1 00:00:00,095 --> 00:00:02,470 The following content is provided under a Creative 2 00:00:02,470 --> 00:00:03,880 Commons license. 3 00:00:03,880 --> 00:00:06,920 Your support will help MIT OpenCourseWare continue to 4 00:00:06,920 --> 00:00:10,570 offer high-quality educational resources for free. 5 00:00:10,570 --> 00:00:13,470 To make a donation or view additional materials from 6 00:00:13,470 --> 00:00:17,400 hundreds of MIT courses, visit MIT OpenCourseWare at 7 00:00:17,400 --> 00:00:18,650 ocw.mit.edu. 8 00:00:28,750 --> 00:00:32,270 PROFESSOR: As it turns out, not by coincidence, the Apollo 9 00:00:32,270 --> 00:00:35,950 program that started after Kennedy's speech was made up 10 00:00:35,950 --> 00:00:38,270 of a few different elements of things that had been sort of 11 00:00:38,270 --> 00:00:40,590 percolating in the background. 12 00:00:40,590 --> 00:00:44,490 And one of them was studies at MIT on guidance, not so much 13 00:00:44,490 --> 00:00:48,290 to the Moon, but to Mars, and coming out of Draper's 14 00:00:48,290 --> 00:00:48,960 laboratories. 15 00:00:48,960 --> 00:00:53,290 So I wanted to say a little bit about that story, because 16 00:00:53,290 --> 00:00:54,290 it's an interesting story. 17 00:00:54,290 --> 00:00:55,970 And it's one you may not have seen quite in this way. 18 00:00:55,970 --> 00:00:58,530 So what we've talked about before and the inertial 19 00:00:58,530 --> 00:01:00,150 guidance and all that sort of stuff. 20 00:01:00,150 --> 00:01:03,990 This is the sort of very clearly posed paradigmatic 21 00:01:03,990 --> 00:01:05,240 picture of him. 22 00:01:07,900 --> 00:01:10,510 And during the Second World War, they started making these 23 00:01:10,510 --> 00:01:11,010 gun sights. 24 00:01:11,010 --> 00:01:12,860 And one of them was called Doc's Shoebox. 25 00:01:12,860 --> 00:01:15,320 And then I think we talked about that too, which allowed 26 00:01:15,320 --> 00:01:16,930 him to sort of lead the target. 27 00:01:16,930 --> 00:01:20,300 You can actually see, if you go into the front door of 28 00:01:20,300 --> 00:01:23,190 Draper Laboratories, they have this prototype that he built 29 00:01:23,190 --> 00:01:26,740 in his garage, sitting there. 30 00:01:26,740 --> 00:01:28,610 And then during the war, as we talked about-- 31 00:01:28,610 --> 00:01:32,590 sorry, during the Cold War, they built a lot of 32 00:01:32,590 --> 00:01:33,380 different kinds of-- 33 00:01:33,380 --> 00:01:34,740 I'm going to skip through these-- 34 00:01:34,740 --> 00:01:37,570 inertial guidance and stuff, particularly 35 00:01:37,570 --> 00:01:39,210 for ballistic missiles. 36 00:01:39,210 --> 00:01:44,155 And a bunch of these guys, including this guy here, Dick 37 00:01:44,155 --> 00:01:49,190 Batten, really did some fundamental work in the theory 38 00:01:49,190 --> 00:01:51,740 of inertia guidance. 39 00:01:51,740 --> 00:01:55,980 It was pretty much literally how do I get a nuclear missile 40 00:01:55,980 --> 00:01:59,850 to land on a Russian city, given the gravity field of the 41 00:01:59,850 --> 00:02:02,260 Earth is not uniform, and there's a lot of calculations, 42 00:02:02,260 --> 00:02:04,650 and we don't have a lot of different things? 43 00:02:04,650 --> 00:02:06,210 And then this group-- 44 00:02:06,210 --> 00:02:09,430 the guy on the left, his name is Hal Laning and he was a 45 00:02:09,430 --> 00:02:11,630 computer guy from the Whirlwind group. 46 00:02:11,630 --> 00:02:15,490 Milt Trageser is the guy in the back there and Dick Batten 47 00:02:15,490 --> 00:02:16,190 is on the right. 48 00:02:16,190 --> 00:02:19,560 Batten actually still teaches a course in Core 16 on 49 00:02:19,560 --> 00:02:20,940 astronomical guidance. 50 00:02:20,940 --> 00:02:23,000 Actually, he may not have taught it this year. 51 00:02:23,000 --> 00:02:24,450 That's more or less a course he's been teaching 52 00:02:24,450 --> 00:02:29,060 continuously since 1945 or so, 1946. 53 00:02:29,060 --> 00:02:33,360 And if you've ever seen when they send a probe-- actually 54 00:02:33,360 --> 00:02:37,390 they just did this with the Messenger probe to Mercury-- 55 00:02:37,390 --> 00:02:43,040 and then they sort of swing it around another planet as a way 56 00:02:43,040 --> 00:02:45,690 to get it accelerated toward its goal, as opposed to just 57 00:02:45,690 --> 00:02:47,635 sending it directly there, that's actually an idea that 58 00:02:47,635 --> 00:02:49,110 Batten invented. 59 00:02:49,110 --> 00:02:53,010 It is now used commonly in all these different space probes. 60 00:02:53,010 --> 00:02:55,040 And they made this Mars probe. 61 00:02:55,040 --> 00:02:57,740 And the idea behind this probe was to go to Mars, take one 62 00:02:57,740 --> 00:03:00,630 picture, and come back to the Earth and reenter 63 00:03:00,630 --> 00:03:01,890 and drop the film. 64 00:03:01,890 --> 00:03:05,130 And this thing, which looks just like the Mercury capsule, 65 00:03:05,130 --> 00:03:06,740 is the film canister. 66 00:03:09,560 --> 00:03:13,590 And they had also in the '50s, they built the guidance system 67 00:03:13,590 --> 00:03:17,080 for the Polaris missile. 68 00:03:17,080 --> 00:03:19,630 And the important thing about the Polaris missile is it was 69 00:03:19,630 --> 00:03:21,890 the first missile that could be launched out of a 70 00:03:21,890 --> 00:03:24,210 submarine, out of the Polaris submarine. 71 00:03:24,210 --> 00:03:29,640 And was a major US, huge project during the '50s. 72 00:03:29,640 --> 00:03:34,110 And basically they wrote a very short proposal, proposing 73 00:03:34,110 --> 00:03:38,150 to use that guidance system to go to the Moon. 74 00:03:38,150 --> 00:03:41,730 And they were awarded a contract August 4, I believe 75 00:03:41,730 --> 00:03:43,120 it was, 1961. 76 00:03:43,120 --> 00:03:46,100 It was the first contract of the entire Apollo program and 77 00:03:46,100 --> 00:03:47,350 it came to MIT. 78 00:03:49,890 --> 00:03:54,920 Which is some indicator a, of how important they considered 79 00:03:54,920 --> 00:03:55,570 the guidance. 80 00:03:55,570 --> 00:03:58,070 Like how do you get there and how you keep from smashing 81 00:03:58,070 --> 00:03:59,360 into the Moon, and how do you get home and keep from 82 00:03:59,360 --> 00:04:02,010 smashing into the Earth when you get home? 83 00:04:02,010 --> 00:04:07,800 And b, it was the only major contract of the entire program 84 00:04:07,800 --> 00:04:09,290 that was awarded to a university, 85 00:04:09,290 --> 00:04:11,940 and not to a company. 86 00:04:11,940 --> 00:04:14,250 And of course, the president was from Massachusetts. 87 00:04:18,570 --> 00:04:20,790 One of the very senior technical people, Bob Siemens, 88 00:04:20,790 --> 00:04:23,040 was an MIT guy, a student of Draper's. 89 00:04:23,040 --> 00:04:25,560 He was essentially the chief engineer at this point. 90 00:04:25,560 --> 00:04:28,710 And a little bit like the Radiation Lab, there's no 91 00:04:28,710 --> 00:04:33,980 question that those influences helped steer this contract, 92 00:04:33,980 --> 00:04:35,580 which was a lot of money. 93 00:04:35,580 --> 00:04:37,520 Although it was less significant for the money in 94 00:04:37,520 --> 00:04:39,050 some sense, than for the prestige that 95 00:04:39,050 --> 00:04:40,020 went along with it. 96 00:04:40,020 --> 00:04:41,420 And it was sole source. 97 00:04:41,420 --> 00:04:42,370 There was no competition. 98 00:04:42,370 --> 00:04:43,260 There was no proposals. 99 00:04:43,260 --> 00:04:44,840 It just came straight MIT. 100 00:04:44,840 --> 00:04:48,760 And all kinds of companies protested it to Congress. 101 00:04:48,760 --> 00:04:51,120 And it turned out it was fully legal, so they 102 00:04:51,120 --> 00:04:54,920 never changed it. 103 00:04:54,920 --> 00:04:56,760 But it was not for a computer, actually. 104 00:04:56,760 --> 00:04:59,360 It was just for this guidance unit and it was over the 105 00:04:59,360 --> 00:05:00,420 course of the project. 106 00:05:00,420 --> 00:05:09,070 And remember, this is 1961 and all they thought about was how 107 00:05:09,070 --> 00:05:10,620 do we get there and back? 108 00:05:10,620 --> 00:05:15,560 And they begin working on this problem in seven years or so. 109 00:05:15,560 --> 00:05:17,520 And the idea was that the astronaut would 110 00:05:17,520 --> 00:05:19,290 sit inside the capsule. 111 00:05:19,290 --> 00:05:25,180 And if you looked at the stars with the telescope, you could 112 00:05:25,180 --> 00:05:29,000 basically align yourself and align the inertia platform. 113 00:05:29,000 --> 00:05:32,580 Remember the inertial platform navigates with accelerometers, 114 00:05:32,580 --> 00:05:33,940 which can be very accurate. 115 00:05:33,940 --> 00:05:37,300 But what's problem with doing that? 116 00:05:37,300 --> 00:05:38,860 It drifts. 117 00:05:38,860 --> 00:05:41,370 So it loses its absolute accuracy over time. 118 00:05:41,370 --> 00:05:44,610 So every once in while, you have to realign it and retell 119 00:05:44,610 --> 00:05:45,480 it where it is. 120 00:05:45,480 --> 00:05:49,280 And you could align yourself by just pointing at the stars 121 00:05:49,280 --> 00:05:51,380 and knowing which way is up, in what's 122 00:05:51,380 --> 00:05:53,880 called an inertial frame. 123 00:05:53,880 --> 00:05:57,280 But if you wanted to actually navigate, you had to do things 124 00:05:57,280 --> 00:05:58,690 or closer to the planet. 125 00:05:58,690 --> 00:06:00,240 So you could look at particular 126 00:06:00,240 --> 00:06:02,110 points on the Earth. 127 00:06:02,110 --> 00:06:04,810 The obvious one, the easiest one is measure the size of the 128 00:06:04,810 --> 00:06:06,610 Earth, because as you go further 129 00:06:06,610 --> 00:06:07,365 away, it get's smaller. 130 00:06:07,365 --> 00:06:09,620 It turns out that's not a very accurate measurement. 131 00:06:09,620 --> 00:06:12,100 But it is one way to measure where you are. 132 00:06:12,100 --> 00:06:15,610 The most accurate way to measure it is star 133 00:06:15,610 --> 00:06:16,070 occultation. 134 00:06:16,070 --> 00:06:19,270 So that if there's a star that passes by the horizon, you 135 00:06:19,270 --> 00:06:22,650 note the timing of that star and that gives you a very good 136 00:06:22,650 --> 00:06:23,790 sense of where you are. 137 00:06:23,790 --> 00:06:26,470 So there's a whole variety of things-- is anybody into 138 00:06:26,470 --> 00:06:29,160 astronomy here? 139 00:06:29,160 --> 00:06:30,230 I never was particularly. 140 00:06:30,230 --> 00:06:32,670 But last year, my brother brought me a telescope that 141 00:06:32,670 --> 00:06:35,120 has one of these motorized mounts on it, where you can 142 00:06:35,120 --> 00:06:37,270 sort of tell it what star it is and it'll go there. 143 00:06:37,270 --> 00:06:40,770 That's basically what the Apollo guidance computer was, 144 00:06:40,770 --> 00:06:42,990 where it would point out different stars. 145 00:06:42,990 --> 00:06:44,500 It had a little database. 146 00:06:44,500 --> 00:06:47,580 And the database of stars that was used in the entire Apollo 147 00:06:47,580 --> 00:06:50,710 program was, I forget, something like 100 stars long. 148 00:06:50,710 --> 00:06:54,230 And it was one junior engineer in 1962, came to the MIT 149 00:06:54,230 --> 00:06:56,990 libraries and looked up all the star coordinates and put 150 00:06:56,990 --> 00:06:57,650 them in the computer. 151 00:06:57,650 --> 00:07:00,760 And that's the one they used all the way through. 152 00:07:00,760 --> 00:07:03,850 And astronauts were then trained to know which stars, 153 00:07:03,850 --> 00:07:05,610 and they each had a number on them, and they could align 154 00:07:05,610 --> 00:07:10,460 themselves with the catalog and make sure they were 155 00:07:10,460 --> 00:07:12,040 getting there. 156 00:07:12,040 --> 00:07:15,900 And eventually, they ended up putting a digital computer in 157 00:07:15,900 --> 00:07:18,900 it, which was a pretty radical step at the time. 158 00:07:18,900 --> 00:07:21,990 Because again, computers were room-size things. 159 00:07:21,990 --> 00:07:25,080 And this was a computer that was briefcase size. 160 00:07:25,080 --> 00:07:27,540 And for those of you who are into computers, it looked a 161 00:07:27,540 --> 00:07:30,570 lot more like a microcontroller than a 162 00:07:30,570 --> 00:07:32,470 full-up, general purpose computer. 163 00:07:32,470 --> 00:07:36,680 It was modest numerical capability, but lots of I/O 164 00:07:36,680 --> 00:07:40,530 and interrupts and things going in and out, 16 bits. 165 00:07:40,530 --> 00:07:43,440 And it wasn't actually a single chip thing, like a 166 00:07:43,440 --> 00:07:45,810 microcontroller today, but it had a lot of those 167 00:07:45,810 --> 00:07:51,000 characteristics, including this interface for the crew. 168 00:07:51,000 --> 00:07:53,580 And it tied into the inertial system. 169 00:07:53,580 --> 00:07:56,560 So you have this combination of the gyroscopes, the 170 00:07:56,560 --> 00:08:01,180 telescopes, the computer itself, and this display 171 00:08:01,180 --> 00:08:03,560 keyboard unit. 172 00:08:03,560 --> 00:08:06,620 And this is-- 173 00:08:06,620 --> 00:08:08,870 actually I forget who this is. 174 00:08:08,870 --> 00:08:10,790 I think it's Jim Lovell. 175 00:08:10,790 --> 00:08:12,360 No, Jim Lovell's the next one. 176 00:08:12,360 --> 00:08:15,170 One of the Apollo astronauts operating this system. 177 00:08:15,170 --> 00:08:19,290 And there would have been a display keyboard like that, up 178 00:08:19,290 --> 00:08:22,480 on the main console, which he's looking at, you can see 179 00:08:22,480 --> 00:08:23,320 at the top. 180 00:08:23,320 --> 00:08:24,080 I don't know if you're familiar. 181 00:08:24,080 --> 00:08:27,430 Has anybody seen the Apollo capsules down in the 182 00:08:27,430 --> 00:08:31,080 Smithsonian Air and Space Museum? 183 00:08:31,080 --> 00:08:32,500 They looked very awkward on Earth. 184 00:08:32,500 --> 00:08:35,580 But they were easier in space, where when they're launching, 185 00:08:35,580 --> 00:08:38,220 this guy, he would be in the launch position, and have his 186 00:08:38,220 --> 00:08:41,470 back on the ground, looking at this control panel. 187 00:08:41,470 --> 00:08:43,539 And then when they get into space, he 188 00:08:43,539 --> 00:08:45,080 actually can move up here. 189 00:08:45,080 --> 00:08:48,460 And this is actually 90 degrees from that position. 190 00:08:48,460 --> 00:08:51,300 And there's a whole station where they navigate from. 191 00:08:51,300 --> 00:08:57,420 And what they would do would be navigate by-- 192 00:08:57,420 --> 00:09:00,140 again, they're not flying the spaceship doing this. 193 00:09:00,140 --> 00:09:02,750 They're just aligning the inertial system. 194 00:09:02,750 --> 00:09:08,340 And then once the inertial system knows where it's going, 195 00:09:08,340 --> 00:09:11,160 they type in where they want to go on the keyboard. 196 00:09:11,160 --> 00:09:16,930 And so every few hours or so, the inertial system would need 197 00:09:16,930 --> 00:09:17,690 to be realigned. 198 00:09:17,690 --> 00:09:20,620 So it would point at where it thinks the star was. 199 00:09:20,620 --> 00:09:22,660 The astronaut would look through these telescopes. 200 00:09:22,660 --> 00:09:25,360 One was a sort of low power, for getting the thing in the 201 00:09:25,360 --> 00:09:25,910 field of view. 202 00:09:25,910 --> 00:09:28,300 And one was a high power, for the precision pointing. 203 00:09:28,300 --> 00:09:30,120 And then he had these two little sets of 204 00:09:30,120 --> 00:09:31,390 toggle switches here. 205 00:09:31,390 --> 00:09:35,840 One of them would just move the mirror in one axis and the 206 00:09:35,840 --> 00:09:37,830 other would actually rotate the spacecraft around. 207 00:09:37,830 --> 00:09:41,330 And between those two things, they would realign the star. 208 00:09:41,330 --> 00:09:43,600 It would usually be off center, just a little bit. 209 00:09:43,600 --> 00:09:45,100 And they would say no, that's not where it should be. 210 00:09:45,100 --> 00:09:46,390 It should be here. 211 00:09:46,390 --> 00:09:48,790 And then enter it into the computer 212 00:09:48,790 --> 00:09:49,860 and say that's centered. 213 00:09:49,860 --> 00:09:52,550 And the computer would take those error corrections, in 214 00:09:52,550 --> 00:09:54,080 what today we call a common filter-- 215 00:09:54,080 --> 00:09:55,130 Batten was one of the inventors 216 00:09:55,130 --> 00:09:57,180 of the common filter-- 217 00:09:57,180 --> 00:09:59,835 and then update its guidance solution. 218 00:09:59,835 --> 00:10:04,120 It turned out you could do that very accurately in 219 00:10:04,120 --> 00:10:05,270 getting to the Moon. 220 00:10:05,270 --> 00:10:08,880 And in Apollo 8, this is Jim Lovell, who is probably best 221 00:10:08,880 --> 00:10:12,480 known for his commanding of Apollo 13, a few years later. 222 00:10:12,480 --> 00:10:14,490 But this is Apollo 8, where they first flew to the Moon. 223 00:10:14,490 --> 00:10:15,900 And here you can see him doing it. 224 00:10:15,900 --> 00:10:18,690 And there's the keyboard unit up there. 225 00:10:18,690 --> 00:10:21,950 And there was essentially no real hands-on of the joystick, 226 00:10:21,950 --> 00:10:23,290 flying to the Moon. 227 00:10:23,290 --> 00:10:25,850 It was much more OK, we've got our unit aligned. 228 00:10:25,850 --> 00:10:29,300 He did this 200 times over the course of the trip to and from 229 00:10:29,300 --> 00:10:30,810 the Moon on Apollo 8. 230 00:10:30,810 --> 00:10:34,170 This is where they orbited, but they never landed. 231 00:10:34,170 --> 00:10:41,440 And he realigned it 200 times. 232 00:10:41,440 --> 00:10:43,790 It had very, very accurate navigation. 233 00:10:43,790 --> 00:10:46,600 And the rest of it would be OK, here's where we want to 234 00:10:46,600 --> 00:10:48,920 go, here's the velocity we want to achieve. 235 00:10:48,920 --> 00:10:52,330 And the computer would just orient the spacecraft, set up 236 00:10:52,330 --> 00:10:54,880 the big thruster to fire, fire it, and then have the big 237 00:10:54,880 --> 00:10:56,720 velocity change. 238 00:10:56,720 --> 00:10:59,280 A bit change in how the astronauts expected 239 00:10:59,280 --> 00:11:00,606 to fly to the Moon. 240 00:11:00,606 --> 00:11:01,538 Yeah, Sarah? 241 00:11:01,538 --> 00:11:03,402 AUDIENCE: Is that a picture of the Command Module or the 242 00:11:03,402 --> 00:11:04,340 [INAUDIBLE]? 243 00:11:04,340 --> 00:11:06,180 PROFESSOR: This is in the Command Module, actually. 244 00:11:06,180 --> 00:11:08,220 AUDIENCE: So there was a DSKY in both? 245 00:11:08,220 --> 00:11:10,330 PROFESSOR: There was a DSKY in both, yes. 246 00:11:10,330 --> 00:11:12,720 So there were two identical copies of the computer, one 247 00:11:12,720 --> 00:11:16,230 was in the Lunar Module and one was in the Command Module. 248 00:11:16,230 --> 00:11:19,100 They had different software running on them though. 249 00:11:19,100 --> 00:11:25,170 And in Apollo 13, that turned out to be a huge benefit 250 00:11:25,170 --> 00:11:28,680 because they could shut the entire Command Module down and 251 00:11:28,680 --> 00:11:37,670 transfer the coordinates from one to the other and keep the 252 00:11:37,670 --> 00:11:40,780 inertial solution of where they were in the Lunar Module, 253 00:11:40,780 --> 00:11:42,370 while they powered down. 254 00:11:42,370 --> 00:11:45,460 So they had a redundant system without really what being 255 00:11:45,460 --> 00:11:47,780 aware of it, in a certain way, without planning it, although 256 00:11:47,780 --> 00:11:49,520 they did sort of plan it. 257 00:11:49,520 --> 00:11:51,690 And the interesting thing is these were two separate 258 00:11:51,690 --> 00:11:52,410 spacecrafts. 259 00:11:52,410 --> 00:11:53,830 How would you do that today? 260 00:11:53,830 --> 00:11:57,220 Well, you'd have some network and you just say, copy the 261 00:11:57,220 --> 00:11:58,080 coordinates over. 262 00:11:58,080 --> 00:12:00,290 And all they could do there was read it off the LED 263 00:12:00,290 --> 00:12:03,820 display and yell it down the tunnel. 264 00:12:03,820 --> 00:12:05,700 And the other guy would type it in. 265 00:12:05,700 --> 00:12:06,700 There was no electrical 266 00:12:06,700 --> 00:12:07,990 connection between two computers. 267 00:12:12,680 --> 00:12:16,610 This is actually-- the picture in the MIT Museum, behind the 268 00:12:16,610 --> 00:12:19,350 prototype of the computer that's sitting there. 269 00:12:19,350 --> 00:12:21,840 It's Davey Hoag, one of the engineers on it, just another 270 00:12:21,840 --> 00:12:22,620 view of it. 271 00:12:22,620 --> 00:12:24,880 Although it's a very interesting thing here, this 272 00:12:24,880 --> 00:12:28,530 was the original block 1 version of the computer. 273 00:12:28,530 --> 00:12:30,520 And the idea was the astronauts would be able to 274 00:12:30,520 --> 00:12:32,500 repair the computer in flight. 275 00:12:32,500 --> 00:12:35,470 And it's sort of like, anybody ever have an old IBM PC where 276 00:12:35,470 --> 00:12:36,080 you could pull-- 277 00:12:36,080 --> 00:12:37,870 I guess PCs are still this way. 278 00:12:37,870 --> 00:12:39,950 You can pull the cards out and put new 279 00:12:39,950 --> 00:12:41,180 cards in them and stuff. 280 00:12:41,180 --> 00:12:42,100 That was the idea here. 281 00:12:42,100 --> 00:12:44,430 You'd be able to pull the cards out. 282 00:12:44,430 --> 00:12:46,790 And they actually talked about the astronauts having tools 283 00:12:46,790 --> 00:12:49,060 and soldering irons and things to repair 284 00:12:49,060 --> 00:12:51,750 the computer in flight. 285 00:12:51,750 --> 00:12:57,220 That turned out to be a bad idea because on a space 286 00:12:57,220 --> 00:13:00,360 flight, and this is still true today, the physical 287 00:13:00,360 --> 00:13:03,180 environment becomes completely contaminated with all of the 288 00:13:03,180 --> 00:13:05,880 exhalations from the human bodies. 289 00:13:05,880 --> 00:13:09,880 And on one of the Mercury flights, one of the 290 00:13:09,880 --> 00:13:11,840 electronics boxes shorted out. 291 00:13:11,840 --> 00:13:16,050 And it turned out there were urine crystals from the 292 00:13:16,050 --> 00:13:18,070 astronaut that had-- 293 00:13:18,070 --> 00:13:21,140 it was vapor I guess, that had crystallized on the circuits 294 00:13:21,140 --> 00:13:22,020 and shorted them out. 295 00:13:22,020 --> 00:13:25,990 And so after this, they ended up going away from this 296 00:13:25,990 --> 00:13:28,640 in-flight repair and just building one sealed box that 297 00:13:28,640 --> 00:13:31,270 was completely hermetically sealed, no possibility of 298 00:13:31,270 --> 00:13:32,240 repairing it. 299 00:13:32,240 --> 00:13:34,670 And how did they have redundancy you might ask? 300 00:13:34,670 --> 00:13:35,760 And the answer is they didn't. 301 00:13:35,760 --> 00:13:37,840 They just tried to engineer it really well, to 302 00:13:37,840 --> 00:13:41,050 make it really reliable. 303 00:13:41,050 --> 00:13:44,520 Anybody know how many on the Space Shuttle to control it? 304 00:13:44,520 --> 00:13:46,170 Five. 305 00:13:46,170 --> 00:13:47,950 Five parallel redundant computers. 306 00:13:47,950 --> 00:13:49,525 There's one in the Apollo system. 307 00:13:49,525 --> 00:13:51,356 It never failed in flight. 308 00:13:54,500 --> 00:13:56,670 This is another view of that process. 309 00:14:02,160 --> 00:14:06,730 Here's where the astronaut in the Command Module can look at 310 00:14:06,730 --> 00:14:08,630 the Lunar Module and they can navigate 311 00:14:08,630 --> 00:14:09,980 relative to each other. 312 00:14:09,980 --> 00:14:12,310 So you could think in your mind about all the different 313 00:14:12,310 --> 00:14:15,790 possible combinations of looking at the Command Module, 314 00:14:15,790 --> 00:14:18,090 looking at the Lunar Module, looking at the Earth, a 315 00:14:18,090 --> 00:14:19,380 million different things you can do. 316 00:14:19,380 --> 00:14:23,270 And it was an supremely flexible program 317 00:14:23,270 --> 00:14:25,630 that they could use. 318 00:14:25,630 --> 00:14:26,880 Here's another view. 319 00:14:30,400 --> 00:14:31,930 I probably just like the way they make these 320 00:14:31,930 --> 00:14:32,680 drawings and stuff. 321 00:14:32,680 --> 00:14:35,420 And you can see in all of them, made by the computer 322 00:14:35,420 --> 00:14:38,120 engineers at Draper, with computers in the center of the 323 00:14:38,120 --> 00:14:40,840 system and the astronaut is like this minor peripheral 324 00:14:40,840 --> 00:14:43,360 that contributes something here and there. 325 00:14:46,510 --> 00:14:49,450 Here's another one, yet another one. 326 00:14:52,270 --> 00:14:54,970 This is actually a view that's kind of interesting to look 327 00:14:54,970 --> 00:15:03,350 at, of the position error at the critical moment where they 328 00:15:03,350 --> 00:15:06,920 left the Earth's gravitational field and they entered the 329 00:15:06,920 --> 00:15:08,690 Moon's gravitational field. 330 00:15:08,690 --> 00:15:14,190 And when they're sighting on the Earth's horizon as they 331 00:15:14,190 --> 00:15:17,460 get that far away, the error is just rising 332 00:15:17,460 --> 00:15:19,100 and rising and rising. 333 00:15:19,100 --> 00:15:22,040 And then at this critical moment, they switched and they 334 00:15:22,040 --> 00:15:26,970 sight on the lunar horizon with these three stars. 335 00:15:26,970 --> 00:15:30,480 And you can see the error just goes down almost to nothing, 336 00:15:30,480 --> 00:15:33,720 compared to the later calculated accurate value. 337 00:15:33,720 --> 00:15:37,840 So they plotted this stuff obsessively and 338 00:15:37,840 --> 00:15:40,140 were extremely clear. 339 00:15:40,140 --> 00:15:41,450 Actually, this is the actual error. 340 00:15:41,450 --> 00:15:47,040 That's the position of where their orbit is and depending 341 00:15:47,040 --> 00:15:48,110 on where you are. 342 00:15:48,110 --> 00:15:50,890 And that was part of the common filter problem is you 343 00:15:50,890 --> 00:15:52,700 only want to take the sightings to update the 344 00:15:52,700 --> 00:15:55,110 solution when it's going to really be better than the 345 00:15:55,110 --> 00:15:57,320 noise in the process to do it. 346 00:15:57,320 --> 00:15:59,520 And Batten was very good at figuring out 347 00:15:59,520 --> 00:16:01,410 when that would be. 348 00:16:01,410 --> 00:16:02,660 Another view of it. 349 00:16:05,840 --> 00:16:08,000 Here's the one in the Lunar Module. 350 00:16:08,000 --> 00:16:10,720 Remember, the crews are standing up. 351 00:16:10,720 --> 00:16:11,480 You can see them here. 352 00:16:11,480 --> 00:16:14,620 Here's the astronaut right there. 353 00:16:14,620 --> 00:16:17,060 Again sort of considering obviously this human and 354 00:16:17,060 --> 00:16:18,850 machine computing thing that I've been 355 00:16:18,850 --> 00:16:20,100 talking about all semester. 356 00:16:24,430 --> 00:16:27,960 Another view of the landing. 357 00:16:27,960 --> 00:16:30,800 So I wanted to show you one little clip about the software 358 00:16:30,800 --> 00:16:32,400 for this, because it's actually really interesting. 359 00:16:36,410 --> 00:16:41,920 The entire work statement for MIT's role in the Apollo 360 00:16:41,920 --> 00:16:45,770 program is 10 pages long. 361 00:16:45,770 --> 00:16:48,580 This is before it became $140 million program. 362 00:16:48,580 --> 00:16:53,640 And one line says, of course MIT will provide the programs 363 00:16:53,640 --> 00:16:58,730 that are required to run them the computer. 364 00:16:58,730 --> 00:17:01,220 And it doesn't use the word "software." The word software 365 00:17:01,220 --> 00:17:02,240 hadn't been invented. 366 00:17:02,240 --> 00:17:04,440 And I'll show a graph in a little bit. 367 00:17:04,440 --> 00:17:07,160 By the end of the program, they were worried they may not 368 00:17:07,160 --> 00:17:08,750 make the landing on the Moon because the 369 00:17:08,750 --> 00:17:10,230 software wouldn't be ready. 370 00:17:10,230 --> 00:17:14,170 So the transition I would say between people bragging about 371 00:17:14,170 --> 00:17:16,430 how big their computers are and Apollo, was the beginning 372 00:17:16,430 --> 00:17:19,250 of people bragging about how small they are. 373 00:17:19,250 --> 00:17:21,760 And then also the transition from idea that hardware was 374 00:17:21,760 --> 00:17:24,020 really unreliable. 375 00:17:24,020 --> 00:17:26,579 But one of the things this computer used was really the 376 00:17:26,579 --> 00:17:29,315 first silicon chips, integrated circuits. 377 00:17:32,030 --> 00:17:38,100 They bought 60% of the entire US production of semiconductor 378 00:17:38,100 --> 00:17:39,630 chips in 1964. 379 00:17:39,630 --> 00:17:43,090 It gave a huge boost to this budding, very uncertain 380 00:17:43,090 --> 00:17:45,280 technology. 381 00:17:45,280 --> 00:17:50,610 And that made the hardware actually really reliable. 382 00:17:50,610 --> 00:17:52,420 And like I said, they never really had a 383 00:17:52,420 --> 00:17:53,730 hardware failure in fight. 384 00:17:53,730 --> 00:17:56,940 But there were all kinds of funky software bugs that they 385 00:17:56,940 --> 00:17:59,520 kind of uncover and the beginning of that software 386 00:17:59,520 --> 00:18:01,260 could kill you too. 387 00:18:01,260 --> 00:18:02,440 AUDIENCE: What was the percentage again? 388 00:18:02,440 --> 00:18:04,580 PROFESSOR: 60%. 389 00:18:04,580 --> 00:18:06,860 So I'll play this little clip, which is from a video that we 390 00:18:06,860 --> 00:18:07,760 made a couple years ago. 391 00:18:07,760 --> 00:18:09,080 [VIDEO PLAYBACK] 392 00:18:09,080 --> 00:18:11,720 PROFESSOR: So one of things interesting about that is that 393 00:18:11,720 --> 00:18:13,810 they wove this core-rope memory. 394 00:18:13,810 --> 00:18:16,930 And we started this class talking about the relationship 395 00:18:16,930 --> 00:18:20,240 of industry in New England, and typically the textile 396 00:18:20,240 --> 00:18:23,400 industry to the foundation of MIT and how many of the people 397 00:18:23,400 --> 00:18:26,200 who had founded that industry were MIT's early trustees. 398 00:18:26,200 --> 00:18:29,540 And here we are a hundred years later, and the textile 399 00:18:29,540 --> 00:18:32,693 industry which is essentially dying or dead by that time in 400 00:18:32,693 --> 00:18:38,250 New England, these workers are still very much using those 401 00:18:38,250 --> 00:18:39,230 same kinds of skills. 402 00:18:39,230 --> 00:18:43,430 Literally, that software was manufactured in an old textile 403 00:18:43,430 --> 00:18:45,210 plant in Waltham, Massachusetts, which still 404 00:18:45,210 --> 00:18:46,800 exists today. 405 00:18:46,800 --> 00:18:49,660 And also we probably mentioned Digital Equipment Corporation. 406 00:18:49,660 --> 00:18:51,860 They started in old textile plants. 407 00:18:51,860 --> 00:18:56,320 If you go today to Manchester, New Hampshire or even Lowell, 408 00:18:56,320 --> 00:18:59,220 you'll find high-tech software companies and stuff in these 409 00:18:59,220 --> 00:19:00,270 old textile factories. 410 00:19:00,270 --> 00:19:02,140 So there is this sort of site all of 411 00:19:02,140 --> 00:19:04,570 industrial decline and renewal. 412 00:19:04,570 --> 00:19:05,880 This is one model. 413 00:19:05,880 --> 00:19:07,390 You can see, those of you familiar with 414 00:19:07,390 --> 00:19:11,850 microcontrollers, it looks very similar to what you might 415 00:19:11,850 --> 00:19:15,110 see today, all the different inputs, radars, and sensors, 416 00:19:15,110 --> 00:19:19,520 and things, and the outputs for the Lunar Module Another 417 00:19:19,520 --> 00:19:21,360 view of the system. 418 00:19:21,360 --> 00:19:26,060 Totally fly-by-wife, so every time any commands from a 419 00:19:26,060 --> 00:19:30,180 joystick, which weren't used very commonly anyway, they 420 00:19:30,180 --> 00:19:33,140 always went through the digital computer before going 421 00:19:33,140 --> 00:19:34,390 out to any of the thrusters. 422 00:19:36,800 --> 00:19:39,790 At first, the astronauts wanted direct wires, physical 423 00:19:39,790 --> 00:19:42,520 wires, to control the solenoids for the thrusters. 424 00:19:42,520 --> 00:19:47,550 But there are 16 attitude thrusters on the Lunar Module. 425 00:19:47,550 --> 00:19:51,800 And you can imagine trying to, in these clusters of four, 426 00:19:51,800 --> 00:19:53,890 imagine trying to control each one of those individually. 427 00:19:53,890 --> 00:19:57,170 It was just not possible to do. 428 00:19:57,170 --> 00:19:59,790 Here's one view that I like to show from the Draper-- 429 00:19:59,790 --> 00:20:02,930 the instrumentation lab, of no automation in 430 00:20:02,930 --> 00:20:04,880 the cockpit, no software. 431 00:20:04,880 --> 00:20:08,460 The crew is completely busy, overwhelmed, doing all these 432 00:20:08,460 --> 00:20:09,420 different jobs. 433 00:20:09,420 --> 00:20:11,030 And then full automation. 434 00:20:11,030 --> 00:20:13,560 They're sleeping and smoking cigars and playing cards. 435 00:20:13,560 --> 00:20:15,860 And all they have to do is look at the abort button. 436 00:20:15,860 --> 00:20:17,910 It's pretty much how the launch happened. 437 00:20:17,910 --> 00:20:19,410 And still happens in the shuttle today. 438 00:20:19,410 --> 00:20:23,890 There's no manual control for the first part of the launch. 439 00:20:23,890 --> 00:20:28,770 And the astronauts were convinced they could fly the 440 00:20:28,770 --> 00:20:32,790 Saturn V rocket by hand, off the launch pad, which turned 441 00:20:32,790 --> 00:20:34,750 out not to be the case. 442 00:20:34,750 --> 00:20:37,350 And they tried to find a kind of middle ground between these 443 00:20:37,350 --> 00:20:38,805 two positions. 444 00:20:41,400 --> 00:20:43,620 Another view of that navigation flow. 445 00:20:43,620 --> 00:20:46,240 This one is a little bit later and much more sophisticated 446 00:20:46,240 --> 00:20:48,590 than the earlier ones. 447 00:20:48,590 --> 00:20:52,490 I'll skip over a lot of this. 448 00:20:52,490 --> 00:20:56,030 That's the DSKY, the display keyboard. 449 00:20:56,030 --> 00:20:58,990 Another view of it. 450 00:20:58,990 --> 00:21:02,820 This type of digit, anybody seen those before? 451 00:21:02,820 --> 00:21:04,010 There's sort of a standard-- 452 00:21:04,010 --> 00:21:06,190 you can still see them today in some places. 453 00:21:06,190 --> 00:21:08,490 They're called 7-segment displays, because it's seven 454 00:21:08,490 --> 00:21:11,420 segments of a number. 455 00:21:11,420 --> 00:21:12,710 They're a little bit obsolete today. 456 00:21:12,710 --> 00:21:15,270 But they're sort of part of calculators and stuff. 457 00:21:15,270 --> 00:21:17,960 As near as I can tell, that display was invented for the 458 00:21:17,960 --> 00:21:19,210 Apollo computer. 459 00:21:29,100 --> 00:21:32,100 There's one guy on the team who claims that he did it. 460 00:21:32,100 --> 00:21:34,270 And I've never found anybody who claims that they did it 461 00:21:34,270 --> 00:21:36,070 before that. 462 00:21:36,070 --> 00:21:39,320 So this kind of numerical output, which became an icon 463 00:21:39,320 --> 00:21:45,660 of the digital age for a long time, was created for Apollo. 464 00:21:45,660 --> 00:21:49,680 Just beginning to see touch-tone telephones. 465 00:21:49,680 --> 00:21:51,765 Anybody ever see a touchdown telephone? 466 00:21:51,765 --> 00:21:53,050 Probably before your time, mainly. 467 00:21:53,050 --> 00:21:54,620 But. 468 00:21:54,620 --> 00:21:55,690 Instead of dial tones. 469 00:21:55,690 --> 00:21:59,220 And this was the layout of the touch-tone telephone. 470 00:21:59,220 --> 00:22:02,650 And although in the end, AT&T flipped it upside down. 471 00:22:02,650 --> 00:22:04,820 And the astronauts had a lot of resistance to 472 00:22:04,820 --> 00:22:05,710 typing things in. 473 00:22:05,710 --> 00:22:09,800 Because again, typing was not something that highly-trained 474 00:22:09,800 --> 00:22:11,260 test pilots were used to doing. 475 00:22:11,260 --> 00:22:14,200 They'd consider that office work and mostly women's work. 476 00:22:14,200 --> 00:22:17,550 And now of course, everybody types all the time. 477 00:22:17,550 --> 00:22:19,820 But at the time, that sort of work was not 478 00:22:19,820 --> 00:22:21,270 their idea of fun. 479 00:22:21,270 --> 00:22:24,860 This guy is the guy you saw in that slide show, Don Eyles. 480 00:22:24,860 --> 00:22:26,430 Here he is with Don Draper. 481 00:22:26,430 --> 00:22:28,640 And I like this picture because-- 482 00:22:28,640 --> 00:22:30,760 and it actually relates us to some of the 483 00:22:30,760 --> 00:22:33,800 readings for today. 484 00:22:33,800 --> 00:22:37,590 It captures the sense that this Apollo project was partly 485 00:22:37,590 --> 00:22:39,660 a product of what was really the '50s, 486 00:22:39,660 --> 00:22:42,200 kind of Cold War mentality. 487 00:22:42,200 --> 00:22:45,770 And by the time it came to fruition in the late '60s, was 488 00:22:45,770 --> 00:22:47,100 a whole different world. 489 00:22:47,100 --> 00:22:50,640 In fact, if you look at the naming of the Apollo capsules, 490 00:22:50,640 --> 00:22:53,770 the first ones are called Eagle and Colombia and these 491 00:22:53,770 --> 00:22:57,010 kind of super high-brow, patriotic names. 492 00:22:57,010 --> 00:23:00,170 And then by the later Apollo programs, the capsules are 493 00:23:00,170 --> 00:23:03,760 called Snoopy and Aquarius, and much more kind of 494 00:23:03,760 --> 00:23:07,240 reflecting '60s hippies culture. 495 00:23:07,240 --> 00:23:12,900 And this guy Don Eyles, he graduated from BU with a 496 00:23:12,900 --> 00:23:16,310 degree in mathematics in 1967, so two 497 00:23:16,310 --> 00:23:17,310 years before the landing. 498 00:23:17,310 --> 00:23:20,560 The Apollo program was already beginning to wind down on the 499 00:23:20,560 --> 00:23:21,600 engineering. 500 00:23:21,600 --> 00:23:24,840 He got a job at Draper Labs and he started working there. 501 00:23:24,840 --> 00:23:27,310 And within a few months, he was writing the code that 502 00:23:27,310 --> 00:23:30,570 controlled the landing on the Moon. 503 00:23:30,570 --> 00:23:33,120 And most of the code was written by him and a few other 504 00:23:33,120 --> 00:23:35,950 people for the actual landing. 505 00:23:35,950 --> 00:23:39,360 And he was 24 years old when Apollo 11 landed on the Moon. 506 00:23:39,360 --> 00:23:40,450 And he was basically a hippie. 507 00:23:40,450 --> 00:23:41,420 He had long hair. 508 00:23:41,420 --> 00:23:45,510 And he talks about, during his lunch breaks, he would go out 509 00:23:45,510 --> 00:23:48,730 in Cambridge to protest the Vietnam War. 510 00:23:48,730 --> 00:23:51,510 And then come back to this super-secret facility that was 511 00:23:51,510 --> 00:23:54,540 working on nuclear guidance and other things and work on 512 00:23:54,540 --> 00:23:55,895 the Apollo program. 513 00:23:55,895 --> 00:23:57,110 And he's this great example. 514 00:23:57,110 --> 00:23:58,270 And he's still around. 515 00:23:58,270 --> 00:23:59,560 And he's friend of mine these days. 516 00:23:59,560 --> 00:24:01,620 And he comes to speak in my Apollo class. 517 00:24:01,620 --> 00:24:03,080 And he's actually a visual artist. 518 00:24:03,080 --> 00:24:06,090 He does photography downtown. 519 00:24:06,090 --> 00:24:11,470 And of the way that, on the campus you have this mixture 520 00:24:11,470 --> 00:24:14,500 of different kinds of cultures and different sorts of 521 00:24:14,500 --> 00:24:18,470 politics, mixing very freely on the campus. 522 00:24:18,470 --> 00:24:21,330 And he never found any problem with that. 523 00:24:21,330 --> 00:24:24,960 And he did his job just fine. 524 00:24:24,960 --> 00:24:28,030 And he worked at Draper Labs for another 25 years until he 525 00:24:28,030 --> 00:24:32,120 retired about 10 years ago. 526 00:24:32,120 --> 00:24:33,340 This is an interesting one because it 527 00:24:33,340 --> 00:24:34,860 also shows you how-- 528 00:24:34,860 --> 00:24:36,810 this thing right here is a paper tape. 529 00:24:36,810 --> 00:24:40,260 And they would literally take the code from the computer, 530 00:24:40,260 --> 00:24:42,950 and very much in the same spirit of the numerically 531 00:24:42,950 --> 00:24:46,070 controlled machine tools that we talked about, feed the code 532 00:24:46,070 --> 00:24:48,590 into this paper tape, into this big machine. 533 00:24:48,590 --> 00:24:51,980 And it would automatically position this frame over the 534 00:24:51,980 --> 00:24:53,680 hole, where she's putting her needle through. 535 00:24:53,680 --> 00:24:57,250 So they did manage to automate it somewhat, where all the 536 00:24:57,250 --> 00:24:59,300 worker had to do was thread the needle, wait for the thing 537 00:24:59,300 --> 00:24:59,920 to move, and thread it. 538 00:24:59,920 --> 00:25:03,020 She didn't have to look up a binary chart and 539 00:25:03,020 --> 00:25:03,910 then find it there. 540 00:25:03,910 --> 00:25:06,025 So that work was becoming increasingly automated. 541 00:25:11,710 --> 00:25:13,950 These are some of the people who were involved in the 542 00:25:13,950 --> 00:25:15,770 thing, some of who you saw in the video. 543 00:25:15,770 --> 00:25:18,390 This guy Joe Gavin was at MIT grad. 544 00:25:18,390 --> 00:25:21,730 He just died, I think less than a year ago. 545 00:25:21,730 --> 00:25:25,100 And he was vice president of Grumman when they made the 546 00:25:25,100 --> 00:25:29,210 Lunar Lander and then became president of Grumman later. 547 00:25:29,210 --> 00:25:30,730 And these other guys are mostly 548 00:25:30,730 --> 00:25:32,400 programmers of various kinds. 549 00:25:32,400 --> 00:25:35,170 And they mentioned Margaret Hamilton, one of the very few 550 00:25:35,170 --> 00:25:38,550 women engineers on the program. 551 00:25:38,550 --> 00:25:43,530 Here's the manpower for hardware, which peaks in '65 552 00:25:43,530 --> 00:25:44,500 and goes down. 553 00:25:44,500 --> 00:25:48,690 And then the software manpower just goes right up and almost 554 00:25:48,690 --> 00:25:50,380 swamps them by the end. 555 00:25:57,080 --> 00:25:59,315 I'll just say a little bit more about the landing. 556 00:26:05,290 --> 00:26:07,820 The landing was the most challenging part 557 00:26:07,820 --> 00:26:09,850 of the entire mission. 558 00:26:09,850 --> 00:26:13,360 And up until this point right here, at about 50,000 feet, it 559 00:26:13,360 --> 00:26:15,590 was entirely controlled by this computer. 560 00:26:15,590 --> 00:26:18,700 So all these programmers down in Cambridge are sitting 561 00:26:18,700 --> 00:26:20,460 there, watching it on the big screen. 562 00:26:20,460 --> 00:26:22,980 And there's a room full of people like this size. 563 00:26:22,980 --> 00:26:25,090 And each one of them has a little piece of that program 564 00:26:25,090 --> 00:26:26,180 that's written in. 565 00:26:26,180 --> 00:26:28,680 And they're all sort of contributing in real time. 566 00:26:28,680 --> 00:26:31,760 While the astronauts are standing there monitoring, 567 00:26:31,760 --> 00:26:33,700 until about 500 feet. 568 00:26:33,700 --> 00:26:37,770 And they have this very clever thing where the computer would 569 00:26:37,770 --> 00:26:41,480 spit out a number, which you could see down here, like 43. 570 00:26:41,480 --> 00:26:44,210 And then the commander, this would be Neil Armstrong, would 571 00:26:44,210 --> 00:26:48,000 look through his window at this number, minus 43, and he 572 00:26:48,000 --> 00:26:51,960 could look out and see where the target point was. 573 00:26:51,960 --> 00:26:54,630 And the idea was then he could change the target point if he 574 00:26:54,630 --> 00:26:56,190 didn't like where it was. 575 00:26:56,190 --> 00:27:00,550 And this is another view of what he might see. 576 00:27:00,550 --> 00:27:03,820 So there would be that the landing site there, just sort 577 00:27:03,820 --> 00:27:06,256 of a passive heads-up display. 578 00:27:06,256 --> 00:27:09,350 And he could actually jog his joystick and just say, I don't 579 00:27:09,350 --> 00:27:10,200 like that landing site. 580 00:27:10,200 --> 00:27:11,530 I want to kick it over there. 581 00:27:11,530 --> 00:27:12,700 It would move by a degree. 582 00:27:12,700 --> 00:27:14,940 And then the computer would recalculate the entire 583 00:27:14,940 --> 00:27:16,720 trajectory and bring them that way. 584 00:27:16,720 --> 00:27:19,830 And he could do that laterally and fore and aft, as many 585 00:27:19,830 --> 00:27:22,330 times as he wanted in the course of the landing. 586 00:27:22,330 --> 00:27:24,280 And in some of the landings, they actually did it up to 18 587 00:27:24,280 --> 00:27:25,430 or 20 times. 588 00:27:25,430 --> 00:27:27,290 And the idea was it would gradually converge on a 589 00:27:27,290 --> 00:27:28,980 perfect spot. 590 00:27:28,980 --> 00:27:31,920 And the interesting thing about that, which is sort of 591 00:27:31,920 --> 00:27:35,650 the topic of another story, is that on all six landings, they 592 00:27:35,650 --> 00:27:39,080 turned that system off at about 500 feet and they flew 593 00:27:39,080 --> 00:27:42,120 it in a more semi-automatic mode, still 594 00:27:42,120 --> 00:27:44,580 very much fly by wire. 595 00:27:44,580 --> 00:27:47,920 But then the story of the Apollo landings became, the 596 00:27:47,920 --> 00:27:51,300 astronaut intervenes, turns off this sort of balky 597 00:27:51,300 --> 00:27:54,360 technology, which is landing him in a crater, and saves the 598 00:27:54,360 --> 00:27:57,260 day by finding the perfect spot and bringing it down. 599 00:27:57,260 --> 00:28:00,220 And it was not really an accurate story because the 600 00:28:00,220 --> 00:28:03,380 whole system was designed to interact with the operator to 601 00:28:03,380 --> 00:28:05,680 allow them to select the right kind of spot. 602 00:28:05,680 --> 00:28:07,680 But they all turned it off anyway. 603 00:28:07,680 --> 00:28:10,630 They just wanted to fly. 604 00:28:10,630 --> 00:28:15,060 Actually Dave Scott, who is an MIT graduate-- 605 00:28:15,060 --> 00:28:16,910 it's about a third of the people who walked on the Moon 606 00:28:16,910 --> 00:28:19,190 had degrees from MIT. 607 00:28:19,190 --> 00:28:25,250 And I teach a graduate course in AeroAstro with Larry Young 608 00:28:25,250 --> 00:28:27,540 on this whole story. 609 00:28:27,540 --> 00:28:31,560 And we often have the Apollo astronauts in to 610 00:28:31,560 --> 00:28:32,390 speak to the students. 611 00:28:32,390 --> 00:28:34,850 And they'll tell you, they were either Larry's office 612 00:28:34,850 --> 00:28:37,410 mates or a few of them were his master's students when 613 00:28:37,410 --> 00:28:39,230 they were here in the early '60s. 614 00:28:39,230 --> 00:28:41,280 And actually tomorrow, Dave Scott is going to come. 615 00:28:41,280 --> 00:28:44,160 He was the commander of Apollo 15. 616 00:28:44,160 --> 00:28:46,420 And he's the only one who I think gave a really accurate 617 00:28:46,420 --> 00:28:48,290 description of what happened in these last seconds. 618 00:28:48,290 --> 00:28:52,790 He said, you can say that you wanted to be in the loop if 619 00:28:52,790 --> 00:28:53,670 anything went wrong. 620 00:28:53,670 --> 00:28:55,490 But you knew that if something went wrong with a computer, 621 00:28:55,490 --> 00:28:57,680 you were dead anyway. 622 00:28:57,680 --> 00:28:59,210 But I came away that whole way and darn it, I 623 00:28:59,210 --> 00:29:00,480 just wanted to fly. 624 00:29:00,480 --> 00:29:04,600 And so they all ended up flying it the 625 00:29:04,600 --> 00:29:06,310 last 500 feet or so. 626 00:29:06,310 --> 00:29:10,260 And then NASA and the public and the press could all say 627 00:29:10,260 --> 00:29:18,610 that sort of traditional American values were confirmed 628 00:29:18,610 --> 00:29:21,190 by these sort of cowboy astronauts landing on the Moon 629 00:29:21,190 --> 00:29:21,870 by themselves. 630 00:29:21,870 --> 00:29:25,280 But actually there was this MIT-designed computer there. 631 00:29:25,280 --> 00:29:27,960 And it's an interesting story too about MIT educations, 632 00:29:27,960 --> 00:29:31,700 because the earlier crews-- 633 00:29:31,700 --> 00:29:35,480 Aldren had a Ph.D. from MIT. 634 00:29:35,480 --> 00:29:38,680 And the earlier astronauts were not 635 00:29:38,680 --> 00:29:40,310 very academic engineers. 636 00:29:40,310 --> 00:29:43,880 But gradually over time, they became 637 00:29:43,880 --> 00:29:45,440 much more highly educated. 638 00:29:45,440 --> 00:29:48,700 And MIT, with the exception of West Point and the Naval 639 00:29:48,700 --> 00:29:51,930 Academy and Air Force Academy, the service academies, MIT has 640 00:29:51,930 --> 00:29:54,660 trained more astronauts than any other school. 641 00:29:54,660 --> 00:29:59,150 And it became increasingly sort of de rigeur to have a 642 00:29:59,150 --> 00:30:01,380 master's or a Ph.D in engineering. 643 00:30:01,380 --> 00:30:05,290 And those crews were much more accepting of these advanced 644 00:30:05,290 --> 00:30:07,810 techniques for the guidance and much less interested in 645 00:30:07,810 --> 00:30:10,270 just flying it on in by themselves. 646 00:30:10,270 --> 00:30:12,400 Alan Shepard, and we mentioned his flight, he 647 00:30:12,400 --> 00:30:13,650 commanded Apollo 14. 648 00:30:17,360 --> 00:30:19,750 And he said, aah, if the computer failed, I could have 649 00:30:19,750 --> 00:30:22,990 just taken over and landed by myself on the Moon. 650 00:30:22,990 --> 00:30:25,750 Which is physically impossible, because you're 651 00:30:25,750 --> 00:30:30,090 looking at a terrain landscape that has zero indicators of 652 00:30:30,090 --> 00:30:31,690 how far away from it you are. 653 00:30:31,690 --> 00:30:33,950 You could actually be six inches or six miles and you 654 00:30:33,950 --> 00:30:35,330 wouldn't be able to tell the difference because there's no 655 00:30:35,330 --> 00:30:40,480 trees, there's no trucks, there are no highways, 656 00:30:40,480 --> 00:30:42,640 anything to give you a sense of scale. 657 00:30:42,640 --> 00:30:47,295 And he had a problem with his landing radar. 658 00:30:47,295 --> 00:30:49,300 It didn't come in until the last few minutes. 659 00:30:49,300 --> 00:30:52,330 But if the landing radar had never come in, he said he 660 00:30:52,330 --> 00:30:53,380 would have landed it anyway. 661 00:30:53,380 --> 00:30:55,175 And he probably would have crashed, if he 662 00:30:55,175 --> 00:30:56,990 had tried to do that. 663 00:30:56,990 --> 00:30:58,710 They're actually using this-- 664 00:30:58,710 --> 00:31:01,010 anybody want to guess what this is, what 665 00:31:01,010 --> 00:31:02,260 this diagram portrays? 666 00:31:05,170 --> 00:31:08,140 It's actually a model of the lunar terrain. 667 00:31:08,140 --> 00:31:10,506 So if you've ever programmed a control system-- we used to 668 00:31:10,506 --> 00:31:13,680 have this problem building underwater robots all time-- 669 00:31:13,680 --> 00:31:17,040 and you want to hold a fixed altitude. 670 00:31:17,040 --> 00:31:19,170 If you're cruising along the bottom of the ocean in a robot 671 00:31:19,170 --> 00:31:21,320 that's programmed to hold a fixed altitude, so it's 672 00:31:21,320 --> 00:31:24,270 sending a sonar beam off the bottom and trying to hold an 673 00:31:24,270 --> 00:31:27,940 altitude, and then you go over a hole, the vehicle is going 674 00:31:27,940 --> 00:31:29,440 to dive down like this and smash into 675 00:31:29,440 --> 00:31:31,720 the side of the hole. 676 00:31:31,720 --> 00:31:33,360 And same thing here. 677 00:31:33,360 --> 00:31:36,640 If you're trying to fly over a mountain range-- 678 00:31:36,640 --> 00:31:39,940 it's not a coincidence that the site for the Apollo 11 679 00:31:39,940 --> 00:31:41,760 landing was called the Sea of Tranquility. 680 00:31:41,760 --> 00:31:45,070 They found the flattest, most even terrain they could find. 681 00:31:45,070 --> 00:31:47,270 But the geologists who were running the science 682 00:31:47,270 --> 00:31:50,220 experiments, really wanted to go into places where there was 683 00:31:50,220 --> 00:31:52,190 a lot of geological activity, so the terrain 684 00:31:52,190 --> 00:31:54,690 was a lot more crazy. 685 00:31:54,690 --> 00:31:58,360 And on the later flights, they really did land very much high 686 00:31:58,360 --> 00:31:59,560 up in the mountains. 687 00:31:59,560 --> 00:32:03,450 And this is just a way, where with five line segments, in a 688 00:32:03,450 --> 00:32:07,310 very, very simple computer model, all it takes is these 689 00:32:07,310 --> 00:32:11,770 five line segments and like six numbers, you could model 690 00:32:11,770 --> 00:32:12,960 that there were mountains here. 691 00:32:12,960 --> 00:32:15,440 And then you expected the terrain to go like that. 692 00:32:15,440 --> 00:32:18,060 And the computer model could eat all that stuff for lunch 693 00:32:18,060 --> 00:32:20,300 and it would keep a very stable trajectory and allow 694 00:32:20,300 --> 00:32:23,030 you to go over the mountains, and then land right at the 695 00:32:23,030 --> 00:32:26,820 edge, which is what they did on Apollo 15, and 16, and 17, 696 00:32:26,820 --> 00:32:28,130 a very clever thing. 697 00:32:31,520 --> 00:32:34,930 That's the view out the window, of Aldrin's side. 698 00:32:34,930 --> 00:32:38,750 AUDIENCE: When do you begin to see the sort of the decline in 699 00:32:38,750 --> 00:32:44,570 the pilot-orientedness of the astronauts 700 00:32:44,570 --> 00:32:47,230 wanting to control more? 701 00:32:47,230 --> 00:32:51,070 Because the thing that your lecture brings to my mind is 702 00:32:51,070 --> 00:32:54,320 the book, The Right Stuff, in which that's a big issue. 703 00:32:54,320 --> 00:32:58,880 But clearly what you're saying is over time, that declines. 704 00:32:58,880 --> 00:33:01,900 PROFESSOR: Hopefully, we're seeing it now. 705 00:33:01,900 --> 00:33:03,280 After the Moon landing, what did they do? 706 00:33:03,280 --> 00:33:06,240 They started sending an airplane into orbit. 707 00:33:06,240 --> 00:33:08,020 It wasn't the wisest choice. 708 00:33:08,020 --> 00:33:10,540 But it was because the astronauts wanted to fly 709 00:33:10,540 --> 00:33:12,860 something that had wings on it. 710 00:33:12,860 --> 00:33:15,310 And it's not a very efficient way to go to and from space. 711 00:33:19,150 --> 00:33:21,890 AUDIENCE: So was Alex Roland right all along? 712 00:33:21,890 --> 00:33:24,140 PROFESSOR: Yeah, in some way. 713 00:33:24,140 --> 00:33:28,060 But from MIT's point of view, again building into the Bill 714 00:33:28,060 --> 00:33:34,720 Leslie piece from today, the Apollo program ends in 1972. 715 00:33:34,720 --> 00:33:37,230 It has this huge triumph in 1969. 716 00:33:37,230 --> 00:33:40,000 And that's right when all these protests end up 717 00:33:40,000 --> 00:33:43,050 happening and the Draper Labs gets spun out. 718 00:33:43,050 --> 00:33:45,840 They were extremely offended and insulted by that move. 719 00:33:45,840 --> 00:33:49,310 And there is still hard feelings about it today. 720 00:33:49,310 --> 00:33:52,950 A lot of the people who worked at the Instrumentation Lab 721 00:33:52,950 --> 00:33:55,050 were happy to work on a civilian project. 722 00:33:55,050 --> 00:33:57,660 And one of the reasons they liked working on Apollo was 723 00:33:57,660 --> 00:34:00,410 that it was a non-weapons kind of project. 724 00:34:00,410 --> 00:34:03,240 And then they felt like they got the bad end of the rap 725 00:34:03,240 --> 00:34:04,690 during a lot of the protests anyway. 726 00:34:04,690 --> 00:34:06,710 There was still a lot of military funding and a lot of 727 00:34:06,710 --> 00:34:10,139 weapons work going on there. 728 00:34:10,139 --> 00:34:14,590 But it's sort of this irony of this moment when this part of 729 00:34:14,590 --> 00:34:20,159 MIT has this world shattering technological triumph in a 730 00:34:20,159 --> 00:34:23,580 way, and then also becomes the brunt of it. 731 00:34:23,580 --> 00:34:26,210 And it just shows you again the ways that all these things 732 00:34:26,210 --> 00:34:27,870 on this campus and in the country were 733 00:34:27,870 --> 00:34:29,159 conflicted at the time. 734 00:34:29,159 --> 00:34:33,530 And somebody wrote recently that at the time it seemed 735 00:34:33,530 --> 00:34:36,060 like the hippies and the Moon landings were the furthest 736 00:34:36,060 --> 00:34:36,650 thing apart. 737 00:34:36,650 --> 00:34:39,690 One of them was the military industrial complex and one of 738 00:34:39,690 --> 00:34:42,400 them was this kind of Utopian vision of the world. 739 00:34:42,400 --> 00:34:45,030 But looking back 30 years later, you can see they have a 740 00:34:45,030 --> 00:34:45,690 lot in common. 741 00:34:45,690 --> 00:34:48,469 They were both sort of Utopian, futuristic visions 742 00:34:48,469 --> 00:34:52,239 that never quite succeeded in the promise that they held. 743 00:34:52,239 --> 00:34:55,210 And yet they both influenced the world in different ways 744 00:34:55,210 --> 00:34:57,139 that people hadn't really expected. 745 00:34:57,139 --> 00:34:58,460 And that was true on campus. 746 00:34:58,460 --> 00:35:01,170 So mentioned in their response paper, I'm forgetting who, 747 00:35:01,170 --> 00:35:08,310 about how it was really the one moment there was major 748 00:35:08,310 --> 00:35:09,930 political activism on campus. 749 00:35:09,930 --> 00:35:11,060 There were other places too. 750 00:35:11,060 --> 00:35:15,100 But in this entire story, we were teaching you the history 751 00:35:15,100 --> 00:35:19,670 of Yale University or NYU or Columbia. 752 00:35:19,670 --> 00:35:22,280 It's a very political story the way the students are 753 00:35:22,280 --> 00:35:25,600 responding to the currents in the outside world and how 754 00:35:25,600 --> 00:35:26,900 those things played out on campus. 755 00:35:26,900 --> 00:35:30,870 And it's a much more muted story here for reasons that 756 00:35:30,870 --> 00:35:31,890 might be worth discussing. 757 00:35:31,890 --> 00:35:35,740 And in the discussion, it's really remarkable and notable 758 00:35:35,740 --> 00:35:39,700 this one moment in the '70s, when it happened. 759 00:35:39,700 --> 00:35:45,455 And whereas, I went to college at Yale in the '80s and there 760 00:35:45,455 --> 00:35:49,080 were students being arrested all the time for protests and 761 00:35:49,080 --> 00:35:51,470 occupying the president's office and things. 762 00:35:51,470 --> 00:35:55,680 It became almost a kind of a cartoon of itself, 763 00:35:55,680 --> 00:35:58,040 it happened so often. 764 00:35:58,040 --> 00:36:04,400 Well, if you noticed in the 150th Exhibit, one of the 150 765 00:36:04,400 --> 00:36:07,960 objects is a tape of the Grateful Dead playing on the 766 00:36:07,960 --> 00:36:09,710 front steps of the student center. 767 00:36:09,710 --> 00:36:13,390 And I think it was April 1970, was that when Kent State was? 768 00:36:13,390 --> 00:36:17,010 And the thing they write about it, it was within a week after 769 00:36:17,010 --> 00:36:18,940 Kent State. 770 00:36:18,940 --> 00:36:21,130 And the campus was extremely tense. 771 00:36:21,130 --> 00:36:23,540 Everybody was very unsure what was going to happen. 772 00:36:23,540 --> 00:36:26,360 And when they played their music, they were very aware of 773 00:36:26,360 --> 00:36:29,240 that and really tried to make it be a sort of soothing, 774 00:36:29,240 --> 00:36:34,980 calming moment as opposed to a fractious, contentious, 775 00:36:34,980 --> 00:36:36,360 potentially violent moment. 776 00:36:36,360 --> 00:36:39,080 And that's one of the reasons Debbie chose it as part of the 777 00:36:39,080 --> 00:36:40,330 150 objects. 778 00:36:42,640 --> 00:36:44,930 AUDIENCE: Very interesting. 779 00:36:44,930 --> 00:36:48,040 PROFESSOR: So that takes us up to 1970 or so. 780 00:36:48,040 --> 00:36:51,220 And then there's four more decades to deal with, between 781 00:36:51,220 --> 00:36:53,360 now and then, some of which we read about. 782 00:36:53,360 --> 00:36:58,670 The Cambridge recombinant DNA story is in some ways very 783 00:36:58,670 --> 00:37:01,200 much a continuation of this sort of story. 784 00:37:01,200 --> 00:37:04,113 Cambridge was, I'm not sure how much it still is, it still 785 00:37:04,113 --> 00:37:07,280 is a little bit, a very left-leaning community outside 786 00:37:07,280 --> 00:37:11,520 of the university, very activist, very concerned about 787 00:37:11,520 --> 00:37:14,900 these big, potentially scary technologies. 788 00:37:14,900 --> 00:37:19,730 That was very much the mood of the time, was what is this big 789 00:37:19,730 --> 00:37:22,670 nasty force of technology doing to our world? 790 00:37:22,670 --> 00:37:24,900 And MIT, it was pretty easy for to become 791 00:37:24,900 --> 00:37:26,460 the center of that. 792 00:37:26,460 --> 00:37:30,040 And John Durant's piece shows you how that played out, 793 00:37:30,040 --> 00:37:35,950 arguably in a constructive way, a very interesting one. 794 00:37:35,950 --> 00:37:38,140 And then in '80s, the military-- 795 00:37:38,140 --> 00:37:40,940 we talked about this a little bit last time-- 796 00:37:40,940 --> 00:37:43,100 got very much going again under Reagan. 797 00:37:43,100 --> 00:37:47,060 And a lot of money came into computer science and other 798 00:37:47,060 --> 00:37:49,250 engineering disciplines. 799 00:37:49,250 --> 00:37:53,840 But also the economics of technology and the challenge 800 00:37:53,840 --> 00:37:55,470 from Japan became a big issue. 801 00:37:55,470 --> 00:37:57,490 That had a lot of influence on campus. 802 00:37:57,490 --> 00:38:01,480 All the while, biotech and biology is sort of creeping up 803 00:38:01,480 --> 00:38:06,138 in importance and dollar value. 804 00:38:06,138 --> 00:38:09,040 And then the Cold War ends. 805 00:38:09,040 --> 00:38:13,840 Chuck Vest becomes president in the late either '89 or '90, 806 00:38:13,840 --> 00:38:15,590 '91 I think. 807 00:38:15,590 --> 00:38:17,660 He became president right about when I came here as a 808 00:38:17,660 --> 00:38:18,930 grad student. 809 00:38:18,930 --> 00:38:21,550 And the big question on everybody's mind was what at 810 00:38:21,550 --> 00:38:22,800 that point? 811 00:38:26,990 --> 00:38:29,690 MIT is dependent on government funding for so long. 812 00:38:29,690 --> 00:38:32,720 Now that the Cold War is over, are we going to just shrivel 813 00:38:32,720 --> 00:38:34,140 up and die? 814 00:38:34,140 --> 00:38:36,600 And the pendulum goes again. 815 00:38:36,600 --> 00:38:39,010 Where do they go? 816 00:38:39,010 --> 00:38:40,210 Back towards industry. 817 00:38:40,210 --> 00:38:41,980 And the '90s is very much-- 818 00:38:41,980 --> 00:38:44,100 I mean the government research support is always there, but 819 00:38:44,100 --> 00:38:46,040 very much industry. 820 00:38:50,030 --> 00:38:52,240 I remember Chuck Vest saying-- it was probably pretty late in 821 00:38:52,240 --> 00:38:54,950 his presidency, soon before he stepped down, in a faculty 822 00:38:54,950 --> 00:38:58,030 meeting, I've spent so many years cultivating industrial 823 00:38:58,030 --> 00:38:59,830 sponsors, it's very difficult. 824 00:38:59,830 --> 00:39:01,320 It's very time consuming. 825 00:39:01,320 --> 00:39:03,570 I would be very happy for the government to start up again 826 00:39:03,570 --> 00:39:04,350 and support-- 827 00:39:04,350 --> 00:39:07,770 I mean it didn't go down ever, as much as 828 00:39:07,770 --> 00:39:09,460 people feared it did. 829 00:39:09,460 --> 00:39:14,890 But it became less of a percentage of MIT's income. 830 00:39:14,890 --> 00:39:18,480 Then the financial crisis in '08 happens and the stimulus 831 00:39:18,480 --> 00:39:20,890 package comes in, more government. 832 00:39:20,890 --> 00:39:23,980 And the government runs out of money, less government. 833 00:39:23,980 --> 00:39:25,770 So this pendulum keeps swinging. 834 00:39:28,540 --> 00:39:30,440 AUDIENCE: The periods are shorter. 835 00:39:30,440 --> 00:39:31,710 PROFESSOR: The periods are maybe shorter. 836 00:39:31,710 --> 00:39:32,960 It could be true.