Why Scientists Can't Rebuild a Polaroid Camera [César Hidalgo]

I'm Cesaralgo. I'm the director of the Center for Collective Learning and I recently completed this book, The Infinite Alphabet and the Loss of Knowledge. The book has two ambitions. The scientific ambition is to establish the scientific study of knowledge by showing that actually it can be organized around three laws. A law governing how knowledge grows in time, a law governing how knowledge diffuses across the space and activity, and a law showing us how we can estimate its value. But it also has a policy ambition which is that when we try to develop the knowledge sectors of our economy, we have to make sure that we incorporate these laws into our policy strategy and design. If we don't do it, we're going to have failed development efforts. So the book also tells a number of stories of failed development attempts that defy the laws of knowledge and that I equate to trying to build a rocket without respecting the law of gravity or understanding chemistry or aerodynamics. We're going to try to understand how knowledge es and flows and what are the speeds and rates at which that happens. You know, what are the functional forms that govern the growth of knowledge over time and how the growth of knowledge over time changes when you move from the scale of individuals and teams to those of industries. We're also going to look at how knowledge crosses mountains and oceans and how it moves between activities and how those movements also satisfy certain laws and principles. And we're going to then try to figure out how we can count knowledge in a world in which the idea of 1 plus one knowledge equal two knowledges doesn't make a lot of sense because knowledge is nonfible. It's made of a lot of unique components and we need to find ways to score them that can help us understand the potential of economies. Many economists try to develop by creating cities of knowledge or science parks and so forth. And usually when they do that, they do it in ways that we could say are boneheaded because they defy these principles and those projects involve vast amounts of money and they end up in failure. So the idea is that well if we understand that knowledge follows principles like other quantities that we have learned to understand in the past like temperature or other things and now we can think of policy h in light of these principles and try to create policies that do not contradict them so that we can develop knowledge in ways that are compatible with its nature. Intelligence is about the efficient acquisition of coarse grained knowledge and you you develop this idea that knowledge is incredibly important and we've become obsessed with it. So we've been thinking well what does it mean to understand something and and in a way we we've developed this incredibly abstract view of knowledge you know almost like it's a probabilistic graphical model or it's a symbolic expression or you know maybe you know maybe the types of things in in a neural network is knowledge. But this this idea that that knowledge as a quantity is really important is something that has been impressed on us. >> Yeah. And I think that's something that you have in common with other disciplines. In my case, I'm coming more from the perspective of economics and social psychology in which we look at knowledge as this sort of quantity that is essential to explain economic growth and the wealth of nations. And this is something that has led to a couple of Nobel prizes. 2018 you know Paul Ror got the Nobel prize for indogenous growth theory. this year you know Aion and how it also got the Nobel Prize and the idea of Ror in particular is the one that is interesting and I think might be different from the way in which maybe a computer scientist thinks about knowledge but it's the following so when you're trying to explain economic growth you're trying to explain output you know so imagine you have 10 carpenters that have access to hammers and nails and boards of wood and they have to produce birdhouses and these 10 carpenters produce 10 birds per hour now you know let's Say that you want to produce 20 bird houses an hour. Well, you might need to double the number of carpenters because if they're doing the same thing, you know, and you want to do more of them, you're going to have to have more compressors, more nails, you know, more hammers and so forth. So, that tells you that labor and capital arrival inputs. And if you want to increase output, you don't increase output in per capita terms. The bird houses per carpenter remain the same. Imagine now one of the carpenters figures out how to build a nail gun that embodies knowledge or figures out a technique to maybe organize the workshop differently that saves them some time and they're now producing, you know, 12 bird houses an hour instead of 10. Well, knowledge has this property of being non-rival that can be shared without being depleted. I can teach you a song, but I still would know the song. If I give you a hammer, I cannot use the hammer while you're using it because it's rival. So what economists figured out in the 80s and in the 90s is that if you wanted to explain economic growth, which happens in per capita terms, the only way that you could do that is by assuming that growth was a consequence of a non-rival quantity, something that could be copied without being depleted. And that was ideas or knowledge. And that became a big revolution in the 90s. In the 90s, everybody was talking about the knowledge economy and the idea that knowledge is the secret to the wealth of nations. But what my book tries to do is to bring that to the next level because in that interpretation from Romer and other people in the 90s, knowledge is still some sort of quantity that you can accumulate in a barrel. It's undifferiated. So my book focuses a lot on the fact that knowledge has another property which is that it is nonfgeible, not also non-rival. And that nonfgeibility is the one that makes it interesting to study because it has all of this categorical you know differentiation that um requires you to use a math and a set of representations that are more similar to the ones that use in machine learning which also deals with nonfible things like language words are nonfible >> this nonfgeibility thing is is fascinating so all of us have this intuition right um you you hire someone and I'm a big believer that knowledge is situated And there is this fanciful idea about knowledge that it's a completely abstract thing that you read a book and you acquire the knowledge and it can just be copied any amount of times. In fact, that's the argument that AI existential risk people make. You know, you can just copy the language model and now you've got a thousand um you know Einsteins instead of one. What we find in practice is that it's quite difficult to exchange knowledge. Why is that? there is a tacit and implicit idea there that knowledge is something that something can have while my view is that knowledge is a much more collective phenomenon. Okay. So and it's not something also that you can put in something like a book. In my opinion the book doesn't have knowledge. The book is an archival record of some ideas that I was able you know to put together in a nice structure. But you cannot have a conversation with the book in the way that you can have a conversation with me in which I can tell you the story of Jachai or the story of Sam Slater or the story of how Sony got started based on you know what we're talking about and and have that dynamic response. So knowledge only can go to work when it's embodied. You cannot throw like you know a bunch of engineering manuals and cement into a gorge and expect to get a bridge because the books don't have knowledge, teams have knowledge, organizations have knowledge and all of that. Now that diffusion of knowledge is something that is is hard and and that's something that we have established really well. What is interesting to me also about this field of study is that people say well there are no really laws in economics. But when it comes to economic geography there are a lot of things that are very well established and that are lawike. For example, the fact that knowledge diffuses more effectively at shorter distance and that that short distance diffusion is explained by social networks has been established not by one or two papers but by dozens of studies if not maybe hundreds of studies that have verified you know those effects. You know the idea that knowledge moves more easily among related activities and that that can come from the complimentarity of the inputs that are required to develop each one of those activities is something that also we call the principle of relatedness and that has been documented by hundreds of studies. So we do have lawike behavior for the growth, diffusion and value of knowledge uh that we're starting to understand and some of that low-ike behavior goes into your question which is that of why it is difficult to diffuse knowledge. >> Just the words in a paper or in a book they are completely different to the actual physical embodied process. But is there a middle way? But are you saying only the physical embodied process is a form of knowledge and understanding or do you think there exists any type of model or representation which we could say is a form of understanding? >> The problem of talking about knowledge and that's something that I addressed in the introduction of the book is that it's a word that we use to mean vastly different things you know and and we we kind of like understand that from context but it's good to classify you know those different ideas and uh there's people that have done that. So one of the ways that you can understand um different types of knowledge is by thinking of a detective novel. So a detective novel or a detective TV show usually starts in a murder scene, you know, and ends in an arrest. And it's beautiful because the writer just needs to fill the space in between. And it starts usually at that murder scene when the detectives come in and they start collecting what we call factual knowledge. Yeah. There is a bullet hole in the wall. There was a call at 700 p.m. last evening. And those facts don't tell you about, you know, the motive of the murder or who was involved, you know, any of that stuff. And factual knowledge is knowledge that is very easy to diffuse. You know, I can tell you that Santiago is the capital of Chile and you can remember that you can transmit that information or or that little piece of factual knowledge very easily to someone else and so forth. Then you have conceptual knowledge which is what usually the the hero of the detective novel would do which is putting everything together in a story where all of the facts are you know little anchors that can be used to validate that story. Okay. So that's story. Okay. You know, the bullet hole was there because when you know the the murderer tried to shoot the this person moved to the side and and and the phone call was placed because someone was trying to warn him and they kind of like figure out the entire story. Now to validate that story, what they need to do is sometimes they need to collect additional evidence that is not factual knowledge but that is collected through procedural knowledge. So they maybe have a little bit of blood and they need to send that to a DNA lab for sequencing. Now the DNA lab has procedural knowledge because it understands how to perform that procedure to sequence DNA and that produce another facts that get put into the concept. So when we talk about knowledge we talk about all of these different things. Now there is another distinction that is extremely important and that I use a story at the beginning of the book to illustrate is that also especially among academics or people that are highly educated we talk about knowledge as these sort of truths that have been validated by the scientific method and so forth. And the book is not about that knowledge. And in economics, knowledge is not just about validated truths that have come, you know, from universities, scientists or or researchers. But there's knowledge in a lot of different things that is much more pedestrian and common. So a car mechanic has knowledge. A baker that has been producing different types of pastries and and and breads, you know, would have knowledge. Everyone has knowledge. And knowledge is highly specific. is not necessarily things that are 100% guaranteed to be true because of the scientific method, but are all of this experience and received wisdom that people have and that allows the world to work because the world works not because everybody's operating according to a scientific theory, but because you know car mechanics know what they're doing, because gardeners know what they're doing, because the guy that comes and clean the pool know what they're doing and they have their own experience. Maybe you know it includes even knowing how to deal with pesky dogs. If you're a pool cleaning guy, they have knowledge on how to deal with that. that comes from experience and it's not what you would find in a book. So it's about that more democratic definition of knowledge. >> Yeah. I mean I I do agree that we should have a a notion of knowledge which doesn't completely depend on humans in the most abstract sense. We might say that it's a form of modeling. So there are certain types of systems which we might say are alive and part of the process of staying alive and you know sort of like minimizing this free free energy you might say um is being able to model the world. Now the only reason I'm bringing this up is is you said okay there are facts which is like you know this is what the state of the world is there's procedural knowledge this is what we what we can do there's conceptual knowledge this is how to think you're talking about this detective film so he he was doing this conceptual type of reasoning where he was imagining possible worlds so he was saying oh you know what if this person um you know killed the person what if this person and that kind of imagination is simulating without direct physical experience so this person had knowledge and what they did was like a jigsaw puzzle. They were just trying different configurations of counterfactual futures. They found one which was plausible and then they generated a hypothesis. So that is a form not necessarily of it being a physical embodied process but also a form of like mentalized internal thinking. >> Yeah. And and I think that's correct. And the reason I think why that is correct is because it involves knowledge that is simple enough that fits within an individual, you know. But the thing about knowledge is that knowledge um can be such that you need multiple individuals to hold it. So yes, there could be a detective that can put all of the pieces together, can generate multiple representations, multiple alternative stories and use facts, evidence uh to decide among those alternative stories. Now, um, when we talk about economic growth and development, we're talking about knowledge that tends to be procedural and that tends to produce products or services that can improve the standards of living of people. So, for instance, manufacturing an aircraft, you know, manufacturing an aircraft is an operation that simply cannot be done by a single individual. No individual has all of the knowledge needed to manufacture a large, you know, jet passenger aircraft. And that knowledge tends to be distributed and embodied in this case in networks that involve humans and machines. They include whether it is printed material from manuals, whether it is an LLM that is helping now retrieve some information, you know, that comes from those manuals, whether it is the experience of people that have worked on that same model in the past and so forth. So the book focuses a lot on knowledge at that collective level. you know, I I run a center that's called the center for collective learning for a reason because I think, you know, of learning and knowledge at um at that scale, you know, and I think that's a very different story than that of sort of like figuring out the right theory in the detective novel, you know, it's a story that I think it's it's not so logical. It's much more experiential and that we do have models for. So, the model that I love in that space, there's this model by Linda Argot. She's a professor at CMU and she says an organization is a network that connects three types of nodes like people you know things and let's say ideas, concepts, procedures and like something you know more intangible. And at any point in time an organization is a network in which some people are working with some other people. Some people are using some tools to produce some goals and so forth. And an organization learns not only by the learning of people. It also learns as that network reconfigures which is kind of interesting because it's sort of like a parallel to like the deep learning type of idea that you're adjusting weights and in an organization we're also adjusting weight. So we discovered that team maybe that's like working with Robert and they don't get along they compete whatever. So maybe he's going to work better with Charles and the moment that maybe management or maybe organically Tim starts working with Charles the organization learns something and maybe maybe Tim was working in marketing but he hates marketing so maybe team wanted to work in engineering and if we assign team now to this different activity or to this different tool then there is learning and there is organizational learning that happens only by the reconfiguring of the same parts in a system that's a model of learning that goes beyond the individual and that has an analog to the types of learning that we're trying to reproduce I think in silicon right now. But still I would say the incilico models are still individual learning you know systems. They're not so much collective learning systems that involve all of these other social relationships and complexities. >> I absolutely love that. I mean, there was a wonderful example in your book. You you're talking about um Barnes & Noble and they're over there in Seattle and they they they went up against Jeff Bezos and they said, "Well, you know, Jeff, we've just launched a website and and we think we can do what you do better than they do." And they, you know, Jeff said, "I don't think so. You know, you might have a website, but you're a completely different type of business to us. We are geared up. We have the logistics. We can send individual things anywhere in in, you know, in in America. They were set up for wholesale and and retail." But but the thing is so you know they they could do the same thing but they were wired differently. >> So that brings in the idea of architectural innovation architectural knowledge. So it's a very interesting concept that was introduced by Rebecca Henderson from HBS and the idea is that when you innovate uh often you have what would be called gradual innovation in which you are changing a component. So for instance one of the classic examples in this literature is um the manufacturer of aircraft. So, if you had propeller aircrafts that had combustion engines, changing one engine for a more powerful engine or a newer engine model was something that you could do relatively easily because you didn't have to redesign the entire airframe. You just brought, you know, the new engine, replace, you know, the the old one with a new one and you were done. When jet engines were invented, uh you needed to redesign the entire airframe, you know, to be able to produce an aircraft. So the companies that were operating with combustion engines, they went bust most of them. And there was a new wave of companies like Boeing, you know, that were newcomers at that time that were specialized on jet engines because they were designing the entire airframe around the new engine. Now in the case of Blockbuster and Amazon, those are very good examples of architectural innovation because you might think that well Barnes & Noble is able to ship you know millions of book to all of these stores. you know, it has thousands, if not maybe tens of thousands of employees that are experts, you know, on the business of books and dealing with clients. And the idea of shipping the book directly to a consumer might look like a small incremental innovation, but in reality was an architectural innovation. And and when I do talks and I present this with slides, what I do is I show the picture of a Barnes & Noble and then next to that I show a picture of an Amazon fulfillment center which looks like kind of like this part of the airport that is you know sorting all of the different luggage and that shows that no that little idea of just shipping directly to consumer require a completely different organizational design and the distance between the Barnes & Noble organization in this network that we were describing before in that model with the to the Amazon model was enormous in reality just because of that change. >> Exactly. And and this is the reason why in my opinion LLMs are not intelligent because they don't have this coarse grained dynamic adaptation of their architecture. But we're getting ahead of ourselves a little bit. So at the beginning of the book you spoke about this concept of a person bite which is roughly how much can one person know and we're a collective intelligence. we work together and you you spoke about this kind of power law learning curve which is basically at what point does our learning as and and and we'll get to that as well but there was one fascinating example you you gave um you're talking about this this ship building company and over the course of I think was it the second world war the first world war they became much more efficient at building ships and was that because of experience or was it because of process >> the first law of knowledge the law of time is divided into several subprinciples and the first one is about the growth of knowledge in individuals and teams. That's a story that starts with Leon Thirststone. He was the first one to, in my opinion, map like a really good learning curve in 1916. Funnily enough, he started as an engineer and then, you know, he, you know, actually produced a camera that um got him an interview with Thomas Alba Edison. He decided not to work with Edison and go to teach at the University of Minnesota. Instead, he becomes frustrated that he's really good at math and engineering and it's hard to teach it to students. So, he becomes interested in learning. He goes to Chicago. He enrolls in the PhD in education and after a year he switches to the program in psychology and there he gets access to a data set that was being collected at the DAFF College of Business in Pittsburgh in which h you had records of how well people learn how to type. So imagine you have a mechanography class you know people are learning how to type. These are 18, 19 year olds that are typing on a typewriter for the first time. And you see every week how many words they're able to type per minute, every four minutes actually, you know, and then you see how many pages they've written throughout the semester. And when you put those two things together, you get a very neat, you know, learning curve that follows this sort of power like imagine like a square root type of shape, you know, uh, in which learning is really fast at the beginning and then it peters out. Then in 1936, you know, that's about 20 years later, Theodor Wright, which is an aircraft engineer in the United States. he was actually important enough to be in charge of aircraft manufacturing for all of the United States at the end of second world war publishes a paper in which he looks at the cost of producing an aircraft you know uh he's very smart he looks at the cost of the last aircraft produced in a batch because aircrafts are produced in batches and he finds also that the number of manh hours as a function of the number of aircrafts in the batch decreases as a power law okay so it's the same result that thirststone got in one case you can look at capacity in another case you can look cost. And then in 1965, Leonard Rapping, an economist, grabs data from the liberty ships. The United States was producing during the Second World War, an insane amount of liberty ships in multiple shipyards. So, he could use the fact that shipyards started at different times to have like a more causal story. You know, economists love kind of like having that extra little hint. And uh he was able to show that this learning that was observed the fact that the man-hour needed to complete a ship were decreasing over time was not a consequence of changes in technology or increase in capital expendure or increase in labor that basically more people was working on the ships but it was a function of experience. So how many ships your shipyard had already built. So that provides evidence of learning. Now what happens is that that phenomena is true only at the level of individuals, teams, firms and so forth. And once you transition to the industry level, you get to Morse curve which is very different. It's qualitatively different is exponential. And part of the book focus on explaining the connection between the two. >> We we know from experience, right, that when we have experience, we get better at things. And we have this this weird I don't know whether it's an illusion that all we need to do is just write down our understanding into a into a wiki document. It's the same thing with what I do on MLST. So I I've I've tried to write down how I edit the videos and how I do the sound design and the video and so on. And it just became more and more and more content. I could probably write an encyclopedia about it at this point. And I realized at some point that a lot of it is tacit and it's very very difficult to transfer to any new staff that that come on. It it is simply just a function of experience and this is really depressing to anyone who wants to start an enterprise or a business because the biggest problem is this knowledge transfer bottleneck. >> Um are you saying that that cannot be overcome in any other way than just having lots of experience and lots of people working? >> No, I think exponential learning is important to transmit that tacid knowledge. And I think you have a stories of people that have that intuition and that have been successful because they have developed careers following that intuition. One example that comes to mind, you know, is did you watch Arnold's documentary? Uh Arnold Schwarzenegger had a fantastic documentary on Netflix like No, no, no, no. A three-part documentary that basically goes through his entire life. So the first part is about him as a bodybuilder, the second part as him as a movie actor and the third part as a politician. And there's a constant in the documentary says like look, I wanted to become the best bodybuilder in the world. If you want to become the best bodybuilder in the world, you have to be with the best. So I figured out that the best were in California. So I moved to California and I became the best. Then you know I wanted to become the best paid actor in the history of Hollywood. So you have to work with the best. So, you know, like I I had money that I had saved from my bodybuilding activities. I had real estate that could keep me alive so I could, you know, be picky about the roles and I wanted to work with the best. And eventually 10 years later, he or 15 years later, he becomes the best paid actor in Hollywood. And he accomplishes everything that he has set his mind to. But he's very conscious that the only way to do there is not by figuring it out on his own on a quiet room on the back of his house. It's by trying to make sure that he's with the best. And when it comes to politics, he was part of the Kennedy family. You know, he he marries Maria Shrivever like early on and he learns from the Kennedys for, you know, more than a decade before he decides to run for governor of California. So again, you know, you learn from the best. The example I have in the book is that of Samuel Slater, which is a local lad, you know, and it's truly a hero. H and this is a a guy that is born in the Midlands at the time that the Midlands were the place that had for the first time figured out how to do water powered cotton spinning. That was a devilishly difficult technology. The first patents for water power cotton spinning are from the 1730s. They tried to build a mill in Birmingham. It doesn't work. Another in Northampton doesn't work either. about 50 years have to pass after that for people like um Arkrite and Strat to develop water power spinning in you know first in Kroford which is a very small town but that had you know water power and then eventually they created mills in Derby and Belper and so forth. Now Samuel Slater you know is born at the time that these mills are first being erected and he joins one of a strat mills at the age of 14. He's very smart, becomes an overseer, and at the age of 21, he says, "Okay, you know, I I know this business, and I know that I'm not going to make it in this business because this technology is just spreading like wildfire. Everybody figured out how to build these meals, you know, but in the US, they have not figured that out. So, uh, he escapes, you know, Belper in the middle of the night without telling a soul. He goes into London. in London. He boards a ship pretending to be a farmer and he lands, you know, in h New York 66 days later. He goes into a mill that was in Manhattan. He immediately sees that the machines were no good. They had no water power. So, he quits, you know, after 4 days. And then he learns from us loop captain that there was a man in Pucket that was trying to develop you know water power cotton spinning technology but they were not able to produce you know yarn of good enough um finess strength. You know the thing about cotton yarn like the one that we have in our jeans is that to resist the tension of the loom it has to be very well spun. And if you manually spin uh cotton, you cannot produce jeans or or fabric of that type because it would just snap under the tension of the loom. No. So he moves to patake it, you know, and eventually, you know, develop the first water powered cotton spinning in the United States within a period of about a year and it starts the American industrial revolution and mills start to spread there just like before. So it's a very good example of that embodiment of knowledge like the the people in Potake had tried to develop water power cotton spinning based on hearsay. There is this story you know from from the book where I where I got that story that there were some Scotsman that had seen one of Arcright's meals and they had told these other guys how they worked but based on that hearsay they were not able to develop it. You had to have someone that had that experiential knowledge, you know, that had worked with the best with Arrite and Strat to come all the way to the US in an act of treason because it was a punishable act of treason to bring that technology to America to eventually be able to build that capability. >> You're strongly asserting that there is a huge physical component that there's an embodiment to the propagation of knowledge. the way it flows es and decays you know everyone can access GitHub people the other side of the world can start playing with software so do you think at some point at least the propagation of knowledge becomes more virtual I think there's two things one thing is um to be precise about what we mean by physical and everything has to be physical because even GitHub you know has to store its data in some sort of hard drive or magnetic field or whatever technology but it's not storing it in nothingness you know so so knowledge Knowledge information always has this form of physical embodiment. Now I think we tend to think about it as non-physical uh because it is a thing that is not a thing which is uh the same as temperature. Yeah. So in in the book I have a a chapter in which I tell the history of temperature. Temperature is kind of funny because today you wake up, you look at your phone and you see the temperature and you decide how you're going to dress and nobody has any doubt that temperature is something that can be measured. But it took about like 2,000 years for us, you know, as a species to figure out, you know, what temperature was and the fact that it could be measured. And there were two fundamental difficulties that I would say made it difficult for us to understand, you know, uh temperature. The first one is that first people thought that hot and cold were two separate things. Okay? So that temperature was like a mixture of the two like when you make green out of blue and yellow. Okay? And it took a while for people to understand that cold was the absence of heat and not that cold and heat were two different quantities that were tempered together. They were mixed. So temperature actually mix means mixture not you know like what we now mean by temperature. The other thing that was very difficult to understand is that people thought that temperature was a thing was some sort of fluid that grabbed onto things. So let's say if you had a steel uh rod that is hot is that steel rod kind of like has this sort of invisible fluid that is heat and they had good reasons to believe that it was an invisible fluid because it could flow. Let's say you could connect that rod to something that was cold and that cold thing was going to warm up because that fluid was going to be flowing in that direction and so forth. So they thought that it had a physicality as a thing. um a brilliant Englishman Jou basically figures out that that is not the case that you know temperature is not a thing and the way that they do it is through this observation in which I don't know if you know how cannons used to be built you know so if you just grab a piece of sheet metal and you make it into a cylinder and you try to make a cannon out of that the moment exactly that you that you shoot the cannon that's going to open up like a flower in a cartoon you know like like you know like a Looney Tunes type of situation So what they would do is they would make these solid, you know, cylinders of metal and they would bore a hole in it, you know, to create the cannons and boring those holes released an enormous amount of heat. So J thought, well, how come all of that heat is there? It's like an infinite amount of heat. If I continue to bore a hole in a piece of metal for an infinite amount of time, I'm going to it cannot be a thing then. And that you know leads him to realize that temperature is actually something that has to live in things but it's not a thing itself is related to the kinetic energy of the particles in the thing but it's not a thing itself. It doesn't have its own particle. There isn't kind of like a temperature particle. Temperature is kind of like a property that matter has and that holds on to things. Knowledge is similar, you know, in that it holds on to you and to me, you know, and and and to the collective to exist, but it doesn't have kind of like a physicality in itself, but it always exists in some sort of physical medium or substrate. So, in that sense, it's always going to be physical. No matter how virtual it gets, it has maybe a different type of physicality. But even electromagnetic waves that are transmitting, you know, data from your Wi-Fi router to your laptop are technically a physical embodiment. There's um an interesting perspective here. So David Krakow, he says that temperature is an intensive property of matter. Yeah. And he says intelligence is an extensive property, you know, and and I think when he says intelligence, he's actually talking about knowledge. So he says that when we look at these complex adaptive systems, he says there's two types of systems in the world. There are the sort of the Roger Penrose symmetry dominated systems and then there's the systems that break symmetries which are these kind of complex systems which are you know things like life and and evolution and um so yeah the the these systems the amount of information they've accumulated in their lifetimes is a good proxy for the amount of intelligence that that they've had and and you're basically saying that that accumulation of of information you know roughly as as a physical property of matter is how we should think of knowledge. I' I've thought a lot about whether knowledge is intensive or extensive or whether complexity as you know is what we measure in in the technical literature is intensive or extensive and the key insight is is the following is that uh let's say you are wondering whether putting together two countries would lead to having more knowledge h than having those two countries separately. Okay. and you're going to proxy knowledge by the specialization that these countries have on the activities that they perform. So now when you put those two countries together, if let's say there are two developing countries in Africa that are specialized in a few activities, they're going to be specialized mostly on activities that have low knowledge intensities and a few that have high knowledge intensities. And when you put them together, you're going to realize that the high knowledge intensity activities, which are the ones that are not in common, maybe get subtracted because now they're not specialized. they're not producing enough to justify the now combined you know area and therefore you know the complexity of those economies when they're put together doesn't go up. So an intensive quantity is one that when you put two units together it averages out and an extensive quantity is one when you put two units together you add it up you know and knowledge has kind of like a little bit of that intensivity that depends on complimentarities you know so if you put two people together that know the same thing you don't get twice the knowledge then in that case it's clearly not intensive because it would be redundant if you put two people together that know the same thing and one is really dumb and that one is okay you don't get a super smart person as a result. You get kind of like a half dump person maybe, you know, you you average them out, you know. So, for that extensivity to kick in, you need to have, you know, a good level of performance, but also complimentarities. You need to be able to put things together that when they're put together, they're more than the sum of the parts. And that's not always guaranteed. You have a lot of examples in which actually putting things together is going to give you like an an intensive result or property. you need kind of like that diversity and those complimentarities to have that extensivity kick in. >> So for example, if I take a let's say a Chinese TV show and I try and launch it in the states, it won't work. It it'll fail because just the the cultural fit doesn't make sense because there's no shared phoggenetic history. So this idea that we can just take random bits of knowledge and kind of stick them together even if they have different histories doesn't make sense. >> No. Yeah, it doesn't make sense. You know what what you reminded me? Um so I I've been in the UK promoting the book for about 3 days now and uh the book is about the laws that govern the growth, diffusion and value of knowledge. But there's one chapter there that is about the laws that govern forgetting. >> Yes. >> Okay. It's the last chapter of you know the time section of the book. >> Okay. >> Yes. And that's the one that has resonated the most with people here and it connects to what you're saying because after you know telling some stories from that chapter I had a colleague you know um from from here from the UK that sends me an email he says this remind me of the is temple story okay what's the is temple story so I looked at it I I read it up and there is a temple in Japan that they rebuild every 20 years okay and by rebuilding the temple every 20 years they basically train the new generation of people that are going to know how to build a temple and it's going to have to train the next generation of people in 20 more years. So if you think about it, you can think of a temple as you know a structure and you might want to preserve that structure like here in Europe we have this historical heritage and so forth. But in this particular example they're deciding not to preserve the structure. It's not that this beam here is 300 years old or 500 years old or no what they're preserving is the knowledge on how to rebuild the structure by keeping the muscle you know h fit you know by by continue to doing the activity over and over again. >> Our current state depends on everything that went before. That's not really true because so much knowledge decays and gets lost. So there was this there's this interesting thought experiment that what if we could have a parallel universe. We could do a simulation and we just skip alchemy. Would we still be in the same place now? And in a sense, do you think that garbage collection, you know, like this deliberate kind of pruning of the knowledge tree is actually a feature of evolution? It >> it's hard to tell. So what what we do know is that knowledge decays and we maybe don't tend to focus on that as much because it's it's a more depressing story but we know that knowledge for instance in the case of the liberty ships was estimated to decay about 3% to 6% per month. You might think 3% sounds like a little bit, but it's 50% a year. Meaning that if an organization were to stop working one day and wanted to resume after a year, they would have lost 50% of the knowledge, you know, uh so knowledge decays really fast. And the reason why we don't observe that decay as frequently is because of course knowledge is being accumulated in a way that offsets, you know, those decays. But you have very good examples of knowledge decay and one of them is the history of Polaroid, you So Polaroid was an amazing company created by a quintessential Harvard dropout and entrepreneur Edwin Land. You know it's called Polaroid because what they did was polarized film. They didn't start with instant photography. They started doing polarized film like basically land figure out a way to create you know polarized filters that were large enough to have an industrial use and application by stretching kind of you know like this this goo with this you know material in a magnetic field so that you could create the polarized filters. And then after the second world war they needed to find a new business model because you know polarized filters were usually sold to the military. So they come up with this idea of doing instance photography which was devilishly difficult involved tons of patterns because the chemicals that you use to to develop a photograph are are extremely reactive. So if you put them on an envelope that people are going to be shaking around and and you're going to be transporting around the country most likely than not you know they're going to react at some point. they might mix and you know your your envelope becomes worthless and they develop extremely high quality instant photography so that photographers like Ancel Adams and so forth they would use you know uh Polaroid cameras now in the '9s Polaroid fails to transition into digital and then in the late 2000s there's one plant left in the Netherlands that is producing polarized film h and there's a man in uh Vienna that has a shop online that sells kind of like this vintage film and so forth. And when he finds out that this plant is going to close, you know, he flies into the Netherlands and he gets kind of like a group of people together to acquire, you know, the plant because Polaro was just simply going to close the plant and say, "Why don't instead of closing it, selling it to ourselves, they they they sold him the plant, you know, and they tried to restart the production of the technology." Now, they had access to the factory. They had access to all of the original equipment because they they didn't dismantle the factory. And I interviewed him when I visited Vienna a couple of years ago. And I asked him, "Well, the people that you hired to work on the plant, you know, did you did you get your pick or did you have to deal with whoever wanted?" I said, "No, no, no. We had the A team. It was the star team. You know, we we couldn't hire everyone back, but for each machine, we hired the top player that we had." And still you know um after they run out of the stock of film that had been produced previously and they have to start selling their own film. It's black and white film. It takes 30 to 40 minutes to develop. It often has aberrations. You know the knowledge was lost and it took you know this impossible project. you know, several years to start producing film of any quality and maybe only like about a decade later they start producing, you know, film that is of a quality that maybe could be considered comparable to the one that Polaroid was producing in the 70s. So knowledge can really disappear rather fast, you know, if we stop exercising it. If you don't use it, you lose it when it comes to knowledge. >> No, and it's it's so fascinating that example you use the term embers of knowledge. So in that particular case, it was almost lost. the embers were there, the flame had gone out and it was possible for them to get the folks back in from the factory and what they discovered was you know supply lines they couldn't access some of the chemicals and some of the things you know so what they had to do painstakingly was almost reinvent some of the knowledge and it is possible to >> you you use this um wonderful term um absorbative capacity there there was the story of um you know the the US after the second world war that they were in Japan and there was this inventor in Japan and he studied studied one of these tape players in great detail and he could go away and he could create a prototype just using like you know a frying pan and like you know like this kind of >> frying pan shellac hemp reinforced baby that's Ibuka who is one of the co the technical co-founder of Sony that basically develops magnetic tape using a frying pan a badger hair brush you know and and he reinforced paper and shellac um after observing you know only a couple of times you know that Exactly. And because this scientist and his collaborators, they were researchers and and they had lots of adjacent experience, it was possible for them to reignite this knowledge and almost reimagine it and resituate it in a different environment. But just in terms of this knowledge decay thing, isn't isn't it amazing that we think that we are at the pinnacle? We think that we've accumulated so much and we haven't lost anything. And there are so many examples like Concord is is is a great one. What if we wanted to build Concord again? Would we just be able to 30 years later pull up the blueprints and just kind of stitch this aircraft together and it would be just like Concord was 30 years ago? Probably not. Another example is in software engineering. There are famous examples of I think IBM publishing their source code online and everyone in the company said, "No, no, you can't publish the source code online. This is all of our knowledge. If someone gets hold of this source code, they're just going to be able to recreate everything that we've done and they'll know they'll know everything." That's not true because the actual knowledge is just in the minds, the interactions, the the the the sort of the the the ecosystem, the organism of the company. And you also said in the book which I thought was fascinating, which is that the purpose of an organization is to retain as much knowledge as possible. >> Yeah. To retain and preserve knowledge. And they're beautiful examples about the role um that large teams play in the development and diffusion of knowledge. So before I told you the story about Sam Slater. Yeah. This you know uh whisk that at age of 2021 starts the American you know industrial revolution. H but there's another story later in the book that is the story of the city of Donetsk. Now most people have learned about the city of Donetsk today in the context of the conflict between Russia and Ukraine. But Donetsk is a city that was actually created by a Welshman h by the name of John Hughes. So John Hughes was a Welshman. He was a successful entrepreneur here in the UK. He was in iron works and he was in the business of ironclading ships. So you have to remember that in the 19th century, you know, wooden ships were being ironcladed so that they would be more resistant to damage, you know, when when you had sea battles. And later in life, you know, this is not a story of a of a 21-year-old like Sam Slater. Later in life, like when he was in his 50s, he goes, you know, to ironclad a ship, you know, for the Russian Empire. And then erh he develops a relationship that leads him to apply for a concession to develop the coal and iron resources that were available in this area of the Russian Empire where in the future there was going to be the city of Donetsk. So he wins that concession. He comes back to the UK and he loads seven ships. seven ships, you know, with more than a hundred men and with all of the equipment that they would need to set up that operation. Now, they grab those ships, they sail all the way to the Azov Z. They unload and then they drag this thing through the mud into what later is going to become Donets. They set up a camp and they start building, you know, the iron works from scratch in that location. Erh and in about 3 years they were producing pig iron and then you know they start you know developing what eventually becomes one of the main you know h iron and still producing regions of the Soviet uh union. H they build you know the schools they build the hospitals eventually and all of that and the city originally was called Yusovka or Husovska because it was John Huges. So actually there were some people that when the conflict started they say we should secede back to Britain because we are British you know this was a city that was created by British but but the whole point is that you see Huges understood that if he went only with the knowledge that he had to that area and he tried to set up you know iron works on his own you know he was not going to succeed because the knowledge had to be embodied on a much larger crew you know a crew that required seven ships to be taken from England land to the Ukraine. >> It's almost as if there's a sufficient embodied carrying capacity for knowledge. It reminded me, I don't know if you've seen the Apple TV series Foundation. It's the the Azimoff, >> you know, Isaac Azimoff foundation, >> the genetic dynasty and all of that. Yeah. It's >> and and they had and again, you know, this is science fiction, but they they still had this possibly wrong-headed idea that all they needed to do was take a library of knowledge and and a small group of people and put them on a planet in the outer galaxy and then they would be able to restart civilization and probably you would argue that there is actually a threshold point where it's just not possible to do that. >> If if that would be true, shipwrecks would be much more successful than they are. So coming coming back to to thirst then so you were describing this kind of you know asmtoic curve of of learning. So you know it's a function of of experience but then um in the next chapter you went on to talk about disruptive technology. So you gave a great example of steel. So you know like to manufacture steel there are roughly four or five tiers of ski of steel. And what happened was that as companies started to develop higher grades of steel, entrance would come in and develop lower tiers of steel that were kind of crap, but they had more upwards trajectory and that would create this disruptive cycle which wasn't this limiting curve but was actually a bit more like Moore's law. It was like an exponential curve. Explain that. >> Yeah. So that that is the idea of disruptive innovation by clon Clayton Christensen and it's the connection in some way between Moore's law and the laws of thirst right and wrapping. So if I can draw in the air here you know we had curves first that are a little bit like a square root you know that they grow fast in the beginning and then they they slow down and peter out and then you have Mor's curve which is an exponential that grows over several orders of magnitude and you have to reconcile the two. Now the way that you reconcile the two is that you realize that when you're operating at the industry level or at a geographic level you know like at a country scale and over long periods of time you don't have the development of a technology that involves a single learning curve but a collection of generations of technology that have multiple learning curves and that M's law is like that envelope that h captures that collection of other individual learning curves you know so for example when you look at manufacturing of LCD panels, they go through generations and each generation is able to produce panels that are bigger, you know, and and that have less defects and so forth. So, uh you have lots of examples of that. And what is interesting about that is that um when you move from one learning curve to the next the next learning curve even though it can soar higher it starts at a lower point because you are at the beginning of it. No. So when the new technologies are introduced they are worse than the incumbent technologies. Like when digital photography was introduced it was much worse than chemical photography. And people in the 1980s that were involved in chemical photography, like the people at Polaroid, they laughed at the digital photography said, "These guys are never going to be able to get, you know, the type of colors and and and and resolution that we're able to get, you know, with our chemistry." And the technology gets laughed at because it was worse. You know, instant photography was also laughed at at some point because it was worse. But then eventually that curve keeps on growing and it gets to like a plateau that goes even higher. And then a new technology comes along and moves into that is higher. Every time you have that intersection coming, you have a window of opportunity because in that window of opportunity, you have a technology that is worse, you know, that the incumbents don't take seriously, you know, and that is going to be able to surpass them. And the moment that those two curves cross, you know, is when the incumbents get, you know, desperate and they think that they're going to be able to get there. But usually those changes also involve architectural innovation, which is what we discover discussed earlier. So moving from let's say chemical photography to digital photography would have required completely redoing and restructuring the entire operation of polaroid because they were not set up to do that. They were like a chemical company that started with doing polarizing film. They were heavy to chemistry not into electronics. >> Yeah. And you also gave the example I think it might have been the the Sony Walkman or a Sony technology where they came in and originally they were much worse than than the you know than than the incumbent and then it had more upwards trajectory >> that transistor radio. transistor radio was did not have such a good sound quality in the beginning as tube radios you know so they were like this cheap alternative you know it was like a radio for you know this security guard that would not be able to have of course like a tube radio you know in their booth so they would have like their little pocket radio that was low quality and then transistors eventually became good enough and and nowadays we can get amazing sound quality from transistors. Yeah. >> Yeah. And and and you spoke about transistor because this is like the mos law thing. So I think was it originally 12 months and then now it's been set to about 18 months or something but it's it's still holding which is amazing and in a way I've got a few questions here. I mean first of all it it sounds like this appeal to infinity that we can have these exponential curves and they just keep growing and growing. Um it's interesting that Moors law has kept on and and you said that the reason for that is they have all of this disruptive technology. They've got larger and larger teams, better and better processes and they can just keep innovating. How long will that go on? And also what what is the reason why we can have these disruptive cycles because what we were saying before about Barnes & Noble is that it's really really difficult for an organization to do this architectural rewiring. So the way that disruptive innovation happens is that usually you have an independent thread. So somewhere else in the epistemic fogyny you have different people with different ideas and different objectives and they come up with a completely different architecture and they innovate. But now we're in this domain where we have these large big technology companies and they might acquire any new startups and that means the new startups will be contaminated with all of the cultural knowledge of of the incumbent. So don't you need to have like independence diversity preservation and competition to keep on this exponential curve? So at least in the history of the transistor I do think we have some good examples uh that the teams required to develop those innovations have been growing over time and this is something that Nick Bloom in Stanford he's a professor of economics there um has emphasized the the fact that there's an increasing cost of innovation that let's say to duplicate again you know we have larger teams and larger budgets like it's getting costly and costly and costly now some people you know have arguments against you know his evidence and so forth but but I think it's the right question to ask because you know definitely there it looks like there's something in in in that direction and you can see it in the history of transistor because the first transistors uh were developed by a team of three people you know originally actually a team of two it was uh Bratain and Bardin and then Shockley got jealous and over Christmas developed a transistor design that replaced the first transistor that was developed by his uh lab assistants Bratain and Vardin and that became the point contract transistor in in 1948. But then you had several different transistor designs. For example, Shockley semiconductor company was not successful at producing transistors. But when Mo and moved to Fairchild then they produced the Messa transistor I think in 1958 and then they produced a planner design in 1959 and then they produced an integrated circuit in 1959. There was also another team in Texas Instrument that produces, you know, an integrated circuit, Jack Kilby. Jack Kilby was someone that had just joined the company and because he has just joined the company, the company was kind of like empty. So during the summer, he had nothing to do and he created the integrated circuit as kind of like an experiment on his own. But you see, it's a team of one. nowadays to produce the next generation of Nvidia CPUs or you know Intel CPUs or or whoever you probably have enormous teams involved in the design in the manufacturing. So I think that might be at some point the limiting factor which is a at some point maybe you know to double again the teams are going to get larger than what we're able to coordinate and if that coordination capacity doesn't get to scale to the amount of knowledge that we would need to generate another doubling you know we might see you know this curve petering out now is that close to happening I I don't know it's hard to you know bet against a a law that has been stable for so long >> I would love it if you read Kenneth Stanley's book, Why Greatness Cannot Be Planned? And he's a good friend of mine and I'm going to tell him to read your book because I think there's a lot of overlap. But his basic idea was that um you know consensus, committee meetings, objectives, they they are actually quite toxic for progress because creativity is about following your your own gradient of interest and and basically preserving diversity and having new ideas about things. So yes, we have a new generation of these large language models. They're getting, you know, the 10 times bigger every few years. and the the bullish people say, "Oh, we're on an exponential curve and it's just going to keep going up." But I think that because there is so much group think and it's fundamentally the same technology and the same people and there's no fresh new ideas that in a sense it's converged and it's not disruptive anymore. You know, I I remember using GP3, not even Chad GPD or GPT2 and all of that and and definitely this has been going for a while like the the the improvement has been sustained not for like five years but for longer. Yeah. like these technologies started at a rather simple um level of proficiency that that has continu before people know about you know instant photography uh those are you know the Ibukas before Sony makes it big with you know the transistor radio and so forth so Yes, I agree that there might be incumbents that are really big. Those are like the Barnes & Nobles and the Walmarts and you know of the tech industry and there might be disruption that comes from teams that maybe have thought of a different way of creating you know artificial intelligence that might overtake those incumbents and that might be disruptive because maybe require doing things in a different way. But we're not going to know until until those curves cross. you know, usually it's very hard to see those disruptors early on simply because they're not yet getting the attention. >> Let's talk about the flows of knowledge. >> Okay. >> So, this was um chapter six of of your book. So, um you had many wonderful examples of let's say migrant flows for example. So, I mean one one example that really sticks out to me was um there was this trade embargo with with Vietnam. So, >> Vietnamese boat people sorry >> tell me about that. >> Yeah. So, that's a good story. So um so when the United States left Saigon in 1975, it was as glamorous as when they left Kabul a few years ago. It was it was rather quick, you know, so they expected to have more time to evacuate. And what happened is that, you know, the Vietnamese army came into Saigon rather quickly and lots of people uh went into boats and went into the ocean there, you know, and they needed to be relocated into the United States. These were people that had cooperated with the United States and and now they were looking for asylum. So the United States had to relocate, you know, hundreds of thousands of people, you know, h within the period of a year. And that relocation can be considered to be quite random or exogenous because it wasn't that where do you want to go? Would you prefer to go to New York or to San Francisco? No, it was like we have, you know, a a small town in Iowa that has a church that is willing to take 10 people. Okay, here we have two families of four and this couple. Boom. Iowa, you know, and they would be relocating people at an enormous speed because, you know, this was a humanitarian crisis that they were trying to solve, you know, and there were different takers and they were being allocated like that. So after the war is over also what happens is that the United States imposes an embargo on Vietnam. So they cannot trade. So what these economists Parsons and Vessina did is then they look at the stock of Vietnamese people that was exogenously allocated through this process and then they look at trade data starting in the year 1995 when the embargo is lifted. They say well you know if this state got more Vietnamese people because of this exogenous allocation did they trade more with Vietnam after the mark was lifted? And the answer is yes. You know there is an effect there. So they show that in this case, you know, the relocation of these people brought knowledge on how to uh trade with Vietnam and they had relationships that they could use to develop that that commerce. >> And there was also an example you you you use this analogy of like monkeys in a forest to talk about not only the geography of knowledge but also like the um the geometry, the structure, the topology of of knowledge. Talk to me about that. >> Yeah. So exactly. So the the second law of knowledge is about diffusion and diffusion has two subp principle let's say one is about diffusion across geography or across social networks and the other one is uh diffusion that is constrained by the geometry of knowledge itself. Okay. So one way to explain that idea is to say look uh the economy involves multiple activities like multiple industries or multiple products and you can think of each one of those activities as like a tree in a forest. Okay. So you have multiple trees in a forest. So so let's say this is a tree that represents shirts and nearby you have a tree that represents you know blouses and those are very nearby trees because shirts and blouses are are are similar products. So it's if you produce one you can produce the other and then maybe far away there you have a tree that represents you know natural gas and then on the other side you have a tree that represents you know tractors or combustion engines and so forth. And you have kind of a geometry or a geography of knowledge itself. So in that representation if knowledge or is represented by the activities that you know an economy can produce and these activities are trees countries are collection of firms which are collections of monkeys that live on these trees. So let's say you have a garment company and all of your monkeys you know all of your knowledge is being harvested from the uh fruit that is being grown in the in the tree of blouses in the tree of shirts in the tree of linens in the tree of you know socks and so forth. You know you have an electronics company you are in a different part of the product space and economic development is the process by which countries move into new trees but the ability of monkeys to jump from a tree to another depends on their distance. So what we have uh shown and after we did it it has been shown by hundreds of papers is that the probability that you would enter an activity that you were not specialized in in the past depends on how many monkeys you have around in other trees. You know this geometry really matters and and the story that I use to illustrate that which I think is the most um telling is the story of Vespa. So everybody knows Vespa you know it was even in a recent Disney movie. is an Italian scooter erh that is iconic but people don't know that it was created not by a motorcycle engineer but by an aircraft engineer. His name was Coradino Dascano. Corin is the equivalent of Theodor Wright but for Italy. He was in charge of the production of aircraft for Italy during the Second World War as well. And he was working in a factory that was from the Pio family that was specialized also in the manufacturing of aircraft. When the war is over, three things happen that change the destiny of the pajos and of kadino. The first thing is that Italy is no longer allowed to manufacture aircraft. Okay, that's forbidden. The second thing is that the factories had been bombed. You know, aircraft factories are primary military targets. They're not hospitals. They're not schools. They're, you know, they're weapon factories. So, they got bombed. So, all of that capital, let's say, was destroyed or or or whatnot. The third thing is that the bridges, you know, the roads had also been bombed and destroyed, you know, it was a war zone, you know, so people in Italy needed a way to move around. They they wanted to arrive at work without being all muddy, you know, and in that process, you know, people start thinking about vehicles that they could create to satisfy that need. So first Corino goes to work with you know Centi was a competitor that you know Centi wanted to do a a motorcycle based on on tubular you know metal and Corino wanted to do sheet metal so they they they don't get along they had a falling out and he goes back to the pages and then they create a motorcycle in which the mudguard you know is the body the engine goes in the back so you can sit with your feet together you know and the wheel in the front comes on and off the same way as the it would come on and off in a helicopter. Now you might think that's a nice anecdote. Look, these guys were manufacturing aircraft. Now they cannot manufacture aircraft anymore. They do motorcycles. But if you go to Japan or if you go to Germany, you see the same example like you know miles and miles away, you see companies like Kawanishi going into light vehicle manufacturing after not being able to produce jet uh sorry fighter aircrafts anymore. Hankl in Germany, the same story. they produce the high turries when they're not able to produce you know in in their case they they were even able to produce jet aircrafts during you know the war. So what that tells you is when push comes to shove and you have to get out of your industry because in this case these guys had no option they could not stay aircraft manufacturers and they had to jump into something else they all jump into something similar. So if they're all jumping to something similar, well, it means that they're moving along the same map, you know, and it's a map in which motorcycles were close to aircraft and aircraft maybe was not close to blouse manufacturing. So they all ended up in the same parts because the diffusion of knowledge is also constrained by the geography of knowledge itself. There are signals for knowledge diffusion. So you know prestige is what is one signal. So if if if children kind of um there there was an experiment where if children see a particular person as prestigious then they're more likely to transfer knowledge. But what I want to get to with with creativity though is a lot of people think that it's completely serendipitous. So YouTube started as a video dating website and you know like the way microwaves are invented you know took quite a divergent and weird and wonderful path and we might we might um conclude from that oh it's just a random walk through you know epistemic space and that's not true at all. um there are like these analogical jumps that that you can make when you perform creative actions and I think the examples you gave just kind of pointed to that. So the jumps are still very much dependent on the history but they are kind of like obvious stepping stones that can be taken from the the previous position. Does that make sense? >> Yeah. Yeah. So I I I would agree that for example a video dating website and a video streaming website are quite related. there trees that are close by in that product space. You know what what is interesting about this story is that since this principle has been so much established in the economic geography literature now we talk about strategies that involve for example uh targeting related or unrelated activities and we know for instance that migrants are better at developing unrelated activities while locals are better at entrepreneurial activities that are related to the current you know economic structure. So we even kind of have a ways to connect these two stories. You know migration is going to be something that is going to help you jump far in the product space like the the migrant monkeys might land in a tree that is not near your trees and might help you develop a new area. >> Yeah. And and actually you you spoke a lot about um migration in general. You first of all um migrants they have a lot of um choice about where they can go because you started the book talking about you know neom in Saudi Arabia and and there was a yachai. Exactly. And these were places that were designed to be kind of epistemic centers where they can do lots of innovation and they can invent lots of things. And actually it didn't really work because it didn't respect the the the outer organism. They were almost so disconnected that no migrants would want to go there and it can't actually be self- sustaining in terms of the creation of of new knowledge. >> The people that have a choice, you know, they tend to be quite strategic about how they exercise those choices. you know um there's a lot of statistics about the role of migrants in innovation and the fact that people that are highly creative or highly innovative tend to be migrants like there there is this statistic I think in the United States if you count a Nobel Prize winners after the year 1970s about 70% of them you know were born outside of the country or or or they migrated at some point in their lives there's something like that >> 60% >> yeah the higher the level of education the the the higher the propensity to migrate you That's something that we know for a fact. But that migration is not random. It's these are people moving to centers of knowledge where they know that they're going to find the complimentaries that they need to develop the things that they're looking to, you know, develop. So being attractive to high-skll migrants, it's a very good signal for an economy. Like I was talking with people here in the innovation agency of the UK, Nesta the other day and and one of the things I was say like look if you want to look at the impact of you know migration in innovation in the UK what you need to count is the number of superstars that you attract because at the end you worry about kind of like that tail end of the distribution and uh you know I know migration is a topic that is quite um controversial right now around the world and I I don't do in my book a call for a single way of thinking about it. But I do think that I show evidence that migration is not all equal. That if you're able to attract people that have a high level of skill and talent, the probability that they would be net contributors to your economy and to your society is going to be much higher. So it's not kind of like a let them all in and we'll figure it out later. But you know how can you become attractive you know so that the best people in the world are fighting to be you know working in your cities in your universities in your companies >> you know like for example if you look at genetic diversity in in Africa it's much higher and and that actually is a proxy for well you know that that that's where the the bowl of of evolution was because so much diversity and and similarly you you you can look at the diversity of innovation and knowledge as as almost a proxy for how uh where you are on on the maturity curve in in in evolution. But there was this really interesting concept and by the way as a point on what you just said I think there was a statistic that something like 50% of the large companies in in the US were started by migrants. I can't remember the the exact statistic but as you say when you have skilled migrants it should be obvious that you have this kind of diversity um acquisition and you need to have diversity to try new and interesting things to not get caught in the basin of attraction that you're currently in. This seems obvious to me. There is a lot of like nasty talk about migration at the moment. But I mean there there is migration is clearly a very very good thing. Maybe you want to comment on that first. >> No. Yeah. No. So so so migration definitely is an important vector for knowledge diffusion and that's one of the best established facts in the economic geography literature you know but those facts we have to keep in mind that they're always established using data that focuses on very high skill individuals. So they're looking at migration by looking at people that patent. what fraction of the population patents know or people that publish papers that's a larger group but still it's a very elite you know group of people we have other examples like that one of the Vietnamese boat people that in that case I I put that example in the book because it's a non- elite example of knowledge diffusion but you know the impact per capita of the migrant might be lesser than that of let's say an inventor that has 20 patents and is going to produce 20 more in the next 10 years. Um so you you do have kind of like that differential aspect. Now I wanted to go into like your diversity um uh in in the context of Africa because I was in in Rwanda a couple of weeks ago actually. You know it's a fantastic country honestly like like I was very impressed of what they've been able to accomplish. The country is clean. People you know are are you know nice and respectful and they're really pushing forward and so forth. But one of the things that you notice when you go to a developing country, I do a lot of development work and I travel around the world quite a lot is that there's many people doing the same thing. So for example in Randa, taxis are mostly motorcycles. Okay, they're moto taxis. Okay, so these guys are going around a motorcycle. They're carrying a couple of helmets and you put on a helmet, you sit on the back and that's your taxi. And you have parts of Kegali that you have tons of motor taxis. So it's a lot of people but if you were to add let's say the diversity of that knowledge based on the different activities that they do you wouldn't add that much because you know it's the same activity and then when you go to other places that are knowledge intense what you have is that everyone is specialized on something different and they tend to be complimentary. So it goes again into this idea of whether knowledge is extensive or intensive. If you put a 100 moto taxi guys, you know, the amount of knowledge that you add from one to 100 is not that much. But if you then, you know, put a 100 people that do different things and that are complimentary, you could get an amazing amount of knowledge. Like whether it is extensive, whether it adds up depends on whether there are complimentarities and differences. >> Very cool. Now in chapter 8, you started talking about Bretonwoods, which was this thing after the Second World War. And um essentially to rebuild Europe they um set it up they set up these institutions and they kind of did financialization to encourage growth and that was seen as such a success that financialization was actually seen as something that could be used to aid development even when it's not a war a war torn country. Tell me about that. the the story of like let's say 20th century western institutions that is told more commonly is that after the second world war is over the United States has an interest on helping Europe you know recover and they generated you know different funds and institutions to help do that there is a German Marshall fund you know the world bank is created the IMF are created all in that context and you know the world bank starts lending money to the Netherlands to you know France to different countries And those development efforts uh do really well because the money that comes into Europe at that moment, you know, translate into investment that helps rebuild Europe and Europe starts to pick up, it starts to grow, it starts to rebuild and so forth. And then people say, "Wow, you know, development is a really easy business. You know, you just throw money, you release the constraint of finance and it just happens." Now, the thing is that Europe is extremely knowledger. So they were you know throwing money in a continent that was extremely knowledge rich. So sure the bridge was not there. The hospital had been maybe you know partly destroyed and so forth but the knowledge was still around. So if you liberated the financial constraint you were going to get that rebuilding happening you know and the GDP was going to you know start to return to what it was supposed to be. Then they started to do the same thinking that you could develop other regions of the world through a similar model and the results were not the same you know so many things happened. On the one hand people started to think well maybe it is about you know institutions and so forth and there were lots of efforts to try to now connect financial support with institutional reforms. So okay we're going to make you this loan but you're going to have to do all of these reforms within your country. And what happened is that a lot of countries would do those reforms and the medicine still did not take you know now there's people that have said that well the thing is they would do these reforms only in form they would not be real reforms it would be kind of like a mimicry of the reform and therefore that's why it it it doesn't take you know but also you know there is people that think that in reality it's not just about institutions knowledge also plays a role because the demand for those institutions that you need to develop comes from the more knowledge intense members of your society, you know. So what I do in in that chapter is is is built a little bit of a parallel because when it comes to economic development, I think the two dominant ideas are that it's about knowledge or it's about institutions and I'm happy to believe that both play a role and that sometimes one is the horse and the other one is the carriage. But there have been periods in which you can clearly see a a change between the two. So I tell a lot of stories about China and how knowledge intense workers in China helped demand institutions of you know intellectual freedom and and entrepreneurship that helped develop Jong Wanjung the main innovation district of Beijing. I tell the story of the printing press which is an excellent example of a technology that enabled institutional change you know about 60 years later with the reformation >> and and just quickly on China you said that the China story was was quite different. So they didn't have the institutions of of of the west. What what happened there? >> The story of China is a story of a country that um institutionally was was rather constrained was very poor but it has pockets of of knowledge you know even all the way back in the 60s and 70s. The story that I used to start my description of of the Chinese growth miracle and in particular Shangh is a physicist that developed the first fusion reactor in you know Beijing in the 1970s. He did his PhD with the Soviets that had invented tokamak technology. this idea you know this cipher idea of confining plasma in a magnetic field you know that is a soviet technology that he had learned during you know his graduate studies he brought to China and he built you know a fusion reactor so you cannot say that was you know a a low knowledge intense worker you know building a fusion reactor probably is a little bit difficult yeah anyway as a as a consequence of that he gets invited to be part of a committee that goes to the United States to see how the Americans were building the fusion reactors so you to Princeton, he goes to Boston, he goes to Stanford, so forth. And and he was expecting that the Americans had these huge factories where they were building the components for their fusion, you know, reactors. And and what he realizes is that they didn't have these huge companies that even though the US was so much more advanced than them, what they had this small companies of professor entrepreneurs. So the same professors or professors that were adjacent to these universities would have a company and the company would be like 10 12 people that would be building some component some electronic device you know some something that would then be sold to the plasma fusion lab. So he goes back to China and he starts advocating for that. He goes back to the US to learn more about this because he becomes obsessed with this idea that professors should be allowed to be entrepreneurs and to have outside activities you know that would help you know develop their ideas beyond you know academic research and so forth. and he goes through hell. He gets ostracized, you know, like like and everybody is watching him because if Shenun Shan doesn't succeed, everybody else that had the same idea maybe with a different application, maybe they were not doing electronics for a plasma fusion lab, but you know, people for example, the ones that did Lenovo or or you know, others were watching that because if he did not succeed, why bother to try? you know, I they cut his head, they're going to cut mine next, you know, and eventually through a set of serendipitous, you know, encounters, you know, Chenu Chan is able to succeed. like one of the persons that was championing him, a very smart woman is is married to a a journalist that would write a briefing for the polit bureau and together with Chenun Shan they write an article about like this successful experiment on entrepreneurship on Jong Wanjung that makes it to the higher spheres where you know now then Xiaoing's people was you know in charge you know remember China had a big institutional change that starts in 1978 they grab onto this as an example of where they want to go they protect him, you know, from the middle management that was oppressing him. And when Shenzhun survives that, then there is this wave of entrepreneurship that gets released. But the point of that story is that the guys that are demanding those institutions of entrepreneurship are not guys that are selling oranges in the red light. Okay, these are guys that are building plasma fusion reactors. So, so that demand for institutions also needs to come from somewhere. And this is a very good example that that demand in this case is coming from knowledge intense workers. So knowledge also generates a demand side for the institutions that you might need to continue to develop that knowledge. >> Yeah. And and in a sense there was a story about there there was this incredible amount of of latent and locked up knowledge and and it became released as a result of this process. But you know there are folks who say you know they're just negative about China. They say China, they just they do corporate um espionage and they just copy the West and whatnot. That doesn't really jive with what you're saying that that there's incredible amounts of talent and expertise in in China and and possibly go going forwards they're going to become the innovation center of the world. But what would you say to people that have that kind of common perspective of China? >> It's it's tough like I like now we're getting into like a more colloquial territory, but um I I I live in France and I have you know conversations sometimes that they surprise me. I've been to China many times. So I'm I'm very bullish in China. I say China is not a country. China is a planet, you know, and the West doesn't understand that because we tend to think in terms of countries, but China is like all of the Americas plus Western Europe put together, you know. So it's a planet, you know, and therefore the diversity that you have internally, it's enormous in terms of like creativity, innovation, and capacity and so forth. But there's a lot of skepticism and I think kind of like like a um you know I wouldn't maybe say xenophobia in an industrial context which I think is is unjustified. These people that usually have those attitudes they have never put a foot on China or sometimes outside of the continent of Europe themselves you know. So they they they might not you know have the experience that you know would have helped them open their eyes to these other countries and how they work. Erh, and I I think China it's, you know, a country that, uh, for better or for worse, you know, uh, it's here to stay. is growing faster than everyone else for a long time has taken so many people out of poverty and and it's a force to reckon with that you want to be friends with you know like so so that geopolitical side of like trying to be too defensive about it I I I don't think it's conducive to the growth of knowledge at the global scale >> yeah and I think it's just wrong-headed I hope the thing that people get from reading your book is that knowledge is not just a bunch of blueprints and things written down on a bit of paper if you're capable of of actually recreating something. Um, I mean, this goes to my theory of creativity. It it's not possible to create something unless you're in possession of the entire generative process. I mean, understanding is is the generative process of of new knowledge and new artifacts. And if you're not in possession of that, then you can't create anything. So, in a sense, it's doing them a disservice if they are creating things that that work. But I want I wanted to get we just touched on this a little while ago which is that um you know certainly during your PhD you were doing all of this sophisticated graph analysis you might have been using something like gey so you were looking at all sorts >> oh okay right so you know you you you were trying to create beautiful sort of partitions of graphs that made sense in in a parasimmonious way and it's a very difficult thing to do and I guess all of this is leading to you're you're trying to come up with sort of like analytical or statistical representations of the accumulation of knowledge and you created this interesting link with complexity and and entropy. I think can you tell me about that? What makes the study of knowledge difficult is this idea that it is nonfgeible. And when you look at the laws of knowledge that have been studied the most and for the longest, that principle is not uncomfortable. When you look at the learning curves of thirststones, well, eventually, you know, it's learning how to type. You don't have knowledge about different things. It's about one thing. You're learning the same thing. When you look at Mo's curve, well, it's transistors. Like there's nothing there about other technologies or you might have the same law in other domains but it's a within knowledge or within domain law. You know when we look at relatedness the story of the monkeys and the trees now we have some specificity of the knowledge. You know you you basically the monkey is in this tree and this tree is close to these other trees and not those other trees. So nonfibility is playing a role. So then you have another question which is well if knowledge is nonf fungeible if it's kind of like the letters of an ever growing alphabet h you need to maybe figure out a way to count the letters that a country or a city might have you know from that alphabet as a way to score you know their you know their their game of scrabble or as a way to to get an estimate of the value of the knowledge that they have. And how do you do that? You know should all letters count the same? You know scrabble they have different values? Yes. like a Q gets more points because it's harder to use, you know, and and and and and you know, being more rare, you know, makes it more valuable and so forth. So, how do you do this? And when what I did is and and I think originally was was a little bit serendipitously, but then eventually we develop all of the formal models to validate it and prove it is that there's a smart way to estimate the number of letters that a country will have in the alphabet by extracting vectors of matrices of a specialization that are carefully normalized. So if you grab a matrix that tells you which country exports which product and you normalize it in such a way that you take away the effects of country size because larger countries well you know there is some extensivity to things you know products that have larger markets yes you know that that column is going to be very heavy is going to have a lot of big numbers you have to take that into account you know because there's a lot of heterogeneity in geography and if you do that you normalize things properly and everything you can then extract a vector that we can show through a deductive model it's a mono estimate of the number of different letters that a country would have you know of the infinite alphabet and what is interesting about that vector that once you estimate that I vector that vector is very good at explaining future economic growth you know so it tells you ah the countries that have more letters than you would expect based on their income are going to grow so for example when you run that model right now what it tells you is that the country to bet on right now is not China because China is already quite rich so you know the the the the the room that they have to grow is is less. They're slowing down. They're going to be growing at 4%. But India should be the next rocket and Indonesia and Philippine are the next rockets. But not other countries that are similarly poor in other c parts of the world because for the other countries, we don't expect that growth because what we we we don't observe the letters based on the products that they export or the industries that employ individuals and so forth. >> Yes. And I don't think we explain this concept of the infinite um alphabet. So we we should do that as well. But um in a sense though the when when you apply this analysis you could look at somewhere like Kuwait and it might have quite low diversity and complexity and then and then like a a place like I don't know India or or the UK um you would you would look at that um entropy almost as a proxy for the productive future potential of that economy. >> Yes. Exactly. So it's a measure of potential. So this complexity is a measure of potential because if you have an idea of how many different letters an economy has, you have also an idea of how they can be recombined, you know, so they would be able to find new products or find new combinations. And that's why it predicts future economic growth. is giving you some sort of like fundamental like um in in economics you have this idea of convergence clubs you know that it's not that economist simply converge because they're poorer and therefore you know capital is going to generate larger returns and and these economies are going to grow faster but that that convergence is going to be conditional on meeting some conditions in this case you know we can show actually that that convergence clubs associated to same levels of complexity so when you look at India according to this measure you say well India in terms of income is here but in terms of complexity it's at the same level as Turkey so India should be able to converge to the level of income of Turkey you know and it's within that club you know but when you look at Liberia Liberia is very poor but in its club there's there are not a lot of rich countries they're none actually you know so therefore it's complexity it's in equilibrium with its income at that moment and when you look at a country like Qatar you get the opposite it's like okay their income is extremely high given their complexity So if they were to run out of petroleum gas then you know their income would have to um come down because they would be in equilibrium with countries that are more middle income maybe like with you know countries that about 7 10 12k of GDP per capita. Can there ever be too much complexity? Because um again to use the the language models analogy and we should also talk about the relationship between embeddings, you know, natural language embeddings that use this distributional hypothesis, you know, you know, a word by the company at Keeps and and I guess you could describe that you're creating something similar to embeddings based on um you know, different types of products being sold in in the same location and economic space. But the point I was making though is that you could just argue that more complexity good. But in a way what that's saying is that there are more paths. So there are more parts in innovation space which are reachable through combinations of the letters that exist. But can too many paths be a bad thing? And too many paths be a a bad thing? I I don't think you have to take all of the paths, you know. So I think you have two different dynamics. the space of possibilities and and the choices that then you make, you know, like so in principle any combination of notes could be considered a song. In reality, you know, you want to find the combination of notes that actually work well together, you know, the C and the G and so forth. So, so you're working, you know, with with combinations that that that eventually become quote unquote meaningful, you know, or or or delightful to to the listener. So, I think having the paths available is probably a good thing. you you don't want to be constrained by possibilities because if you're constrained by possibilities then there are things that are simply not a choice for you. So in the case of a developing country with low complexity there are a lot of paths that are not a choice for them. Even if they would want to go there the probability of success is is extremely low and they need to gradually get to a point in which those paths open to them. So I would say in general, yeah, having, you know, more letters of the infinite is is a good thing. >> A virus in some ways is more intelligent than us because it has the ability to delete strategies. So almost keeping your options open too much will lead to you exploring bad parts of innovation space and kind of like not good parts if that makes sense. >> No. Yeah. But like but but I I I think one of the thing is to have a wide space of options and another thing is is the ones that you are exploring which is a strategy. So I would say like a country like the UK or you know China or you know the United States, Singapore, Switzerland they have a wide set of options you know and that doesn't mean that that exploring all of them. There might be a lot of bad options to explore like you you you let's say Breaking Bad is a business model. there's a bad option you know you can you can use your skills as a chemist you know to go down that path you know but the same skills that Walter White had as a chemist can be used also to you know find cures for diseases which is you know another application so the the options I think are there and the choices are the ones that can be good or bad but not having chemist in your country I think you know it's a really constraining you know capacity >> absolutely and and we should also factor in knowledge decay So in a sense, the reason why this is such a good um proxy for future potential is because any bad strategies would have already been deleted by knowledge decay because we would have let things die out. So in a sense, we're kind of like we're we're trimming and we're pruning as we go. And if these convergent strategies still exist, they probably exist for a good reason because they have utility. >> No. Yeah. and and I think like knowledge also gets replaced by um other representations sometime encompass the previous one. There is a an example that I use in the last chapter of the book uh to illustrate disruptive innovation which is about the way in which we calculate pi you know uh is the calculating pi using polygons or calculated pi using Newton's serious formula and and and you know the idea of calculating pi using polygons was an idea that that people used for more than a thousand years and the last person to use it extensively was a um I think a Dutch mathematician that dedicated his life to calculate pi to like about 32 digits and it took him 20 years to do so and then Newton was able to do that you know in a matter of days using this formula h so therefore you had kind of like now disruptive innovation in an example in which the new curve didn't even start worse it already started better and and and and so to enormous you know heights and in that context I think it's fine that we forget how to calculate you know pi using you know a polygon with a million on sides inscribed into a circle because it is useless. it is contained within a better way of doing things you know so so we have a superior knowledge uh representation that replace you know that that that inferior one >> so based on our conversation today do you think large language models have knowledge >> so I don't think of knowledge as an individual phenomena I think of knowledge as a collective phenomena you know uh a baby born in an empty island it's not going to grow to be smart even though it has the capacity to develop language because that's a genetic you know gift that we have it's not something that is socially learned. So I do think that LLMs, you know, contribute to this ecosystem of knowledge. Some people would have said in the past, I have all of the knowledge in the world in my library. You know, I I I would say books do not have knowledge, but books contribute to that collective capacity to know and to learn and to communicate knowledge and so forth. So if knowledge is a collective phenomena, the question is not whether LLMs, you know, have knowledge or not because none of us have knowledge only exists in that larger context. The question is well if LLMs within that larger ecosystem are increasing our collective intelligence or our capacity for collective learning. >> There is obviously two schools of thought. One is you know just what you've said that it's a cultural technology and and the the locust of cognition thinking understanding and and action comes from us. The other school of thought is that they are agentic you know cognizant machines in in in their own right. And I just I don't think that's true >> for me. like what matters is that they're useful. I I I'm I'm more pragmatic when it comes to that, you know. So, are they useful? Yes. Are they perfect? No. Nothing is perfect. And a lot of things are useful. And I think they fall into that category, you know. I I learn a lot by talking with LM. For example, I I moved to France, you know, five years ago. There's a lot of things that I don't know about the local rules that I can more easily consult using an LLM than a Google search, you know. So I want to know about you know tax law you know in France you know there's a lot of details well I I explain my situation I get those answers and then I have an accountant and when I meet with my accountant to discuss that I'm much better informed and I can make better questions so I think I'm becoming smarter through those conversations because I'm learning faster thanks to those interactions. Is it because the LM has knowledge? Is it because I have knowledge? or is it because we are wiser when we are together? >> So then there's the question of um how efficient your book is at knowledge diffusion. So um I I absolutely loved reading it. I actually read it in in the space for 24 hours and I I gobbled it all up. But but it it's overwhelming. I want to read it again several times just to get all the value out of it. But in the afterward you said that I think it was about 10 years ago you wrote a book I think it was called Why Information Grows. and you know you've done your PhD and you were using very abstract thought experiments and now you've had time to think about it and you've told it using um stories you know like um you you found really interesting examples that exemplify all the different ideas and it makes it so much easier to understand. Tell me about that journey. There was an an excellent point that was done by Annabelle Haxley. Um she works at Penguin and and she was the one that basically you know parade me around London the last three days. She's actually you know funny fact part of the family of th Hxley you know th Hxley like Darwin's bulldog. So you know I felt so honored to go around you know the London science scene with someone like her and she said one of the things about your book which is interesting is that your book is about the fact that knowledge is extremely specific. The book starts with the story of Charlie which we haven't told but it's a story that shows that knowledge can get extremely nuanced but then you go and communicate all of your ideas about the principles that govern the growth in fusion and value of knowledge using these very nuanced ideas but these nuance ideas in this context are not just decoration because they help exemplify that exact point that knowledge is extremely nuanced that the story of Ibuka absorbing knowledge on how to produce magnetic tape and the story of you some hours later bringing cotton spinning manufacturing to the United States are full of details and that in those details you know is where knowledge hides you know uh so that's something that I thought it was quite interesting because honestly I had not realized it until she mentioned it that way that that the book made that point about specificity and that therefore the specificity of the story like that detail was not a simple literary resource but was a way to hammer on on that point. >> Beautiful. Uh, Professor Hadalgo, this has been absolutely amazing. Thank you so much for joining us today. >> Thank you. Has been a delight. >> Um, folks at home, I really recommend you read this book. I' I've enjoyed it so much. Genuinely, it's really amazing, especially if you're a fan of some of the content we've been making recently. Check this book out.

César Hidalgo has spent years trying to answer a deceptively simple question: What is knowledge, and why is it so hard to move around? We all have this intuition that knowledge is just... information. Write it down in a book, upload it to GitHub, train an AI on it—done. But César argues that's completely wrong. Knowledge isn't a thing you can copy and paste. It's more like a living organism that needs the right environment, the right people, and constant exercise to survive. Guest: César Hidalgo, Director of the Center for Collective Learning The Big Ideas 1. Knowledge Follows Laws (Like Physics) Just as temperature and gravity follow predictable rules, so does knowledge. César outlines three laws: - Time: How knowledge grows (fast at first, then it plateaus) - Space: How knowledge spreads (it's way harder than you think) - Value: How we can measure a country's "knowledge potential" 2. You Can't Download Expertise The most memorable stories in this conversation prove that knowledge is embodied—it lives in people, teams, and organizations, not in manuals. 3. Why Big Companies Fail to Adapt César explains "architectural innovation"—the idea that small changes (like shipping books directly to customers) can require a completely different organizational structure. 4. The "Infinite Alphabet" of Economies Every skill, every industry, every capability is like a letter in an alphabet. César's research shows you can actually predict which countries will grow by counting their "letters." If you think AI can just "copy" human knowledge, or that development is just about throwing money at poor countries, or that writing things down preserves them forever—this conversation will change your mind. Knowledge is fragile, specific, and collective. It decays fast if you don't use it. The Infinite Alphabet [César A. Hidalgo] https://www.penguin.co.uk/books/458054/the-infinite-alphabet-by-hidalgo-cesar-a/9780241655672 https://x.com/cesifoti Rescript link. https://app.rescript.info/public/share/eaBHbEo9xamwbwpxzcVVm4NQjMh7lsOQKeWwNxmw0JQ --- TIMESTAMPS: 00:00:00 The Three Laws of Knowledge 00:02:28 Rival vs. Non-Rival: The Economics of Ideas 00:05:43 Why You Can't Just 'Download' Knowledge 00:08:11 The Detective Novel Analogy 00:11:54 Collective Learning & Organizational Networks 00:16:27 Architectural Innovation: Amazon vs. Barnes & Noble 00:19:15 The First Law: Learning Curves 00:23:05 The Samuel Slater Story: Treason & Memory 00:28:31 Physics of Knowledge: Joule's Cannon 00:32:33 Extensive vs. Intensive Properties 00:35:45 Knowledge Decay: Ise Temple & Polaroid 00:41:20 Absorptive Capacity: Sony & Donetsk 00:47:08 Disruptive Innovation & S-Curves 00:51:23 Team Size & The Cost of Innovation 00:57:13 Geography of Knowledge: Vespa's Origin 01:04:34 Migration, Diversity & 'Planet China' 01:12:02 Institutions vs. Knowledge: The China Story 01:21:27 Economic Complexity & The Infinite Alphabet 01:32:27 Do LLMs Have Knowledge? --- REFERENCES: Book: [00:47:45] The Innovator's Dilemma (Christensen) https://www.amazon.com/Innovators-Dilemma-Revolutionary-Change-Business/dp/0062060244 [00:55:15] Why Greatness Cannot Be Planned https://amazon.com/dp/3319155237 [01:35:00] Why Information Grows https://amazon.com/dp/0465048994 Paper: [00:03:15] Endogenous Technological Change (Romer, 1990) https://web.stanford.edu/~klenow/Romer_1990.pdf [00:03:30] A Model of Growth Through Creative Destruction (Aghion & Howitt, 1992) https://dash.harvard.edu/server/api/core/bitstreams/7312037d-2b2d-6bd4-e053-0100007fdf3b/content [00:14:55] Organizational Learning: From Experience to Knowledge (Argote & Miron-Spektor, 2011) https://www.researchgate.net/publication/228754233_Organizational_Learning_From_Experience_to_Knowledge [00:17:05] Architectural Innovation (Henderson & Clark, 1990) https://www.researchgate.net/publication/200465578_Architectural_Innovation_The_Reconfiguration_of_Existing_Product_Technologies_and_the_Failure_of_Established_Firms [00:19:45] The Learning Curve Equation (Thurstone, 1916) https://dn790007.ca.archive.org/0/items/learningcurveequ00thurrich/learningcurveequ00thurrich.pdf [00:21:30] Factors Affecting the Cost of Airplanes (Wright, 1936) https://pdodds.w3.uvm.edu/research/papers/others/1936/wright1936a.pdf [00:52:45] Are Ideas Getting Harder to Find? (Bloom et al.) https://web.stanford.edu/~chadj/IdeaPF.pdf [01:33:00] LLMs/ Emergence https://arxiv.org/abs/2506.11135 Person: [00:25:30] Samuel Slater https://en.wikipedia.org/wiki/Samuel_Slater [00:42:05] Masaru Ibuka (Sony) https://www.sony.com/en/SonyInfo/CorporateInfo/History/SonyHistory/1-02.html [01:01:45] Corradino D'Ascanio https://link.springer.com/chapter/10.1007/978-3-319-09858-6_38#:~:text=6%20Conclusions,%2C%20comfort%2C%20and%20technical%20performance. [01:16:00] Chen Chunxian https://thebhc.org/sites/default/files/tzeng.pdf Event/Place: