History of Science and Technology Q&A (January 24, 2024)

Hello, everyone. Welcome to another episode of Q and A about history of science and technology. I'm still not back in my natural habitat, so things may not be quite as technologically perfect as sometimes. Let's see. I see a bunch of questions here. Memes asks, what are the key requirements in place for past scientific revolutions? You know, I think it's my impression is that sort of what launches a scientific revolution is some kind of new methodology, either new experimental methodology or new theoretical methodology. And on the experimental side, what often happens is that there is sort of an ambient growth of technology having nothing to do with science, just better lens grinding for people to make glasses, or better. Or electronics amplification of electronics or something like this, or computers for that matter, being built for some completely different purpose. And then at some moment it gets realized, gosh, we can use this technology for a method of doing science, typically maybe some experimental method or some modern computer experimental method or something like this. And then there's sort of this, this sudden moment when lots of things get opened up and there's kind of a lot of change in the science. I mean, I suppose an interesting question for today is if we look at the whole neural nets AI story, is that going to be something where. I think I talked about this, in fact, last time, is that going to be something where kind of that suddenly opens up sort of a new opportunities in science that didn't exist before because methodology has been developed for reasons having nothing to do with science itself? I think that's an interesting question. I consider that that's still up in the air. I've been working on a little bit of, you know, to what extent can you expect to use this new methodology to do things that have been of scientific interest before or, and this is often the case when sort of you open up new kinds of experiments or something or new methodologies, that the questions you end up asking end up changing as well. Because it could be the case that there are these, oh, there are these mysteries in science and nobody can solve them. Suddenly some new method comes along and it solves those mysteries. Or it could be that some new method comes along and you realize there are a bunch of new questions that you never thought to ask before. You can now ask. And we see that, for example, with the LLMs, questions about sort of the structure of language, the structure of knowledge and so on that we didn't really have, wasn't something that it seemed to make sense to ask before this, but now it does. I think one of the things that always helps a Scientific revolution is when the place where it's going into is fairly desolate. That is, if there's a field that is where lots of stuff is happening and everybody's very excited that it's all being solved and then some new method arrives, typically people are like, oh, we're making progress anyway. We don't need this new thing. And so there's less enthusiasm for it than there would otherwise be. And so I think that it helps when sort of everything else is burnt out and then suddenly this new spark of life, so to speak, that can then take off and can expand sort of into a vacuum rather than having to push back existing kinds of ideas in that area. So that's what be my impression about what's involved there. There's a question from Pierre about the, I think, the effective accelerationism movement, what history led to it or influenced it. I don't know all the details of this. I mean, I think the general sort of pattern of whether it's. There's sort of these general patterns of how can we make the world a better place? And there's sort of the question of can science help solve the question of how to make the world a better place? And I think that's a. The history of such efforts is not particularly good. I would tend to think that the reason that it's not good is because of my. My friend, computational irreducibility. The idea that the, the. The concept that you can organize the world, society, whatever else, in a certain way by putting. By sort of saying, we know how it's going to work. We just put this piece in place and that piece in place, and then we get to this point. The fact is that like other systems that have computational irreducibility, when you set down the system of rules, the system sort of has an irreducibly complicated behavior that it will follow that will always lead to unexpected, surprising things happening. So in other words, you say, I've got this great idea, it's this utopian scheme, and it's going to be, you know, and everybody's going to do the right thing and et cetera, et cetera, et cetera. But it turns out that the world in the. In, you know, sort of the fact that there's computational irreducibility to, for example, things like what happens in society means that your perfect scheme will always have a bug, and that bug will come to have all sorts of bad things happening. Now, it's sort of interesting the extent to which there are, in a sense, schemes like, I don't Know capitalism or democracy or something like this, which have been reasonably robust over the course of time, not perfect, but reasonably robust, and haven't, for the most part, splurted out into sort of complete horror. And it's sort of an interesting question of why. Why some of these sort of approaches that feel a little bit more inclusive, in a sense, a little bit less brittle, seem to work better than approaches that are more constrained and more like we know what's going to happen. I mean, I think that there's a sense in which. In which sort of the. Kind of the. The more. I don't know, quite. I suppose it's a little bit like when you have a material and the material can be very strong, but it's brittle, you force it a bit and eventually it cracks and just breaks. Whereas if it's ductile, if it flows, you can kind of, you know, things change a bit and the thing doesn't as dramatically break. It's more able to withstand change. And maybe that's a criterion to some of these approaches to sort of the organization of society and so on. I mean, I think that it would be a. It's kind of a. These questions about where sort of, you know, there have been. There have been approaches from, you know, Plato had approaches to, you know, this is how to set up the perfect system of government and so on. And he had things which today we would be like, no, no, no, that's horrifying. You shouldn't do it that way. And the philosopher kings and so on, and his version of the aristocracy, so to speak, of the power of the best, so to speak, which is not what aristocracy came to mean later on. But there are many things there which we now would think, well, that doesn't really work that well. And in more recent times, I don't know, Thomas Moore's original utopia and all kinds of approaches of this is how we'll organize things and everything will be perfect. That they haven't tended to end very well. I think, in this question about another feature of history is the. And this time it's different belief. I think that history. One of the features of history is that certain aspects of it do have a habit of repeating themselves. There's a question of why. I mean, in other words, some part of the repetition of history, presumably, is because we're the same species as we were 100,000 years ago. And there are certain features of us humans and the way that we interact and make societies and so on that have been unchanged over that period of time. And insofar as those are the drivers of the sort of patterns of history, those things will repeat. I suppose there are other. I mean, there's a certain. And perhaps that's the dominant effect. I mean, there may be other effects like even the fact that people know history and so have a habit of, of going back into the same ruts as happened in history. But I suspect that it's much more the dynamics of our species and sort of the necessary features of how the society gets built that lead to that kind of repetition. I mean, it's always interesting that there's sort of quantitative places where people make use of the repetition of history, particularly in quantitative finance, where you go collect a bunch of data on how things moved anytime in the last hundred years of, you know, this got more expensive and that got cheaper and this and that and the other. And somehow those things managed to be sort of successfully useful as detailed predictors of things today. I think that, you know, those kind of lessons of history that's a very microscopic version of how sort of lessons of history tend to tend to be repeated. I suppose one of the questions that always comes up is can there be a theory of history? In other words, could there be some kind of approach that says given that this and this and this is happening, it is inevitable that such and such a thing will happen. I mean, people say that about sort of the, you know, as empires get older, they build up more bureaucracy and that leads to certain kinds of things and so on. I don't know to what extent those are. Those are real effects and to what extent those are various kinds of sort of measurement biases and so on. I think it's certainly the case. One thing you can see in terms of the pattern of history is it's kind of interesting to see we have, in Wolfram language, we have all sorts of data on historical country borders, but pretty much all the ones known in history. And you can kind of make this animation of what did, for example, Europe look like as a function of time. And you do see this kind of very sort of physical or biological looking processes going on. One country will puff out over some period of 50 years or something and take over lots of neighboring ones. And then there'll be some moment where it pops and goes to practically zero as there's a big war and everybody overtakes that country. And then things keep going and so on. And it gives one at least the sense, I don't know if it's the qualitative sense that the dynamics of what's happening in that History are in some sense kind of like what you would expect in physical systems, like what you would expect in something where you could imagine some kind of predictability. But I suppose all these questions about sort of what's the perfect way to set up society. I mean, you know, even if you agree on what the objectives are, and that's always a difficult thing to do, in fact, not really a possible thing to do in the end, I think that sort of computational irreducibility bites you in terms of the ability to sort of poke this or that. But I was also going to talk about the. And this time it's different question, and the issue of whether sort of the technology we have today is whether whether sort of the future of society is kind of going to be sort of whether the changes that are happening today are somehow very qualitatively different from ones that happened in the past. My tendency would be to think. Not my tendency would be to think until the human condition has changed, which might happen if, you know, effective human immortality is achieved or sort of digital uploading of humans is achieved. Those kinds of things that will dramatically change the sort of common experience of the human condition. And that could make things really profoundly different. I tend to think that the motion of, well, this job becomes more important, that one becomes less important. There's sort of less work of this kind to do. I mean, if you look at the course of history, sort of manual labor work became less a thing that people had to do. Other kinds of work became more important, the things that count as work. You know, maybe it counts as work to be professionally playing video games or something, which is not something that would have counted as work at some point in the past. So I think as time goes on, what counts as work, what people think is worth doing, what is sort of scarce in some sense, changes. And I don't think that creates a qualitative change in kind of what one can expect for society. So a few comments there. Let's see. Boy, you guys expect me to know an awful lot of stuff here. Let's see. RBS is asking about binary code, its historical role in programming, connection to physical laws and so on. Well, let's talk about binary. So the idea that numbers can be. We're used to writing out our numbers in base 10 with decimal digits, but it's been known in one way or another. Certainly it's been known with some clarity since the 1600s, that you can as well write out numbers in terms of powers of 2 instead of powers of 10. And then the digits are just 0 and 1. I think that the I Ching and the hexagrams and so on of ancient Chinese kind of tradition was certainly an example of where kind of a binary like thing was constructed. Remember, when you're setting up kind of the binary numbers, if you set them up in a fixed number of digits, you have kind of this tree structure where you're saying, well, it's a one branch or a zero branch. And you keep doing that a few times and you've kind of constructed the set of possible numbers corresponding to the possible paths of the tree. And that's a little bit what happens in the hexagrams, I guess, of the I Ching, if I remember those kinds of things correctly, are the various different sort of binary configurations that you get from doing that. And that's been something that's been sort of out and about since antiquity, I think the time when sort of binary numbers kind of became notable. Like Gottfried Leibniz in the late 1600s became an enthusiast of binary numbers. Whether he knew about them from missionaries who'd gone to China, I don't think is very clear. He might have done. He famously, he was very proud of the binary numbers idea. And he had a medal that he wanted some duke or something to strike for him that you can find, I found in his archives, kind of the plan of the medal. And it says something like, Gosh, let's see if I can do it in Latin, X. Unum, or something omnia ducendit, something like that, which means, from one everything can be descended, so to speak, everything can be produced. And then he has just a table on this kind of metal thing of the numbers in binary written up to, I don't know, up to, I don't know, 15 or something, written out in binary digits. And he was very proud of this, actually, a number of years ago. I've been a Leibniz enthusiast for a long time. I know a number of other Leibniz enthusiasts. One is my friend Greg Chayton. And for, I think for his 60th birthday, I decided I would try and get him a reproduction of this medal. So we actually finally struck the Leibniz medal, so to speak, as hadn't been done in Leibniz's time, and put a thing about Greg's work on the reverse side of it. But so I had the opportunity to take our sort of struck version of the Leibniz medal, and when I was visiting the Leibniz archive a number of years ago and stick it on a page right by where Leibniz had drawn the original version of the medal. But so anyway, Leibniz was aware of the idea of binary numbers. They kind of flapped around with nobody paying much attention to them. The only place where they were notable was in kind of sort of recreational math. I think, particularly in the late 1800s, there were a bunch of kind of math puzzle books. And they quite often talked about sort of things you can do with digits and binary. For example, I mean, I know even when we were first building Mathematica in the late 80s, mid to late 80s, that there were mathematicians who were sort of buzzing around what we were doing. And they were like, well, what do you mean? You can have a function that extracts the digits of a number. Nobody cares about the digits of a number. They said numbers are just numbers. The fact that they're broken into digits is sort of irrelevant. Now, that idea. Well, as a matter of fact, that very idea was sort of thrown into incorrectness by things in dynamical systems theory. So starting around, well, 1900 or so, this idea emerged of sort of sensitive dependence on initial conditions. It was an idea that had existed in the 1800s. People like James Clerk Maxwell were aware of this idea. For example, if you take two spheres and you bounce them off each other, if they hit exactly on axis, they'll bounce back on axis. If they hit a little bit off axis, they'll bounce back more off axis than they hit, so to speak. And if you keep bouncing the back and forth, you'll amplify the small perturbation of how far off axis they hit. And that amplification is an exponential amplification. It sort of, let's say, doubles every time or something. Well, that same phenomenon, Henri Poincare, around 1900, observed that same phenomenon in celestial mechanics, that you can have a situation where if you change the positions of idealized planets just a bit over time, the positions that the planets will end up will become exponentially divergent. And then the idea of Lyapunov exponents. Lyapunov in the Soviet Union, kind of in studying control systems, sort of had this idea of you can sort of measure the rate of exponential divergence. Given that you make a small perturbation, what is the effect of that? And when things are stable, there'll be a negative Lyapunov exponent. The thing will damp out any perturbation you make. When it's positive, it will expand it. And that this idea. So why is this related to digits? Well, the point is that if you have two possible initial conditions and they're very close together. It's as if you have, if you write out the digits of the number, it's as if what's happening is when there's exponential divergence, it's as if you're excavating more and more digits, lower and lower order digits, so to speak, smaller and smaller digits in the number, so to speak, as time goes on. And so in the 1930s, in the origin of this thing called symbolic dynamics, particularly Marston Morse was one of the big people in that idea. Later Garrett Berkoff these are American mathematicians. The idea, one of the things that was studied quite a lot was the shift map, which is something where you're saying just write out a number in base two, for example, and at every step, let's say the Number is x, compute 2x mod 1. So in other words, find the fractional part of twice the number. And so what that's doing is it's gradually shifting the number one binary place to the left at every step. And so it's gradually excavating more and more digits. So that's a place in sort of the mainstream mathematics where digit sequences became a thing. And the shift map, the binary shift, the shift map on a binary sequence was well studied. That's in mathematics quite different direction. When people were starting to think about making computers, when Charles Babbage was doing it in the 1830s, 1840s, his computers were all decimal computers. So his computers involved little wheels that represented a decimal digit and he didn't have the idea of using anything other than that. I think it's sometimes claimed that Conrad Zusa, who was a German, well eventually sort of computer entrepreneur who did some early work on computers during the Second World War. It's a little bit unclear exactly what he did. I've tried to track it down as sort of the mystery of a computer that he built in his parents house. And the house was bombed, but the computer had probably been moved to an air raid shelter. Or maybe it's in the collapsed basement of the house and could be sort of archaeologically excavated. It's not clear. There's a single photograph of this computer. It's a mechanical computer, but it's claimed to have been a binary based mechanical computer. There's a reproduction that Conrad Zusa made with his son before he died and that claims to be the computer, although it doesn't look a lot like the picture, the one remaining picture that exists of that computer. But that's a claimed place which I don't think connected very well to anywhere else. It was a project done partly during World War II. And I don't think that was sort of what connected to other things. I think by the time computers were being made, things like, well, Atanasoft's computer at Iowa State and then the ENIAC done in Philadelphia, I guess, near Philadelphia, those were, by the time there were actual sort of digital electronic computers being made, those were definitely using binary as a way to represent numbers. I'm not sure that any moment anybody said, you know, dramatically, oh, binary is the thing we're going to do, that there were still ideas about using things other than binary. For example, there were some Russian computers made, I think in what's now Ukraine, I think in Kiev, that were ternary computers. And that was even done in the 1950s where there were computers where you could, you know, you would have plus 1, where you'd have 0 1, 2 or I think they were minus 1, 0 1. And I think that was supposed to have the advantage that when in a standard binary computer there's always the question, how do you represent negative numbers? The most common thing to do for integers is two's complement arithmetic. That sort of become the dominant approach to it, where the highest order bit in the number is flipped to represent negative and the rest of the number, the zeros and ones are reversed. But there is some awkwardness in that. And I think this ternary computer avoided some of that awkwardness in dealing with negative numbers. It didn't catch on sort of. It went to binary computers. I have to say that when I was a kid in the 1960s, there were, you know, kind of, you know, I remember one sort of, sort of, here's some Cool Things about Math book. This must have been 1967 or something like that. And that book kind of had, as a, as a, wow, look at this. It's different, the idea of numbers in different bases and, you know, pictures of octopuses and so on and for base 8 and things like this. So at that time it certainly wasn't the case that in sort of the everyday, you know, that everybody knew about number bases. It was still at that time kind of a slightly obscure piece of knowledge. Something that was sort of part of the engineering of computers, but not part of sort of everyday, the everyday world. I'm not sure at what point in educational process these days people learn about numbers in different bases, but there's certainly, I think people tend to be more familiar with the idea that their numbers can be represented by sequences of bits and so on. Let's see. I mean, the whole question of sort of the relationship between bits and sort of physics. Well, that has its own story. I mean, this question of whether physics is discrete or continuous has existed since antiquity and sort of got answered for matter at the end of the 19th century and for light at the beginning of the 20th century. Those are discrete. People kind of thought space might be discrete back in those days. I keep on learning about more examples of ways in which people sort of thought space might be discrete. It was very hard to make it consistent with relativity theory. And people kind of gave up and said, well, I guess space isn't discrete. And that sort of had happened by the 1930s or so, I suppose. My own efforts started again in the 1990s, realized that sort of the idea of space as a grid wasn't going to cut it, but space as a network most definitely was going to cut it. And that that was a way to think about things. Now, there had been intermediate ideas. John Wheeler was fond of this bit from bit kind of aphorism that was a kind of create all of reality from something which is fundamentally. Well, what I would say is computational. I don't think John Wheeler would have said computational. I'm not quite sure what he sort of. When he said a bit. I think it was more sort of a binary distinction. I mean, that's a thing that reappears many times in both science and actually sort of more mystical pursuits and so on. This idea that sort of the binary choice is what leads to everything, that was kind of the idea of the I Ching. That was the idea of. There were sort of subsequent efforts. Ramon Lull had some, although they were not quite so binary. But this notion that a sort of sort of a single choice distinction somehow makes everything. I know George Spencer Brown with his laws of form from the 1950s, 60s, I guess that's a form of essentially Boolean algebra is again kind of the sort of this distinction of 1 or 0, a distinction makes everything. It's worth mentioning in the midst of that. George Boole in the 1830s with Boolean algebra and this kind of mathematicization of logic. And logic, after all, had also had this kind of binary distinction of true or false. Now that binary distinction of true or false has led to all kinds of craziness and complexity in logic. Because, you know, the characteristic question is, is it true that not. Not P for any P is true? So in other words, if I say it's not not raining, does that mean it's raining? Well, it's complicated. That is the law of excluded middle. That is a kind of a thing that people argue about because if the thing you're talking about doesn't exist, what does it mean to say not, not P is equal to P? That's this kind of constructivism sort of question in mathematics. Can you assume a law of excluded middle? You know, back in the. In the 1920s and so on, the validity of the law of excluded middle became some kind of bizarre political issue. And for example, the Soviet Union as it was kind of. I don't know my dialectical materialism well enough, but there's kind of an incompatibility, I think, between some of that and this notion of the law of excluded middle. So there was sort of a we don't believe in that. And that became kind of a logic meets politics kind of kind of thing. I mean, when, you know, people always sort of hope that there are some things about what we as humans do that kind of transcend the details of kind of human society and politics and so on. And things like mathematics and logic are often held out as examples of that. And that's a case where that wasn't really the case. But, you know, this general notion of there is this binary distinction, true versus false, one versus zero. I mean, George Boole was the person who kind of merged the idea of true versus false and 1 versus 0 by inventing Boolean algebra, which was kind of like multiplication is like or, and addition is like, and. Except that true plus true, if that's one plus one, it doesn't equal two, it equals one. And George Boole had to kind of smoosh algebra to the ordinary algebra of ordinary numbers to kind of deal with this Boolean algebra. But there are all these places where kind of this notion that you can get everything from, this one sort of distinction arose, I don't think with great kind of precision. I think that the statement that yes, you can, you know, Leibniz saying you can get all the numbers from 1 and 0, kind of George Boole saying you can get kind of in the end, one said one can get sort of all the Boolean functions from. Well, that's sort of obvious, This idea that sort of everything can be got with that one distinction. I suspect that's what led to John Wheeler's sort of it from bit notion. But I don't know that history very well. I met John Wheeler only once when he was in his mid-90s. It was kind of a bit of an unfortunate meeting in a way, because he really was quite old. And I suppose in a piece of rather sad Moment. I mean, we were sort of talking about some of my science and so on, and kind of looks and says to me, you know, you know, who you should talk to about this stuff? There's a chap over at the institute, I was visiting him in Princeton, New Jersey, over at the Institute in Princeton. He said, there's a chap over there named. He said it kind of slowly, you know, John von Neumann. And I was like, yes, I know he died before I was born. And I think this is a thing that even at my not terribly advanced age, I think one of these things about sort of remembering what happened when and how many years passed, and is this person. That was a severe offset of kind of dates, so to speak. And I can see how that would happen. Let's see. Gosh, lots of different questions here. Eli asks, can I talk about the history of cybernetics, second order cybernetics, and its current connections to the observer and the ruliad? Well, so cybernetics was kind of originated. That word was originated by Norbert wiener in the 1940s. And kind of the notion was it was all about controlling things, this how do you control stuff. Cybernetics comes from the, well, Latin word gubernator, meaning a helmsman. It's kind of like, how do you steer the ship, how do you keep things under control, so to speak. There had been this notion of a governor that had come into existence, particularly with steam engines. I think James Watt had invented a form of governor where things are spinning around quickly and you have weights that, as they spin quickly, the weights sort of get pulled out by centrifugal force. And that pulling out causes a valve to close and that slows things down. And so you have this notion of having a control system, having a way of keeping things at a particular level. Kind of like in modern times, a thermostat might switch on air conditioning or heat, trying to keep you at a particular temperature. Now, what's always tricky about those things is if the thermostat just says, oops, it's 71 degrees, I better switch on air conditioning. Oops, it's 69 degrees, better switch on heat, the thing will start oscillating like crazy. And so a lot of kind of the technology of control theory has to do with making controllers that sort of are sensible with respect to those kinds of things. But this notion of controllability and how do you set things up so that they sort of stay within a certain required boundary? How do you make an autopilot that will keep the plane at A certain altitude and so on. That whole set of ideas had originated with steam engines and things. I think James Clerk Maxwell worked on these things also in the late 1800s. There had been kind of a bunch of engineering done on that. The kind of theoretical ideas behind that I think really originated During World War II, I think particularly in connection with anti aircraft guns. One of the issues there is the plane's flying and you've got to, you know, you shoot something which you're trying to hit the plane, but the projectile is moving at a certain speed, the plane's moving at a certain speed, you have to kind of aim in front of it. And there's sort of a feedback loop that is needed to achieve that. And I suppose that's the big thing about these control systems is the notion of a feedback loop. The notion that you observe what happens and then you use that to control what you should do next. You observe that you put the heat on, the temperature goes up, okay, it went up this amount, then switch the heat off and so on. So the theory of these kinds of things became very much a theory of complex analysis using complex numbers and so on and doing all sorts of things that Norbert Wiener had learned as a mathematician. But when he got kind of pulled into the anti aircraft gun thing in World War II, got to apply those methods and sort of started making a theory of control and that theory of control for sort of when the thing is very close to being in control, you can use a linear theory. You can assume that there's small deviations which can be approximated as linear changes. And. And that gives you a whole set of sort of doable math that became control theory. Meanwhile, there was also a much kind of more sort of ambitious version of cybernetics. That was kind of the idea that somehow this notion of control would be also the how animals and machines sort of figure out what to do in general, the general kind of planning, what to do, how did they respond to stimuli, these kinds of things. Not the narrow sort of linear control of we've got a thermostat and we're trying to maintain a certain temperature and so on. But that led to, by the 1950s, this whole cybernetics movement of we're going to figure out how everything gets controlled and works. It sort of segued into what I would say in more modern times might be related to kind of complexity theory and things that I sort of developed in the 1980s, although in a rather different way. And there were a series of famous conferences or the Macy Conferences Organized well, funded by the folks who did the Macy's department store. That was a time when. That was probably 1952, 3, 4, that kind of timeframe where lots of people getting together to talk about this problem of sort of how do all these systems, like animals, get to do what they do? People like Warren McCulloch, Walter Pitts, the inventors of the neural net from the 1940s. Were there people like Margaret Mead, anthropologist, I think John von Neumann was there at some of these. There's a whole collection of characters. And there's also people like a chap called Ross Ashby, who was sort of these. There were people and a guy called Conrad Waddington, I think, was part of this picture. Some of those people were people who were more on the sort of let's make a mathematical theory of biology kind of side of things. Then there were people who were more on the. Well, I suppose neural nets was more in the psychology, psychiatry direction at that time. And then there was some kind of theory of automata, kind of Turing machine, finite automata type people. Those were all sort of mushed together in this kind of cybernetics movement. And it didn't really work out very well. They didn't really figure out a lot. The part of it that was control theory branched off and became sort of a very practical thing that's used all over the place to do. You know, there's all kinds of stuff in Wolven language for doing control theory. And it's just a method that's used in lots of engineering disciplines and so on. But I think that was kind of the main ambition of cybernetics, had kind of died off by, oh, gosh, I don't know, Sometime in the 1960s, probably. I know when I was a kid, there were. People were still sort of talking about cybernetics. And obviously the term cyberspace and cyber everything came from that same route. I remember there's a program, maybe it's leaving on. There's this Doctor who program in British television which might be classified as science fiction, might be classified more as comedy, I suppose. But I think it's changed over the course of time. But I remember this must have been, oh, gosh, 1965, 1966, something like that. They had a whole. Some of the villains for a season or something. Were these characters called the Cybermen, who were kind of cybernetic, robotic kind of future instantiations, I think, of our species or something like that. Or perhaps they were from outer space or whatever, not sure. But in any case, that was a Kind of a typical of the time kind of cyber cybernetics. You know, we will. We will computerize everything, we will mechanize, automatize everything. But I think by the. I would say by the 70s even, certainly by the 80s, by the time I was working on kind of how complex behavior arises from simple kind of rules, cybernetics wasn't really a thing. There was a branch of it that survived in the Soviet Union and there were sort of institutes of cybernetics there which I would say were on the more mathematical end of things and ended up probably doing more surviving work than some of the other parts. One of the pieces that came out of that whole cybernetics discussion was this idea of second order cybernetics. Sort of original Cybernetics is kind of a. I suppose as a system that. Well, second order cybernetics is kind of a system that is a. That has the observer of the system somehow inside the system and that as opposed to sort of first order cybernetics where you've got this, I suppose, where you've got a system doing its thing and you've got a controller of the system trying to keep the system under control, that's sort of first order and second order is the controller, the observer of the system is somehow inside the system. And there are all sorts of lovely pictures of eyes looking at eyes and so on. John Wheeler had drawn some of those. I don't think he really was thinking so much about second order cybernetics. I don't think second order cybernetics got very far. I mean, I think it was mostly a sort of a concept of. That's the next thing one has to think about. Certainly in recent thinking that we have about the ruliad and the idea of sort of the observer being part of the ruliad. I've tried to sort of dust off some things that sort of historical documents about second order cybernetics. I have to say there's a particularly weird book. It's a very big book. Who's the author? Oh, gosh, maybe Gregory Bateson may have been one of the people involved, I'm not sure. But it's a book about kind of cybernetic everything. It's a very weird book to me at least. And it has sort of pictures of second order cybernetics, but I don't think it really got anywhere. I don't think there was ever a sort of a formalism. And by the way, this whole idea about sort of control and observers and all this kind of thing, this idea of feedback control and so on. It was sort of rediscovered in what's now called reinforcement learning, which is sort of a big thing in modern machine learning and so on those things. It's kind of funny because it often happens in science that there are ideas which end up with sort of parallel paths of evolution. And there are completely different names for the exact same concepts in these two different fields. And that's happened in that case. Let's see. Here. Well, Rather different. Let's see. Hold on a second. Was the cyber definition before the definition of robotics? I don't think so. I think robotics came from this word for a worker in the Czech language in the 1920s, late 1920s, this movie, I think Rossum's Universal Robots or something. Robota. I think there's the word in the original language and I think that was a thing that. So that word existed by the 1930s. Robots. The term robotics, I think was invented much more recently. I think a chap called Diebold invented that in connection with kind of factory automation. And I think Isaac Asimov kind of popularized robotics. I'm not sure he ever even knew the person who'd invented that term. And maybe he independently developed it from the original robot word. As far as cybernetics is concerned, I'm pretty sure that was 1940s wiener. And I think that was a quite separate origin from the whole robot thing. The term automaton is much older. The automaton word comes from probably existed even in the 1600s. Certainly the notion of an automaton made from kind of clockwork, this kind of toy that would do these things in a music box that would have, you know, figures that went around or, or, you know, some, some. I know there was a. I think it was in the maybe 1700s, a duck that was. Made the rounds, I think in France that had, you know, the duck walked and quacked and things like this and had a lot of duck like behavior. And it was kind of described as an automaton. And that was. I think the term automaton was used. Gosh. I think it was used in some philosophical works by the 1700s for sure to describe kind of what would happen if sort of it was mechanically determined how a system would work. So that term is older than these terms robots and cyber and so on. Lazaro is saying you are forgetting George Spencer Brown for second order cybernetics. Yes, you are right. I'm remembering that George Spencer Brown did talk about this. George Spencer Brown was the originator of so called laws of form. I think I mentioned those earlier it's kind of a notational reformulation of logic of Boolean algebra that sort of packages a bunch of things into the notation of writing, sort of hook things and so on. But it's very big on this idea of start with the distinction, start with this kind of true false type distinction or just a distinction and sort of build things from it. I once talked to George Spencer Brown on the phone and he was a wild character. He's a person who, you know, if you read his resume, he was kind of, you know, he was a pilot, he was a psychiatrist, he was a electronic designer, he was a. Oh, gosh, so many different things, I'm not sure. And you know, one of the things he produced was this very kind of very, very clean set of ideas about laws of form. Very sort of thin book and that, that led to quite a. A movement. I'm not sure that it's engaged as much as it could have done with sort of the formal machinery of working things out mathematically. I mean, I've tried to do that a little bit with what he did. I've not found it particularly useful because it really is equivalent to Boolean algebra. And Boolean algebra is a fine, convenient thing to work with. I am remembering that George Spencer Brown worked on second order cybernetics, but. But I am not immediately remembering what he did there. So I'm sorry about that. Yes. You're asking, have I used the ideas expounded in Laws of Form? Not really. I mean, I think it's sort of a formal system like other formal systems like you could define Boolean algebra. You know, one of my little achievements back in 2000 was to find the very simplest axiom system, single axiom, actually, that leads to Boolean algebra. And George Spencer Brown in effect has an axiom that leads to the equivalent of Boolean algebra. It's much more complicated than the one I found, but that's the thing there. Let's see, there's a question from Robot Watcher. What was the significance of the ENIAC computer? So the ENIAC was, what was it? Electronic numerical, something rather integrator and something I think was what it stood for. It was produced, I think, at the Moore School in Philadelphia. And it was somehow in connection with the Aberdeen Proving Ground, which was a ballistics research operation very active in World War II. But the ENIAC was produced, I think, in 1946, right after the end of the war, and it was J. Press Eckert and William Mauchly, I forget Eckert and Mauchly were the two sort of principal people who were kind of engineers and then they kind of pulled in as a sort of consultant. John von Neumann, who was sort of a mathy person, and he was a person who kind of could mathify anything. And so he ended up writing up nice, fairly good sort of summary of sort of how to think about sort of this electronic computer idea in sort of mathy terms. And it's kind of charming. He writes about. He really describes things a very biological way. He's always talking about organs, by which he means kind of gates, what we would now call gates, electronic gates, he's talking about as organs. And he has a very biological nomenclature for things. But sort of this idea that you can kind of connect Boolean algebra to electronics. That idea had existed like Claude Shannon's, maybe master's thesis at MIT had been about that topic of connecting switching theory, electronic switches to Boolean algebra. That was slightly earlier than that. But so what emerged from in those times was sort of these design for sort of the blueprint for a stored program electronic computer. Von Neumann certainly made many kind of mistakes, in a sense, conceptual mistakes. I think that it's always interesting to study how smart people make mistakes because it's a. It's a lesson to all of us, I suppose, in what you might think at one time that might not be true in the future. So von Neumann, one of the things he said is the US will only need five computers, and he could only foresee the kinds of things you might do with a computer to the tune of there might need to be five of them in the U.S. obviously he didn't think of a lot of the things that eventually got used for computers got used for. The other thing he famously said was there'll never be a program written. That's more than, I forget the length, you know, 200 lines or something long, which of course isn't true either. That one I find interesting. I'm pretty sure that what von Neumann had in mind there is that he couldn't think of mathematical proofs that ran to anything much longer than that. But the mistake he made was he didn't think of subroutines. He didn't think of. Of sub functions, so to speak. He didn't think. Now he should have thought of that, because if you've been thinking clearly about proofs, proofs, the proof of one theorem depends on other theorems or it uses lemmas, those are kind of like the subroutines of proofs. And so the unrolled thing with all of Its lemmas and so on could be incredibly huge. You know, it could be sort of the giant textbook of all of mathematics or something. So he should have been able to think of that. And that one, I fault him for a little bit more than. There won't need to be more than five computers in America, but I think from the eniac, the basic design of the eniac, while John von Neumann made this computer at the institute in Princeton, I think the Johniac, I think it was called, I think there was a computer. There started to be these small numbers of computers that got made in different places. IBM really got into that business along with Univac and Sperry. Oh gosh, trying to remember, Honey Honeywell was also in that business of these were, I think Univac was the direct descendant of the ENIAC project. That was the company. And meanwhile IBM also got into the business and people started making computers. But they were pretty artisanal things. Back in the early days in the 1950s, it was like there were a limited number of these things. And there's one at Los Alamos, there's one in this place, there's one in that place. And it wasn't really a popular thing. And what came much later, I mean after IBM started sort of more mass producing computers then, you know, there were these moments when it was like the first computer in India, the first computer in this place, that place, you know, those were. It was a big deal when the first computer arrived in some particular country and so on. And I think the thing that, so my impression is that the sort of design, the ENIAC was made with vacuum tubes. It was only a few generations later when the first transistor based computers were made. There were, I think the memory of the ENIAC was I believe was mercury delay lines. There were also British computers, by the way, that were being made at the time. I mean, it was an interesting thing because there were some countries that were making computers, the us, the uk, These were making computers. I don't know what point the Soviet Union got into making computers, but sort of the, you know, it's one of those features of business history, I suppose that the US really made it into a significant sort of computer powerhouse. And for example, the UK did not. And nor did Germany, which is another place that had some computer production. And the, I think the thing that. So one of the things that was sort of an issue for these computers is how what was their memory? What was their storage mechanism? You know, in modern times we use semiconductors and we have, you know, semiconductor Memory and so on. But in those days that, that didn't exist. One of the methods was the, the, the mercury delay line where you would fill a tube with mercury and you would send a sound wave through that tube to represent a 1. For example, you'd send a pulse through and it took a while for the pulse to travel down the delay line. You know, this mercury, this sort of thing of mercury. And then you'd pick it up at the other end, you'd amplify it and you'd send the pulse back through. And so you kept your memory. It might be a little bit like the way that neurons keep, you know, that we keep our short term memories, I don't know, in brains. You keep it by sort of circulating it through this thing that a certain delay. And that was kind of the early approach to memory in a bunch of early computers. What got invented later by Jay Forrester was core memory. And that was the idea of having magnetizing little donut shaped little toruses of magnetic material and having these wires that would go through those toruses and be able to either magnetize or demagnetize the torus and be able to also read whether the torus was magnetized by what happened to that electrical pulse. And for example, the first computer I used in 1972 had core memory. And that was sort of a thing until semiconductor memory became the standard approach. But so I think, you know, if you were to trace, I haven't ever seen this, but I'm sure one could do it sort of the genealogical tree of early computers. I think you'd find that the ENIAC was the source of a whole bunch of branches of that tree going into, well, IBM and Honeywell and et cetera, et cetera, et cetera. I mean, there subsequently was the usual kind of clash of titans, business interactions. There were also many sort of smaller computer companies because as I said at the beginning it was an artisanal business, so to speak. And it only became a mass production business somewhat later. And I think that during the time when it was kind of artisanal, it was readily possible for small companies to make computers. Only once there were serious production lines did it become more of a big company kind of thing. I mean, I remember when the first workstation computers existed, that was 1981 or so. You know, the first ones, Sun Microsystems, for example, I remember visiting them right at the beginning. They were wire wrapping, you know that they had these boards, they had microprocessors, but they had to actually sort of build the computer almost Artisanally at that time. Let's see. So question here from Gorgola saying Isaac Newton was known as the father of modern physics. How might he view advancements since his time? Would he have anything to add? I think he was a bright fellow. So I would think there's always the question of how long would it take old Isaac to learn physics as physics has emerged. I think the thing he would have the hardest time with is the abstraction in mathematics. The fact is, in Newton's time, Newton was very much a cleaner up of kind of mathematical presentation. I mean, he really was the person who kind of nailed down the idea of variables in mathematics, you know, x and y and things like this. And he also, well, he invented a bunch of notation like fluxions, the X dot kind of notation to mean a time derivative. Some of his notation there didn't really catch on. Leibniz is sort of, in some ways arch nemesis. The notation, which in some ways was worse of the D by DX type notation did catch on. But the thing. So Newton was, in a sense, the thing that Newton was very successful at was figuring out just what could you mathematicize. There were many questions in science. There were questions about the orbits of planets, there were questions about turning lead into gold. There were questions about all kinds of things. And what Newton managed to do in his laws of mechanics particularly was he got this little slice that was mathematicizable and it's not obvious that that would be there. Like, if he had chosen to study fluid mechanics, he would have had a really hard time coming up with sort of simple mathematics to represent what was going on. I mean, that whole idea of using simple mathematics to represent sort of physical things. Galileo had been an early user of that in discovering things like the quadratic relationship between distance gone and time for an object falling under gravity, things like this. But Newton really took that to the next level and by sort of carving out this one area of mechanics where there was sort of a thing not acted on by any other force. The idea that you could even identify a thing not acted on by any other force wasn't obvious. That was something that he had to make, that idealization and then go with it, so to speak. I think in gosh, what would Newton make of more modern developments? I mean, I think if I were going to try and explain, you know, if you imagine the sort of time travel thing of you meet Isaac and it's probably, I wonder what he would have been like. He was, apparently people viewed him as quite dour. I mean, he was a bachelor all his life. And although in his later years when he moved to London, his niece, Catherine Barton kind of kept house for him and so on. But I think the, and I think that that was in fact her, her husband, I think was. Was quite well connected in the government. And I think that was how Newton ended up getting his gig as Controller of the Mint. And. But in any case, the thing that. I think he was a rather, you know, he's said to have been kind of a Dawah fellow and. But yet it's also said that he went to the local kind of pubs and was very jolly there and I think, you know, and mixing with the townsfolk type thing. So, I mean, maybe he was one of these people for whom, I mean, he was an early professional academic. He was the Lucasian professor of Mathematics. He was like the third one in Cambridge. It was a fairly recently established endowment to pay for a person who would spend their life studying mathematics, so to speak. I mean, the concept of a professional academic was not that well established. And I don't know what the, the interaction with his sort of fellow academics at Trinity College, Cambridge, what they were like. I mean, many of those people will have been theologians. Newton developed his own kind of theological ideas that I don't know much about, but which were somewhat heretical in their time. And I don't know what his interactions about those things were. And maybe he just didn't like the academics and he was just fine with the local townsfolk, so to speak. I'm not really sure. But I think in terms of what. It's an interesting question, what ideas? If you were explaining modern notions of physics, I think, for example, the idea of atomism, the idea that atoms exist and that there are discrete things in the world, I don't think would have been any surprise to, to Newton. I think that he had kind of got sort of proto versions of kinetic theory of gases, although he made some mistakes there to kind of invent a universal law analogous to his universal inverse square law of gravity. He had an inverse fifth power law, I think, for interactions between molecules. That turns out not to be correct and confused people for a while. But I think that the idea that there would be discrete things wouldn't have been surprising to him. I think the kind of the. Well, so what would he have thought of computers, for example? It's an interesting question. I think if one had sort of, you know, it's kind of a fun sort of thought experiment. Imagine you show a laptop to people at different times in history what would they have thought it was? I mean, from the, you know, Pythagoras, who might have thought it was, you know, disembodied human souls in a box type thing. So Newton, and I think his best model for something complicated going on would probably have been chemistry. Probably it would have been, you know, this must be a chemical process. The idea that this was ultimately driven by electricity would be surprising to him. And I think the idea of sort of precision electronics, precision manufacturing, and the idea that you can pack a lot of stuff into that little microprocessor, that would have been quite weird to him. I mean, this was obviously before Newton lived, before the Industrial Revolution. It was at a time when there wasn't a lot of. I would say, well, the sort of. The machinery that was built, I suspect, was in large measure, sort of. You have a thing, it's a chariot, you know, it's a carriage or something. It's got wheels. It does this. Rather than you have one machine that is responsible for making another machine, that's responsible for making another machine and so on. I think that's a more recent concept. I think that's more of an Industrial Revolution concept, which is sort of after Newton. I think the notion that you can kind of set up a formal system, which is sort of something like Wolfram language, where there's sort of a formal set of rules for how the thing operates, and then you sort of see it run, or something like a cellular automaton, where you've got some set of rules and then you see it run. I don't know how Newton would have responded to that. I think that sort of wasn't the kind of thing he was thinking about. I mean, he was much more thinking about. I mean, Leibniz, for example, was much more into those kinds of things. Newton was much more of a geometrical kind of thinker. And he certainly modeled his Principia Mathematica very much after Euclid's presentation of nuclear's presentation. It was like, there are these axioms at the beginning, and from this we deduce geometry. So Newton very much wanted to structure his deductions about mechanics from certain axioms, certain laws of mechanics. He would then deduce kind of everything else about mechanics. And he presented things as scholiums and theorems and things like this. So he was not. I think even though the. The intellectual structure of that is you start from rules and you deduce things, it isn't the same kind of discrete, formal, writing it down with notation kind of approach that Leibniz was, I think, closer to. Oh, my. All sorts of questions here. Well, let's see. Ivy asks a question. What was the history of small business versus big business? Was there a shift in history where one overtook the other? Well, you know, government's a big business in a sense. And a lot of the early sort of. There were early joint stock companies from the 1600s and so on. In fact, the aforementioned Isaac Newton famously invested in the South Sea Company, right. Which was a. Which was an attempt to. It was a stock, you know, a company where you could invest in it as a. And it was to sort of. It had kind of a license from the government to kind of exploit the things in the South Seas with all kinds of. All kinds of terrible stuff. But it was also kind of a stock fraud. And I think one of the more famous things that happened was that the king in England and remember was sort of in on the whole thing. And I think maybe was even the chairman of the South Sea Company. And at some moment South Sea shares were going up so well that they were like. They were the great thing to invest in. And so the king kind of people had treasury bonds and where they were lending money to the government, but there was a lot of debt from the government. And the king kind of very kindly offered to let people exchange their treasury bonds for shares of the South Sea Company. And needless to say, the South Sea bubble burst and everybody lost that money, including Isaac Newton lost some money in that. But the sovereign debt of the country had been sort of spirited away into South Sea Company shares. I have to say, it reminds me of some things that people have proposed with cryptocurrency in more recent times, but in the story of history does repeat itself. One can see that. But in any case, I think that these sort of governmental companies, which, by the way, there are certainly countries in the world today where their governments operate much more like large companies than sort of controlling a lot of industrial production and so on. I mean, that's the story in some places. And sometimes there was kind of nationalized industries where the government controlled the post office, the telephone system, whatever it is, the railway system, whatever. And then sometimes those got sold off into. Into private companies. But I think this, this notion that sort of there can be a big company, it was, it was, you know, back from the, the 1600s, for example. That was a thing that was very much entwined with kind of licenses from the government and the government letting you do it and the government kind of being involved. The, the idea that you could be sort of a thing independent of the government. I think Lloyds of London Insurance Company is one of the older kind of independent sort of you can make a company. It was, I think, a pretty big company from the beginning because I think it was, you know, in the insurance business, the goal is to spread the risk over a lot of different assets and so on. I think that. So that was a thing, but. But in the. I mean, the other thing to realize is the idea of a company in which people could be investors was. Was not, you know, that was probably what the 1600s, that was invented. I think that that was a possibility. Wasn't. You know, it was either the state, the government, or it was just, you know, a family owning a blacksmith shop or something like this. Not a thing where you could have, you know, a private equity firm that had bought up all the blacksmith shops in the. In the. In the town or something like this by having sort of investors owning things and so on. That really wasn't a thing. I think people were just doing business for themselves, so to speak. There were also things like the guilds, which provided some, well, both kind of a protection and training and so on, and sort of a gaggle of if you're going to be a licensed blacksmith or these days it'll be called licensed blacksmith, but I mean, what's it called? Gilded blacksmith. I'm not sure that you had to be part of this thing. But those were not companies. They didn't have, I think, ownership. They may have had some dues that were paid, but they weren't the same kind of thing. So I think kind of at this, that what enabled larger companies was this idea of there being investors and that one could issue stock in these companies and so on. And there were lots of terribly scurrilous things that happened, which in the US led to things like the SEC being created in the 1930s and so on. I mean, I think there were. There were sort of. It was kind of a huge scam, huge kind of potential for scams and so on in that kind of world before that. But I think this thing that tends to happen in industries of sort of. They usually sort of things aggregate into bigger companies and then there are little things that get left over. I mean, it's a little bit like the map of the world. You know, countries get bigger, but there are still little countries that are left over. And maybe the big countries, you know, don't survive forever and so on. I mean, that's become less of a trend of sort of take over large parts of the physical world. It's more take over large parts of the kind of commercial world. But I think, you know, so I think that's a. Is there, will there be a trend back to smaller companies? You know, I think I've seen in recent times, you know, I've had a company that's been sort of an independent, no investors type company for the last 36 years, which allows me to do all kinds of crazy things that I think are interesting because I'm not sort of beholden to a collection of purely financially motivated investors. It's, I think the trend among technology companies for example, has been get investors, get venture capital, kind of build it up big that way and that has a certain set of constraints associated with it. I have noticed in the last few years that more and more people are asking me, sort of Silicon Valley folk and so on are asking me wait a minute, how do you build a company that doesn't have those, that isn't an investor based company and that maybe stays much smaller and doesn't kind of puff out to be sort of the biggest gorilla it can be, so to speak. I think that there's sort of a question of what drives big versus small companies. I would say one thing is sort of the cost of getting started. That is, if the thing you're going to do involves building a factory and the factory is going to cost X number of tens, hundreds of millions of dollars. Well you can't be that small if the thing you're doing involves in modern times doing $10 million of neural net training. Well, you can't be that small and still do that. But at different times in history and as there's more automation and so on, I think it becomes much more realistic to have small companies again. In other words, at times when there isn't so much automation where you have to have a lot of investment in, in stuff and people and so on, you can't have such small companies. When it's just a look, I can just, you know, with a one man band, so to speak, I can operate all these things and set up all these pieces of software and so on. I mean in a sense our company is 800 people or so. But, but is, you know, what I've tried to do is to automate as much as possible so that with a fairly modest number of people we can do as much as possible. And that's kind of a, you know, as you, as you scale that in different directions, once you make use of the fact that there is automation for Many kinds of things. You can scale down the size of companies. Now there are forces that scale up the sizes of companies, like just the way that the investment markets work and so on. And the fact that, oh, it's not really worth raising venture capital unless you, you know, unless it's a 5 million dollar blob or a $10 million blob or something. The fact that, you know, by the time you, if, if you're kind of setting things up, you can, you know, and you've got the money to do it, you can hire lots and lots of people. And that sort of creates at least the appearance of a, kind of a barrier to anybody else entering such a market. So I think, I think probably automation drives down the sizes of the needed sizes of companies. Same is true with kind of the machinery of the world. Like, like if you needed to have, well, without credit cards, for example, if you needed to take payments where everybody is calling the people on the phone and sending letters and getting letters of credit in different countries and all this kind of thing, then you need more people to do that and you have a sort of certain minimum size. But by the time it's all, you know, credit cards with online processing and so on, you don't, you don't need a, you know, there's no, there's no lower limit to the size of, of that. You need to be able to operate those kinds of things. All right, maybe one last question and then should wrap up for today. Lots of questions here. They're a bit complicated to answer. Well, let's see. Okay, there's one question here from Koiki. Can you talk about the history of traveling to space? How come moon visits aren't more frequent? Because it's hard. At least right now it'll get easier. I mean, I think that the original, you know, the first visit to the moon was like a Russian sort of thing that was just sort of thrown at the moon that arrived like October of 1959. I think that was, you know, a couple of years after Sputnik and was just like, let's, let's, you know, throw this thing at the moon and it'll just go splat on the surface of the moon. And then there were soft landings a bit later than that. And then kind of, you know, in the US for sort of, I would say, political and national pride reasons. Kind of when John F. Kennedy sort of launched the let's go to the moon type thing. That was, in a sense, it's kind of like a let's build a pyramid Kind of story. It's one of those things that actually is pretty good for national pride and so on. And in that particular case, probably, I don't know, probably building pyramids led to all kinds of technology developments in that time. But certainly launching the Apollo program led to all kinds of, or became the Apollo program became, you know, created all kinds of technology that was useful. And then the big demo happened in whatever July 20, 1969 or whatever of, you know, land a person on the moon. Interest waned very quickly after that by Apollo 13, you know, by Apollo 11, everybody was like watching and cheering. And it was sort of a, it was interesting because it was kind of a worldwide pride event in the end, which was, you know, there aren't so many of those. But already by Apollo 13, I think people had kind of lost interest. It's like, okay, we sent somebody to the moon, great, we're done. And only when Apollo 13 had its problems, then people were like, yeah, it's exciting again, let's see what happens to the spacecraft kind of thing. So the enthusiasm for going to the moon just wasn't very great. And it's like, well, why would you go to the moon? Well, even now I've asked people this who are involved in modern space efforts all the time. Why go to the moon? Moon, well, stepping stone to Mars. Why go to Mars? Those are other questions. The thing that seems to be a popular thing last couple of years is let's put heavy manufacturing on the moon so we don't pollute the Earth. And that's a possible answer. But I think let's mine this or that on the moon. Doesn't really seem very convincing if you're a scientist. Let's put radio telescopes on the far side of the moon so they aren't messed up by radio emissions from the Earth and things. Okay, that's a thing. But, you know, it's not completely clear why go to the moon, so to speak. And you know, I know one time somebody I knew was promoting the idea of, you know, bringing back moon rock and treating it as a very valuable thing, like diamonds are treated as valuable. That would work well if moon rock was kind of as cool looking as diamonds are. I don't think it is. So that's kind of a, you know, again, that plan isn't totally convincing. I mean, it's kind of interesting that the, you know, moon rock came from the moon. Probably, you know, people don't really discuss that much possibly because the answers are a little bit messy sometimes. You know, where did the Diamonds come from? How did the diamonds get produced in the pressure in the Earth and so on and so on. But I think. So the principal question about why haven't people been to the Moon more? Well, one, it's hard and two, why do it? I mean, an indication of the fact that it's hard is that even in modern times a number of efforts to go to the moon have ended up with spacecraft that didn't make it to land softly on the moon. They just went splat. That was happened with this Beresheet lander, Israeli private, Israeli project. Oh, what was it? Three, four. Gosh, maybe I'm losing track of time. Four years ago, maybe a friend of mine had managed to put some little sort of artifacts from our civilization on that spacecraft, including some nice little micro versions of some things I've written and so on, which was sort of a cool thing. Have your archive your material on the Moon. Unfortunately, some problem happened. Some breaking rocket didn't fire correctly and that spacecraft crashed into the moon at high speed. In fact, subsequent to that, I mean, just as a sub story to this, we were wondering what happened to the thing. It crashed into the moon at what, 2,000 kilometers an hour. I think the it was, it kind of was almost still in orbit around the Moon. I, I forget. The orbital speed around the moon is maybe 5, 000 miles an hour, is that right? Something like that anyway. And you know, in order to land, you have to not be going around in orbit. You have to not, you know, it's not like on Earth. You can't orbit, you know, a mile above the ground because there's an atmosphere and, and you'd be at 17,000 miles an hour of low Earth orbit velocity, you'd be burning up. But on the Moon you can. The only problem is if you're still going at that speed and you hit the surface of the moon, you're not soft landing on the moon. You're ramming yourself into the surface of the Moon. And I think so there was sort of a question of what happened to this thing. And I'd done this little calculation that if the thing had crashed into the Moon at the speed it had. So I guess one knew two things. One knew the last picture that it had taken and we could line that up with the craters that, you know, pattern of craters on the Moon. And then one knew at what time. How did one know that? I think one knew at what time it had impacted the Moon. I'm not sure how one knew that, but, or maybe there was telemetry Actually, yeah, there was radio signals from it, which included the velocity and things like this. So I knew it when it hit the moon and it was going at a very high speed relative to the surface, not so much in the downward direction as in the direction parallel to the surface. So the question is, what happened to it? And I asked a bunch of people and the initial people I asked said things like, imagine you're firing the main gun of a battleship at a piece of concrete. That's what will happen. Thing will vaporize. But it turned out that, well, then there was other data from the Lunar Reconnaissance Orbiter taking photographs of that site. And you know, you could kind of see that there was a difference of what the lunar surface looked like at that spot before the, the, the landing and after the landing. So you could tell there was something there. But anyway, then I, I actually ended up making contact with a geologist I know and who died actually subsequently, a person called Jay Melosh, who was an expert in cratering. And so he said, well, actually if you are at that speed relative to the surface, the thing will not just ram itself into the surface, it will skip along the surface. And actually I had worked out previously that it would take a couple of kilometers to come to rest just based on friction and so on, which really surprised me. It seemed very implausible. But apparently it's true that you can kind of skid along the surface of the moon unless you run into a big boulder or something. And so then it turns out if you look at the pictures from the Lunar Reconnaissance Orbiter, there is a array of kind of light colored lunar regolith that goes out in the direction, exactly the direction you would expect from the vector of the velocity of the spacecraft when it hit the Moon. So probably the things that have bounced along the surface of the moon and quite possibly the little payload is still intact there on the Moon. But anyway, that was just the story of that particular mission was just a reminder that it's actually hard to correctly soft land something on the moon. And so I think that's one of the main reasons. Now my guess is that the sort of engineering of landing things on the moon will be fully solved in the next few years and it will become perfectly feasible to do that. Then the big question is, okay, if it can be done, and it's even fairly cheap to do it, what will be the consequences of that? What will you be able to do on the moon? And will, in fact, will it be a giant story of putting factories on the moon, or will it be something quite different. I'm not sure. Anyway, a few thoughts about those kinds of things. Well, thanks for asking lots of interesting questions and we should wrap up for now and until the next time. Bye for now.