WEBVTT - Ep144 "How do things last?" Part 2: Millennia with Alexander Rose 0:00:05.120 --> 0:00:08.800 What is a ten thousand year clock and what is 0:00:08.880 --> 0:00:12.399 the Y ten k bug? What do ancient ceramics have 0:00:12.480 --> 0:00:15.080 to do with the way that we build ball bearings 0:00:15.120 --> 0:00:19.040 and satellites. Have we entered a digital dark age where 0:00:19.040 --> 0:00:22.720 we're losing more knowledge than we're preserving. Why do some 0:00:23.079 --> 0:00:27.000 organizations last millennium? What does this have to do with 0:00:27.320 --> 0:00:32.240 bristle cone pine trees, or symbols in drum sets, or 0:00:32.280 --> 0:00:36.440 a hotel that's still running that started in the sixth century. 0:00:37.080 --> 0:00:41.320 If humanity disappeared tomorrow, what from our era would still 0:00:41.360 --> 0:00:44.920 be legible thousands of years from now? Join me today 0:00:44.920 --> 0:00:48.320 for thinking about ourselves on a ten thousand year timescale 0:00:48.640 --> 0:00:56.160 with guest Alexander Rose. Welcome to Intercosmos with me David Eagleman. 0:00:56.280 --> 0:00:59.080 I'm a neuroscientist and an author at Stanford and in 0:00:59.120 --> 0:01:02.880 these episodes we dive deeply into our three pound universe 0:01:03.200 --> 0:01:06.640 to uncover some of the most surprising aspects of our lives. 0:01:23.000 --> 0:01:26.679 Today's episode springboards from last week's episode on the topic 0:01:26.720 --> 0:01:32.240 of persistence about things lasting through time. Last episode, we 0:01:32.319 --> 0:01:35.600 talked about why things last, and we saw that some 0:01:35.720 --> 0:01:39.520 do because they are optimized, like the body plan for sharks. 0:01:39.880 --> 0:01:46.680 Other things endure because they constantly repair themselves, like Roman concrete, 0:01:46.680 --> 0:01:50.320 which self heals when it gets cracks. Some things endure 0:01:50.400 --> 0:01:55.160 because they are memorable or they seem explanatory, like urban legends. 0:01:55.440 --> 0:02:00.680 And some things persist because they're fragile but infinitely copyable, 0:02:01.120 --> 0:02:04.200 like paper or DNA. But this week I want to 0:02:04.200 --> 0:02:09.440 turn to much longer time scales, specifically millennia. So today 0:02:09.480 --> 0:02:12.000 we're going to do something that's a little unusual for 0:02:12.120 --> 0:02:17.560 a species that thinks in election cycles or fiscal quarters 0:02:17.919 --> 0:02:20.880 and all of our short term deadlines. We're going to 0:02:20.919 --> 0:02:25.680 step into deep time, beyond decades, beyond centuries. We're going 0:02:25.760 --> 0:02:29.800 to stand on the long arc of civilization where the 0:02:30.320 --> 0:02:34.640 unit of measurement is millennia. Now here's the puzzle we're 0:02:34.680 --> 0:02:38.840 going to tackle today. Human beings keep trying to pass 0:02:38.919 --> 0:02:43.280 things forward, whether that's knowledge or culture or technology, but 0:02:43.520 --> 0:02:49.200 most of what we produce vanishes almost immediately. Our digital lives, 0:02:49.240 --> 0:02:56.120 in particular, are surprisingly fragile. Hard drives fail, formats get 0:02:56.200 --> 0:03:01.240 quickly outdated, servers go bad. A surprisingly large portion of 0:03:01.280 --> 0:03:07.720 the twenty first century will be archaeologically invisible. But Roman 0:03:07.840 --> 0:03:13.080 concrete still holds and a bristle cone pine tree standing 0:03:13.120 --> 0:03:16.480 on a mountain ridge is still alive in counting rings 0:03:16.520 --> 0:03:21.120 after five millennium. So today we're going to launch off 0:03:21.400 --> 0:03:24.080 the question that we started last week, which is what 0:03:24.480 --> 0:03:27.840 actually lasts? And this week we're going to ask can 0:03:27.960 --> 0:03:32.840 we learn to build things like institutions or organizations that 0:03:32.960 --> 0:03:36.720 don't disintegrate as soon as we stop paying attention to them. 0:03:37.480 --> 0:03:40.880 To explore this, I'm joined today by Alexander Rose, who 0:03:40.920 --> 0:03:45.040 goes by Xander. He's the former executive director of the 0:03:45.280 --> 0:03:48.800 Long Now Foundation and organization that I love and I 0:03:48.840 --> 0:03:51.600 sit on the board for for over twenty years. Xander 0:03:51.600 --> 0:03:56.440 has been immersed in projects that stretch our time imagination, 0:03:57.120 --> 0:04:00.360 like the ten thousand year clock that he's been with 0:04:00.440 --> 0:04:02.640 Danny hillis. If you don't know what that is, hangtight, 0:04:02.680 --> 0:04:05.440 because we're going to talk about that today. Xander is 0:04:05.480 --> 0:04:09.360 also involved in the Rosetta disc which acts like a 0:04:09.480 --> 0:04:14.160 linguistic time capsule, and in long bets, which force us 0:04:14.200 --> 0:04:17.839 to put our predictions on the public record. Xander is 0:04:17.880 --> 0:04:22.159 one of the most talented machinists and engineers that I know, 0:04:22.520 --> 0:04:26.440 but also one of the best long term thinkers, and 0:04:26.480 --> 0:04:29.480 by long term I mean millennia. He has spent his 0:04:29.560 --> 0:04:33.800 career asking what it takes for knowledge to persist, what 0:04:33.960 --> 0:04:38.919 makes institutions survive across generations, and how we can design 0:04:39.120 --> 0:04:43.880 for a future that we will never personally inhabit. So 0:04:44.080 --> 0:04:47.720 let's step into the long now and see what it 0:04:47.760 --> 0:04:55.240 takes for anything to endure. Xander, So, you and I 0:04:55.279 --> 0:04:58.560 originally met because you, for twenty six years worked with 0:04:58.640 --> 0:05:02.360 the Long Now Foundation. You ended up as the executive 0:05:02.360 --> 0:05:05.479 director of that for a long time. Tell us about 0:05:05.520 --> 0:05:08.240 the long Now Foundation and what the thinking there is. 0:05:09.080 --> 0:05:12.679 Well, the long Now Foundation was started by Stuart Brand, 0:05:12.680 --> 0:05:15.360 who had started the Whole Earth Catalog. Danny Hillis who 0:05:15.560 --> 0:05:19.320 had developed some of the fastest supercomputers back in the eighties, 0:05:19.320 --> 0:05:23.359 originally with his company Thinking Machines, and other people Brian, 0:05:23.440 --> 0:05:26.159 you know, Kevin Kelly, a lot of kind of early 0:05:26.240 --> 0:05:30.560 digital illuminaries, and they I think they saw, maybe earlier 0:05:30.640 --> 0:05:34.760 on than others, that we were as a society kind 0:05:34.800 --> 0:05:39.680 of really fetishizing speed and only thinking that the things 0:05:39.720 --> 0:05:42.160 that happened fast were the things that mattered. But obviously 0:05:42.200 --> 0:05:46.040 there are issues like climate change or hunger or you know, 0:05:46.120 --> 0:05:49.400 the education system that have returns on investment that are 0:05:49.560 --> 0:05:53.320 on much slower cycles and still need to be addressed. 0:05:53.360 --> 0:05:56.200 And if society was only doing things that could be 0:05:56.240 --> 0:05:59.200 done quickly, those things were not going to be addressed correctly. 0:05:59.320 --> 0:06:04.000 So they thought that some kind of balancing corrective was needed. 0:06:04.080 --> 0:06:06.520 And so Danny Hillis had this idea of a ten 0:06:06.560 --> 0:06:10.960 thousand year monument scale, all mechanical clock, and originally that 0:06:11.040 --> 0:06:12.840 was what I was hired to start working on. 0:06:12.920 --> 0:06:14.880 And the idea is a clock that would last ten 0:06:14.920 --> 0:06:16.240 thousand years exactly. 0:06:16.279 --> 0:06:19.000 It would last the design life was ten thousand years, 0:06:19.480 --> 0:06:21.279 and it would have you know, it wouldn't have like 0:06:21.320 --> 0:06:23.160 a twelve hour dial. It would have all these kind 0:06:23.160 --> 0:06:25.279 of dials that matter over ten thousand years, things like 0:06:25.360 --> 0:06:28.479 astronomic cycles and and things like that. 0:06:28.480 --> 0:06:32.400 And things like once a century it would ding R 0:06:32.480 --> 0:06:34.000 and once a millennium. 0:06:34.200 --> 0:06:35.000 Yeah, I think that. Yeah. 0:06:35.240 --> 0:06:37.400 The original kind of poetic version of the clock that 0:06:37.480 --> 0:06:39.760 Danny wrote about this is like a nineteen ninety five 0:06:39.800 --> 0:06:43.760 Wired essay, was that it would you know, tick once 0:06:43.800 --> 0:06:45.839 a year, it would bong once a century, and a 0:06:45.880 --> 0:06:48.080 cuckoo would come out once a millennium. 0:06:48.440 --> 0:06:49.920 So I signed on to that project. 0:06:49.960 --> 0:06:54.520 My background is an industrial design, But Stuart Brand was 0:06:54.560 --> 0:06:57.120 the one who's like, well, you need an institution alongside this, 0:06:57.400 --> 0:07:00.600 and that's in a way, one of the most difficult 0:07:00.600 --> 0:07:03.640 problems of making something last is not you know, you 0:07:03.680 --> 0:07:06.600 can make an object, especially one that maybe doesn't need 0:07:06.640 --> 0:07:09.480 to be interacting with humans very much, last for that long. 0:07:09.560 --> 0:07:11.080 It's it's not that hard. 0:07:11.480 --> 0:07:14.120 But if you want to make something that's that is 0:07:14.200 --> 0:07:16.800 relevant and is changing the way people think about time, 0:07:17.000 --> 0:07:19.480 you actually need the institution that's alongside it. And so 0:07:20.120 --> 0:07:22.920 they became these kind of two parallel projects. One is 0:07:22.960 --> 0:07:27.600 the engineering project and the other one is the institutional side. 0:07:27.680 --> 0:07:29.000 So I want to zoom in on the clock for 0:07:29.040 --> 0:07:32.120 a minute, because you said something extraordinary there, which is 0:07:32.160 --> 0:07:34.800 it's not so hard to make an object a machine 0:07:34.800 --> 0:07:37.000 that lasts ten thousand years, but no one has ever 0:07:37.080 --> 0:07:40.560 done that or anything like it. So you, having worked 0:07:40.560 --> 0:07:42.520 on it for so many years, you feel like, maybe 0:07:42.520 --> 0:07:46.880 it's not so hard. But tell us about that thinking, well, I. 0:07:46.880 --> 0:07:49.600 Mean often we think that, you know, maybe it's a 0:07:49.640 --> 0:07:53.840 material science problem, but really it's actually it's a it's 0:07:53.840 --> 0:07:57.320 an environment problem. So you know, we have we have 0:07:57.640 --> 0:08:00.680 leather shoes that are five thousand years old, were found 0:08:01.200 --> 0:08:03.600 just in the right kind of pH of soil. Right, 0:08:03.640 --> 0:08:05.640 We've got the Dead Sea scrolls that are on the 0:08:05.760 --> 0:08:08.119 order of you know, many thousand made out of paper 0:08:08.280 --> 0:08:10.760 papyrus that are that have lasted just because they were. 0:08:10.640 --> 0:08:11.640 In the right environment. 0:08:12.720 --> 0:08:14.880 And an early when I was working on this project, 0:08:14.920 --> 0:08:18.239 early on, a material scientist told me that the best 0:08:18.320 --> 0:08:21.400 way to think about this is that basically everything is burning, 0:08:22.080 --> 0:08:25.360 and meaning everything is oxidizing, right, and it's just at 0:08:25.360 --> 0:08:28.400 a different rate. And so you can choose your rate 0:08:28.520 --> 0:08:32.040 at which things are going to oxidize and and and 0:08:32.320 --> 0:08:35.599 mostly you do that by by sealing something in a 0:08:35.640 --> 0:08:38.560 way that it won't oxidize. And the problem with that 0:08:38.640 --> 0:08:40.760 is if you're trying to make an object that or 0:08:40.880 --> 0:08:43.800 in a machine especially that needs that humans want to 0:08:43.840 --> 0:08:44.480 interact with it. 0:08:44.520 --> 0:08:45.600 You want to change the way. 0:08:45.720 --> 0:08:48.960 You know, have this experience of walking through you know 0:08:49.280 --> 0:08:52.520 that this monument scale thing where you're walking through the gears, 0:08:52.559 --> 0:08:55.240 and it really inspires you through the whole experience to 0:08:55.240 --> 0:08:57.679 come out the other end and go, wow, you know, 0:08:57.920 --> 0:09:00.760 time is big, but I have a relevant in it, 0:09:01.360 --> 0:09:04.280 and I want to change maybe some of the ways 0:09:04.320 --> 0:09:06.640 that I operate in the world and that I think 0:09:06.679 --> 0:09:09.880 about the world. To do that, you really need to 0:09:09.960 --> 0:09:13.640 think about material science and human interaction design in a 0:09:13.640 --> 0:09:16.240 way that I think very few people ever have. 0:09:16.320 --> 0:09:18.480 And that was what really attracted me to the project. 0:09:18.600 --> 0:09:20.199 So what does that look like? I mean, how do 0:09:20.280 --> 0:09:22.200 you think about building a machine? And this isn't like 0:09:22.280 --> 0:09:25.120 shoes buried in the right pH this is a thing 0:09:25.160 --> 0:09:28.000 that's turning and moving. All our buildings that we make, 0:09:28.320 --> 0:09:30.480 we say, hey, these are stable, and these will be 0:09:30.520 --> 0:09:32.000 here for a long time. But by long time we 0:09:32.080 --> 0:09:34.719 mean whatever a few hundred years or something. How do 0:09:34.760 --> 0:09:36.840 you think about doing something at a ten thousand year scale? 0:09:36.920 --> 0:09:39.240 I mean, yeah, so we have some things that are 0:09:39.280 --> 0:09:42.360 like you know in Turkey that have lasted on this 0:09:42.400 --> 0:09:46.000 scale that we're man made kind of foundations, and Jericho 0:09:46.080 --> 0:09:49.440 has eight thousand year foundations. We have the Pyramids and 0:09:49.520 --> 0:09:52.880 stonehandjet five thousand years forty five hundred years. So we 0:09:53.000 --> 0:09:55.600 you know, we've built some buildings, but like you're right 0:09:55.679 --> 0:09:58.000 in saying that, a machine is a different thing like 0:09:58.040 --> 0:10:01.480 that has working parts and and this was a this 0:10:01.679 --> 0:10:05.320 was an early problem that we really looked at, which 0:10:05.400 --> 0:10:08.920 was how do you, for instance, have bearings that last 0:10:09.080 --> 0:10:12.920 for ten thousand years, right, And when I started on 0:10:12.920 --> 0:10:15.959 this project in the late nineties, I found the right technology, 0:10:15.960 --> 0:10:19.000 which was basically a ceramic on ceramic bearings, and they 0:10:19.440 --> 0:10:22.240 were developed for satellites that could operate in space with 0:10:22.320 --> 0:10:27.480 no lubrication, and so they're near diamond hard ceramics. And 0:10:28.280 --> 0:10:30.160 the other thing that they do, which is really great 0:10:30.200 --> 0:10:33.000 for in our case, is that since they're non metallic, 0:10:33.320 --> 0:10:36.920 they also separate dissimilar metals from each other, and so 0:10:37.760 --> 0:10:41.320 dissimilar metals in any situation basically cause corrosion and they'll 0:10:41.480 --> 0:10:45.280 build weld shut because of oxidization. They attract what's called 0:10:45.360 --> 0:10:49.760 galvanic corrosion by having a different kind of electrical potential, 0:10:50.559 --> 0:10:53.560 and so by these bearings solved all those problems. But 0:10:53.679 --> 0:10:55.959 at the time, in like nineteen ninety eight when I 0:10:56.000 --> 0:10:59.920 found them, they were made in very sparing sizes, forced 0:11:00.040 --> 0:11:02.000 out of lights, and they cost fifty thousand dollars each. 0:11:03.080 --> 0:11:06.320 But over the course of this project they have they're 0:11:06.360 --> 0:11:09.959 now in fidget spinners and they cost like five dollars, right, 0:11:10.000 --> 0:11:12.400 so you can get them, you know, their roller blades 0:11:12.440 --> 0:11:15.200 and you know, bicycle bearings and all this stuff. So 0:11:15.360 --> 0:11:18.080 we got very very lucky in the timing around when 0:11:18.120 --> 0:11:21.040 we were developing this project. So the clock itself uses 0:11:21.120 --> 0:11:24.120 all ceramic bearings throughout the entire thing. 0:11:24.320 --> 0:11:26.600 There's there's no metallic bearings at all. 0:11:27.200 --> 0:11:29.320 And that was really one of the kind of key 0:11:29.440 --> 0:11:32.680 things that is different about this than any other large 0:11:32.679 --> 0:11:33.800 machine that's ever been built. 0:11:33.840 --> 0:11:35.880 But a fidget spinner, if somebody builds it, they're not 0:11:35.920 --> 0:11:38.240 intending to last a long time. So how do they know. 0:11:38.520 --> 0:11:41.400 I mean, no one really knows if a ceramic bearing 0:11:41.440 --> 0:11:43.360 will last ten thousand years, right. 0:11:44.320 --> 0:11:47.920 I mean we know that, yes, these modern ceramic bearings 0:11:47.960 --> 0:11:49.760 have not been around for that long, but we do 0:11:49.840 --> 0:11:53.360 know ceramics have been around for We have ceramics for 0:11:53.480 --> 0:11:56.560 many tens of thousands of years, right, so just low 0:11:56.640 --> 0:12:01.400 quality ceramics that we found in places or on the 0:12:01.480 --> 0:12:05.320 order of forty thousand years old, so very high quality 0:12:05.400 --> 0:12:07.960 modern ceramics. There's no reason to think that they wouldn't 0:12:08.040 --> 0:12:11.640 last this long. And but but to your point, I 0:12:11.679 --> 0:12:14.319 think there are unknown unknowns when you do this kind 0:12:14.360 --> 0:12:16.760 of project, and I suspect, you know, there will be 0:12:16.840 --> 0:12:19.199 things that cause problems in the clock. 0:12:19.280 --> 0:12:21.200 That we that we don't really anticipate. 0:12:21.280 --> 0:12:23.040 And it could be something as simple as like something 0:12:23.080 --> 0:12:26.560 building a nest in the workings of it. Right, we've 0:12:26.559 --> 0:12:28.600 already like at the we have the site we're building 0:12:28.679 --> 0:12:30.840 the clock, and we've already had to deal with rodents 0:12:30.880 --> 0:12:33.280 and things like that because you know that the whole 0:12:33.280 --> 0:12:37.400 thing isn't sealed up perfectly during construction times and all that, 0:12:37.480 --> 0:12:40.880 so there will be things. And one of the things 0:12:40.880 --> 0:12:43.520 that we we did design into the clock is that 0:12:43.559 --> 0:12:46.720 it is designed for maintenance and it is designed to 0:12:46.760 --> 0:12:50.000 be changed over time. There's most of the clock actually 0:12:50.000 --> 0:12:54.400 sits quite still while nobody's there, and so you can 0:12:54.559 --> 0:12:57.040 do maintenance on it even while it's working. And then 0:12:57.240 --> 0:12:59.720 it's only when people wind it that a lot of 0:12:59.720 --> 0:13:02.760 the things that operate every day, something like the chimes 0:13:02.880 --> 0:13:06.480 or the dials, update only when someone is there to 0:13:06.520 --> 0:13:07.319 wind it locally. 0:13:07.440 --> 0:13:09.959 Oh, I see, And so it is now a daily chime, 0:13:10.040 --> 0:13:10.520 is that right? 0:13:11.160 --> 0:13:14.400 It's sort of, it's a daily chime if someone is 0:13:14.440 --> 0:13:16.840 there to wind out. So basically it's a reward if 0:13:17.120 --> 0:13:20.720 the clock is gets its power from to keep track 0:13:20.760 --> 0:13:23.320 of the time, from the temperature difference from day tonight 0:13:23.920 --> 0:13:26.000 that's harvested up at the top of the mountain through 0:13:26.200 --> 0:13:30.800 just differential air tanks that are up there. And then 0:13:31.760 --> 0:13:33.960 but if what it doesn't do is it doesn't show 0:13:34.000 --> 0:13:35.720 you the time when you arrive, it shows you the 0:13:35.720 --> 0:13:38.439 time of the last people that were there. So when 0:13:38.480 --> 0:13:40.760 you get there, it could have been yesterday and you 0:13:40.840 --> 0:13:42.960 might have to wind it just a teeny bit. And 0:13:43.240 --> 0:13:44.920 but if it was one hundred years ago or a 0:13:44.960 --> 0:13:47.320 thousand years ago, you could be there for hours to 0:13:47.440 --> 0:13:51.520 days updating the dials by walking around this capstin and 0:13:51.520 --> 0:13:55.960 winding things up. And so you get two pieces of 0:13:55.960 --> 0:13:59.440 information how long it's been neglected as well as you 0:13:59.480 --> 0:14:03.520 also get rewarded by having these unique chimes that happened 0:14:03.520 --> 0:14:06.920 that Brian, you know, and Danny Hillis designed this kind 0:14:06.920 --> 0:14:09.080 of pattern of ten bells that could be rung in 0:14:09.120 --> 0:14:11.520 a different sequence each day for ten thousand years. 0:14:12.600 --> 0:14:15.240 And for listeners who aren't familiar with the clock, give 0:14:15.280 --> 0:14:17.720 us a size of the scale and how it's buried 0:14:17.720 --> 0:14:18.320 in a mountain. 0:14:19.160 --> 0:14:22.600 Yeah, So we started with prototypes that were smaller, like 0:14:22.640 --> 0:14:24.440 the first eight foot tall prototypes is at the Science 0:14:24.480 --> 0:14:28.800 Museum in London. There's some later prototypes at the Interval 0:14:28.800 --> 0:14:31.800 here in San Francisco, but in two thousand and five, 0:14:31.840 --> 0:14:34.840 we started working with Jeff Bezos, who funded the full 0:14:34.880 --> 0:14:38.480 scale version of it, which is kind of if you've 0:14:38.520 --> 0:14:41.640 ever seen a Blue Origin launch from their Texas site, 0:14:41.720 --> 0:14:44.760 not the one in Florida, but the Texas site. The 0:14:44.800 --> 0:14:47.680 mountain range right behind that launch site is the Sierra 0:14:47.720 --> 0:14:50.880 Diablo Mountains and. 0:14:49.920 --> 0:14:50.920 That's where the site is. 0:14:51.000 --> 0:14:54.520 And we built it into a mountain, and we actually 0:14:54.560 --> 0:14:58.480 developed special diamond chainsaws and things. 0:14:58.280 --> 0:15:00.360 Like this to work under ground. 0:15:00.400 --> 0:15:04.240 And we built over two thousand linear feet of tunnel 0:15:04.280 --> 0:15:07.680 and five hundred vertical feet of tunnel that the clock 0:15:07.800 --> 0:15:10.920 goes into and cut these special stairs that allow you 0:15:10.960 --> 0:15:12.000 to walk through the whole thing. 0:15:12.480 --> 0:15:15.320 And were those cut by Was that robotic? This yair cutting? 0:15:15.480 --> 0:15:19.280 Yeah, So we adapted a kind of these diamond chainsaws 0:15:19.360 --> 0:15:22.000 or belt saws that are used for cutting marble like 0:15:22.040 --> 0:15:26.160 in Coorra Italy, but they've never really been roboticized. So 0:15:26.240 --> 0:15:29.480 we built a special robot with these amazing folks up 0:15:29.480 --> 0:15:32.400 in Seattle that have been kind of roboticizing all things 0:15:32.400 --> 0:15:35.520 for stone cutting. But they helped us make this custom 0:15:35.600 --> 0:15:38.360 robot that was like twenty six thousand pounds. 0:15:38.640 --> 0:15:39.680 It had a reach of like. 0:15:39.680 --> 0:15:44.320 Thirty six feet and could cut through solid rock about 0:15:44.320 --> 0:15:46.240 as fast as you would expect a chainsaw to cut 0:15:46.240 --> 0:15:49.680 through rock, and it slowly over the course of two years, 0:15:49.760 --> 0:15:53.840 cut a spiral staircase that is over four hundred feet tall, and. 0:15:53.800 --> 0:15:55.960 So just so people can picture this, it's a tunnel 0:15:55.960 --> 0:15:58.520 that goes straight down and the stairs are running along 0:15:58.520 --> 0:16:01.560 the outside of the cylinder, right. 0:16:01.720 --> 0:16:05.880 Yeah, So we cut the initial cylinder itself with a 0:16:05.920 --> 0:16:06.720 mining tool. 0:16:06.480 --> 0:16:08.280 That already exists called a raise boar. It's like a. 0:16:08.320 --> 0:16:12.240 Tunnel boring machine that's pulled up through a mountain. So 0:16:12.280 --> 0:16:15.040 you drill a small hole through a small by meaning 0:16:15.040 --> 0:16:17.360 eighteen inches, and then you hook up the giant drill 0:16:17.400 --> 0:16:19.680 bit to the bottom and you pull that up and 0:16:19.720 --> 0:16:22.080 then you excavate all the stuff that falls down out 0:16:22.080 --> 0:16:23.880 of the bottom. And it's a very efficient way to 0:16:23.920 --> 0:16:27.040 make a very smooth bore. And usually it's used for 0:16:27.080 --> 0:16:29.840 like ventilation shafts for a mine or for a tunnel 0:16:29.920 --> 0:16:32.160 or something like that, but we used it as the 0:16:32.200 --> 0:16:34.400 main shaft that all the clockworks were going to go in. 0:16:34.440 --> 0:16:36.600 But in order to have the people be able to 0:16:36.680 --> 0:16:38.640 walk through all that, we needed to cut a staircase 0:16:38.680 --> 0:16:42.320 around it and we actually wanted that staircase to start 0:16:42.360 --> 0:16:45.040 wide and get narrower and narrower, and so every single 0:16:45.080 --> 0:16:48.640 cut of that staircase was different, and in order to 0:16:48.680 --> 0:16:52.960 do that, a robot was definitely the right piece of machinery. 0:16:53.040 --> 0:16:55.640 Cool So you've got this huge bore that goes down 0:16:55.680 --> 0:16:58.400 into the mountain and at the bottom you install the clock. 0:16:58.440 --> 0:16:59.760 And how large is that clock. 0:17:00.480 --> 0:17:02.520 Well, it's not at the bottom, it's through the entire bore. 0:17:02.600 --> 0:17:06.159 So that clock is stretched out from the very bottom 0:17:06.200 --> 0:17:07.760 of that five hundred feet all the way to the 0:17:07.800 --> 0:17:10.600 top where there's a cupola that harvests sunlight and does 0:17:10.640 --> 0:17:14.119 give us a solar synchronization event. So every year around 0:17:14.160 --> 0:17:16.960 the solstice time, if on any sunny day around the 0:17:17.040 --> 0:17:20.760 first about two weeks around the summer solstice, if we 0:17:20.800 --> 0:17:23.800 get a sunny day, sunlight is focused down in this 0:17:23.920 --> 0:17:26.280 very Indiana Jones moment where it goes down in one 0:17:26.359 --> 0:17:29.240 hundred and fifty feet down into the largest sapphire lens 0:17:29.240 --> 0:17:31.600 ever ground, and then that goes into a thing that 0:17:32.400 --> 0:17:35.280 basically hits a black piece of metal in a chamber 0:17:35.320 --> 0:17:38.280 that expands and says, this is solar noon. And so 0:17:38.520 --> 0:17:41.199 if the clock has been drifting over the year or 0:17:41.240 --> 0:17:44.560 so that it's been operating, it needs this moment to 0:17:45.560 --> 0:17:47.760 correct itself. Now a human could also do that correction, 0:17:47.840 --> 0:17:49.560 but this is a way it can do that with 0:17:49.840 --> 0:17:50.880 just solar alignment. 0:18:06.359 --> 0:18:09.720 And so you mentioned that the clock is made so 0:18:09.800 --> 0:18:12.520 that maintenance can be done on it, but in theory 0:18:13.480 --> 0:18:16.119 it could survive ten thousand years on its own and 0:18:16.200 --> 0:18:17.560 keep functioning. Yeah, that's right. 0:18:17.600 --> 0:18:21.960 We did everything we could possibly do in order to 0:18:22.119 --> 0:18:26.080 make the materials and test all the things that move 0:18:27.119 --> 0:18:30.240 way beyond their design life of number of cycles, and 0:18:30.359 --> 0:18:32.480 as far as we know in material science, that will 0:18:32.520 --> 0:18:33.879 last as long as it needs to. 0:18:34.160 --> 0:18:36.800 Okay, So this is a great segue to the question 0:18:37.080 --> 0:18:42.879 of institutions. Human institutions, not just machines that can last 0:18:43.000 --> 0:18:46.280 a long time. So we've got many examples of this 0:18:46.400 --> 0:18:49.120 sort of thing, which you have been researching for years 0:18:49.119 --> 0:18:53.760 and years. You've been finding what human things last and why, 0:18:53.920 --> 0:18:55.840 So tell us about that. Yeah. 0:18:55.880 --> 0:18:57.800 So, I mean, as you might imagine, I kind of 0:18:57.800 --> 0:19:00.639 made a hobby of like figuring out things that have 0:19:00.760 --> 0:19:03.120 lasted on this timescale. And initially it was really about 0:19:03.160 --> 0:19:06.280 the objects, because that's what we're engineering. This clock or 0:19:06.480 --> 0:19:08.720 buildings and so you know, I went to this seed 0:19:08.760 --> 0:19:11.080 vault in Svallbard that was designed to last for a 0:19:11.160 --> 0:19:14.240 thousand years. I went to the Mormon genealogical vault in 0:19:15.000 --> 0:19:16.760 outside of Salt Lake, which also. 0:19:16.560 --> 0:19:17.959 Designed for a thousand years. 0:19:18.480 --> 0:19:21.040 I've been to historical sites, the pyramids, all these things, 0:19:21.760 --> 0:19:25.280 you know. The nuclear waste repository sites multiple the ones here, 0:19:25.359 --> 0:19:28.439 the ones in Oncolo designed for one hundred thousand years actually, 0:19:28.600 --> 0:19:32.040 and there's in both Finland and Sweden they have these 0:19:32.080 --> 0:19:33.920 sites that designed for one hundred thousand years. 0:19:34.359 --> 0:19:36.360 Once I cantanded, how do you design a site like that? 0:19:36.520 --> 0:19:38.119 It's just thick cement. 0:19:39.080 --> 0:19:42.000 Each one of those sites is uses very different principles, 0:19:42.000 --> 0:19:44.719 like the Yucca Mountain site in North America that we 0:19:44.800 --> 0:19:47.720 have designed for storing nuclear waste, which so far is 0:19:48.000 --> 0:19:51.159 currently shut down because of I think largely political reasons. 0:19:51.200 --> 0:19:54.720 But we've done an amazing amount of engineering and digging 0:19:54.760 --> 0:19:56.960 to build a site for a nuclear waste that we 0:19:57.000 --> 0:19:59.639 have not used, but that one is designed has actually 0:19:59.720 --> 0:20:01.360 has a it's a law on the books that it's 0:20:01.400 --> 0:20:05.560 a ten thousand year repository because I think it was 0:20:05.560 --> 0:20:08.720 because the problem the nuclear kind of waste problem was 0:20:08.760 --> 0:20:10.879 over a quarter million years, so they thought they'd be 0:20:11.040 --> 0:20:13.399 They're like, well, that's too long. So we'll say that 0:20:13.440 --> 0:20:15.080 we at least have to keep it safe for ten 0:20:15.119 --> 0:20:18.320 thousand years, which I think is interesting that it's exactly 0:20:18.359 --> 0:20:20.320 the same as this clock, which is, you know, about 0:20:20.320 --> 0:20:23.080 as long as you know. Also to just say a 0:20:23.119 --> 0:20:25.919 little bit about that timeframe, that it's not meant as 0:20:25.960 --> 0:20:28.760 a forever clock, which is, you know, you get into 0:20:28.760 --> 0:20:32.119 these kind of astronomic time scales or even geologic time scales, 0:20:32.119 --> 0:20:34.919 which are millions of years, but ten thousand years is 0:20:34.960 --> 0:20:39.120 about how long we've had agriculture and cities as humans. 0:20:39.160 --> 0:20:40.879 And so that was the idea that this is our 0:20:41.000 --> 0:20:45.600 human entropscene moment is ten thousand years in the past, 0:20:45.760 --> 0:20:48.160 and so we should be looking at least ten thousand 0:20:48.240 --> 0:20:50.399 years in the future. And if we think of ourselves 0:20:50.800 --> 0:20:53.320 more broadly as in the middle, at least in the 0:20:53.320 --> 0:20:56.840 middle of a twenty thousand year story, rather than at 0:20:56.920 --> 0:20:59.080 the end of a ten thousand year story, we might 0:20:59.160 --> 0:21:01.880 think about, you know, how we would be more responsible 0:21:01.880 --> 0:21:02.639 towards the future. 0:21:02.720 --> 0:21:05.479 And one second tangent, which is long now foundation uses 0:21:05.720 --> 0:21:09.600 when it marks years, it uses five digits instead of four, 0:21:09.920 --> 0:21:13.120 so you might say, oh, it's oh twenty twenty six, Yeah, 0:21:13.119 --> 0:21:13.720 that's right. Yeah. 0:21:13.840 --> 0:21:16.400 Very early on, when when Danny Hillis and I were 0:21:16.400 --> 0:21:19.760 designing that the dials for the clock, we realized that 0:21:19.800 --> 0:21:21.520 the dials are going to have to read with an 0:21:21.520 --> 0:21:25.080 extra zero in order to read past the year, you know, 0:21:25.240 --> 0:21:29.840 ten thousand, and so that we kind of used that 0:21:29.960 --> 0:21:33.719 as a mechanism to show that how far ahead we 0:21:33.720 --> 0:21:35.960 were thinking. And one of the thing Danny Hillis quickly 0:21:35.960 --> 0:21:39.800 found that there's a bug in Microsoft Excel because he 0:21:39.880 --> 0:21:42.000