History of Science and Technology Q&A (December 27, 2023)

Hi everyone. Welcome to another episode of History of Science and Technology. Q and A. And I see we have a number of questions saved up here. One from David. If eyeglasses were developed in Italy in the 13th century, why was the invention of the telescope delayed until 1608? Yeah, I wondered about that too. I know, for example, Kepler wore eyeglasses and how could it be the case that nobody had thought of putting a converging lens and a diverging lens near each other and noticed that that gave you a telescope? How could that have not been done over the course of hundreds of years? I think the answer is that the lenses just weren't quite good enough to have it work. That the lenses, if you got long sighted when you got older, which meant that you couldn't read things up close, then that had been a thing that until having even the most basic kind of lenses that allowed you to keep reading. If you were a monk or something or somebody who was into sort of things of letters or a merchant or something like this, things that could keep you reading as you got older, that was a big deal. And the lenses didn't have to be that good. But to make a telescope, it's an interesting question how you would kind of characterize the quality of lenses to know what will make a telescope that would actually be useful or not. So I think that's the answer, that the quality just wasn't good enough. Even though the concept was there now, why people hadn't internalized the idea that yes, in principle you could make a telescope, but the lenses just weren't ground well enough. I'm not sure. I think that lens grinding became kind of a, was kind of a big deal. And I think in the Netherlands, where telescopes were invented, presumably lens grinding was quite the thing in the 1600s. And I don't know, for example, Spinoza famously was a lens grinder by profession and a philosopher as a kind of hobby. And that was kind of the, the high tech thing of its time, I think, was lens grinding. I mean, it's kind of like people often say, oh, you know, Albert Einstein went off to be a patent clerk when, you know, after he got his physics Ph.D. and you know, was sort of toiling away in obscurity. The truth is that at that time being a patent clerk was kind of like being an operative in Silicon Valley would be today was kind of the happening thing in high tech. And I think lens grinding was that in the 1600s. But this whole question about why things that are in principle inventable don't quite get invented. It's often this kind of. Does it really do well enough to actually work? I mean we had a great example of that in the end of about a year ago with the large language models of. Yes, people had known about language models for a long time. I mean back into the. Even in the 1960s people were thinking about kind of the statistical structure of language. But from Claude Shannon's 1948ish paper on information theory, that was kind of a statistical approach to thinking about language. And people had been kind of constructing better and better statistical models of language. But the idea that you could have sort of an AI that would write essays and answer questions and things, that was really a matter of. Not of the principle of the thing, but the details of whether it really worked. And I think the same is true. I don't know. Pick lots of different things that sort of were there but not there. I mean, another example would be drones. The idea of a quadcopter was a thing that had been there, but one didn't quite have the battery density, one didn't quite have the kind of control, fast control system and so on to keep it stable. So it's these things where in principle you can do it, but in practice it doesn't quite work. Or sometimes in principle you can do it but it isn't cost effective enough that it gets sort of economically developed and gets to be good enough that it's really practical to use. And I think it's always very hard to predict that. I mean, I know, oh, one of sort of a story from my own life. I It's a story my wife is fond of telling of a time this must have been early 1990s. We were building some piece of some house and there was a question of could one have a niche for a television that was only four inches deep. And I was like, yes, definitely. Televisions will be 4 inches deep. Now why did I say that? Well, I saw that, I said that because I knew that one could make flat screen televisions. In fact, I'd even seen a flat screen television from in the mid-1980s. And it was kind of my assumption that it wouldn't be very long before there were sort of flat screen televisions everywhere. But it turns out it wasn't true. It took another decade for there to be flat screen televisions. What did I get wrong? What I got wrong was that the rate of defect pixels in a flat screen television was the thing which was very challenging to reduce. If you make a memory chip or something and one of the memory elements is bad. You can kind of route around it. But if you have a television where you've got all these pixels displayed on the screen and one of them is bad, it's kind of like you're going to see the bad pixel on the screen. It's going to be super distracting. And so to get the yields up was a complicated engineering problem which took another decade to happen, but that was very hard to predict. So I think. That's one of these places where you can see what the general arc of what's going to happen is. But knowing when the details will come together is unclear. And then there are things where, like, for example, in space, for instance, there's plenty of examples where it didn't quite happen because there wasn't quite the political or economic force to make it happen, even though it could have happened. And that's, in a sense, one of the things that makes predicting the future hard. I mean, another thing is that you just don't have the idea of something like that being relevant or important or something people will care about. And then it turns out that they do. But in any case. But yes, I've always been curious about the glasses versus telescopes, but I think the answer is lenses just aren't quite good enough. Let's see. Gorgola asks, are there possibilities of researchers purposefully misinterpreting history to support their current research? I often hear some things are open to interpretation, but is there a case of this causing false experiments? Actually, I have to say, well, there's plenty of history being rewritten for political purposes. I've not. I'm trying to think whether there is history that's been rewritten for scientific purposes. I think there has. In the case of medicine, there are cases where people will talk about, oh, how some particular medical practice was, which. Yeah, I think there are cases there. And in physics, for example. It's an interesting question. I think the, you know, I suppose a case that I've encountered recently that's kind of not really in that direction is I've been very much involved in the idea of discreteness of space. And our big physics project, which I'm happy to say has made rather spectacular progress over the last few years, understanding the fundamental theory of physics, kind of starting from the idea that space is fundamentally discrete. Well, turns out I didn't really know this, but in the early part of the 20th century, kind of all the serious physicists were sure that space was discrete, and one by one, they just couldn't quite make it work and they kind of went off in different directions and invented big chunks of physics because they wanted to avoid having to address this question because they didn't know how to make it sort of technically work. So that's a case where, although I didn't know it, the history of physics is sort of very much in. Well, people really did think space was discrete. And this idea that a few academic generations later people say, no, no, that couldn't possibly be right, is an interesting case where the history was deeply forgotten and where it kind of went in a weird direction. There are other cases, like, for example, there's this thing that I think has been most popular in the last probably 20 years of kind of remembering people who invented things and tacking names onto names for things. You know, unfortunately, names for things are often not correct. And I think that as a person who is definitely, you know, has had both things where it's like, really, you're going to name that after me? And I, you know, I think I was involved in it, but it wasn't, wasn't particularly my thing versus things where, look, I invented that. And you're calling it something different kind of thing. It's the, the kind of, the, the way that history works is, can be quite weird. But there are certain cases where people are, you know, often for ideological kinds of reasons, say, really, we should make sure, you know, this particular person. We should make sure to tack the name on. Sometimes it's because the history gets done better. Sometimes it's because there's a point that's being made. I ran into this recently, the Fibonacci numbers, which are kind of a thing that were in a book by Fibonacci in 1200 AD roughly. But they weren't invented by Fibonacci. Almost certainly. They were things that were kind of known kind of among all sorts of people thinking about math. They probably were particularly known to folks in India at a date that's not very well known because some civilizations are obsessive datas of things. Like this document was produced in the 11th year of the reign of Queen so and so and so there's a definite date. And some civilizations just didn't date things. So Indian and Chinese civilizations, I think generally documents would circulate, but they just didn't have dates on them. And so one may be wrong by Hundreds of years, 400 years or more about when something was actually produced. But there's some reasonable claim that there were versions of Fibonacci numbers known in antiquity in India. And I think one could argue that there might be similar things in other civilizations. We don't really know. I don't think we found any Babylonian tablets that have Fibonacci numbers on them. So that's probably not probable, probably not a thing there. Although one has to remember that when it comes to finding things, things can be very confusing. Like for example, the Antikythera device, this clockwork computer that dates from sometime between the first century BC and the first century AD was found around 1900 in a shipwreck of small Greek island. And it's the only example we have of kind of computerish technology, mechanical computer technology from antiquity. The next example we have is, comes from the 1640s. But just that one discovery kind of makes it clear there was a whole strand of technology that existed basically 1500 years earlier or so earlier than we had imagined. So sometimes somebody might one day find some single Babylonian tablet that makes it clear that the Babylonians understood this or that thing that we didn't know about. I mean, I certainly have often thought in some of the things that I've discovered about cellular automata programs with very simple rules, there's really nothing about those things that couldn't have been done by the Babylonians. And I kind of wonder if one day some tablet will be unearthed that is kind of just those same patterns that I've made in the, In the, you know, that existed, you know, 3,000 years ago or something. Now, you know, I have to say, when I was trying to design the COVID for my big book, A New Kind of Science, which came out in 2002, one of the inspirations for that cover was kind of this whimsical idea of imagine a cellular automaton that had been carved in stone, so to speak, thousands of years ago, or maybe even in Paleolithic times and was found in a cave. And then somebody was kind of looking at this cave with a flashlight and had this flashlight and they illuminated this pattern which turns out to be one of these cellular automaton patterns. That was the original inspiration for what became kind of this rather nice, I think, cover design of a cellular automaton pattern with a sort of shaft of yellow light. But it's kind of the whimsical sort of storyline that that came from, was this idea that maybe all these things which I'm sort of describing as a new kind of science in this book were things that were really actually known 3,000 years ago, but where people had sort of not joined the dots to understand why that kind of phenomenon would be relevant. And Important for sort of connecting to other areas of science. So let's see. There's a comment here from Pierre, I think, about attribution. The same for LLM hallucination today. Yeah, I think that's the. You know, one of the things about sort of the way that, oh, I don't know, academic science tends to be done. It's. There have been different periods of time in terms of the way that sort of history of things is recorded. I mean, it is a very typical thing for the last, I don't know, probably 40, 50 years now. Just throw a reference in. Okay, let's. And people accumulate these sort of databases of references. And it's just like, I'm writing a paper about this. Let me wheel in my references about that kind of thing. And I see these papers routinely now, particularly in some AI kinds of things, where there'll be two pages of paper and three pages of references. And it's like, well, that's interesting for bibliographic purposes, but it's a little weird. And I think what I've tried to do for the last, I don't know, at least 30 years or so is I try to actually write the history of things, which is something where you're actually trying to narratively put together. And I suppose this live stream is an example of kind of an outgrowth of that effort is, you know, try and actually figure out the narrative of how did this discovery get made? And that's. To me, that's both intellectually interesting. It's also very revealing often in the actual nature of the discoveries. Why did people come to think this? Were they actually right? Were they misled? You know, who was the person? It's nice to recognize the person who came up with a good idea. Often when something's come up with, you know, there are fields of science, for example, where the winner is the person who proves the theorem, for example. Whereas the real, in my view, intellectual heavy lifting might be even imagining that such a theorem was out there. But there'll be a field where people will just say the winner is the person who puts that particular dots, that particular I, so to speak. And I think what is more interesting and more kind of useful in terms of understanding kind of the arc of what happened intellectually is to actually write these histories. And I've obviously done a fair amount of that. It's a lot of work. And it's always the case that at the beginning, it's like there are all these pieces and, you know, certain elements and actually threading it all Together is not trivial. And sometimes there are pieces of that thread where there are very famous things that turn out to be not quite as they were advertised. And there are things that are very obscure where you have to dig quite hard because the person who did was a sideshow to what they usually did, and they didn't really tout it very much, or they died the year after they did it, or for one reason or another, or the person was sort of ideologically out of favor in some way and they were canceled, so to speak. And so nobody heard about them or just a zillion different things. And it's often significant work to go and figure out sort of what really happened. And more than that, when there are big ideas, they tend to be ideas that exist within a certain context. And sort of figuring out what the precursors of those ideas were without the context can be really hard. It's like, look, this person said it might work this way, but they didn't have any context around it. And so that became a sort of isolated claim that was really not. It didn't really connect to anything. And had one actually talked to that person. And in some cases, I've had the opportunity to do this in the case of some of these things from the past. Sometimes you talk to them and you realize, yeah, this person really understood the core point. They just didn't communicate in a way that had the context that connected to the things we understood later. And sometimes you're kind of shaking your head because it's like, wait a minute. They have this one thing that they were pushing, but it wasn't really connected. They didn't really understand how it connected to anything else. I noticed this a bit. A chap. I wrote a long, complicated obituary, probably the most complicated obituary I've ever written. Person named Edward Fredkin, who died a few months ago. Yes. In the middle of this year. He was a person who had a bunch of interesting ideas, and he didn't really connect them. They didn't really connect. And I understood. I knew him fairly well. And he would always say, this is an important idea. And it's like, wait a minute. For example, he was very keen on the idea that a cellular automaton could be made that would reproduce the universe. And I'm like, that can't work as such. And yet he was sort of very keen on this idea and sort of had various kinds of. But it didn't contextualize in a way that connects it particularly usefully to kind of lots of other ideas about discrete space that pre existed and that sort of I've constructed and so on. It was this one sort of pronged that was kind of a separate thing which had its own consequences that were quite interesting for reversible computing and things like this. But so it's really hard to. It's often very hard to sort of do the history and put the pieces together. I think that there is a tremendous tendency in modern academia to just say, let's throw the references in references 5, 6, 7, 8, 9, 10, and there's some random paper. The person hasn't even read the paper. I mean, I always find it amusing that people will regularly detect that there was a reference which was an incorrect citation, the page number was wrong, or something like this. And it propagates kind of like a defect in the genetic code, so to speak, or some weird feature in biological evolution. It will propagate for many sort of generations of papers, so to speak. So that's kind of a funny thing. The, let's see, Prior was commenting on the replication crisis and how many PhDs would be flagged for plagiarism. Yeah, I don't know. I think I have to say, I would say that the answer is not as many as you might think. Because I think this phenomenon of kind of industrialized sort of high pressure science that gets people to do things like copy paragraphs from, you know, from somewhere else. I think maybe I'm wrong, but I think that's a more modern phenomenon. I think that people who bothered to write a PhD back in the 1960s kind of really wanted to write a PhD. They were not doing it for the sake of getting to the next rung of the ladder, so to speak, which I think is what leads to a lot of this. Look, it doesn't really matter, you know, and I just have to fill the thing up with something so that I can get my PhD certificate and go on to the next thing type thing. Maybe I'm being too cynical about that, but my impression is that I think that less of a thing now. One of the things that one always thinks is that, oh, in the past everybody was terribly clever and produced wonderful books and things like this. What one has to remember is that the good stuff survives, the bad stuff does not. And so if you look at what books were actually being published in the 1800s, let's say about science, there were some good ones which we've heard of today we say, look at those wonderful things that were produced in the 1800s, but there was also a lot of kind of very junky stuff. Very derivative, very kind of polemical kind of nonsense, et cetera, et cetera, et cetera, that also existed at that time. It's just people didn't keep it somehow. It's like the used bookstore phenomenon or something of how much do you. There are books where they show up a lot in used bookstores and you say, wow, a lot of people must have this book. And you realize those are the ones that people threw out. Those are the popular science books that show up because people threw them out. I have to say I've always been rather flattered that my book A New Kind of Science is comparatively rare in the used book world and considering how many copies of it I know we sold. So I've always been rather flattered by that on the grounds that people actually kept it rather than it's a book that has value in the used book trade but hasn't shown up there as much as others. And that's kind of a sign of something, I suppose. But I have to say that the whole story of sort of skullduggery and science and there's plagiarism where it's like for the sake of saving yourself work and there's plagiarism for the sake of stealing something from somebody else. And I think those are little different. And I certainly, when I used to be in the middle of kind of academic science back in the late 1970s, early 1980s, particularly at that time, I just. Skullduggery of people stealing things was absolutely rampant. Now I have to say I might have been in one of the more cutthroat fields that was particle physics at a time when there was a lot of low hanging fruit discovery to be made. And it was kind of one of these things where sort of new discoveries were being made every couple of weeks. And so it's like, who's going to grab that piece of fruit? Who's going to grab that other piece of fruit? For myself, it was something where I kind of decided I would work on things which are a little bit less in the fray, so to speak, because I find it more interesting to do things where I don't feel like if I don't do it, I can feel like if I don't do it, nobody's going to do it for 50 years rather than if I don't do it, somebody else is going to do it next week type thing. But certainly the instances of skulduggery were absolutely rampant. And it's one of these things that's poorly policed, I suppose, in Business. There's lots of skulduggery that can happen too. But in the end, there's much more in terms of contracts and the legal system and God knows what else. That tends to be kind of looming in the background. That I think in my sort of imperfect anecdotal evidence, I would say the density of skulduggery is higher in academia than it is in, for example, the tech industry. Although not to say there isn't any in the tech industry, but it's somehow of a different character and somehow more, perhaps more insidious in the academic sphere. I mean, it's complicated. And I have to say, in my own experiences of this, I realized only, oh, gosh, decades later that perhaps one reason I was more exposed to this than I might have been is that most people, they operate in a particular, very narrow area of academia. They just know all the people in that area. They keep operating within that one area. And it's kind of a single uniform social clique. And one of the things that I've done quite a lot is to sort of enter fields from the outside, so to speak. And that means one's not immediately, I have to know the people in that field, at least some of them, often the leading people in the field, but I don't know the whole kind of social clique there. And I think perhaps that sort of entering the thing from the outside, which I hadn't really internalized, was such a rare thing to have happen, is a reason why one ends up seeing more skulduggery. Because I remember one time, one very, very egregious situation of a person who's no longer alive, actually, but the very egregious case of just bizarre, where eventually the whole thing had sort of come crashing down. And I asked this person why this was just an egregious thing. I mean, why did you think you would get away with this? And the person said, after some pause, they said, because you're not in our field. Okay, it's interesting. That was sort of an interesting kind of. I don't know whether that was. It wasn't really intended as a justification. It was intended for a, you know, why would one get away with this, so to speak? By that point, this. This person and their collaborators sort of admitted that this was a piece of pure, pure skulduggery, so to speak. So I thought that was interesting. The. And the whole thing was very annoying as well, let's be clear. But at least given that it's going to be annoying, one might at least have some human and sociological interest from the whole thing. The, let's see, Ogolo is asking what is best for historical research? Photos, text, artifacts. Depends on the period. I mean, when I am sort of trying to understand how I came to understand things. A very self centered view of history. I have good records and I have good computerized records. And so if I want to know how did I discover this thing, I can often pin down to the second, when did I generate that picture that caused me to do this? And I can see I did this recently actually in looking at the history of some things I'd done related to, well, in the end, the second order of thermodynamics. Where this was, oh gosh, must have been sometime in the 1990s. I was generating orphan language notebooks, Mathematica notebooks, and they have timestamps in them. And so I was able to say I generated this picture. And I could see in the next three minutes after that that I got very excited about this and I was generating more pictures and so on. So I could really see the chain of thought that happened there. And then an hour later I went to sleep or whatever. And so that's a case where you can really see the chain of thinking at a very granular level, right down to the second when this particular picture was generated and so on. I think if you look at different periods in history, for example the 1800s, I will give a shout out to the Internet Archive, my friend Brewster Kahl, who runs that, for all of their scanned material from that period. That's just amazing for historical research that you can just go and find kind of, you know, this obscure journal, these obscure kinds of things. So that's a place where read the primary material, don't read the. I mean, you know, sometimes there are good historians who've really threaded things together in very interesting ways. Those are worth knowing about. But I have to say in my own experience, go read the primary stuff. The primary stuff, you've got to read the primary stuff. If you want to do any serious history research, you've got to read the primary stuff. And sometimes that's in other languages. That's not such an issue these days. Sometimes it's in Gothic script in German. And now there are finally OCR systems that can deal with that and there are good translation systems, et cetera, et cetera, et cetera. And I think then you have sometimes the challenge that people explain themselves in ways that are completely alien from today's ways of explaining themselves. And that's difficult to Deal with mathematical notation, for example, since pretty much the early 1700s, has been pretty understandable to us today. A few weirdnesses. But I would say you can understand. You can even understand it back to Leibniz and people like that. I would say that when you get things that are less formulaic, so to speak, less formalized, it becomes more difficult to understand what's going on. And sometimes there are papers that are very beautifully written where you really can understand them. Sometimes they're really incredibly badly written and just very obscure and very poorly constructed. And that doesn't help, but it's always really worth reading the primary literature. I mean, I think that's the. You really can't do history seriously without doing that. Now the next level is going to visit archives. Sometimes you can request material from archives and, you know, you pay a few dollars and some archivist will make you a scan of something. I actually find it much better to go visit the archives when I can, because usually it's like, okay, just give me box number such and such. And you can go through box number such and such. And you often find things that you never would have guessed were there. I mean, many archives have these finding aids which are kind of just rough. This is a letter from person A to person B. But you really don't know what it is unless you look at it. And so I think that's an important thing to do. And somehow one of the things that seems to happen to me almost every time I visit an actual archive, the archivists will say, oh, yeah, you're in the tech industry. When will somebody make a really good system for maintaining online archives? And I've been slowly working towards that. I intend to do it for my own archives to begin with, just as a test case. But it's been a surprisingly slow process. I mean, I think the technology we have now, between our notebook technology, a bunch of machine learning technology that we have, a bunch of computational technology, a bunch of actual data on history that's useful for kind of pinning things down and making timelines and so on. We actually have very good technology in our technology stack to make kind of a great archiving system, kind of my plan. There is, for example, for me, I have, oh, I don't know, 400 boxes, I think, of primary material that's all scanned. And then the issue is, you have that material, you can look through it. It's in boxes that are more or less categorized. The categories sometimes are a little bit wrong. Sometimes there's a document where the beginning and end of the document aren't the same thing, so to speak, where they got smooshed together in some way they shouldn't have been. And sometimes, particularly from early years for me, there's handwritten stuff which is not yet OCR able, but it will be. And we've tried over the years actually to train machine learning systems on my handwriting, since I have a bunch of that. And I also have cases where we have training sets where I have a handwritten version of a paper and then I have a typed version of that paper. So we actually have a kind of supervised learning kind of training process to be done there. But any case that hasn't yet been successful, but it will be. But I think the thing that I've wanted to do is to be able to sort of take the raw data from these just boxes of stuff to write narratives, what was going on. Because I often will remember, you know, I'll see some documents and I do remember roughly what was happening. And it's kind of like, let me put together these documents in a form that actually makes sense, put them together in a notebook, have some way in which the things you put in that notebook are appropriately linked back to the original source documents and so on. I think that's the way to sort of make essentially an overlay of an archive that becomes the beginning of sort of a computational history, so to speak. And by the way, you can get lots of machine learning things to help you in sort of piece together timelines and so on and figure out when was this letter, was it before or after this letter, these kinds of things. So that's the kind of coming attraction of what's possible in computational history. But I would say from the 1800s, the primary thing is Internet archive and scan books and so on, and scan journals and then go visit the archives themselves to find things. In terms of photographs, they are certainly of more human and whimsical interest. They're less important, typically in terms of tracing the history of ideas. Although sometimes there are funky things like there'll be photographs of a computer or a machine and it's like, what was this thing? And sometimes it's really pretty hard to decode what it was. Sometimes there'll be photographs of people and it's like, well, okay, those people were in the same room at the same time. And that tells you something. Although I have to say, I've certainly been in plenty of situations where somewhat frustratingly, one will be. It's like, oh yeah, I want to talk to that person. I'm going to be on some events, and I'm going to be sitting on a stage with them at some. Some random event. And then you realize you just sat on the stage for an hour next to the person. You didn't exchange a single word with them, and you never do. And so it will be misleading to say, well, there's a photograph of these people. They're sitting right next to each other. They must know each other. Well, not necessarily, but I think what I found in all the stuff I've done in history is that there is always a narrative. There's always a. You know, it went from here to there to there to there to there. There are really no miracles. There's no. And particularly in intellectual history, you really don't find a case where somebody has a big idea that just comes to them one day and that's all there is. It's always a story, a decade of development or at least five years of development or something where people are kind of moving in that direction. They're contextualizing things so that they are ready, so that maybe they see something. And then when they see that thing, they realize, aha, that's this. That really means something big. But they have to be primed for that. And without that, they just don't notice it. And when you go back and you try and do sort of attribution in scientific history, it gets very difficult because somebody may have seen something but have no idea what it was. And it's like, does Robert Brown, who saw Brown in motion in 1830 or something, you know, get credit for the discovery of molecules? Well, not really, because he saw Brownian motion, but he didn't know what it was. That wasn't known until around 1900. And it's, you know, so it's these things where you can see something but you don't know what it is. And that's a feature of, I think, with sort of. And then you can ask the question. And this is an argument for writing history rather than just quoting references. When you write the history, you can say, brown saw this thing, but it wasn't understood until 1900. So it's kind of like if you're trying to say, well, who was really responsible? Who should get the Brownie points for that? It's kind of distributing those in a way that is accurate rather than just saying, oh, we'll tack a reference in here. Let's see. Oh, there's also a question from Gogollo about artifacts. Yeah, I mean, that's an interesting issue because that affects whether you could have an archive, for example, which is a pure scanned archive. Whether photographs of things are enough or whether you have to have the physical artifact. Most of the time, for me, the scans have been enough. I have to say. I was writing historical stuff about my big book, New Kind of Science, which in the end discussed the actual production of the book. And I had to go look at the printing plates, which I still have, or at least I have a few of the printing plates, and look at them under a microscope. And that was to figure out how we managed to get certain things to work and so on. That was something I would never have expected. And I'm glad I still had the actual plates so I could pull them out, look at them under a microscope, because when I scanned them, they were scanned at whatever resolution, a good resolution for the human visual system, so to speak, but not good enough to be able to say, oh, let's use a microscope to look at this in more detail. I would say that in terms of artifacts, in terms of learning the history of science. Well, we did a project, in connection, in collaboration with the National Museum of Mathematics in Manhattan that I've long been involved with. We did a project, you can find it on the Web, History of Mathematics project or something, which is sort of a precursor to a potential gallery for Museum of Math that will be a collection of mathematical artifacts. And so we were trying to sort of inventory what are the significant mathematical artifacts from history. And that becomes sometimes even more difficult to interpret than written material. Like, you'll have some artifact that is, I don't know, some strange board that has markings on it and so on. And it's like, what really was this? What did people understand from it? And so on. And so it can be difficult from artifacts. I mean, we see this in the most extreme form in archaeology, where we'll see some object from 5,000 years ago. And it's like, what was this? We have no idea. Is that if we have a context that connects to things we do today, we might say, oh, we understand what this was. It was a place where they kept, you know, grain or something, because it seems it is, you know, we can visibly tell that it has that functional form of a function that we still do today. But if it was, you know, the altar for the God of ants or something, and had some special sort of ant, you know, ant paths in it or some weird thing like that, I just made that up. But it's like, we would never know, because the God of ants is not really a thing that is in wide circulation Today, and so that would be a thing. Very hard to tell what it was. And we see that there are extreme versions of this in cave paintings and so on, where there, for example, are these, you know, handprints, black and black and red handprints, and it's like, what on earth were these? I mean, what were they? Was it just kind of for fun or was it. Was it very significant? Was it, you know, oh, if you are elected to the distinguished, you know, cave society or something, then you get to put your hand on there. What was it? It's not. Not clear. So, you know, with artifacts, that can be an issue that it really is. There is even less of a thread of connection than there is in written material. I would say that it's an interesting question whether there is written material from. Yeah, I think the amount of written material that's just completely not understood is not great. I mean, there are examples of that. You can certainly see that in kind of secret and sacred writings, Kabbalah kinds of things and so on, where it's like, it's not really clear what this meant, but I think for the most part, written material, it does tend to be somewhat, you know, one can sort of decode what it's talking about. Isla asks the interesting meta question. What do you know about the history of archiving? When did archiving begin? That is an interesting question. Well, I can tell you a few things. I mean, look, the whole idea of writing things down and storing records that originated when cities originated, when it became necessary to have sort of organized ways to keep land records, things like this, that was ancient Babylon, that was 4,000 years ago or something now. And we still have plenty of the tablets because they happen to write on clay tablets that got baked in the sun and so on. And you can buy them on ebay for a limited amount of money. There are lots of these tablets around. I have one, for example, little thing that is the receipt for some, it's actually kind of a barley, an option for buying barley at some point in the future, and some kind of receipt for that option. And that's a piece of sort of record keeping that we still have, you know, 4,000, 3,000 years later or whatever. Other civilizations didn't keep their records even if they had records. They kept them on, let's say, papyrus or something like this. And most of those are gone. Usually in many civilizations, kind of the sort of in the monumentary buildings, so to speak, there would be inscriptions carved, you know, you know, whatever it is, you know, Darius is the king of Kings and, you know, then there'll be a whole description of all of the amazing things that Darius has done. That's a real person from Persia, where the fairly famous sort of list of the achievements of Darius and that was a fairly common thing to have sort of carved in stone and so survives in a sense. That's a. I don't know if that's a kind of archiving or not, but it's the way we know a bunch of things about people, I suppose another form of archiving is funereal archiving, whether it's burials of people buried with things that were characteristic of their time or of the person or recorded their deeds and so on. And there were different belief systems of whether it was the Egyptian come and be weighed in the balance type thing, the heart being weighed against the feather type thing. And. And, you know, you had to have sort of the ammunition of your good deeds to help with that. And that meant that those things were recorded and kept as part of the funeral material, so to speak. I mean, I always, I suppose a striking thing for me, years ago now, you know, one certainly knew that the purpose of all of this sort of fancy burial stuff was sort of for the afterlife. And in a sense, it seems like a big fail in the sense that people go to all this trouble, they put all this stuff there and then it's just sitting in some stone enclosure in a pyramid, sitting there forever, so to speak, or until it was stolen by some grave robbers type thing. But I was realizing this. I was at some museum somewhere looking at these Egyptian artifacts and I realized, well, in some sense they were absolutely right. They have an afterlife because they're sitting in a museum. And now 4,000 years later, we know about their lives that we absolutely would not have known about had they not archived that material in a pyramid or something like this. So that was, I think, an early form of kind of archiving beyond life, which had at the time a nominal belief system having to do with sort of the success in the afterlife. But in fact, the true afterlife, in a sense, is the intellectual afterlife of us getting to know what these folks did 4,000 years ago or whatever. Now, in terms of other types of sort of maintaining records, governments have maintained records. Roman Empire, they maintained records they had for purposes of taxation and things like this. Lots of records are maintained for purposes of kind of recording what happened in government, so to speak, for purposes. Sometimes people would. So, I mean, those types of things have been preserved. I think that the question of when. Well, when did things start getting sort of preserved in organized places. The Library of Alexandria, for example, was a place where apparently there was a giant collection of scrolls from all over the world that recorded that had all sorts of interesting material. It was a big library. I mean, when it's called the library, it was really a university. I mean, people like Euclid, for example, taught there. And it was a place where people would come to learn things and they would learn them from humans. And they would presumably also go pick the scroll off the shelf or whatever and go read this scroll. You know, unfortunately, the Library of Alexandria was destroyed in a fire. Actually, there were several different pieces of the library and they were destroyed at different times. And some were destroyed when, what was it, Cleopatra and Mark Antony and Caesar and so on were having their big kind of punch up. And I think that was one time when part of the Library of Alexandria was destroyed by fire. There's always this, and this is where history gets so messy because, you know, there's this whole sort of, sort of anecdote of history that says that when the. Islamic Muslim invaders to that part of the world came in, that they were kind of, you know, the oft quoted line is, you know, anything that was a scroll in the Library of Alexandria, if it's in the Quran, we already know it. If it's not in the Quran, we don't care, so to speak. And therefore they burnt the other scrolls for heat. This story is almost certainly complete nonsense. But it's one of these stories that gets told for reasons of sort of history, ideology, I think, more than anything else, and repeated because it's a good story. But the actual historical facts may be much more complicated and may involve, yes, there was a belief that certain kinds of things, they didn't want these kinds of scrolls there. They believed in this, they believed in that. There was sort of pressure of this kind and this, who knows? But it's these kind of. The quick versions of history are usually not right. But any case. So Library of Alexandria was another, was sort of an early example of a place where people were trying to collect kind of knowledge, archiving, so to speak. I would say that the real period of time when that really got started was things like the Bodleian Library in Oxford and oh, what is it? Bibliotheque Nationale something in Paris maybe. I think at the time when the universities were getting started in Europe in the 1200s, I think there was sort of the beginnings of that kind of thing. I think, I'm trying to remember, maybe the Bodleian Library dates from the 1600s, I'm not sure, but there was kind of this idea of keeping material was something that I think by the time, okay, so the British Museum was founded, I think in the early 1800s. And so there was a big sort of period of museumification that happened by the mid to late 1800s. Things like the Natural History Museum, the in London and so on. But you know, things like the British Museum which were keeping artifacts, I think in the period, in the late 1700s, early 1800s, there was sort of an interest in looking back at history. I mean, you know, at different times in history there have been sort of political regimes which say sort of forget all of history, we can invent everything for ourselves. And the modern is the only thing that's good. And there have been other times when it's like we are rooted and our credibility comes from our rooting in history, so to speak. And so both of these things are forces that once seen in sort of political history. And I think people like Napoleon, for example. Napoleon was very big on Egyptology and kind of in a sense the culturification through sort of the connection to this grander culture of the past, so to speak. And Napoleon was responsible for bringing a whole bunch of Egyptian artifacts to Paris. And then when the British kind of won those wars, then a bunch of that stuff wound up at the British Museum. And these Cleopatra's needles that show up and there's one in London, there's one in New York, I think there's one in Paris. I think there were quite a lot of these things available in Egypt. I think you didn't. Although kind of the concept of, oh, we're just going to take this giant, very many ton rock and just sort of take it out of here. It's hardly as if that was done. It must have involved a lot of donkeys. And a lot of these are not easy things to sort of pick up and put in your backpack type thing. But in any case, that was a period. I think the 1800s was. There was late 1700s, early 1800s was a big period where people got interested in sort of maintaining history in that way. I mean, certainly the historical sites even were often just like, oh, well, you know, in Rome, for example, lots of historical buildings, people would scavenge the rock from them, you know, the stones from them and so on. It wasn't until much later that that was sort of a. There was an effort to preserve that history. I would say that in, in terms of kind of the archiving of things like papers that people have produced. I Mean, usually papers get preserved in complicated ways. So for example, it's always amusing though there will always be this term, the so and so, nachlass German term, which means, I think, suitcase. And sometimes it really is literally that. For example, Charles Babbage, early sort of engineer, originator of mechanical computers, lived in the first half of the 1800s mostly. There was a thing was always called, there was a bunch of material from him that was in. I happened to research his history quite a bit was in the cowhide trunk. Okay, what was the cowhide trunk? Well, I actually got to see the cowhide trunk. It's in the Museum of History of Science in Oxford. And well, what had happened to it? What happened was when Babbage died, his lawyer had wound up with this material from Babbage, a bunch of papers, a bunch of plans of the Analytical Engine and so on that were in the cowhide trunk. Well, what happened to the cowhide trunk? Well, it was actually in the playroom of the kids of the lawyer for pretty much a generation. And at some point somebody opened the cowhide trunk and said, oh, there's a bunch of drawings in here. I don't know exactly what happened next, but eventually they wound up at this museum. So there are weird things like that that happen. I mean, for example, in. Oh, there are people who will contact me, for example, and say, oh, my grandfather or great grandfather or something with some well known scientists. And I've got tons of boxes in the attic that were from this person. And it's like, okay, that's great, they've been found. There are probably good places to keep them. It can also be very complicated. There are things where things will get scattered to the winds and then will show up again in very complicated ways. So, for example, the Babylonian astronomical diaries, the longest running science experiment in history. 700 years of astronomical observations. We have about a couple of percent because they were daily observations. So we know how many there were taken. We have a couple of percent of those tablets. Where do they come from? They came because people bought them and random antiquities dealers around 1900 and they presumably were in one place. Maybe they were in one place, maybe they were in several places. But by the time maybe they had been excavated maybe one 200 years earlier, but then they'd been scattered to the winds and people were just buying these, oh, here's a cool Babylonian tablet. I don't really know what it says. And then eventually people, particularly a couple of people, really tried to go and find all these Babylonian tablets and piece together the ones that were the astronomical diaries. So they kind of got reconstructed that way. For example, I. Well, this is a story of my own. I wrote about this in some. Some piece I wrote 2019, I think I was a former high school teacher of mine who. Well, let's see, how did this go? There was a former high school teacher of mine who died, who gave a book to another former high school teacher of mine with the statement that it should get given to some appropriate person in due time, so to speak. And so the second high school teacher of mine gave this book to me in 2019. And it's a book. It's a quantum mechanics book. It was originally owned by a chap called Norman Routledge, who was a high school teacher of mine, but had been a friend of Alan Turing's. So, okay, so I get this book, and I actually have another copy of this book. It was Dirac's quantum mechanics book. And I'm leafing through this book as the person's handing it to me at breakfast I was having with them. And I leaf through it, and there's a piece of paper in it, and it's like. And this is a book about quantum mechanics, and it's a book. Oh, I should explain that this book had been owned by Alan Turing and had gone from Alan Turing to Norman Routledge and from Norman Routledge to this other teacher of mine, and then to me. So it has this piece of paper in it, and this piece of paper had something which I could recognize was mathematical logic. So that's pretty weird that you have this book that was Owen Ballanturing, and then it ends up being a book about quantum mechanics. And it then has this page of mathematical logic, and it's like, did Alan Turing write this page? And so I look up the sort of handwriting of Alan Turing, and I couldn't tell where it really matched. I know a handwriting expert who said she didn't think it matched. But then I kind of went into the sort of deep dive of, what could you tell about this book? Well, it had a bookmark in it, and the bookmark came from a bookstore in Cambridge. And the bookstore in Cambridge had a telephone number on the bookmark. And the telephone numbers changed at different times, so you could tell when this bookmark was from and so on. And then there was a. Oh, that's right. There was a piece of paper that had a watermark that was from a German journal that had been published, only, you know, had started publication at a certain time anyway, a whole long chain of events. In the end, the piece of paper wasn't written by Alan Turing. It was written by a person called Robin Gandy, who was a student of Turing's. But in any case, that was a case where it's like, you know, how do you actually go sort of this archive? In a sense, that piece of paper wasn't an archive, it's just free floating. And there are, you know, I have a bunch of documents, for instance, that, yes, I have pretty good archives myself, but they're documents that are sort of fragments from archives that, where they're quite substantial archives somewhere else. Like I have a bunch of material that Dick Feynman gave me which is, you know, sort of my little fragment of something which would be a larger archive of his work. But, you know, so I think the. And things get very complicated with archives because, for example, there are questions like who owns this material? So, for example, one that I happen to know a little bit about is Ada Lovelace's archive, some part of which wound up at the Baldean Library in Oxford. And there's a complicated, there was, I think it's now resolved, sort of a complicated mess going on between Ada Lovelace's descendants and the library. And it had been stuck in this library sometime in the late 1800s and it's been sitting there for 100 years or more. And it's kind of like, okay, but what happens to it? Who owns it? Well, maybe it's actually owned by the descendants and it's probably worth a lot of money. And so then the question is, well, can you just sell it? Maybe it's worth, I don't know, $20 million, let's say, can you just sell it and then somebody gets the $20 million and it goes to some happy collector's home rather than to a kind of a general long term library. Well then, of course, things get more complicated because there's a lot of forces at work. So it's like, well, most of the collectors are American or maybe Asian. And then the British government can declare that these are documents that are in the national, they're sort of national treasures, they can't leave the country. And so that then puts a whole constraint on the thing and then it becomes a whole negotiation. I'm not sure what happened in that particular case because I was looking at that material in 2015, I haven't looked at it since. But any case. So, gosh, let me see, Let's see. Baby is commenting that what we've been talking about today pinpoints how critical it is for scientists to know the history, history of their Field as well as become excellent communicators. The. Yeah, they're commenting that I'm correct in my observation that people just have the sort of the bank of references and wheel out the references to sort of fulfill an obligation. You know, it's a funny thing, the history of scientific publishing, because back when scientific publishing was first really being developed, which was essentially the Proceedings of the Royal Society and edited by, I guess, Edmund Halley originally, that was in the 1650s, 1660s or something. And it was really, if you read the early volumes of the Proceedings of the Royal Society, it was very kind of whimsical. It was very much, you know, I saw a goat that had, you know, horns that were shaped like this, or I saw a flash of green right at the moment when the sun rose over the Alps. And then it had some extra color because, you know, but my view was obscured by the fact that, you know, I don't know, a herd of cows came out at that time or something, but it had this kind of whimsical extra color in it. And I think what happened over the years is that in sort of the professionalization of science, there was sort of this belief that things. And perhaps also a little bit, you know, science had many influences. You know, in various periods of time. Germany was sort of the top country, with France as sort of runner up and the UK as an even, you know, as a further runner up. And then the US kind of entered the scene in the 20th century. And, you know, each of those countries had a certain cultural. Left a certain cultural imprint on the way that science was communicated and done. And I think to some extent, the sort of ultra professionalization of, let's make it sound very official perhaps that was, you know, the French style is a bit on the formal side, particularly in mathematics. The American style could be something of the. Just a little bit funny for a country that was sort of from so many pioneers. But it became at some point the very sort of ultra professionalized version where things had to sound very official and where even the most basic sort of ideas were couched in these complicated words and jargon and so on. And that's become something that is a feature of science. It's a feature of other fields. Even philosophy, for example, went from kind of a time when people wrote things in sort of plain language, that a time when they were righted in terms of layers of jargon, so to speak. I think that often that is done for, you know, for the appearance of professionalization, so to speak. And more cynically, it's sometimes done because people actually don't know what they're talking about. And if they were forced to explain those things in simple terms, they would fall apart. I mean, it just wouldn't be. They wouldn't be explainable. I know in my own sort of story of trying to explain things, for me it's an important constraint that I try to explain things as sort of simply as I can. Because once I can do that, I actually have confidence that I understand the things myself. And any time where I fall back on, oh, there's a technical, formal way to describe this, it's. And that's the only thing I can say. It's a super bad sign. And I know myself well enough to know that's something I don't really understand that well. And I think that. But it's somehow, if you know that formalism, it can be easier to be at that level rather than saying, let me break it down into sort of these simplest component parts. But I think it's a big shame in many ways that sort of scientific communication evolved to the point where so much. I think there's been a little bit of a trend backwards in more recent years where people don't just write a paper that just says Smith and Jones showed X and Y, now we're going to show Z. That's a small increment of that. But they're not really explaining why they're doing this, why anybody would care. There's a bit more sort of. There's some pressure I think from some journals and so on to say why would we care about this? And that I think is very worthwhile my own form of communication. I sort of have decided that in my efforts to sort of eke out a certain amount of time to write things and I end up writing a fair amount. But, you know, I've done that in a very free form way where I kind of just can have defined a way of writing things that lets me write kind of technical things, right. Lets me write narrative kinds of things. And I don't feel any compunction not to make the whimsical point right in the middle of some quite technical thing because I'm not, you know, I feel like I don't. I'm not trying to prove how professional it is, so to speak. I don't think that's really the point. I'm just trying to communicate what I'm trying to communicate. I think also in terms of the history side of it, what I found is that as I say many things I've done. I've really tried to dig into the history sometimes in a kind of a. I suppose it's not really a. It's a variant of the wheel in lots of references I've made bibliographies of fields like did that for combinators and for post tag systems and things like this where I really tried to go through and just get the complete story. What were all the papers published in this field in the last hundred years since the thing was originated? And in these sort of fairly small areas one's able to do that. And I think it's sort of useful to see the flow of those bibliographies. I think it's less useful than seeing the narrative history, but it's sort of useful. But when you have bigger fields where there are just, you know, 10,000, 100,000 papers, whatever in this field, it's just hopeless. I mean the only way you're ever going to understand things is to try and actually piece together a narrative history. Huh. It's an interesting question from Conformal here. What has history taught us on how to deal with a science community that is misguided? That's an interesting question. I mean, I can think of many misguided moments in science, some quite unsavory, like the eugenics movement of the early 20th century and, or phrenology, you know, the shapes of heads and things and as a, as an indicator of human characteristics, things like that. I think, you know, it's complicated because there are plenty of things where people have been viciously attacked for what seemed like nonsense, like the person who, gosh, what is their name? And maybe Austria or Germany who kind of sort of understood the germs theory of transmission of disease and was a doctor in a hospital and they were like, you really should wash things, so to speak. And people didn't. And eventually there was such a sort of this person is non. Is crazy. You know, they're talking about unseen, you know, things that carry disease that, you know, this has to be, you know, this. They have to be nuts. And of course it turned out they weren't nuts and in fact they were quite right. But you know, and there are plenty of examples of people who sort of see a truth, so to speak, and don't and other people don't see it. I mean, I have to say in my own life and times I've had plenty of examples of that where it's kind of like, you know, there's something which, you know, because of things I've tried to figure out or whatever. Seems kind of obvious to me. But yet people say, no, no, no, it can't be true, you know, because we know it isn't true. Because there's this whole stack of scientific papers that assume it isn't true. It's not like it's proved it isn't true. It's just we assume it isn't true and people have built a lot of on the assumption that it isn't true. I think, you know, Max Planck, I think, who had his own issues, I would say was, I think has a quote that the progress of science is measured in a procession of funerals, so to speak. By which he meant that people hang on to some belief about how things work and they'll hang onto it their whole career. And it's only when the next generation comes in that that can change. And maybe that's a little bit of the answer. You just have to wait. And it's a little bit. It may be, it perhaps is damaging for society that they have to wait another 50 or 100 years. It's disappointing for the originators of the ideas because they're dead by the time anybody says, oh yeah, that was a clever idea type thing. I mean, that's a feature of inventing things that are sort of ideas from the future is typically you won't be around at the time when people realize, oh yeah, that really was a good idea. It's a shame, but it's, in a sense, it's at least fun to be able to see the future, even if you don't get a pat on the back for being, for being right. I think that. Well, let's see what are some examples of. I'll give you an example. I mean, I think I would say that, yeah, it's complicated. I mean, there are things even that I've originated where people have run off and done things that I thought were kind of silly with the things that I've originated. I mean, that happened a bunch in kind of complexity theory and so on, of things that were kind of sort of ideas and discoveries of mine that then got applied to things where, like, that's not a very good application, that's not a sensible thing. I don't think I could have a good thing to do with that other than, I don't think. And it's very complicated because if you just say, oh, it's all nonsense, kill it all, that's not really a positive thing to do because you've got a field that is a, A young field that's sort of Struggling to survive. And it's doing a few kind of crazy things. But it's like you don't want to kill off all those things. You want to allow some degree of kind of craziness, because maybe some of those crazy things will turn out to be right. Also. Just the very fact that they're being done under the banner of this field is important for the future development of the field. So I think the only. Perhaps you just wait, but you might have to wait 100 years for sort of the misperception to be kind of worked through. I think the other thing. Well, the other thing realistically is just do what you believe, so to speak. And the fact that lots of people are telling you, oh, it couldn't possibly be true, because, you know, there's this whole orthodoxy that builds up about how that couldn't possibly be true. If you've got a good reason to believe it and you really, you know, you believe in your own sort of tower of belief, so to speak, then I think, you know, go for it kind of thing. And it, you know, what does it take to sort of move the orthodoxy in that direction? I'm not sure. I mean, I've been in the middle of that a bunch of times and I think it's sort of a gradual increase in the number of people who can do things with the particular direction you've defined. And then that eventually starts taking over. It's very rarely the case. You would think with all the sort of story of falsifiability and science and so on, you get to a point and some experiment will be done and it's like kaboom. That's the definitive experiment. It goes this way or it goes that way. That isn't so often the case because the experiments are difficult and their interpretation is not so easy. I mean, there are cases where this has happened. The discovery of the cosmic microwave background was one example where it's like kaboom. There really was a big bang, hot early universe type thing. The steady state theory really isn't the story. That was a case where I think that really did. I don't know that history perfectly, but. But I think that was a case in 1964 where it really did sort of an experiment which wasn't an experiment where people were doing it. They weren't doing the experiment to say, are we going to find theory A or theory B. They were trying to communicate with satellites and finding out that there was a background noise that they couldn't get rid of and then were sort of smart Enough to realize, to have the context to think about. Is this relevant to cosmology, so to speak. But I think that you might think, well, we got theory A and theory B and we're just going to do the definitive experiment and we're going to figure out which theory is right. That doesn't play out as cleanly as you think most of the time because the experiments are difficult. The experiments are gotten wrong. The people who do the experiments are sort of gunning for a particular result, and they'll keep tweaking things until they get that result. And often when the experiment is difficult, it's right on the edge and it's right at the, you know, the thing you can only just slot, start to detect, so to speak. You know, I think most people, I think by this Point believe the 1919 validation of Einstein's general theory of relativity by Arthur Eddington and friends was probably not really right. That that experiment was kind of like, you know, tweaked to show the theory was correct because they wanted to show the theory was correct. And that can happen very often. It's very difficult to. So I think this idea that we'll just do the experiment and that'll prove it to them and then everybody will be happy. It often doesn't work that way. And so I think what other things about the scientific community, I mean, know there are all these things that happen where people, unfortunately, there are often forces from beyond science that impinge on. On science that are very difficult for science itself to deal with. I mean, there are things where, whether it's, you know, politics, usually it's politics of some kind, where there's sort of a, oh, we have to believe this because of this, or because we have to believe this because it disagrees with some religious belief, or we have to believe that because it's better for the government winning in this way or that way and so on, those forces from outside of science are super difficult, I would say, for science to deal with. And most scientists are pretty naive in trying to deal with those things. And I think in some places, there's plenty of places which I could start describing, but they're often sort of politically charged things which I personally try to stay out of. But there are plenty of things where from a science point of view, it's like, look, I'm pretty sure this is just nonsense, but nevertheless, it has a big weight of kind of political force on it that says you can't say it's nonsense, and it's difficult to deal with that. And I Think that sometimes really you just have to wait for often quite long periods of time for those forces to dissipate. I mean, certainly in the. There were forces arrayed against, for example, relativity theory that were partly scientific. I mean, this is back in the early part of the 20th century, partly scientific, partly kind of political, effectively of what was the. Was it good German science or was it Jewish science or was it this? Was it that? I think it's a very. I mean, my own feeling is that it's a. One of the nice things about science is that it does have a certain internal kind of validity to it that really is quite independent of who made it up. Sometimes it's really important to understand who made it up and to understand kind of the context and the historical lineage, to understand even what the science is. But in the end, the science is the science and it kind of rises or, you know, stands or falls on its own merits. It doesn't really matter who made it up and how shocking the person was who made it up or how you know, politically, you know, sort of supported they were or whatever else. But it sometimes takes a long time for the kind of those influences to kind of dissipate. And usually what happens in the course of history is that in the end all that people remember is the science sometimes attributed to the wrong person, but all they remember is the science. The bloodshed associated with whatever produced that science is long forgotten. And so that, that was, you know, so that, that ends up being sort of the ironic final twist is that in the end the sort of validity of science and its own, and its own, you know, standing for itself is the thing that survives the centuries, so to speak, and all the details of who, you know, who was the, you know, sort of who won and who lost in the French Revolution or whatever else is all completely disappears. Let's see a couple more things here, maybe I can try and do. And then we should wrap up here. What does the modern history of science tell us, asks ERI about non academic contributors and the future of these types of contributions? Well, one thing to say is that if you look at major advances in pretty much everything, they rarely come from the center of heavily professionalized fields. Not never, but maybe, I'm going to say half the time maybe, but some large fraction of the time they come from something that came from outside that field. You know, a field gets this big inertia of this is what we do in our field. And there is a machine that makes incremental progress. And over the course of decades that incremental progress can really add up to quite a lot. But when it's a dramatic kind of, oh, there's this exciting new thing, it usually comes from introducing something, a set of ideas, a way of thinking about things that wasn't part of the big inertia of that field. And so it's often the case that sort of new ideas come from people who are sort of a little bit at best interstitial and sometimes from outside of field X. Now, are they from outside of academic science? Sometimes I would say that there's been a sort of constant drumbeat of that throughout history. It's back in a few hundred years ago, it was hard to make a living as a professional scientist. Some people did. Isaac Newton was a professional scientist. Lucasian professor of Mathematics in Cambridge, at least until later in his life when he upscaled, so to speak, to be the controller of the mint in England, a better paying, more prestigious job. But that was, I think it was, you know, most people were not, you know, Leibniz was not a professional scientist. He was a professional courtier, so to speak, who would do kind of consulting about history or this or that or the other for dukes. And he did science on the side or mathematics and so on and philosophy on the side. Most. It's, it's only in, in quite recent times that there really is a track for being a professional scientist. And it's worth remembering that there's only a track for being a professional scientist in the professionalized fields of science. So for example, there are plenty of areas like things I've worked on, like what I now call rul, the study of the effect of simple rules in the computational universe and what they do. That's an interesting one because it's something that kind of. I really pushed in the early 1980s, a bunch of people got involved in it. There's probably 500 people who've made contributions to that. And they span all these different academic areas as well as non academic areas. And their contributions may be in papers, they may be on websites, they may have not been published in a form where other people can find them. It's very, very scattered because that field is not yet maybe it will become so a heavily professionalized field. There'll come a time when people are professors of that get degrees in that, where there are professional societies of that and so on. And at that point the picture will probably change. But you know, in one of these young fields, it's inevitable that there are non professionals because the field doesn't really exist. So I think the now, in terms of people who make a living as software engineers or quant finance people or something else who make significant contributions to science, that absolutely happens. It happens with varying degrees of success in different areas. So, for example, as I said, it's much easier for that to happen in a young field. You know, there was a period of time in the late 1800s, for example, in natural history when, you know, it was all amateurs of various kinds, but it was, you know, I think in today's world, it's much more, you know, there's a professional expedition to the depths of the jungle and so on. More so than an amateur kind of, oh, I'm just going to go to the jungle and pick up these things of beetles and so on. I think it depends on the field. In astronomy, there's been a long history of amateur astronomers discovering astronomical phenomena. My guess is that will come to an end because there are beginning to be very good systematic sky surveys every day. And the amateur who notices that comet or that variable star or some other thing that there's just a big sky to cover and there haven't been automated systems to cover it, that will come to an end. I think that in something like mathematics, it's pretty hard to make a frontline discovery in mathematics if you're an amateur. Now, there are different kinds of amateurs because there are amateurs who are fully trained, but they make a living doing something else. And there are people who didn't sort of go through and learn the existing canon and are now trying to contribute to something. I think that second category, it works much better in a younger field. It's certainly possible in ruliology to be absolutely of that kind, because there isn't a deep canon of work in something like some obscure area, well, not obscure, some deep area of mathematics like, I don't know, number theory, for example. It's pretty difficult to make a frontline contribution, sort of coming from the outside. I would argue if you use computers and you do mathematical experiments, it might be a bit easier because that's something that hasn't been well picked over by the main sort of main bulk of mathematicians. But I think, and this is a place where things get sort of complicated. I mean, I don't know. If you look at people who are making contributions in machine learning, for instance, they come from all different places. A young field. Well, it's not that young a field, but in its current form it's a young field. But the people who are doing that are coming from all over the place, and many of them you would say, well, it's sort of a science contribution, but they're absolutely not sort of professionally employed to do that kind of science. I think of people who are sort of trained but don't do science for a living, there aren't as many as you might hope. And I was thinking, I know many of them, I mean, because such people, I suppose I'm an example of this. One of the features is that I can do science that is quite ambitious and doesn't follow the incremental steps because I'm not going out and writing grant proposals to support my next incremental step that the peer reviewers will understand, so to speak. I can kind of do whatever I want and try some much more ambitious direction in science. And I've been lucky enough to be able to do that for many years now. And that's been critical, I mean, to my success in discovering things has been that I can do that. And that's a consequence of the fact that I don't make my living sort of producing science professoring, whatever else. I think that the number of people who are in a situation like mine where they sort of professionally know the science and they're doing science, but they're not doing it as a profession, so to speak, that number is rather small. And I think I know in for example, physical science, I probably know most of, maybe I'm wrong about that, but I certainly know a good collection of such people. I would say that it really varies with the field. I mean, in medicine, for example, there are lots and lots of practicing doctors who do a bit of research on the side. And that's a place where I don't know whether that counts as amateur or not. I mean, they're MDs, but they're not PhDs and they're not at a research facility, they're just doctors at a hospital somewhere. But they're doing some research on the side. That's a pretty common thing. And I don't know how much that counts and I don't really know. I think it really depends. I mean, it would be hard to be an amateur experimental particle physicist because if your equipment costs a billion dollars a pop, it's just not something amateurs can do. At least very few amateurs can do that, and I think so. I would say that the respect for non academic contributors to science has varied over time. If you looked at fellows of the Royal Society back 100 and something years ago, you would find, well, in the 1800s you would find many people who were in absolutely no way professional scientists, whereas I'm sure if you look today, the vast, vast majority. I was noticing this, a couple of different things. Some honorific thing that I got elected to, I was noticing that, you know, if I looked at the list of all the people there, I was almost the only one who wasn't employed to be a scientist. And so, you know, maybe it's quite rare in modern times. I'm happy to be able to do that because, as I say, it allows one to do much more ambitious, big picture kinds of things. And because one's only, you know, one's taking one's own resources and betting on those ambitious kinds of things rather than saying, well, we're going to use government money for this, or something like that. Yeah. Pontius comments. It's not just science where new ideas take time to settle. Van Gogh famously sold only one painting during his lifetime. Yes, I mean, indeed, it happens in all areas. And the challenge is that the good ideas not get lost. Because, you know, there can be good things that are done and they're just washed away in the tide of history. I mean, one person I've studied quite a bit is Ludwig Boltzmann, kind of originator of many things in statistical mechanics and things related to the second order of thermodynamics, and also a big believer in atoms and molecules, discreteness of matter. And people didn't believe in the discreteness of matter in the 1800s, when Boltzmann mostly worked. And Boltzmann, who became kind of a philosopher in his later years and wrote quite eloquent things famously was like, I'm going to record what I have to say about atomism. This was around 1900, because, you know, he says, you know, one man cannot withstand the tide of history. But I'm going to write down what I know about atomism because when these ideas are re discovered in the future, hopefully not too much will have been lost. So he said. And Boltzmann sadly killed himself in 1906. Not, I think, because people were attacking his theories, but for psychiatric kinds of reasons, not related to kind of exogenous forces. I think that it was sort of sad in many ways, but particularly because 1905 was when Einstein wrote his paper about the interpretation of Brownian motion in terms of molecules. And that was the time when the tide really started to turn on the atomic theory of matter. And Boltzmann died just too early to see that. And then indeed, Boltzmann was right that his work on the molecular theory of matter was rediscovered, so to speak, because his, for example, lectures on gas theory which was one of his things had not been gone for long. People still knew about it. There was still a generation of people who'd known Boltzmann, known his work, and so on. It wasn't kind of lost for five generations. I think there probably are things which are lost for five generations or five academic generations, and then it's pretty hard to reconstitute them. I mean, I think. Well, you know, for example, I've often wondered whether Leibniz came how close Leibniz came to inventing the idea of universal computation. I even went through his archive trying to look for evidence of that. I don't know. I don't think he came that close. But had he come close, that would have been an idea lost for centuries before it was refound. Another idea that I know was lost is the idea of fractals and nested patterns, which was originated, certainly there were people, one particular group of people, mosaic layers, who were creating such patterns in the 1200s, early 1200s, and that idea was kind of lost for basically, well, close to 600 years. So that's really the issue is, do the good ideas actually survive, or are they rediscovered independent of their original form? And sometimes then when people do the history, they say, oh, my gosh, this person had this idea back in the day, and nobody knew that. And so it's really a fresh discovery, so to speak. See to off in a moment here, I think. Adam asks, how and why should we document science in the future when AI starts to lead progress? Well, I think. The idea of AI leading progress is a complicated one, because there are an infinite number of directions for progress, and the choice of which ones we take is not a choice that is ultimately going to be a human choice for us. It's kind of. The AI can mechanize how we do it, but what we choose, if the AI chooses, we're going to say. And indeed, something I've done a lot is have. Have computationally just go out into the universe of possible programs and see what's out there and see what they do. And one of the things that happens is most of what's out there is really alien to us humans. And we're like, yeah, that's kind of an interesting picture. I don't know what it means. And until there's a kind of human connection to these things, it's kind of like the AIs are going off and exploring this infinite world of possibilities, most of which we just shrug our shoulders and say, why does anybody care? Now, I think that, say, in a Sense, this is already a problem for us that in this field of ruliology that I'm talking about, about studying simple programs and what they do, there's an infinite number of those to study. Just like in mathematics, there's an infinite number of theorems one could prove. But one question is, which of these are actually ones that we humans kind of care about and are ones that we are going to bring into our kind of arc of intellectual history? That's a thing that is, I think, not something. It's intrinsically not something that AI can do. It's just going to go out and explore everything. For example, one project I have recently is to see whether AI can tell whether a mathematical theorem will be considered interesting by humans. It's like you've seen the 3 million theorems humans have written down. Now AI tell me, this random theorem that I produced, will the humans find it interesting? So far, we haven't managed to get very far with that. So far that's been a big flop. It's really hard to tell. It feels very historical, so to speak. Okay, so, all right, one last. There's a question here from Incomprehensible Handle. What are the positives of reformulating history? You know, history's hard, and really understanding the narrative of what happens is, I think, very revealing for understanding the richness and depth of ideas. And sometimes you don't get the history right first time around. Sometimes the, I mean, this need to reformulate history, sometimes it's a purely, you know, political, ideological, kind of thing, but sometimes it's just because, for example, one common reason is at some point we really understand something. We understand some scientific points. Then we can go back and see, oh, now I can see why this or that fell this way. Because we now understand the science for real, so to speak. But we didn't understand it before. When that history was first being formulated, people didn't know that that particular aspect of it was important or what that really meant. So that's a reason that knowing what we know now, we can go back and see how all these pieces fit together we couldn't have seen earlier in time. Also, it just can be difficult. I mean, I know even history that I myself have been involved in, it's kind of often challenging to see what was the bigger picture of what actually happened. I'll give you a couple of examples. I mean, for example, I started working on kind of simple programs in the computational universe and how they produce complex behavior and so on. I started working on that right after this is 1981, 1980, right after I'd spent a bunch of time developing my first big computer system thing called smp, forerunner of Mathematica and Wolfram language. And so I, you know, it's like. It's an interesting coincidence in time that I was working on this computer system and then I start working on this kind of computational way of thinking about science. And I realized after the fact that before I worked on that computer system, the science I'd done was science, where you start from the natural world, you try and drill down and see what it's made of. What I did in building that computer system was something purely constructive, sort of starting from nothing, build this structure. And what I realized is it was doing that that allowed me kind of the intellectual freedom to say, let me just invent a theory from nothing and just look at the space of all possible theories and see what they imply. The idea of just looking at a space of all possible theories was not an idea that it occurred to me when what I was doing and what I thought I was doing in science is to start from the world as it is and drill down to see kind of the primitives from which that's made. So, in a sense, that. And it took me, oh, gosh, I don't know, decades after that time to kind of thread together those pieces of history. Even the construction of smp, which now has, well, many, many resonances with the construction of Wolfram language, even that took me years to understand that, in a sense, SMP was constructed very differently from the way that people had constructed kind of computer languages in the past. And I realized that again, many years after the fact, I kind of realized that. Why was that? That I'd done something differently? Well, the answer was because I was used to doing natural science where what you do is you take a set of phenomena and you try and drill down and find their primitives. And that was essentially what I was trying to do for computation. Take the computations people want to do and drill down and find their primitives and build those into the language. But it took me many years after the fact that to realize that was the history, so to speak, rather than, oh, you know, I just happened to develop this and, you know, I happen to have these ideas. Wasn't the case at all. And I think, why is it useful to reformulate things in those ways? It's useful because then you really understand what was the flow of thought that happened there. And that's useful to me, at least in understanding. Well, you know. Know what are the flows of thought that, you know, are realistic to have today. I mean, you know, when we see, oh, I don't know, the arrival of the LLMs, the arrival of kind of generative AI and so on, you know, what, how do we think about that? What's really going on here? What will be the, you know, in the long term? What will be seen as the significance of all of that? I think it's particularly what's significant is what it tells us about the structure of particularly human language and so on. That's the big sort of science discovery from that. And, you know, you can kind of, I don't know. So understanding some of these sort of what fits with what in history, in the past helps one to sort of get a better idea of the big picture today. Well, I should wrap up here and Prabh is commenting that saying happy New Year. And yes, I'm not sure, let me just see. When are we next on here? I think we get one more live stream for the year on Friday afternoon, I believe. And after that we're onto a new year. And I have to, I have, yes, there are all sorts of interesting things that I'm hoping will happen in the new year. And I certainly, I have a lot of work to do of things that I'm trying to make happen in the new year. But anyway, we should wrap it up for now here. Thank you all for joining me and see you again another time. Bye for now.