving as in street-fighting: Rules are for fools! 00:00:15.364 --> 00:00:18.798 (Laughter) 00:00:18.798 --> 00:00:24.278 (Applause) 00:00:25.602 --> 00:00:29.397 Let's see how far we can go by bending rules 00:00:29.397 --> 00:00:32.634 as we estimate the fuel efficiency, 00:00:32.634 --> 00:00:36.551 the miles per gallon of a 747. 00:00:36.551 --> 00:00:40.255 The fuel is used to fight drag, 00:00:40.255 --> 00:00:41.899 the force of air resistance, 00:00:41.899 --> 00:00:43.260 what you would feel 00:00:43.260 --> 00:00:45.841 if you stuck your hand out of a moving car -- 00:00:45.841 --> 00:00:47.814 don't try this at home -- 00:00:47.814 --> 00:00:51.191 or try to run in a swimming pool. 00:00:52.141 --> 00:00:55.271 There are at least two ways that you can use 00:00:55.271 --> 00:00:57.981 to figure out the drag. 00:00:57.981 --> 00:01:01.531 You could spend 10 years learning physics 00:01:01.531 --> 00:01:04.565 and you write down the Navier–Stokes equations: 00:01:04.565 --> 00:01:08.257 the differential equations of fluid dynamics. 00:01:08.257 --> 00:01:12.284 And then you spend another 10 years learning mathematics 00:01:12.284 --> 00:01:14.749 to solve for the pressure. 00:01:14.749 --> 00:01:16.734 And whereupon you find 00:01:16.734 --> 00:01:18.710 that actually there's no exact solution 00:01:18.710 --> 00:01:21.176 for the flow around a 747, 00:01:21.176 --> 00:01:23.486 or, in fact, for most of the situations 00:01:23.486 --> 00:01:25.546 which you want to know. 00:01:26.436 --> 00:01:29.068 Rigor, the rigorous approach, 00:01:29.068 --> 00:01:33.016 the exact approach has produced paralysis, 00:01:33.016 --> 00:01:35.064 rigor mortis. 00:01:35.064 --> 00:01:37.582 (Laughter) 00:01:37.582 --> 00:01:40.269 We need a different way. 00:01:40.269 --> 00:01:42.696 The street-fighting way, 00:01:43.856 --> 00:01:46.673 which starts with a home experiment. 00:01:46.673 --> 00:01:48.977 Chair please. 00:01:54.075 --> 00:01:55.801 Props please. 00:01:55.801 --> 00:01:58.289 (Laughter) 00:02:03.859 --> 00:02:07.469 Small cone, big cone. Coffee filters. 00:02:07.469 --> 00:02:09.585 They're the same shape, 00:02:09.585 --> 00:02:13.221 but this one has one-fourth the area. 00:02:13.221 --> 00:02:16.247 This one has four times the area, twice the diameter, 00:02:16.247 --> 00:02:18.989 but otherwise the same shape. 00:02:18.989 --> 00:02:20.995 When I drop them, 00:02:20.995 --> 00:02:24.791 how fast do they fall relative to one another? 00:02:24.791 --> 00:02:31.579 Is the big one roughly twice as fast? 00:02:31.579 --> 00:02:34.784 Are they comparable in speed? 00:02:34.784 --> 00:02:38.094 Or is the small one roughly twice as fast? 00:02:38.094 --> 00:02:42.863 Take ten seconds and think. 00:02:42.863 --> 00:02:45.553 What do you believe? What does your gut tell you? 00:02:45.553 --> 00:02:48.280 And then we'll take a vote. 00:02:51.291 --> 00:02:53.325 Check with your neighbor. 00:02:53.325 --> 00:02:56.086 (Laughter) 00:03:00.870 --> 00:03:05.348 (Crowd murmuring) 00:03:10.616 --> 00:03:13.220 OK, let's take a vote. 00:03:13.220 --> 00:03:15.491 You don't have to agree with your neighbor. 00:03:15.491 --> 00:03:16.850 (Laughter) 00:03:16.850 --> 00:03:19.375 That's the beauty of democracy. 00:03:20.815 --> 00:03:25.674 So, cheer if you believe that the big cone 00:03:25.674 --> 00:03:29.158 will fall roughly twice as fast as the small cone. 00:03:29.158 --> 00:03:30.750 (Faint cheering) 00:03:30.750 --> 00:03:32.790 OK, I hear a few. 00:03:32.790 --> 00:03:35.845 Cheer if you believe that they'll be roughly comparable. 00:03:35.845 --> 00:03:39.239 (Louder cheer) 00:03:39.239 --> 00:03:41.678 And cheer if you believe the small cone 00:03:41.678 --> 00:03:44.097 will be roughly twice as fast. 00:03:44.097 --> 00:03:47.157 (Loudest cheer) 00:03:47.157 --> 00:03:49.020 A lot of cheering for that one. 00:03:49.020 --> 00:03:53.376 OK, well, as Feynman said and believed, 00:03:53.376 --> 00:03:56.755 in science we have a supreme court: experiment. 00:03:56.755 --> 00:03:59.280 So, let's do the experiment! 00:04:00.770 --> 00:04:03.565 One, two, three. 00:04:05.495 --> 00:04:10.249 (Cheering) (Applause) 00:04:13.181 --> 00:04:15.501 They're almost the same. 00:04:16.561 --> 00:04:19.447 Within experimental error. 00:04:20.287 --> 00:04:21.724 So what does that mean? 00:04:21.724 --> 00:04:24.463 What can we use that experiment to tell us? 00:04:24.463 --> 00:04:26.322 Well, 00:04:27.992 --> 00:04:31.105 the cones fell at the same speed. 00:04:31.105 --> 00:04:33.739 They fall in the same air. It has the same density. 00:04:33.739 --> 00:04:36.156 The same properties. The same viscosity. 00:04:36.156 --> 00:04:38.818 The only things different between the two cones 00:04:38.818 --> 00:04:40.770 is this one has four times the area, 00:04:40.770 --> 00:04:43.333 the cross sectional area of this one, 00:04:43.333 --> 00:04:46.057 and their drag force is different. 00:04:46.057 --> 00:04:47.381 How different? 00:04:47.381 --> 00:04:50.487 Well, the drag force is equal to the weight. 00:04:50.487 --> 00:04:54.018 Because they were falling at a steady speed with no acceleration. 00:04:54.018 --> 00:04:55.916 So the drag and the weight cancel. 00:04:55.916 --> 00:04:57.393 So we have a very sensitive measure 00:04:57.393 --> 00:05:00.287 of the drag force without any force sensors. 00:05:00.287 --> 00:05:02.768 All we do is measure the weight. 00:05:02.768 --> 00:05:05.368 So this one has four times as much paper as this one. 00:05:05.368 --> 00:05:08.681 So it's four times heavier, four times the drag. 00:05:08.681 --> 00:05:11.433 Only change, four times the area. 00:05:11.433 --> 00:05:14.918 The conclusion: drag is proportional to area. 00:05:14.918 --> 00:05:17.439 Not square root of area, not the square of the area. 00:05:17.439 --> 00:05:19.010 but just the area. 00:05:19.010 --> 00:05:21.293 That's the result of our home experiment 00:05:21.293 --> 00:05:25.410 without the rigorous rigor mortis method. 00:05:25.410 --> 00:05:26.800 How can we use that? 00:05:26.800 --> 00:05:30.191 Well, that one constraint, 00:05:30.191 --> 00:05:33.392 along with the next street-fighting tool 00:05:33.392 --> 00:05:37.405 of dimensional analysis, solves the drag force. 00:05:37.405 --> 00:05:39.125 We match their dimensions. 00:05:39.125 --> 00:05:43.961 We match the dimensions of force, drag force on one side 00:05:43.961 --> 00:05:46.091 with what we have on the other, 00:05:46.091 --> 00:05:49.472 which is area, density, speed and viscosity. 00:05:49.472 --> 00:05:52.606 But we already know how to put in the area, just one of them. 00:05:52.606 --> 00:05:55.963 That gives us length squared, meters squared. 00:05:55.963 --> 00:05:58.899 Now we look and we say, "Oh, there's kilograms over here, 00:05:58.899 --> 00:06:00.607 we have to get a kilogram over here." 00:06:00.607 --> 00:06:02.677 The only place to get it from is density. 00:06:02.677 --> 00:06:07.301 Speed and viscosity, the kinematic viscosity, have no mass in them. 00:06:07.301 --> 00:06:10.778 So we put in one density. 00:06:10.778 --> 00:06:14.792 Now what we need still is meter squared / second squared, 00:06:14.792 --> 00:06:16.165 out of speed and viscosity. 00:06:16.165 --> 00:06:19.460 The only way to make it is speed squared. 00:06:20.790 --> 00:06:23.226 So there is our drag force. 00:06:23.226 --> 00:06:25.067 One experiment for a constraint. 00:06:25.067 --> 00:06:27.606 Dimensional analysis for the rest of the constraints. 00:06:27.606 --> 00:06:31.988 Drag Force = Area x Density x Speed squared. 00:06:31.988 --> 00:06:34.196 How can we use this? 00:06:34.196 --> 00:06:37.413 Well, the fuel consumption is proportional to the drag force. 00:06:37.413 --> 00:06:41.617 So, let's compare the fuel consumption of a plane with a car. 00:06:41.617 --> 00:06:45.490 Rather than calculating the plane from scratch, compare it to a car. 00:06:45.490 --> 00:06:48.901 Another street-fighting technique. 00:06:48.901 --> 00:06:52.149 So there're three factors in the comparison, in the ratio: 00:06:52.149 --> 00:06:57.670 the area, the air density and the speed squared. 00:06:57.670 --> 00:06:59.562 Do them one at a time. 00:06:59.562 --> 00:07:03.330 So, the area. Well, in the old days of plane travel, 00:07:03.330 --> 00:07:05.388 you could lie down on three seats 00:07:05.388 --> 00:07:07.106 and there were three sets of those seats. 00:07:07.106 --> 00:07:10.194 So three people wide. Plane is about three people high. 00:07:10.194 --> 00:07:12.720 So it's nine square people. 00:07:12.720 --> 00:07:17.026 A car: Well, from nocturnal activities in cars 00:07:17.026 --> 00:07:20.342 you know you can sort of lie down in cars a bit uncomfortably. 00:07:20.342 --> 00:07:21.520 (Laughter) 00:07:21.520 --> 00:07:24.062 And you can stand up. So it's one square person. 00:07:24.062 --> 00:07:26.022 So it's roughly a ratio of ten, 00:07:26.022 --> 00:07:27.355 maybe nine or ten. 00:07:27.355 --> 00:07:30.732 So the plane is 10 times less fuel efficient for that. 00:07:30.732 --> 00:07:32.175 What about air density? 00:07:32.175 --> 00:07:34.394 Well, the planes fly high, about Mt. Everest. 00:07:34.394 --> 00:07:36.004 So the density is about one third. 00:07:36.004 --> 00:07:38.154 So that helps the plane. 00:07:38.154 --> 00:07:40.226 But they fly about ten times faster, 00:07:40.226 --> 00:07:42.518 600 miles an hour versus 60. 00:07:42.518 --> 00:07:48.122 That means planes pay a factor of a hundred, 10 squared. 00:07:48.122 --> 00:07:51.270 The result is planes are 300 times 00:07:51.270 --> 00:07:53.728 less fuel efficient than cars. 00:07:53.728 --> 00:07:57.567 Oh, no. By flying here, did I damage the environment 00:07:57.567 --> 00:08:00.123 300 times compared to driving? (Gasp) 00:08:00.123 --> 00:08:02.111 What saves it? 00:08:02.111 --> 00:08:04.789 300 people on my plane! 00:08:04.789 --> 00:08:07.185 So the conclusion is planes and cars 00:08:07.185 --> 00:08:09.331 are roughly equally fuel efficient. 00:08:09.331 --> 00:08:11.047 (Laughter) 00:08:11.047 --> 00:08:12.506 All from that. 00:08:12.506 --> 00:08:17.054 (Applause) 00:08:20.864 --> 00:08:24.430 So let's say the plane is 30 miles per gallon. 00:08:24.430 --> 00:08:28.524 Crossing the country back and forth 6,000 miles, 00:08:28.524 --> 00:08:30.900 30 miles per gallon, 2 dollars a gallon. 00:08:30.900 --> 00:08:33.246 That's 400 dollars of gasoline. 00:08:33.246 --> 00:08:36.374 That's not that different than the price of my plane ticket, 00:08:36.374 --> 00:08:40.660 which may explain why airline companies teeter on bankruptcy 00:08:40.660 --> 00:08:42.577 and why they charge us for peanuts. 00:08:42.577 --> 00:08:45.226 (Laughter) 00:08:45.226 --> 00:08:52.436 So connection between the 747 and the cones. 00:08:52.436 --> 00:08:55.377 They increase our enjoyment of the world 00:08:55.377 --> 00:08:58.268 and expand our perception. 00:08:58.778 --> 00:09:01.235 And that, making connections here 00:09:01.235 --> 00:09:03.152 was enabled by street-fighting reasoning, 00:09:03.152 --> 00:09:06.869 by getting away from rigor mortis. 00:09:06.869 --> 00:09:08.746 Making connections is so important 00:09:08.746 --> 00:09:11.798 because it builds ideas and isolated facts 00:09:11.798 --> 00:09:14.100 into a coherent story. 00:09:14.100 --> 00:09:16.071 Imagine each dot is an idea 00:09:16.071 --> 00:09:19.052 and the lines are the connections between them. 00:09:19.052 --> 00:09:21.229 As I increase the fraction of connections 00:09:21.229 --> 00:09:24.566 from 40% to 50%, to 60%, 00:09:24.566 --> 00:09:27.429 the big story, the red connection network, 00:09:27.429 --> 00:09:29.192 grows to fill the whole space. 00:09:29.192 --> 00:09:31.087 That's the long lasting learning. 00:09:31.087 --> 00:09:33.144 That's what we want to build in our thinking 00:09:33.144 --> 00:09:35.081 and in our teaching. 00:09:35.981 --> 00:09:38.430 The goal of teaching should be 00:09:38.430 --> 00:09:42.262 to implant a way of thinking that enables a student 00:09:42.262 --> 00:09:45.934 to learn in one year what the teacher learned in two years. 00:09:45.934 --> 00:09:48.317 Only in that way can we continue to advance 00:09:48.317 --> 00:09:51.310 from one generation to the next. 00:09:51.310 --> 00:09:55.035 In fifty years, all education 00:09:55.035 --> 00:09:58.249 will, I believe and dream, be based on this principle. 00:09:58.259 --> 00:10:02.283 Richard Feynman, I think, would have agreed. 00:10:02.283 --> 00:10:03.661 Thank you. 00:10:03.661 --> 00:10:08.516 (Applause)