History of Science and Technology Q&A (December 13, 2023)

Hello, welcome to another episode of Q and A about history of science and technology. Let's see. Looks like we have a bunch of questions saved up here. One from Aaron. Can I talk about the history of gazettes? That is the recording of official information? Well, I can tell you a little bit about what I know. I think that the idea of sort of recording in some tangible form information about what's happening in a city, state, whatever else, I mean, that was kind of pioneered, I think, in antiquity, in Babylon, maybe in Egypt. I think what one sees there is a lot of. Okay, what does one see? One sees lots of kind of contracts between people saying, you'll pay me and I will deliver barley to you within three months or something like that. Those are on little clay tablets with cuneiform sort of markings from certainly 1500 BC or probably earlier than that. There's also kind of the official sort of description of the king achieved this and that and the other, and the big kind of sort of governmental displays of I am so and so king of kings, et cetera, et cetera, et cetera, and the long list of the achievements of the king. I took over this territory. I crushed my enemies. Whatever else, I think my impression is that by Roman times there was a fair amount of information that was being collected about from typically tax collectors going out to the empire, so to speak. And I think that ended up getting stored. There were definite places where that kind of information was stored. I don't know if on tablets or papyri or whatever. I think not much of that survives. There certainly were things like official price lists in the Roman Empire, things like this that did survive. I don't think that the, you know, who paid what for to whom necessarily survived. I know that the kind of classic place where there was the origin of kind of statistics, properties of the state, so to speak, was a thing during much, much later in the 1600s in England, a guy called John Graunt, I think Gaunt Graunt, produced this thing called the Bills of Mortality, rather grimly during the Black Death, probably 1665, I would guess, in London was. This is kind of a summary for the city of who died how in this particular week. And that was an early example of kind of a delivery of a kind of a global piece of information about what was going on. I think in terms of sort of the collection of information and its organization. I mean, another famous example of this, this is already the 1800s, was John Playfair, I think, Produced this thing. What was it? Statistical breviary, Maybe it's A book which has pretty good visual representations of the time series effectively of what amounts to what we would now call GDP of countries and things like this. That's much later. I think that the idea of producing, for example. Well, okay, the question of when do people start doing censuses? Well, there clearly were censuses from long ago. There were censuses in antiquity, in Roman times and so on. Certainly in. For somebody like me growing up in England, you always learnt about the Doomsday Book from just after 1066 when the Norman Conquest of England happened. Somebody went around and tried to inventory what did we just conquer and produce this kind of summary of what were all the towns and villages and so on in the countryside and even down to what houses there were. And people nowadays are proud that their village was recorded in the Doomsday Book from whatever it is, 1068 or something. And perhaps even their house is marked there or a thing that was on the site of their house was marked there and so on. So certainly that kind of thing existed. The idea that one would use the kind of the census was I think largely collected in most places for purposes of taxation or for purposes of, okay, you have X number of people there. You should send this number of people to be part of the army or something like this. I think when the US was founded, the kind of idea that the census would be sort of a central part of the way the government was set up and defined and how many representatives came from different places, I think that was a new idea, I'm not sure to base representation, so to speak, on the results of a census. Certainly around. Well, by that time, well, there were, I don't know what happened, for example in France and the time of Napoleon and so on. Napoleon was very organized fellow among all sorts of other things, but set up a lot of kind of structures that probably many of which still survive today that were kind of census like things. And I don't know, I would strongly suspect that was pretty organized by that point in places like France. So that's kind of what I know. I think the concept that the census is public information, that's a pretty US centric concept. I think at least in its origination. The idea that it's. I think it comes with the government, by the people for the people type thing that, okay, government produced documentation becomes in the public domain for the public. I mean, I know when as collectors of lots of data about countries, it's sometimes been a challenge for us as we collect data for our knowledge base. You know, that's probably A decade ago I remember hearing from the European analog of the faa. The FAA distributes certain kinds of data freely as a government organization in the U.S. but the analogous organizations in Europe are like, well, you know, we don't have the same setup as the U.S. you know, we're trying to make money from our data type thing where, you know, that's, that's what we do as a government kind of thing rather than it's kind of by the people, for the people type thing where all of that government data is freely available. That seems to be a particularly US centric. Same is true in England for example, things like the maps, the so called Ordnance Survey mapping of the country was absolutely not freely available. In fact, when OpenStreetMap, Steve coast, who was a former intern of ours who ended up creating OpenStreetMap, his original goal was sort of free. The maps that the UK government could claimed were crown copyright kinds of data. And eventually sort of the maps were effectively reverse engineered. And that's what gave us OpenStreetMap, which is very widely used as kind of a mapping, a foundation for mapping. We certainly use it in Wolfram language and Wolfram Alpha and so on. But that was so the convention, the idea of what was the property of the government and what was sort of for the people different in different places. And the US was very much on the for the people type thing. So a few things to say about that. I would say that in terms of usually information gets collected by governments and things when there's an obvious kind of tax or other related purpose in doing that. Although somehow the US government particularly has been particularly big on just collect data on all kinds of things. And particularly in the early days of Wolf Malpha, back well before we released WolframAlpha in 2009 and in a few years after that as we were probing more and more different sources of data, it was pretty common for us to find US government data collection operations where people would say to us, thank you for asking for this data. We've been collecting it for a hundred years and almost nobody ever asks for this data. And so there's a lot of that that's gone on. And I suppose it's sort of interesting that there's a lot of data that there'll be a time series that goes back to let's say 1970. There's a whole bunch that go back to about 1970. I think that must have been a time, for example, when it comes to economic data, financial data and so on, that must have been a time when Things got computerized to the point where sort of there's surviving data that's been kept to give you examples of other kinds of data. Like in financial data, there are stock prices. And the stock prices have been published in the newspapers since, well, certainly sometime by the late 1800s. That was happening probably earlier than that as well. But the sort of digitizing all of that stuff and lining it up and figuring out, oh, this company with this ticker symbol then turned into this thing. Now that's been very thinly done. Not a lot of that has been done. The times when that has been done and people consider it quite valuable data is by some quantitative finance places where their sort of fundamental principle is history repeats. And if one saw some pattern of stock prices back in 1912 or something, that was the grain price went up and, and the pig price went down and whatever else it is, that by golly, in some sort of modified form, that same thing will repeat today. So it's really useful to have even that very distant data. Let's see, question from Jamie here. Why are some older coins or bills worth more than the original face value? Well, I guess it's because people. There is a certain gravity that comes with age. One likes to think of that for people too. I think when I got to a certain age, I could rightfully call myself an antique, so to speak. I think the thing that's always interesting about old stuff versus new stuff is, let's say, old furniture, for instance. People say, oh, it's wonderful. I think there is a sense in which, yes, it was better built and built in a more artisanal way in the past and built with better materials and so on. But there's also another phenomenon which is the stuff that's junk gets thrown away. The stuff that's good survives, and people keep it. And so the things that have sort of survived 200 years, it's like somebody had to bother to keep it for 200 years. I have to say, when I, you know, in England, for example, I certainly knew people who had, you know, their families had been in the same houses for 400 years and things like this, and they would find junk from 400 years ago that was kind of, you know, unjust stuff lying around. I did remember at some point when we were had a company, TouchPress, that made iPad ebooks and so on, and we had all sorts of different assets for different things. There was a time when some folks associated with Touch Press were poking around at Windsor Castle, which I guess is a thousand years old, roughly, and they were sort of shown to some storage area there which really did have junk that went back, you know, sort of throughout the history of. For a thousand years. And probably when things have been just, oh, let's put it here, we'll just store it here and nobody paid attention to it. There can be not such high quality stuff that survives a long time, but by the time it's passing through many hands, somebody's going to throw it away if it's junk, so to speak. And I suppose you see the same thing with books. For instance, if you say, oh, but back in the day people were writing such wonderful books. Well, actually if you go look at the Internet Archive or something and you look at scanned books where, where somebody has scanned sort of everything that was out there in 1865 or something, there's a lot of junk. There's a lot of things where it's like, this is a book, you look at it, it's kind of somebody's very ponderous essay on something that says absolutely nothing and it's just pure fluff. But yet our impression is there were great books back in the day and that's primarily because the great books survive and get reprinted. The fluff dies out. Of course, there will be some good stuff that dies out because people didn't understand that it was good at the time when it was sort of soon after the event, so to speak. And so you have to kind of. So it gets lost. And that's why I don't know whether some works from antiquity were lost, because in the period right after, let's say, the fall of the Roman Empire, people were like, eh, this isn't very interesting, who cares? Or the monks who are copying things or whatever just were like, we don't happen to care about this or we think it's shocking or whatever else. Let's not preserve this. And then perhaps those things get rediscovered. And this is a remarkable thing that sort of seems to happen with some regularity, that you would think that the past, if we didn't know what happened back then, we'll never know what happened back then. But actually things start showing up. Somebody will notice that, oh, by the way, somebody wrote some new thing on the back of a manuscript that was a copy of something that Archimedes had done that was a thing that happened maybe what, 40 years ago, something like that, that was sort of a new. I'm trying to remember there was also a Leonardo one where that happened. But that's a thing that seems to happen. Things seem to show up that one didn't know were there. One example, I know the Babylonian astronomical diaries, which were a record, talking of what records were kept, was a record made every day somewhere in the Babylonian empire, maybe of the state of the heavens, the weather, the price of grain, the height of the river. Actually, the height of the river probably tells when it was in Babylon that this was done, because I think there was a specific river there that probably the Tigris or Euphrates, and I should know which one, but I don't. In any case, the thing that those astronomical diaries were made every day for 700 years, and presumably copies were made of them. But what survives for us today is a couple of percent of those diaries. But those diaries were survived because about 100 years ago, people started buying random Babylonian artifacts from random stores in places like Paris and London and so on. And they found, you know, these tablets were among the tablets that were for sale there was. And so it wasn't like somebody just found a cache of tablets in some archaeological dig. It was these were assembled through something where they were very dispersed over the course of 3,000 years or more than 2,000 years, and had wound up as just things that were preserved because they were kind of cool and then they had to be reconstructed much later. So I think that this question of does something end up being worth more because it's old, so to speak? When it's cool and it's old, then it tends to be worth more. When it's kind of thought to be junk and it's old, maybe it isn't worth more. So I think that would be my naive version of the answer to that. Let's see. Gosh, many, many questions here. Let's see if I can just handle some of these in. In order. Brady asks, is there a connection between Galileo's equivalence principle and Einstein's equivalence principle? Well, let's see. I mean, Galileo, one of his big discoveries about gravity is that, you know, you drop a, you know, you drop two different cannonballs or something made or two different things made of different materials off the Leaning Tower of Pisa, which was at that time the tallest building outside of the pyramids that existed in the world. It wasn't. It only the. The leaningness was not a design objective. The Leaningness was a design mistake. But it was originally it was just a very tall tower, very tall for its time tower. I think it wasn't until, what was it, the Empire State Building in New York, and things got built that there were sort of structures that beat the pyramids and beat the tower of Pisa and so on for height. But I think so Galileo happened to be in Pisa and used that tower as his experimental platform, one could say, quite literally, and was dropping things and seeing how long it took them to hit the ground and discovered that it didn't matter what the material was, it would still hit the ground at the same time. Of course it did matter if you dropped a feather and a lead ball or something, they would hit the ground at very different times because of air resistance for the feather. And famously, I think one of the astronauts on maybe Apollo 16 or something brought a feather and something else, a ball or something, and dropped them on the moon where there's no atmosphere. And sure enough they hit the ground at the same time and made some cute comment about how if this hadn't worked, our spacecraft wouldn't have made it here type thing. But you know, this idea that the material doesn't matter is kind of the generalization of that for Einstein was not only does the material not matter in terms of the pull of gravity, you also can't tell the extent to which you can't make a distinction between a gravitational field, the pull of gravity, and the effect of pure acceleration, the purely kind of motion related effect of acceleration. So I suppose those are things that. And I think the kind of historical progression that got from one to the other goes through people like Ernst Mark, who was thinking about kind of what's the origin of inertia? Why do things have a mass? Well, kind of one of the theories that Mark had was, well, maybe things have a mass because they are gravitationally somehow attracted by everything in the universe. And there's a lot of sort of confusing stuff about rotating. You know, what if the universe is rotating, there's an effective force, kind of the centrifugal force that's kind of a force that's only the result of the fact that there's motion and so on. So a lot of things that I'm not sure I know the I'm not sure I've really traced through the kind of the detailed history of views about sort of what mass is, what inertia is and so on through that period of time. I mean, I think it's sort of interesting for me at a meta level. You know, we'd made a lot of progress in our physics project in understanding a lot of these fundamental principles of how physical universe gets put together and what gravity is and all those kinds of things. And it's very hard work to go back and see what were the antecedents of These ideas, you know, I did that for the second law of thermodynamics. Part of the book that I wrote about the second law recently is a big effort to kind of untangle the history of the second law, which is really complicated. And I have to say that I was surprised, you know, when I started writing that. I was like, well, let me see if anybody else has written this. I didn't think they had, but I was looking around and really people have written individual pieces, often quite good individual pieces, but not the whole arc of what had happened. And I think there are several such historical things that should be done, the whole bunch, actually, that sort of eventually land in things that we figured out in our physics project. And there are antecedents and it would be interesting to trace those through. I think it's a complicated trade off as far as I'm concerned, seeing kind of how much of one's time I enjoy doing history. That's why I do live streams like this and so on. But somehow I think the priority has to be sort of building the future rather than looking back at the history. Although for me, looking back at the history is often important in building the future. One reason, for example, in the case of the second law of thermodynamics was that I was kind of trying to roll out kind of my version of how the second law works, which is a little different than what people have said before. And I wanted to understand, am I really right about what I'm saying? I think so. But to know how to be. To feel justified saying it's different from what people have said before, I felt like I needed to understand why did people. What exactly did people say before and why did people say it? And I think I got very much to the bottom of that particular issue. But I haven't done that for some of these other areas. And that would be interesting to. See. EPI comments Important work from bright minds gets shelved and then we discover the work decades later and put it to use. Have you thought of ways to reduce this happening so much? Look, the issue with that tends to be a different phenomena. One thing that can happen is the person is just in an obscure place and not well connected to getting the word out. Let's put that aside for a second. The more common case is there is something which somebody discovers but they don't really have a context for the discovery that never really gets critical mass. The context there never really is a big story. It's just this particular phenomenon was discovered and it takes often a long time for the context of that discovery to be clarified. And then sometimes what really happened is somebody discovered something and then nobody understood it, including the person who discovered it in a broader context at that time. And it took that broader context developing before you could go back and say, well, actually they discovered this good example of this. Moses Schonfinkel, 1920, something I've studied quite a bit, invented these things called combinators, which are kind of a, a symbolic representation of what we would now call computation. He thought of it as kind of an idealized model for mathematics, essentially, and for the processes of mathematics in some way. He didn't write that much about what he really thought it was. Well, that was 1920, 1936. Alan Turing invents Turing machines and has a much clearer enunciation of the idea of universal computation. The idea that you can have one fixed piece of computer hardware and you can feed it different programs which will do sort of all possible different computations. Looking back from what we now understand with Turing machines, we can say, yes, Moses Schoenfinkel already discovered universal computation. But at the time, that wasn't an obvious thing. There wasn't really a context into which to put that idea of universal computation. Now there are other things that happen. We're talking about the second law of thermodynamics. A kind of a sad story is the story of Ludwig Boltzmann, who by the 1860s was talking about how molecules must exist and they must be the cause of things that we see in the properties of gases and thermodynamics and so on. He's somewhat overclaimed in some ways, sort of, oh, you can prove the second law must work based on thinking about molecules. But by 1905, Boltzmann had sort of given up people. In fact, Ernst Mark was one of the people who argued strongly against molecules. He said, matter is not discrete, there are no molecules. And that was around, well, the last decade, probably of the 1800s and so on. And it was. So it was kind of the, you know, it seems that molecules don't exist. Well, you know, Boltzmann, who became something of a philosopher, a professional philosopher, professor of philosophy in his later years, after having been kind of a really a card carrying theoretical physicist much of his life. But he has a place where he talks about, one man cannot withstand the tide of history. I'm going to write down, he wrote one sort of last book about the molecular theory of gases, saying, I'm writing this down so that sometime in the future when people rediscover this, not too Much of what was figured out will have been lost. It was a very bad luck. Boltzmann ended up, well, actually killed himself in 1906. You know, sometimes it's sort of presented as, oh, because people didn't believe in his theories, that that's really not what happened. He was, well, in those days, what was called neurasthenia, kind of, you know, what would be diagnosed differently as some kind of bipolar thing in modern psychiatric practice and so on. I don't think it could be. I don't think that Boltzmann's problems can be attributed to sort of the fact that the physicists didn't believe in the molecular theory. I mean, Boltzmann actually was quite a cheerful fellow in many ways. He kind of would go through cycles of cheerfulness and not so cheerfulness. Actually, a year or two before he died, he'd visited California, was a German scientist, and he wrote a book called A German Professor Visits El Dorado and Talks about the Americans and says, I think one of his nice quotes is, I think they have a bright future, even though I've seen them at their worst doing calculus. And he visited, actually the Berkeley University of California in Berkeley, when it was a very young university just at the beginning of the 20th century. But any case. So he was, you know, he. And he was pretty pretty. Certainly his presentation in his book was pretty. Pretty cheerful at that time. But anyway, so Boltzmann is an example of something where a large body of work was like, oh, it's molecular theory. Nobody cares about molecular theory. He had bad luck because in 1905, Einstein kind of figured out that Brownian motion was something which could be accounted for and was an indication of the molecular structure of matter that had been figured out by Smolachowski in 1900, I think, as well. And in fact, Brownian motion, the phenomenon of look down a microscope, see these little grains that come out of pollen, and see the fact that they get kicked around, as we now know, by water molecules. But that phenomenon had been discovered by Robert Brown, maybe by others in the 1830s, I think, and nobody knew what it was. And then by 1905, it's like, okay, it's molecules. And that sort of proves the molecular theory, which had had a lot of development with people like Boltzmann. And it was one of these things where there wasn't quite the push to have to believe in it because there wasn't sort of the ambient context to do it. I can relate a story of my own history. This is 1978, 1979, I was writing a paper with a friend of mine named Rocky Kolb, who's a cosmologist now at University of Chicago, and we were writing this paper about the expansion of the early universe and the fact that if various kinds of particle physics phenomena occurred in the early universe, there would be effects on the expansion rate of the early universe. And we noticed that if there was a. If the universe sort of super cooled, just like if you have water and you cool it really below 0 degrees centigrade at ordinary pressure, if you cool it below zero degrees centigrade, you expect ice is going to form. But actually, ice doesn't form until there are nucleation sites that cause the solid to start building up. So in principle, you can supercool water and end up with something where there hasn't been any. There's no ice that's formed. But when there's ever any sort of imperfection in the water, that becomes a seed on which ice forms. So the question was, in the universe, could you supercool the universe? If you could, you would get all kinds of interesting effects. In particular, you would get exponential expansion of the universe and all kinds of things like this. Well, we didn't think. I didn't think that the universe would be super coolable by what turned out to be 10 orders of magnitude to be a significant effect. So we stuffed this in a footnote. He said, if the universe was super cooled by this huge amount, then you would get exponential expansion. And that would solve various problems in cosmology. Okay, put that there. A couple of years later, maybe a year later, I forget, other people came along and said, you know what? This idea that you can have this exponential expansion period, the things that it solves are so important that even though it seems very implausible that you could get that kind of super cooling, we're going to make a whole theory based on it. And that became the inflationary model of early cosmology, so to speak. I think in the end, in our physics project, that all gets sort of washed away as we see the dynamics of dimension change in the early universe and so on. But that's a different story. The point there was that that was a phenomenon that, yes, one could see the phenomenon. Yes, we could plant a flag and say, yes, we discovered that scenario, but we didn't make the kind of noise about it that subsequently got made. And that made people sort of forget the fact that there were these horrendous assumptions that were being made and say, yes, this is important. We're going to build A theory around that. And that's not an uncommon thing to see happen. I mean, in my own work, looking at kind of how simple rules in the computational universe produce complicated behavior, having discovered that phenomenon in things like Rule 30 and so on, one can go back and see the sort of signs of that phenomenon all over the place. I mean, for example, John Venn, inventor of Venn diagrams, had a book in, oh, in the 1850s maybe, where he wanted to show a random walk. It was the first sort of drawing of a random walk one knows about. And what did he use to make the randomness in the random walk? He used the digits of PI, and the digits of PI are very much the same kind of story as something like Rule 30. There's a rule, but yet it makes apparent randomness. He didn't really comment on that. People realized that there was sort of apparent randomness in the distribution of prime numbers, but they mostly commented on the regularities that were sort of mathematically accessible. There wasn't really a framework for thinking about complicated things and thinking that complexity as a phenomenon was something worth studying. And so that was something. Well, I had the fun of figuring that out, but that was a thing which needed that kind of context. And that's the fact that it needs this big chunk of context is what led me to eventually make this big book called the New Kind of Science, that is about the context needed to understand the science around what you see in the computational universe. And that's that 1200 pages is the chunk of context that I think you need to get an idea of what's going on there. Even that was not enough for many people, or some people, at least. I think over the years, that's tended to be something which people have felt, yes, it really does communicate what it wants it to communicate, which is nice to see, even though it takes a few years for it to be realized. I know in the preface to A New Kind of Science, I say right there, I say, as of right now, people will find a lot of the things in this book surprising and hard to believe. Eventually in time, they will come to seem obvious. And 22 years later, now, 21 and a half years later, we're definitely on the. We're rolling down the obviousness curve. To some younger people, like some people who've worked on a physics project, many of the phenomena in NKs new kind of Science, like computational irreducibility, are like, it's obvious, of course, it's got to work that way. But for people coming into that from the other Side, so to speak, not having grown up with these ideas, it's like, how can that work? How can it be that way? That's a very common phenomenon. I think this question about how to sort of prevent things which were worth discovering. How do you. I think bringing discoveries that need context, sort of highlighting this is really hard. I mean, you. You just have to. I suppose the best thing I can say about this is that when I see things, discoveries get made in science. There are many incremental discoveries that do not need conceptual development. They are particular experiment, a particular technical innovation, these kinds of things. They are a discovery where the overall concepts don't have to be redefined. It is merely. And now we know there's a higher temperature superconductor, and now we know there's a particular biological signaling pathway or something like this. But the conceptual framework is the same. I mean, I see the same thing with language design. In Wolfram language, there are things where there is a conceptual framework for mathematical functions. Let's say, oh, we've got another mathematical function, it's the appel F2 function or something. There's a conceptual framework. We're just wheeling in another function that fits into that conceptual framework. And then there are things that really require new conceptual development. And that's happened a whole bunch with our physics project and with the RULIAD and other things like this. They're really new conceptual development. And I think that for many people in science, that's very unusual. It's almost unheard of that people can go through their whole career with not much change of the conceptual framework. And I noticed this, I was realizing this recently as I was thinking about podcasts I've done and things like this. And it's like, well, I do ones with physicists, I do ones with philosophers. And also when I've talked to groups of people and things, it's sort of interesting because the physical scientists often are like, they are used to technical complexity. They're not used to conceptual complexity or conceptual change. And they get very confused by it. It's like, but aren't you saying something which is conceptually the same as has been said before? Well, no, actually, it's like, well, it must be conceptually the same, even though it's technically different. Well, no, actually, whereas for the philosophy folk, it's much more par for the course that one has something conceptually different because their world is about sort of conceptual change, so to speak. So I think that's a, you know, the willingness to kind of embrace conceptual change is, I Think that's probably the number one thing that affects whether sort of discoveries that were made ahead of their time get recognized and sort of integrated into the mainstream is can one sort of break through this kind of conceptual, this new paradigm type thing? And I think that sort of the learning new paradigms thing is hard for everybody. I think that the willingness to embrace new paradigms, that's the thing you have to look for. And for example, as science gets more institutionalized, it becomes more difficult to accept new paradigms because it's like, well, we've got to, you know, every biology paper must have a this or this or this or this or this. Oh, whoops, it's theoretical biology that doesn't work that way. Well, wait a minute, how do we deal with that? We don't have a way to deal with that. You know, we don't have the same value system of did you do a double blind clinical trial? That's not really relevant. If what you're doing is a computer experiment, you could apply that methodology, but it would be silly. So know kind of a field which has developed to the point where it's heavily institutionalized and where its paradigm is institutionalized, where in a sense it's kind of like every paper has to hit the, this, this, this, this, this of the paradigm. That's kind of how you prevent the possibility, how you, how you end up with submerging things that require a new paradigm to understand them. So I don't have a great answer to that. But you know, I think this, this willingness to accept conceptual change is something that is important and it goes along with the general thing that I suppose might be a mantra of mine is kind of learn to think rather than just learn to do things in a particular mechanical track. Well, let's see, There's a question here from Giovanni about the development of string theory and whether its inception was within mathematics. Has it yielded tangible technological advancements? So string theory is a bizarre story of scientific development. Let's see if I can get it right. So after World War II, sort of the physicists who worked on the Manhattan Project went to populate particle physics and a bunch of money got allocated to building particle accelerators. And the result of that was that lots of protons crashing into protons and so on and producing lots of particles. And one of the surprises was in the 1950s that there were more and more particles that were discovered. You know, first it was, well, some particles have been discovered in cosmic rays, pions, the muon, the kaons, the lambda particle, the hyp the sigma particles, these kinds of things. But then there started to be particle accelerators and there was a whole zoo of particles. Soon there were hundreds of particles. This was my, when I was a kid, one of my sort of favorite, all time favorite books was this little booklet, the little particle properties booklet. I started getting them back in 1972 when I was 12 years old. There's a summary of the properties of all these particles you could look through. And there was the F meson with a mass of 1270 MEV and the, the 8. Is that right? I think that's right. You know, the A2 meson with a mass of 1200 ish, maybe the Rho meson with a mass of 760 meV, I think. And see, this is a, this is a good memory test for me. Let's see if I can do some of the others. The sigma particles, I think they're 1385 MeV. The lambda particle, I remember that one. That was 1115 MeV. Although I think it changed because back when I knew about these things it was always 11:15 MEV. But now I looked it up more recently and it's got four decimal digits after that. And I think it might have flipped over to 11:16 now. But in any case, whole zoo of particles, it's like, what were these things? Can we understand what's going on? There were a couple of. Well, there were, there were a couple of hypotheses. But one of the things that was. Well, there were really two. Well, okay, there was. Okay. In the history of particle physics, there had been great success in quantum electrodynamics, the theory of the interactions of electrons with photons that was making use of quantum field theory. Quantum field theory had been developed in the late 1920s, really on the, on the, on the tail of the development of quantum mechanics in general, the idea of how would you represent not just a single particle in quantum mechanics, but a field where there can be particles produced and annihilated and so on and varying numbers of particles. So that development had. By the 1940s, it was a pretty elaborate mathematical development. The idea of renormalization, which kind of dealt with the infinities that developed there. And so on came in and Feynman diagrams got invented as a calculational method. And by the 1950s, people were calculating away and you could work out the precise probability for an electron to scatter against a positron or the exact anomalous magnetic moment of the electron. Things like this, you could do that by calculating by doing these very Elaborate calculations with Feynman diagrams, with more and more elaborate diagrams, trying to work out more and more precisely the predictions of quantum electrodynamics. And it worked really well. And there were a bunch of things discovered which said yes, it agrees with quantum electrodynamics, we've got a great solution there. The problem was that the calculations of quantum electrodynamics relied on the fact that you emit a photon, you another photon. The chance that you emit two photons is much smaller than the chance you emit one photon. And there's a quantity, the so called fine structure constant value about 1 over 137 denoted alpha. And roughly the fine structure constant tells you the probability to emit another photon. And so you emit one photon, probability alpha, emit two photons, alpha squared three alpha cubed and so on. So you can make this kind of power series of how many photons did you emit? And it's a rapidly. The numerical contributions, at least nominally are rapidly decreasing. Okay, footnote to that, actually the true formula is that in the kth order of so called perturbation theory, the typical contribution in quantum electrodynamics is square root of k factorial times alpha to the k. And so for large k actually the pre factor wins and everything blows up. And that's still not well understood. Let's put that aside. People didn't pay attention to that back in the 1950s when calculations in quantum electrodynamics were happening. And it looked really good. Looked like you really calculate things. Problem was, not everything had the kind of very weak interaction strength that electrodynamics has. In particular the strong nuclear force that binds atomic nuclei together. And that was Yokawa Hideki Yokawa, 1935 had suggested that there was this thing called the pion which became called the pion, which was the sort of a particle that got exchanged the analog of the photon, but a thing that would bind nuclei together. Problem was that the analog of the alpha being 1 over 137 in quantum electrodynamics was at best a coupling constant, as it's called, of order one. So you couldn't really make this series where you say, oh, there's one photon effects two photon effects three photons and they get progressively smaller. It's like everything was all put together. So the question was how to deal with that. And that was one thing. Then there was this kind of zoo of particles being discovered and there were different hypotheses. So one was, let's figure out a way to do something like field theory. Something like what was done in quantum electrodynamics for the strong nuclear force and various kinds of limits of zero mass pions and things were studied that led to one branch. The other branch was so called S matrix theory, where it's like, let's not think about the mechanics of how the interactions happen. Let's just say particles come in, something happens in the middle, particles go out. Let's just look at the comparison of what comes in and what goes out. Let's just look at the overall effect of the interaction. The S matrix was an idea that had been introduced by Werner Heisenberg. Actually I just recently learnt the origins of the S matrix idea, which is something I absolutely didn't expect. So Heisenberg back in the 1920s had been quantum mechanics was in. Everything looked like it was being discrete. There were all kinds of the. All kinds of discrete energy levels and atoms, discrete spin values for particles, lots of discrete stuff. And so the assumption got made that space would also be discrete. And that's a thing that. In fact it seems like I'm sort of accumulating the list of people who thought that. I recently learned that Niels Bohr thought that, that Werner Heisenberg thought that that space would be discrete. It's sort of ironic because a big piece of the origin of our physics project is the idea that space is discrete, which 100 years later was completely shocking until we started to sort of work through how that really works out. But in any case, Heisenberg wanted to make a model of discrete space and had this kind of cellular model of discrete space. Really kind of interesting. I didn't know that he'd done that until just a few weeks ago. And apparently in the effort to do that and to make that consistent with relativity, which was very hard to do, he sort of threw up his hands and says, let's not look at the mechanics of how the particles interact. Let's just say how did they come in, how did they go out? That led to the S matrix idea. So then people had the idea of, well, let's just look at the properties of the S matrix. We know, for example, that as we kind of change the momentum of the particles that are coming in, the S matrix should vary. The probability of interaction should vary as some function of the momenta. It should have be a function that has certain properties, it should have certain analyticity, certain smoothness properties, it should have certain invariances under so called crossing swapping particles and antiparticles in the initial and final states and so on. Let's just not worry about what's happening inside. Let's just say it's just an S matrix and it has certain properties as a mathematical function, so to speak. And that was sort of one side of things was the S matrix approach. The other side of it was the field theory approach. Let's see the mechanics of what's inside. Sort of in the middle of that was, what are all these particles? Do all these particles have sort of a mechanical explanation? In the end, it turned out they were all made of quarks. That was a thing proposed by Mari Galman and George Zweig around 1964. But that was the. And there had been precursors of that. A person called Sakata had a precursor model of that and a bunch of other precursors to that. But any case, the thing was kind of, are they mechanically made of stuff or not? A person called Jeff Chew in Berkeley had kind of the democratic theory of particles. It was kind of a time, I suppose this was heading into the 1960s, when there was a big push for kind of changes in the world order. And the idea that there shouldn't be a particle in charge, so to speak, a top particle, the idea that there could be sort of a democracy of particles, was, in a sense, ideologically and politically, kind of a pleasant thought for the Berkeley folks of that time. And so this was an idea that sort of all particles were equal. And it was kind of this notion of the bootstrap where you would have a different. And quite different from the bootstrap in statistics, because something quite different came much later. This was an idea where you would say, well, you could think about particles. You could think about the fact you have a particle. You could think about a force that came from the exchange of particles. Somehow all these things were related. And when you put them all together, this was sort of the only thing that was consistent was the particles you would have. Okay, so that was another piece. Another piece to this picture was a chap called Tullio Reggie, who I unfortunately never met. I could easily have met, but I never did. A very creative scientist who, well, had, I don't know, a somewhat complicated trajectory. But back in the 1950s, he invented these things called regitrajectories. And what he was doing, he was studying things like the scattering of one particle from something that scattered it, like a particle being scattered by an atomic nucleus or something like this. And he was asking the question. So in quantum mechanics, there's this quantization of angular momentum. And he asked the question. And so you would have angular momentum, j equals 1, 2, 3, 4, et cetera, et cetera. Reggie asked the question, well, what happens if J is a complex variable? It's a continuously varying thing. What happens to these so called scattering amplitudes? And that led him to this idea that, for example, you could have a scattering amplitude as a function of momentum that had, well, where you looked at this angular momentum in the complex plane. And that led him to questions like, well, maybe all these particles that we have are just sort of features of a mathematical function. Each particle is a pole, a 1 over x, a place where the, where the function blows up when X goes to zero. That they're all sort of poles in a function. And then that you would have something where you could, and I should say, by the way, that the S matrix as a mathematical function, when there is a. When you have the energies of the incoming particles being aligned exactly so that you produce just exactly one of this zoo of particles, then the scattering amplitude in the probability of interaction would be infinite were it not for the fact that the particle you produce doesn't live in infinite time. So it is a pole in the S matrix, an infinity in the S matrix associated with the production of a single particle. If that single particle lived in infinite time, if it lives only a finite time, it kind of gets smoothed out a bit. But any case. So the idea of Reggie poles that corresponded to these particles led to these sort of diagrams where you would plot the mass of the particle against its spin, its angular momentum, and you would see that roughly, particles lay on straight lines. The particles that were being discovered lay on straight lines in these Reggie plots, so to speak. So that was another piece of sort of the picture that was kind of on the S matrix side. Okay, so into all of that came, I'm trying to remember the order in which all this stuff happened. The. There was a particular form of mathematical function that was sort of a convenient version of the S matrix that had convenient mathematical form involving gamma functions, beta functions, and things like this, that had, gosh, I lived through this history. I should remember this stuff that was due to a person called Gabrieli Veneziano, who invented this particular form of the S matrix. And let me think, the whole point was that there was an interpretation of that in terms of strings. What is a string? Well, in standard quantum mechanics, and particularly in relativistic quantum mechanics, you're always talking about particles. You're talking about a point particle. Like, there are equations like the Dirac equation is the equation for a point particle like an electron with spin a half. And there are these various equations that describe point particles. And when you think about a point particle and what it does, it has a world line. In space time, the particle can move, and you kind of trace out its world line and in space time, and it's just a line. Okay, so people have been interested in, well, how do you generalize that, even sort of at a mathematical level? How do you generalize that to a world sheet where you have a whole string that is moving through space time? Moving through space and time. And one thing that was true about particles was that you could define kind of what world line a particle would make using a principle of least action, using a kind of optimization principle, the particle will follow the trajectory that corresponds to the path of least action. And so you could imagine generalizing that to this extended object and its world sheet. Instead of the world line has a particular form, the world sheet has to have a particular form. Now, I'm trying to remember the. And I'm sorry, this is what happens when I get close to a piece of history that perhaps it's harder to disentangle it. The Veneziano amplitude, the form. Of the S matrix, that had an interpretation in terms of strings. And then this idea of sort of this physical layout of strings propagating through spacetime. And I remember when I was. You see, this is fragmentary memory, I'm afraid. When I was probably. This is 1975, I was 15 years old, I happened to visit physics group in Cambridge, England, and a couple of the people there were working on what became string theory. And I remember they were explaining to me this thing about world sheets and the minimization of some analog of the action for world sheets. So that must have already been a thing by 1975. Those were sort of the leading people, some of the leading people who invented that stuff. So maybe it wasn't terribly old by that time, might be six months old or something. But in any case, the original idea of string theory was it was part of a theory of strong interactions. And it came out of this kind of very mathematical construction. And it was something that was supposed to be kind of a way of thinking about the S matrix, which would be a way of thinking about strong interactions and the kinds of processes that happen in nuclei. And then, oh, it got very confusing because in 1973, the discovery of asymptotic freedom, this is the property of quantum chromodynamics, the theory of quarks and gluons, that if you kind of kick the quark hard Enough, its effective value of alpha alpha sub S for the strong interaction will be low enough that you can use perturbation theory, that you can use this series expansion thing. So when you have interactions which kick in hard enough, so called deep inelastic scattering, where you have particles that are very inelastic in their scattering, they don't just bounce off, they kind of bounce at a big angle and a loss of momentum is transferred. In those cases, the effective interaction strength is lower. And that means that you can use quantum field theory. You can use all the apparatus of quantum field theory. And that was the thing that was discovered in 1973 and really sort of turned the tide of quantum field theory versus the S matrix approach. This mechanistic quantum field theory of how strong interactions worked, at least at high momentum transfer, high energy was something that then really took off. It was something I was much involved in in the, a few years after that. But the, the thing that, but that was sort of a discovery made in 1973. And after that time, kind of the, the fate of the S matrix and, and kind of that approach to things, it sort of degraded and quite rapidly. You know, particle physics was at that time a really sort of a very energetic field full of the best and the brightest, so to speak. But the best and the brightest definitely as a herd rushed off to whatever seemed like the brightest thing. And that wasn't S matrix theory. So that died very quickly at that time after the discovery of asymptotic freedom. And somewhat confusingly, even in qcd, the idea that when you pull quarks far enough apart, they would have kind of this string of gluonic stuff between them, that was kind of a confusing thing. That was sort of a little bit like strings. But it wasn't the same kind of strings. But it sort of was, but it wasn't mathematically the same. All a bit messy there. Okay, so that was thing number one was strings as a mathematical theory of the strong interactions and so on. And that kind of died off by the 1970s and that was. Yeah, so I mean, I think except as kind of a mathematical thing like studying world sheets and so on. Okay, then what happened next? By the late 1970s, there was kind of a rebirth of this in the form of. That came out of super gravity. Boy, this is a really shaggy dog story. I mean, string theory got re, sort of retooled multiple times for multiple purposes and kind of as a theory of gravity, something which sort of would be a theory that would sort of where the vibrations of the string would correspond to all possible particles and correspond to effects of gravity. But that was only consistent if the whole theory kind of fell apart unless the strings were in either 10 dimensions or 26 dimensions, which isn't what we observe in physical space. And then that was sort of explained in terms of, well, the extra dimensions are curled up very small and we don't detect those. And then that led to a bunch of development of kind of the expectation that there would be a bunch of particles that would be additional particles, the super partners of existing particles. And those weren't discovered. And so that was kind of a downer for that. And what's sort of ironic is that the ideas of S Matrix theory are back in. There's the thing called the conformal bootstrap, which, which is very similar to the. In fact, it's the same thing as the original bootstrap idea that was part of the democracy of particles back in the 1950s and so on. And the mathematics is all the same. And so things are sort of coming back along those lines. So I mean, I could fill in more of this history, but I think it's ornate and it's going to take me a little while to. It takes actually describing some of the physics of what's going on to see how this stuff, this stuff works. And maybe we shouldn't do that right now. But in terms of kind of what's happened. Well, okay, this question of sort of how do you make string theory, the theory of kind of world sheets and all that kind of thing, how do you make that kind of thing consistent? Turned out to be pretty hard. And that's what led to this. 10 dimensions, 26 dimensions. Otherwise you get sort of wiggles on the string that go faster than the speed of light and other sort of pathologies unless you have those conditions. And then as one added sort of more internal symmetries to the strings and so on, it was kind of like, well, we can kind of get these different situations in which we get different kinds of theories. And eventually there was things called M theory that developed in the 1990s. I guess that was kind of the ultimate sort of put in all the symmetries, put in all the conditions. It's the ultimate kind of unified thing that is kind of a representation of, of, of string theory. And that, that was, that looked like, oh, it's so, it's so pretty that it must be the theory of how, it must be how physics actually works. But it hasn't, you know, that hasn't been, hasn't been possible to kind of show that that's the case. My guess is that there is a. A strong way of thinking about string theory as a limit of the models that we've developed. And roughly the way that would work is in our models. Okay, physics moment here. In our models, kind of, there's this giant graph, hypergraph, that represents the structure of space. And one of the things that's notable is this is just this big graph. And you go from one node in the graph to another node in the graph, and you can think about yourself as moving physically in physical space as you do that. But also it's possible that what's happening is that as you move from one node to the graph to another, you're sort of moving in some internal space. It's like in the case of qcd, one thinks about there being a color space where the particles can be sort of red, green, blue, not really red, green, blue, but that's just an analogy for mathematics. But really there's some angle that's an internal degree of freedom of the particles and something internal to the particles, an internal feature of particles. And in our models, all of that stuff has to be encoded in this graph. So as you move from one node to another, you can be either moving in physical space, or you can be moving effectively in the internal space of these properties, like color of these particles. And so teasing apart what corresponds to physical space, what corresponds to this internal space, that's not an easy thing. We've been trying to do that and understand how to do that. But my guess is that there will be particular ways of sort of factoring the physical space and the internal space that will be particularly elegant cases that will probably correspond to string theory. And whether that's, you might say, well, that's these isolated, elegant cases, they're probably not what actually happens. But it's useful to be able to see those cases as kind of versions of the model that you can explore in more detail. Let's see. All right, let me just have a quick look here. Oh, my. I gotta go soon. Let's see. I see a comment here about going back to the question about why are some old things gain in value? The comment being that some old things are worth more due to nostalgia, eventually it will lose its value again. It's sort of interesting, isn't it? I mean, for example, Egyptian antiquities were not really such a big thing. People just had pieces of giant stones lying around and so on. And then folks like Napoleon were like, this is really cool. We're really establishing ourselves as Sort of important emperors for the ages type thing by having our Egyptian antiquities hauled into our central square type thing. And as a question of do these things actually go down in importance again? Is there a. Is there a moment when history is forgotten? And clearly there are some cultural and political ideologies where sort of the idea is history is irrelevant. Let's start afresh. We're going to invent things that are going to be the great thing going forward, history be damned, so to speak. And it does seem to be a feature that you do see these cycles of, oh, know, let's make the new. Yes, we're going to make, figure out all this stuff, you know, the quotes this time it's different motto and, you know, we can forget everything that happened before. I would say that the main lesson of history is that there are lessons of history, so to speak. That is that perhaps it's because we're still the same species with the same human characteristics. Perhaps it's just because there are certain necessary features to sort of how things build on other things, sort of causal connections between things that lead history to have some degree of repetition. I mean, it's, of course, the meta question of history is, is there a theory of history? Are there things that you can sort of formalize and say historical progression has to work this way? I mean, people will say things like when an empire gets well developed, it gets giant bureaucracies and eventually it falls over in its own inefficiency, so to speak. People say that. I don't know if that's really true as a trend across all of history, but there certainly are things where you see certain kinds of repeating patterns. And to say there's a theory of history, in part, it's like, well, let's take these particular repeating patterns and say, do we see these? And then on another side, let's say, is there a predictive theory of history? Kind of like Asimov's psychohistory idea, which. I mean, my principle of computational equivalence and computational irreducibility kind of denies the possibility of that in the end. But there might still be some pockets of predictability in history. If it works this way, then this is how it must come out. In a sense, quantitative finance tends to be all about finding those pockets of predictability in history. So we know there is some of that. Gosh, people commenting here about my recent efforts in what I'm calling observer theory and oh, gosh, all right, this is one that. Oh, lots and lots of interesting questions here. There's a question about the history of observer theory, yeah, it's long and complicated and full of philosophy, and I don't know it. And it's a significant amount of work to track it down. I mean, I am certain that from Plato and his cave to Descartes and his kind of views of I think therefore I am and so on, to more modern kind of versions of. Whether it's in control theory or systems engineering, where you actually have the notion of observers and measuring devices and so on, that are sort of part of the theoretical framework. There are just many, many, many places where this kind of overall idea of making a theory of observers has come up. The kind of particular knitting together that I've tried to do, I think has not been possible before because one hasn't had this overall framework that's come out of our physics project and the RULIAD and so on. That lets one kind of paint this bigger picture. But I think there are antecedents of what I've tried to do in observer theory. There are lots of antecedents, and really piecing those together would be very interesting. It's a lot of work, and I think some of it will be, in a sense, back projected history, that is. Now we can see with some clarity what's going on. We can project back and say, well, Descartes was more or less saying this. If we translate the Latin the right way and kind of we try and project back into his context, very different from our context. You know, I think it's interesting to see this, and I think there will be probably some insights that will come from understanding that history certainly will be useful in terms of seeing, you know, how does it relate to various kinds of Eastern strands of thought, as opposed to Western strands of thought, these kinds of things. Those are really interesting questions, which, as I say, I have not taken the extra few months to try to dig into that. In fact, in general, I would say that sort of, for me, I'm sort of hoping other people, and some people are picking this up, of looking at the sort of philosophical implications and the historical philosophical implications of a bunch of the things that we're doing in our physics project and so on. I think those are very interesting and they really help to provide context for what we're doing. But it's a lot of work. As I say, I just did it for the second law of thermodynamics, and it was a lot of work there. And I think some of these other things are perhaps a bit more difficult than that, because in that case, there's sort of a mathematical result that one is aiming for, in the case of something like observer theory, the question of kind of what one is, you know, how does one understand this notion of, you know, the complexity of the world being projected onto a mind and sort of the things that. Many things have been said about that. The particular kind of way of understanding that. That I developed, as I say, is very much born of modern paradigms in the physics project and so on. But seeing how that relates to things that people have thought about. About that before is very interesting and I just haven't done it, so. All right, I think I better go back to the day job here, but thank you all for giving me a chance to think a bit about history and I look forward to going on another time. I think if we look at the schedule here. Well, we're about to run into the holiday period and we may miss a few of these, but I'll be. For me, these are. I will be trying to do them as much as I can. All right, thanks for joining me. See you another time.