MM: So I read this book called Red Notice by Bill Browder — it’s his story of investing in Russia and taking on the oligarchs of Russia, and it’s this non-fiction thriller. It turns out his father, Felix Browder, was the chair of the math department here for a while, so I was fascinated by the connection. So I’m curious to hear about your interaction with him.

CK: I was very fortunate to meet Felix Browder, so maybe I’ll say a few words as to the circumstances of getting to know him. I was very young, 19 years old. I had left Argentina at the time of the political upheaval there, where the university had ceased to function. Some of my teachers suggested that I come to the United States, and they wrote to Alberto Calderón, who was at that time a professor at MIT and a close friend of Felix Browder.

And Calderón then communicated with Browder about me coming, and I didn’t have a degree. I never got an undergrad degree. And Browder was very kind and agreed to meet me. This was September 1973. So I went to his office, and he said, well, let’s see what you know and what you don’t know, so he gave me an impromptu exam. And he asked me lots of things; I knew some, but I mostly didn’t know them. And since I had no degree, he said the only thing we can do is have you be a student at large for one year, and then we’ll see what happens.

But he spoke with people at the admissions office, and they said that since I had no degree, I had to also take some courses that were not math. So I took the freshman sequence in humanities and the freshman sequence in social sciences and I took three math classes at the same time, and I had to work very hard.

At the end of the year, things went well, so I went for the summer to Argentina, and I received a letter from the department saying that I was admitted to graduate school as a graduate student, and they got a fellowship for me. And this was all done by Felix Browder.

And through the years, I had interactions with him. After my post-doc at Princeton, I went as a tenure-track faculty to the University of Minnesota. And I was there as an assistant professor. And after a year, I was promoted to associate professor and then to professor. And then around ‘84, Browder approached me to see if I would be interested in Chicago. I said yes, I would be interested. Eventually, the department made me an offer. I accepted. In the fall of ‘85, I came to Chicago.

At that point, Felix had resigned as chair of the department and moved to Rutgers. And after that, I saw him a few times, once at Rutgers, when there was a conference in his honor, and I was very happy to see him.

But when I was at Chicago as a student, I was close to Bob Fefferman, who was a junior faculty and I was a graduate student, and we were working in very close fields. And he was very close to Felix and his family, so I got to know both Browder boys. I got to know Bill and his older brother, who’s now an astronomer in Hawaii.

MM: You said you came to the US when you were 19. Were you always interested in math? Can you reflect on your interest at a young age?

CK: Yes. So I first became really interested in math when I was 12. This was my first year of high school. You know, the educational system is different in Argentina, there’s no middle school. You go from elementary school to high school.

In my first math class in high school, I had an excellent teacher. He had a Ph.D. in math, he was from Chile, and the class consisted of Euclidean geometry and he taught us to prove things. So the class was about proofs, facts about triangles. And then I realized that my mind worked the same way as mathematical proofs. So from that point on, I got very interested.

But when I finished high school, my family and my math teachers all told me, you shouldn’t get into math because you will starve. The jobs in math were very ill-paid in Argentina. It was very difficult to have a career at that point in math in Argentina. So I started studying engineering at the University of Buenos Aires. But at the same time, I also studied math. And after the first semester, I was completely disgusted with engineering because it wasn’t rigorous. So I stopped engineering after my first semester; the second semester, I just continued in math; and then I had the third semester. And then after that, I came to Chicago.

MM: I want to hear about your experience at Chicago. You took classes with Alberto Calderón and Antoni Zygmund. Can you reflect on your experience with that? And more broadly, it is known as the Chicago School of Analysis. For someone who is not familiar, what was that?

CK: When I first met Zygmund, he was 73 years old — almost my age now! And in my first year, I took a measure and integration course from him. He wrote — at that point it wasn’t even a book — it was lecture notes that eventually became a book, and I enjoyed it very much. And he was very kind to me.

And when Calderón came back to Chicago the following year from MIT, I took a class from him on some recent work of his in harmonic analysis. And then I became interested, so I started going to what was then called the Zygmund Seminar. It was once a week at 3:45, which is the time that we’re keeping to this day. There were two seminars: one was expository and the other was research. And of course, at the beginning, I didn’t understand anything. But slowly I began to understand.

Let me say a few things about the Calderón–Zygmund School. So Zygmund was one of the premier analysts in the first half of the 20th century. He worked mostly on trigonometric series, Fourier series, and real analysis, but only in one variable, so his aim was to try to find the theory in higher dimensions.

In the late 1940s, he went on a visit to Argentina. What happened is after the war, the US State Department wanted to promote scientific development in the underdeveloped world. They sent Zygmund to Argentina, where he met Calderón. He was his teaching assistant in a class he taught. And Zygmund was of course extremely impressed with Calderón, so he arranged for him to come here to do his dissertation.

Calderón was an engineer before this. He was an amateur mathematician because his father was dead set against it. But he came to Chicago and after 2 years, he finished his PhD, which consisted of 3 new research articles that opened the way to analysis in higher dimensions. That’s how the Calderón–Zygmund collaboration developed. They wrote many papers together and opened up a whole new school of mathematics. It’s been influential for many years.

MM: Can you reflect on the Honors Analysis course that UChicago Math has become famous for?

CK: Yeah, that is the creation of Paul Sally, who was one of the young faculty members when he first came to Chicago in the mid-70s. Eventually, towards the end of my studies, he became the chair of the department. And he was somebody larger than life, he was a real character; the students loved him and feared him.

And he developed this class, which is called Honors Analysis, which was to attract the top undergraduates and teach them really high-level math and very advanced things. And, in fact, my next-to-last year as a graduate student, I was the teaching assistant for that class, which I enjoyed very much. Jerry Bona was the teacher, he was a young faculty member. And I remember one time when I got stuck on homework problems that we had assigned, I had to ask for his help and he was extremely amused about it.

MM: Can you reflect a bit on what does mathematical research involve? You have made so many tremendous contributions, oftentimes in collaboration. What is the collaborative process like? Do you remember any of the moments where you had an epiphany, and what does that moment look like?

CK: So after my post-doc at Princeton, there I had my first really serious collaboration with David Jerison, who is still my collaborator now. We just published a paper together, and he’s exactly my age. And I enjoyed that very much. I was an instructor, he was a graduate student, because he had taken 2 years off to go to Paris to learn math there. Anyway, so we worked very hard together, and we got some success.

But when I went to Minnesota, I collaborated with, and was sort of mentored by, Gene Fabes, who was a former student of Zygmund’s at Chicago. And he was a sweet person. And in his work, he always worked with others. He was a very sociable person, and he enjoyed the process of discussing with somebody else. And I learned this from him and I’ve done it ever since.

When you’re in math as a researcher, you have to go really deep into the subject and spend many hours thinking and working. But if you’re doing it with somebody else, then you have a social outlet for your work, and that is a very nice situation. And the other thing is that in conversation with somebody else, you sometimes generate ideas that you didn’t know you had, and they come out. And you don’t know how they came out, but they come out. And that’s part of the pleasure of the collaboration.

I have had moments of inspiration, yes, and it comes as a flash. It’s not a conscious thing. I’m wondering about something, and then suddenly I don’t know from where some idea comes. When I was younger, it used to happen during the night, or sometimes first thing in the morning in the shower. Now I’m too tired, and so it comes later on.

I remember one of my collaborators and co-authors was in Spain, and we were working with a friend, Gustavo Ponce — this collaboration started in 1987 — and we had been discussing with Gustavo, and Gustavo had to go back home. And Luis and I started discussing, and then suddenly, I don’t know how or why, I had this vision of how to do it. And then I could see the whole problem, and it was really extremely satisfying. One moment you’re stuck, and then the next minute I saw the path.

MM: You mentioned these social interactions and collaborations. You were president of the International Math Union and vice president of the AMS before that. Could you explain what your role was? And you were chair of the Fields Medal Committee, so what was your role there?

CK: The role of vice president of the American Math Society is basically you attend meetings of the Council of the American Math Society, and you give your opinion on the matters that are discussed.

Being president of the International Math Union is a job of tremendous responsibility. It’s supposed to steer the development of math interaction all over the world in both research and education. And, of course the decisions get voted on, but you are the generator of the initial proposals.

So, during my period as president of the IMU, there were two very cataclysmic events. The first one was COVID. The IMU is built to generate the interaction of mathematicians from different countries. COVID was the anti-IMU. On the other hand, we were very lucky that Zoom was developed, and there was a possibility for online interaction. So that saved the day. It meant that we could do many, not all, but many of the activities that we had planned.

But when finally COVID more or less was resolved, we had the centenary of the IMU. So there was a celebration in France at Strasbourg, which was the place where the IMU had first been founded, and we had a meeting to celebrate this. People came from various countries, and the meeting was finally held in 2021.

It should have been held in 2020, but we had to postpone it due to COVID. And people were so happy that we were finally able to meet together and discuss, make math presentations and talks. We had the idea of having 15-minute talks on various different topics. It was a great success, and everybody was just extremely happy.

The next cataclysmic event was the Russian invasion of Ukraine. Because this happened in 2022, and the International Congress of Mathematicians (ICM) was due to be held in St. Petersburg in the summer. So we had to decide what to do on the spot.

I had contemplated beforehand what we would do in this case, because I had a feeling it was going to happen. So I discussed the plan with the Secretary General of the IMU at 4 in the morning our time. We had a virtual meeting with the Executive Committee. Then we had the meeting, and one of the members of the Executive Committee was Russian. And he was a good friend of mine. And he was beside himself, so it was a very difficult meeting.

So I presented the plan, the Executive Committee approved it, we immediately wrote a statement from the Executive Committee, and then we were confronted with the fact that all the money for the meeting came from Russia. So we didn’t have access to it. And second, we had to reorganize the meeting with all the speakers, and we had to hire somebody to do an electronic version of the whole thing. The Secretary General had connections with a European foundation, and he secured some funding to be able to pay for everything.

We had to personally write to all the speakers to make sure that they were on board, that they understood how it was going to work, and so on. I had to chase many of them. And in the end, we got all but one. And this one is a very famous man who has layers and layers of protection — Yann LeCun — I couldn’t get through to him, the layers of people to get to him made it impossible. But every single other person was on board, on the order of 150 people.

As chair of the Fields Medal Committee, we had to select the members of the Fields Medal Committee. It had 12 members at the time. So that involves various considerations: field, country, gender, and all the standard things. And we formed an excellent committee.

We had a seminar over the summer before the final decision in which one member of the committee would make a presentation on the work of each of the finalists. Some of the most impressive contributions. So we had a very serious methodology. And of course, we wrote to hundreds and hundreds of people, got hundreds of opinions on each one of these candidates. And after that, we narrowed it down through our own discussions. And finally, we had the seminar. And then after the seminar, we had a further discussion. And then we had the final vote.

MM: I am very curious to hear whether you are, or were ever, interested in physics — for example, the applications of functional analysis (Hilbert spaces) to quantum physics.

CK: So I am sort of in principle, as an outsider, interested, but I don’t have the technical knowledge to actually be able to do anything about that.

MM: I am very curious to hear your thoughts on how AI can accelerate or influence math research.

CK: This is a big topic right now, not just in math. In general, what would the effect of AI be? Would it even be safe?

This is a very difficult topic, from being incredibly useful to potentially dangerous for the world. In mathematics, as far as I can tell so far, it is a useful tool for some things. But what hasn’t been seen to happen, as far as I know, is for AI to create an entirely new idea. This we haven’t seen. On the other hand, there are sometimes computations that are too large for humans. So, this certainly looks like a great tool. It also looks like it could be very dangerous. So, it is something we have to watch and look at. But the genie is out of the bottle; nobody can put it back. It is there.

What I hate is what it is doing to education. It is pointless to assign homework. Not pointless to assign it, but pointless to grade it. Because all of you guys are using AI to solve the problems. And this is extremely short-sighted. Because it means you don’t learn how to reason by yourself. And the students who are doing this are hurting themselves incredibly. They will be in trouble. It is a danger to the entire education system.

MM: Are you familiar with the book Gödel, Escher, Bach? It is my favourite book.

CK: I only have a vague impression of the book. But it seemed like something very interesting and very much fun. I think the analogy between art and math is a true analogy. It is an interesting thing to uncover.

In math, I know when I find something beautiful, and when I find something ugly, I also know it. But I don’t have any kind of scheme that I can use to explain it. I would say it is more visceral to me. I don’t analyze that kind of thing. But there are certain things that attract me.

MM: I would love to hear about your personal interests or hobbies, or do you have favourite books, favourite movies?

CK: I have many. There’s not a time when I don’t read. I read a lot of books. I love the opera when it comes to music. I don’t know how to sing. I never played an instrument, but my daughters have a lot of talent. They play the flute, they play the piano. But I enjoy the arts very much.

I liked, growing up, the short stories of Jorge Luis Borges. I have a personal experience with him because when I graduated from high school, I got some kind of prize for a math contest, and he handed it to me. And he and my advisor Calderón knew each other.

Nowadays, I tend to read a lot of mysteries. Right now, I’m reading a series of mysteries by a Cuban author whose last name is Padura. My friend, Gustavo, always gives me these books. He enjoys them, so he sends them to me.

When I was a kid, I read Jules Verne, Agatha Christie. Lots of books.

Subscribe now

MM: You graduated with a First Class degree in modern history from Oxford — what was that experience like? What were your favourite classes, favourite professors, favourite tutorials?

SM: Oxford gave students a broad choice in what they focused on. My concentration was British and European history from around the mid 1850s to 1939. Of course, the main political drama in that period was the contest between socialism, democratic capitalism, and fascism. I became fascinated by the question of why the left was a big factor in France, Germany, Italy, Spain and obviously Russia. But Britain had no Marxist movement to speak of. The reason lay in the malleability of the British establishment. By gradually extending the franchise and offering some measure of economic inclusion, Britain turned the revolutionary left into the pragmatic, labor union left. Pragmatism bred pragmatism, in other words. It’s a lesson we might recall today in these polarized times.

MM: From my observations it seems like the species of macro hedge funds (once thriving with George Soros, Jim Rogers, Paul Tudor Jones etc.) has gone extinct or at least endangered in recent times. For example, Peter Thiel’s Clarium Capital suffered losses post 2008 and was shut down; Ray Dalio’s Bridgewater has underperformed the S&P. Why do you think this happened? Can it be revived — if so, how?

SM: Macro hedge funds grew up in response to the end of the dollar-gold peg in 1971, which caused currency fluctuations that could be traded. Macro trading then entered its prime in the 1990s, when governments tried to tame currency volatility by soft-pegging their currencies. The soft pegs set up easy targets for hedge funds, which could sell so much of a weak currency that the pegs would eventually break, handing the traders a fortune. Breaking a soft peg was how Soros and other macro traders profited from the devaluation of the British pound in 1992, and versions on the same drama played out in Asia later in the decade. These days macro trading is somewhat subdued, because there are fewer ill-considered currency pegs to target. Most currencies float, and the ones that don’t (e.g., in East Asia) are generally being artificially held down rather than being propped up. When a currency is being held down, it does not present a target for speculators because the government can always print and sell more of its own currency to keep it down. This is the main reason that macro funds are less successful these days, though there are still opportunities to trade currencies, interest rates and so forth.

MM: What do you think about the species of philosopher-investors? George Soros comes to mind for his theory of reflexivity. Similarly, Peter Thiel for his synthesis of Girardian and Straussian ideas into his investing. How do such people operate? Is their philosophy an attempt to intellectualize or is it deeply intertwined with their investing? And what do you think about being a philosopher-investor in the age of AI?

SM: Hedge fund investing is a highly intellectual pursuit, so not surprisingly many practitioners are also interested in philosophy. In the special case of George Soros, the theory of reflexivity opened his mind to the boom-bust nature of financial markets: reflexivity’s insight is that opinions of what will happen can be self-fulfilling, and that the feedback loop between belief and reality can drive a market so far from fair value that it will eventually reverse violently. But in general I would doubt that the philosophy of hedge fund investors has a lot to do with how they actually trade.

MM: Renaissance Technologies hires the best and brightest PhDs in physics, math, astrophysics and computer science. Evidently, they have done extremely well at generating returns as a quant hedge fund — but what do you think about the societal cost of redirecting top scientific talent into finance (“brain drain”)?

SM: The diversion of scientific talent into quant trading often raises concerns, but I don’t think there is a strong case for worrying. I say this for six reasons.

    First, the hundred or so PhDs at Renaissance Technologies are extremely smart, but one hundred is a tiny number relative to the number of PhDs employed at a tech company. For example, Alphabet is thought to employ several thousand PhDs.

    Second, the hundred-strong team at Renaissance Technologies is eliminating inefficiencies in a very broad range of financial instruments—currencies, stocks, commodities—and they are doing this globally. So the world is getting a lot of additional efficiency from a relatively small number of quants. This efficiency ensures that average retail investors can sell their positions any time without worrying that they are selling at a bad moment when some sort of liquidity distortion or other temporary dislocation will cause them to be ripped off. It’s hard to quantify the social benefit of this financial efficiency, but it’s non-zero. We don’t really know whether those same hundred PhDs would have contributed more to society by staying in academia or going into other industries. But we can’t be sure that their contribution would be greater.

    Third, the discussion of where or how quants contribute most assumes that there is a fixed number of quants. But the rise of quantitative trading creates an incentive for young students to get a quantitative training. The success of Renaissance Technologies has probably increased the supply of Americans doing graduate degrees in fields such as math, computer science, or physics. So, a bit like the famous “lump of labor” fallacy in economics, there is a “lump of quant” fallacy in this debate.

    Fourth, Jim Simons, the founder of Renaissance Technologies, created various philanthropic initiatives to fund math education and scientific research. When my daughter did an astrophysics PhD at UChicago, the Simons Foundation was funding a lot of the activity in her field. (She had no idea that this Simons was the same person as the founder of the hedge fund that I had written about.) Likewise, DE Shaw, founder of the eponymous quant hedge fund, also created DE Shaw Research, creating research jobs for several dozen PhD scientists. A Shaw Research alum, John Jumper, won the Nobel Prize in 2024.

    Fifth, free societies should allow quants to choose for themselves where they want to work. If we start second-guessing their choices, we could second-guess a lot of other ones. Should we allow people to work in the fashion industry, which labors to extract large sums of money from clothing whose functional value is indistinguishable from cheaper clothing? Isn’t the fashion industry stoking envy and insecurity? Shouldn’t we require smart business leaders to work in other fields? (Obviously not!)

    Finally, and more generally, it’s often argued that the growth of employment in the financial sector is a disturbing trend. Do we really need all these spreadsheet jockeys? But as economies become more technically specialized and complex, it’s reasonable for the financial sector to grow, because the role of finance is to allocate capital wisely to all these technical and specialized industries. We know from experience that central planning is a bad way to make these allocation decisions. So we need a sophisticated financial system to do it.

MM: With the proliferation of VC funds, it seems like capital is becoming a commodity. What really differentiates one VC shop versus another? Is it track record? Value added to investee companies? What are some examples of real differentiation?

SM: Track record certainly helps. Most founders would be happy to take money from one of the top funds because the brand is valuable: saying “I’m doing a startup” will elicit a shrug, but saying “I’m doing a startup backed by Sequoia” confers instant credibility. But VC partnerships also get into deals because a particular partner has a good reputation. A strong VC partner may add value to a startup by bringing engineering judgment, hiring experience, go-to-market skills, introductions to potential customers, a network in another country, or even just emotional solidarity and counselling. Also, there are many stories in venture about startups that get turned down by famous VCs, eventually take money from a non-famous one, and then do well—PayPal was initially backed by Nokia Ventures for example. So it’s not just deal access that matters in venture. Deal selection is another differentiator.

MM: From Napoleon Ta of Founders Fund to the team at Renaissance Technologies, some of the best investors prefer to stay under the radar. When you start a new book, what exactly helps you get access — emails sent, mutual friends tapped, documents requested? A concrete walk‑through would be invaluable to aspiring long‑form writers.

SM: It’s all of the above. You have to try everything and eventually you may get access. Expect to have the door slammed in your face a few times. But in general it’s a combination of getting to know people who know other people who then eventually intro you to the people you need to interview, and being very well prepared at each step of the way. Don’t expect people to spend time with you if you haven’t studied all that’s public about them already. When I go see someone, I often have ten pages of typed questions that I’ve worked on for days. If I’ve put in the time, the interviewee is more likely to put in the time. If I can complete their sentences for them because they are repeating a point they have already made on a podcast, this will encourage them to dig deeper and say something surprising that they have not aired with any other writer.

MM: What are the most fascinating use cases of AI in investing that you have seen, across all stages from VC to public markets?

SM: I actually think that being a venture capitalist is one of the jobs that is least likely to be disrupted by AI. Being a VC is about bonding with human founders. It is about judging whether a brand-new product idea will be of interest to human customers. It is about imagining the future—and the future is not in an AI’s training set. On the other hand, there are other types of investing that have been using AI for quite a while—algorithmic market making and so forth. Between these two poles, there are funds that haven’t historically used AI but that are now starting to do so. I hear of lots of use cases: for example, a merger-arb hedge fund that can ask an LLM for help analyzing how an anti-trust review is likely to come out.

MM: What are 3 books you would love to read that nobody has written? What are 3 books you most highly recommend reading?

SM: I’m fascinated about how, in common law systems, private lawyers can invent novel legal structures that change the way that capitalism functions; and yet they are anonymous, have no democratic mandate, and most people don’t even realize that the rules are created this way. So I think a great book on the industry of commercial law would be worthwhile. It’s grown like crazy in New York and London. It’s powerful. It’s secretive. What more do you want? (It would be a huge task to gain access!)

Good books to read: I loved The Money Game by Adam Smith, an old classic about the market frenzy of the 1960s. I also thought Too Big To Fail by Andrew Sorkin, a fly-on-the-wall account of the 2008 financial crash, was wonderful. I am looking forward to reading Sorkin’s new book on the 1929 crash.

MM: What book are you working on next?

SM: By luck, I negotiated deep access to DeepMind, the Google AI lab, in November 2022—the same month that ChatGPT turned AI into a cultural obsession. Since then, I’ve interviewed around a hundred AI scientists inside and outside the lab, and spent more than thirty hours with Demis Hassabis, DeepMind’s principal founder. Hassabis is an extraordinary character, and AI raises profound questions about the nature of human cognition, the future of human society, and the dilemma of the scientist who creates a potentially existential technology—what I think of as the “Oppenheimer conundrum.” Friends who have read early drafts of the manuscript tell me this is my best book so far—but then those are my friends! The book will be out in 2026.

Here is the transcript of our email exchange.

MM: In Cryptonomicon, Neal Stephenson writes - “As it happens, Alan has become fascinated by the incarnations of pure ideas in the physical world ... Turing is neither a mortal nor a god. He is Antaeus. That he bridges the mathematical and physical worlds is his strength and his weakness”. What, to your mind, are the most beautiful examples of the physical manifestations of a pure math idea?

CH: Computation in general is pretty cool, and widely underappreciated. Any computer can simulate any other computer is kind of mind boggling.

MM: From a first-principles standpoint, multi-junction solar cells seem like the obvious successor to monoPERC and TOPCon for breaking past the Shockley-Queisser limit. Yet they remain a niche, space-only technology. In your view, what is the single biggest physical or manufacturing bottleneck preventing their terrestrial dominance, and what non-obvious breakthroughs are needed to unlock a cost-competitive learning curve for them?

CH: I don’t know much about emerging PV technology. The biggest obstacle for PV in general is permitting reform. For perovskites and other more exotic PV technology, the roadmap is clear - get to positive unit economics and then ride a cost curve that is at least as steep as the incumbents.

MM: The 2023 Terraform white paper (version 2) says that synthetic fuel will use 80+% of all solar generation long term — this implies that in the long term, converting photons to molecules will outcompete storing electrons in batteries (and discharging when needed). Could you walk me through the reasoning for why you believe >80% of solar power will be used for DAC fuels and not to charge batteries for battery energy storage?

CH: The leading order term is that chemical synthesis is intrinsically inefficient, so that's a 3x boost right there.

MM: What is the first principles reason for why 1 MW solar module produces 6,500 cubic feet of CH4? In that chain from solar panels to electrolysis to the Sabatier reaction, what are the key bottlenecks that prevent an even greater throughput of CH4 per MW of solar generation?

CH: 1 MW solar array produces about 6 MWh per day (sun is underground half the time, and pointing in the wrong direction another half of the time). 6 MWh is equivalent, energetically, to about 20,000 cubic feet of CH4, but the synthesis process is not very efficient. First, most of the power goes to making H2, at perhaps 60% efficiency. Then half that hydrogen just gets converted into water instead of methane. But you get a little bit of efficiency back because the carbon atom that’s released to become CH4 from the O2 that becomes water, also has positive enthalpy of combustion. So all up, about 30-35% efficiency. There are known ways to push this into the high 40s, and physically 60% efficiency is probably possible - but it would not be close to cost optimal.

MM: There are several emerging trends in AI data centers - e.g., bitcoin mines being converted to AI data centers, direct liquid cooling to the chip, orbital/in-space data centers, modular/containerized data centers etc. Which of these (or any others) do you find most promising, and why?

CH: I think the most promising is direct DC HV systems, including racks powered in series. All the other stuff is either too exotic, or I’m not qualified to speculate, or both.

MM: In your podcast with Dwarkesh, you spoke about the potential for integrated solar with chips (power with compute), since both are fundamentally made of the same material — silicon — albeit at different purity levels. Could you paint a picture for what such an integrated solar-GPU might look like and how TSMC might build it?

CH: That’s a hypothetical about the distant future state of humanity. With our current manufacturing tech, the chips cannot be larger than the lithography field of view, which is too small. So let your imagination run wild.

MM: Like yourself, I am majoring in physics and math. What are the most profound and elegant concepts you learnt in college that fundamentally rewired your thinking? Could you convey that elegance and beauty to motivate why one must study them? (E.g., in freshman year I found Fourier transform and Tychonoff's theorem the most beautiful).

CH: Probably the earliest “aha!” moment I can remember was showing that central inverse square laws give rise to “orbits” broadly construed that map perfectly to the conics, which is to say projections of cones and planes. Something quite profound about that duality, and it provides a strong hint of similarly deep principles, such as holography, that I encountered later.

Subscribe now

JOS: Even today, by the way, if you go to various financial sites or providers and look at the P/E for, say, Apple, what you’ll find is that if you go to Reuters, you’ll get one P/E. If you go to Compustat, you’ll get a different P/E. If you go to Bloomberg, you get a third P/E.

So one of the things that we undertook at O’Shaughnessy Asset Management that my team, I still think hasn’t forgiven me for was, a multi-year project of cleaning data. And that is kind of the blue collar version that quants are kinda like, wait a minute, I gotta spend my entire day figuring out what Apple’s actual P/E is? It’s garbage in, garbage out.

And I still marvel. I read academic papers. I don’t read them as much as I used to since launching the new company, but they’re just riddled with errors. I could put somebody on staff at OSV to simply write up all of the errors that that we find on a daily basis. And, of course, now we have our own in-house AI because the commercially available stuff, a lot of it is very highly nerfed, and we want the unvarnished version.

So we’re also building out an AI suite at OSV. We’ll be able to dump, for example, basically an unlimited number of papers into it and then look for all the errors. And I’m not a conspiracy guy, so I don’t think these errors were done intentionally. I think they were just done like humans. I mean, if you wanna find the weakest link in any system, it’s us humans.

So it’s a great aspiration to write a book as magical as Gödel, Escher, Bach. What would your theme be in the book?

[Context: one of my goals in life is to write a book that will be as magical to someone else as GEB was for me.]

MM: I’m still figuring that out. But I think it will certainly have a mathematical, rigorous component. It won’t be all right brain for sure, but it will also not be a textbook that is all left brain. It will be a balance of both.

I feel very often the ideas that I find very elegant are mathematically elegant results. For example, in my honors calculus class, it’s a proof based class. You don’t actually do calculus. It’s about proving the fundamentals of calculus. And one of the things we proved is that the real numbers are unique.

And I know that Gödel, Escher, Bach talks a lot about isomorphism, and we studied isomorphism in a slightly different meaning of the word, which is that, anytime you have 2 sets, if you have a 1-to-1 correspondence between those sets, that is called a bijection. And if the bijection is order-preserving, then it’s called an isomorphism.

The point being that if you have an order-preserving bijection, then for all effective purposes, the sets are exactly the same with the elements renamed. So, basically, the sets are exactly the same, except that you have called the elements different things.

We proved that if you have any set that has certain properties, like countable dense subset and connected, it has to be isomorphic to the real numbers. So in that sense, the real numbers are unique. And I thought that was a very, very elegant result. Even in physics, we did the first equation of Maxwell, which I thought was very elegant.

Even a much simpler thing like e^(iπ) = -1.

JOS: Yeah. That’s always a fun one.

MM: I also think it should have some artistic or creative component. One thing I’ve been thinking about is ‘intellectual fiction’. I think Neal Stephenson is a prime example. I think Cryptonomicon is clearly a standout example. It’s a fictional book, but it literally has math theorems and graphs and all these fascinating things.

JOS: You might like 2 books that I just finished. One is called The MANIAC, which is a fictionalised biography of John von Neumann. And it’s really enjoyable. I came late to that book, and I loved it. And then his first book, he’s a Chilean author, so these are the only 2 that are translated into English, something we hope to address with our book publishing arm, Infinite Books.

We’re hoping to be able to, in short order, simultaneously publish a book in all languages. It would be the electronic version, obviously, because that’s where it’s easy. But, the other book that he wrote that I finished just last week is When We Cease to Understand the World, and again, it’s also fictional and it’s looking at the various rivalries and various ways of trying to understand quantum physics and mechanics.

And it’s just so alive because he can take all the liberties he wants because it’s fiction. And yet it maps back to the actual history of the origins of quantum physics and Heisenberg and Schrodinger and how much they hated each other. These rivalries are a bit like Newton and Leibnitz, and it’s interesting how much they really informed what was going on.

And so it reminded me very much of what happened to poor David Bohm, who did the implicate and explicate order. Anyway, he was a student of Oppenheimer’s, and it came down during the McCarthy era that he was a communist or had communist sympathies. And so the way that these geniuses like Oppenheimer dealt with David Bohm is very much like the movie Mean Girls. Literally there was a thing that Oppenheimer wrote and said to his colleagues saying, “If we cannot disprove Bohm, then we must agree to ignore him.”

And that always fascinated me because people often forget, that, at least for now, a lot of this stuff is also informed by our very human emotions, likes, dislikes, etc. They affect even the purest math. And When We Cease to Understand the World really makes that point beautifully. And again, you couldn’t do it if it wasn’t fiction.

MM: I’m also a big fan of Claude Shannon, by the way. Like, huge, huge fan. I find it fascinating how he did all these different things. He was at Bell Labs, information theory, invested in the stock market, and made a killing on Teledyne and Motorola and all these great companies. I think he compounded at north of 25% CAGR for a couple of decades.

JOS: I love Shannon, and it still mystifies me that he is not more famous. Because, like, almost everything that we use on a daily basis needed Shannon to figure it out.

The Teledyne story is a really interesting one. Henry Singleton was an absolute genius at issuing shares to use as a cheap currency to acquire other companies, but then having the intuitive sense to know when he should buy them back.

I would bring Teledyne up in groups of asset managers, and and I’m talking big asset managers, right? And if I was sitting with 10 people, like, 8 of them had never studied Teledyne. I was always shocked by that.

So fictional version, but, also including mathematical formulas, beauties, etc. Interesting. And what would be the art part that you would bring in?

MM: I’m trying to figure it out. So maybe language and writing and that sort of thing, maybe photography. Could be astrophotography.

JOS: Oh, very cool. My wife is a street photographer and has a published book, and we collect art, and so I would say probably 30% of our collection is photography, 70% oil, sculpture, that kind of stuff.

But I’ve always enjoyed doing the rounds of the various photography exhibits with my wife because I ask her, why is this photograph so appealing? What about it is inviting me to be interested in it and commanding my attention? And I always figure there’s got to be a mathematical formula underneath it, and that is not always the case, but it is often the case.

It’s like in literature, the well known hero’s journey. Joseph Campbell was the first to really outline the actual step-by-step process of the hero’s journey. One of the things we’re programming our AI lab to do, is to be able to put a manuscript in it and rate it against the various underlying story structures.

One of our fellows is an architect, but he’s building an architecture for written works. And we think that’s really a fruitful endeavor, in terms of looking for kind of the pattern underlying why are some stories absolutely devoured by all of humankind, and others are just ignored? And what you do see is the underlying story structures give you a much greater insight into if you’re in the business of being a publisher and, you know, it’s a for-profit company.

That might be a very helpful piece of software for us to have, in terms of deciding whether we’re gonna go with this one book or one author over another.

MM: That brings me to something I wanted to touch on. I’m a voracious reader, but also I helped 2 authors with their books, and my name is in the acknowledgments of both. They are both bestsellers.

One thing that I had observed since then and, I had written in my notes, thought a lot about this, is that, I think there is a huge opportunity for a publisher that really changes the model. When I saw the process of creating these books, and I saw the whole journey very closely, I found it very dissatisfactory and inefficient. Penguin gives the author 10%. And I just felt it is a very, very inefficient process.

I think there’s a huge scope for a publisher that replicates the brand and distribution of say Penguin, has a quality filter, and does 80% royalty to author, 20% to publisher. And I know you do something similar at 70% to author, 30% for Infinite Books, which is basically the same. And so I want to dig in more and get your insights on that.

JOS: Listening to what you just said, that’s basically the goal of Infinite Books. Having written 4 books myself, I was very dissatisfied with the publishers. I had a variety of publishers. I had McGraw Hill, Broadway, and then yet another publisher. And in every case, it didn’t matter. It literally didn’t matter who the publishing company was. They all treated me the same. All of the line edits that they made actually detracted from the idea I was trying to express rather than amplifying it.

And so, when we launched Infinite Books, it was as a redress to everything that I and our editor in chief, Jimmy Soni, had really despised about publishers. Many books in America are still decided by people who hear something at a Manhattan cocktail party. If you wrote a great book right now and you submit it to the big publishers of the United States, the probabilities are very, very high that they you’ll get a rejection letter.

Whereas at Infinite Books we’re building tools that build in all of our quality filters into AI. So rather than it ending up in what traditional publishers call a slush file, we put every manuscript through the AI with the quality filters and everything.

And let’s say we get a thousand manuscripts — at a traditional publisher, they end up in that slush pile, and they have sometimes junior editors or readers who are meant to go through all of the manuscripts or the outlines, but it’s an incredibly laborious process, and they don’t do it well. And so, if you submit it to Penguin and all of the big publishers, Simon and Schuster, you’d probably get rejected.

If your book was really good and you submitted that manuscript to Infinite Books, we’d probably publish it if it passed all of our quality filters and everything else. But what we’re also trying to do is build a publishing company that is incredibly author friendly.

I joke that the big publishers in America, are using best practices, but only from 1925, not from 2025. If you were sitting at the top of Simon and Schuster and you had a huge back catalog — by the way, that’s where they make a lot of their money — and you were taking 90% of it and the author was getting 10%, or if it’s a really well known author, 15% of the royalties — are you gonna try to innovate that and do a 70, 30 split? I don’t think you are.

And that’s the challenge that we see with all of the traditional publishers. Large organizations tend to be very static and kind of set in their ways. So even if Simon and Schuster said it was gonna do a Skunk Works, side project to to use all the tools that we’re using at Infinite Books, I suspect that it would take them years and years and years and it would be so watered down because of the multiple committees it would have to go through.

And so a business model that is free from all of the constraints of the legacy companies can often break through. We did it with OSAM, for example. We sold O’Shaughnessy Asset Management to Franklin Templeton in 2021 because we had created this idea of custom portfolio creation, custom indexing, custom everything. And classic example, they saw that customization was the future, but they couldn’t build it internally because of all of their legacy issues. There’s a lot of psychological reasons why it doesn’t happen, a lot of human OS reasons why it doesn’t happen, but it’s applicable to other industries as well.

So what happens is a much smaller entity that has the right vision, and they do not have the bottlenecks of committees and other vicious things that stop innovation often dead in its tracks — then the big company simply watches, looks at all of the players in the field, and tags the one that they think is the best and acquire them. And that’s what happened to us.

If you get a bid like we did from Franklin Templeton that was at a price to magic ratio, you end up taking it. And so that gives us a a double interesting thing with Infinite Books and the other verticals that we’re building at OSV is, first off, we're doing it because I love books. I love authors. I love readers.

And it is a category that is ripe for disintermediation, ripe for arbitraging all of the inefficiencies out of it. So in either case, we think we’re gonna have a lot of fun, but also give voice to authors who have traditionally been ignored for a variety of reasons by the publisher.

MM: You mentioned Jimmy Soni, by the way. And I have interviewed him and Rory Sutherland for my Substack.

JOS: What did you take away from your interview with Jimmy?

MM: He mentioned a few stories about PayPal, and, like, the simultaneous chess game. And he said that many of the best founders are, intense game players. He said Travis Kalanick of Uber was the world’s top Wii tennis player. And these people just go very intensely into whatever they’re into.

I think he has spoken about this model of writing upwards versus writing downwards. And writing downwards is you’re an expert in the field, you are kind of writing down to the reader from up there. Versus writing upwards is like, I am very curious about this topic, and I am no expert in it, but I am very curious to learn more. And on that journey of curiosity-driven learning, patching things together, and presenting it to the reader. And I think he does that very well.

JOS: And the reason I made him editor-in-chief and CEO of Infinite Books was that very reason, but also he’s probably the most entrepreneurial author I’ve ever met, in that he really loves business. He really loves business models, all of those types of things. So I think he’s gonna be great in that spot at Infinite Books. And Rory Sutherland is on my advisory council. What did you get from Rory?

MM: He’s a gem of a guy. He always has the most unique insights. He said you can you can use 2 dishwashers, and you don’t need any storage for plates. The level of creativity that he has is unbelievable.

JOS: Did he tell you the story about how he solved British Airways’ big problem with all the complaints they were getting when they would text their passengers that a flight was delayed?

So his company was brought in by British Airways, and they were gonna assign a fairly large mandate to his company to help them reduce the amount of irate customers, from the delays that they were getting on flights. And Rory said, can you show me what you send? Like, if BA flight 10 is delayed, will you show me what you send?

And they said, we send a text and an email saying, BA flight 10 is delayed. Right? And he goes, I can solve your problem for you right now.

And they’re like, what? And he goes, just add by how long, the estimate, how long it's gonna be delayed. And he goes, it isn’t the fact that the flight is delayed that annoys your passengers. It’s the uncertainty about what they can do with the time that they actually have to spend waiting for the flight to take off.

And he said, so if a flight is delayed 20 minutes, they’ll be, oh, it’s only delayed 20 minutes. That’s not a problem. But if a flight is delayed 2 hours, they’ll say, I guess I can have go and have lunch and maybe do a bit of shopping and then return to the gate. And so they tried it and the volume of their complaints just shrank dramatically.

And then another one, when I was chatting with him one time, he just kind of abruptly said to me, why is Uber so successful? And I gave the kind of standard asset management answers for that. And he’s like, no.

Because people can see exactly where the car is, they’re good. They’re fine. It doesn’t matter if it’s a 12 minute wait time or a 2 minute wait time. As long as they can look at that little graphic of the car coming to them, they’re absolutely fine.

Feedback and reading recommendations are invited at malhar.manek@gmail.com

MM: I want to begin by asking you about GEB and Douglas Hofstadter, how you first met him, etc.

SN: I was around 12 when I first met Douglas. Our family was visiting Stanford, where he lived. His father was the Nobel Prize winner, Robert Hofstadter. And I remember that Doug and my brother were all excited about mathematics and logic. And I was the younger kid looking up and not really understanding any of the words that they’re using, but sensing their excitement that this must be very deep stuff. But I hung around and listened because that’s what younger brothers do, I think.

And I remember we worked on some problem that they invented, and it escapes me now exactly what that problem was. But it was something to do with a recursive function, where we get the next term by looking at the previous terms. And it had some interesting properties which they were playing with.

And they were excited about this but I was just a young kid brother in the way. So, that was the first time we met, around 1960. Doug says something about that in his foreword to Gödel’s Proof.

And then we ran into him every once in a while, I guess, but then I saw him more when he was a graduate student at the University of Oregon. He was working on the problem that became what’s known as Hofstadter’s butterfly.

That was his thesis topic, as I recall. And then I ran into him, and he described some of that to me. I didn’t know that he was putting it all together in a book. So then a variety of these things all came together in GEB.

It was a lot about self-reference: Bach with his fugues, Escher with these fun drawings and Gödel with the proof, but tied in with these other kinds of questions which somehow referred to themselves.

MM: And I think he had some sort of Scientific American contest about the Prisoner’s Dilemma. He sent out a letter to 20 people and you had to explain whether you choose to cooperate or defect. He writes about this in Metamagical Themas,

Sidney Nagel was very displeased with his conclusion. He expressed great regret: “I actually couldn’t sleep last night because I was thinking about it. I wanted to be a cooperator, but I couldn’t find any way of justifying it. The way I figured it, what I do isn’t going to affect what anybody else does. I might as well consider that everything else is already fixed, in which case the best I can do for myself is to play a D (defect)”.

SN: Yeah, it’s a good problem. Why did I choose to defect? Well, if the job is to just do the best for yourself, and no one knows, then that’s the solution. The question is whether you believe everyone else could actually cooperate, but cooperation needs you to talk to people. You can’t cooperate alone. So it’s a dilemma. It shows you the world is a complicated one.

MM: On the first day of our class, you had mentioned that you were an English major, and you took this class on E&M. And, chapter 5 of Purcell’s E&M textbook compelled you to switch to physics. Can you recollect some of those memories and experiences, and how that happened?

SN: I started off as an English major, or that’s what I thought I was going to be, I was excited about literature. But then I wasn’t interested in what they said about the literature, and then I somehow also realized, oh, you have to be able to write well. So there were those things, and then I was looking for something else to be interested in.

And I took the honors physics class, which uses the same textbook of Purcell which we are using, albeit an earlier edition. It was a lovely book, and I hope people still find it lovely. The part that made it special for me was at chapter 5: the analysis of how the electric field lines transform under relativistic effects and what this looks like.

I liked that, but I didn’t understand it very well. You know, I struggled with it and I didn’t know how to study physics very well at the time. And it wasn’t as if I did great in the course, but I thought hey, maybe I'll take another physics class. And so I didn’t switch over immediately, but then I had another class on E&M, and the teacher was wonderful, he really inspired me.

MM: You also said that Faraday is your hero, could you say more about that?

SN: Part of it is that of all the laws we’ve come up with in E&M, Faraday’s Law is perhaps the prettiest. It just makes you go ‘Wow, nature does that. Wow, look at that. Isn’t that neat?’

And then you stick around physics, and we started looking at physics and dynamics of sandpiles. And lo and behold, Faraday had looked at that too in his own way. So he was just a very curious physicist whose mind was open to all the phenomena at the time, and it didn’t matter whether it was one subject or another. We hadn’t pigeonholed things at that time into this sub-field or that sub-field. Faraday looked at all these different things.

And then I also appreciated the fact that he was kind of a self-taught person. I don’t think he had a great mathematical education, but his way of dissecting a problem was in an experimental realm, which is beautiful.

But by looking and looking and looking very carefully he was able to come up with these deep concepts such as a field. And the reason we know about fields is Faraday.

MM: You also said once in class that the optical properties of an object, how it appears, relates to its thermal and electrical properties. Could you say more about that?

SN: [Holds two forks, one metal and one plastic]. What are these? They’re two forks. But I’m looking at them, and this one reflects back at me, and the other one I can see through.

Just by looking at it, I know something about it. If I stick [the metal fork] into an electric outlet, I will kill myself. If you have something that looks like this [plastic fork], if I stick it into an electric outlet, I will not kill myself.

So just by how this thing looks, I can tell the difference. So this tells me that the electrical properties are somehow related to the optical properties. And then you also know that if I stuck this into boiling lead, [with the plastic fork] I won’t burn myself, whereas with this [metal] one, I will.

Oh, so that means that the optical properties also affect the thermal process. So these things are all related. And when I was writing up my PhD thesis, I remember thinking about that, so I wrote that down in the introduction to my thesis.

MM: Your office is a beautiful collection of many different things, I’m curious if you have any hobbies outside of physics.

SN: Yes! I enjoy woodworking. I just think carpentry is beautiful, and it’s nice and it smells good, it feels good. So I was trying to build things. And when research isn’t going well and you have a hobby like that, then you think about that instead of your research, and it gives you an outlet for still trying to be creative.

Then as part of my fluid mechanics research I got very interested in photography. So I spent a lot of time trying to really make the images as special as I could because it’s not just the intellectual question of the science. We are not just scientists but also humans. And we humans respond to the science, of course, but also to the aesthetic qualities.

Then, unfortunately, the world went digital, and digital photography doesn’t appeal to me as. With the analog stuff you would go into your dark room and you get all these chemical smells on your fingers and I love that.

And then, during the pandemic, we were all stuck at home, and so I tried to write poems about physics. And again, like all my hobbies, that went on for a while and then stopped. And now I’m hoping that something else will get me excited.

In terms of writing, the one writer who I would like to emulate, if it were possible, is a biologist who was around in the 60s and 70s. His name is Lewis Thomas, and he wrote books and essays about biology. Many books are trying to teach you something. But what I liked about Lewis Thomas is that he simply helps you appreciate something.

MM: In your interactions with all these leading physicists, are there any anecdotes that stand out?

SN: Well, the anecdote that stands out to me is about Purcell, whose textbook we are using. I got to meet Purcell. He came to Chicago to give a talk once. I arranged a dinner with him and my research group.

By the way, I had a cat who I had named after Purcell. Purr-cell is a good name for a cat. I didn’t tell him that I named the cat after him, because I wasn’t sure whether he would be happy or offended. But the story he told us over dinner was very good.

So Purcell’s main research was on nuclear magnetic resonance (NMR). And then there’s another physicist who was actually my teacher, for the second E&M class I took, who is also a Nobel Prize winner, Polykarp Kusch. And Kusch was also an experimentalist, and he got the Nobel for precision measurements of the magnetic moment of the electron and muon. These were very high-precision experiments.

And one time Kusch went to see Purcell’s lab in Harvard. And there was a vacuum can in the middle with a bunch of wires. So Kusch was looking at this, and there were red globs on it. Back in those days vacuum technology wasn’t what it is today and there was a tiny leak in the can. We would use something called Glyptal, which was a kind of red paint that would somehow help clog up the leaks.

And Kusch was looking at this, and this was magnetic resonance, remember. And Kusch asks, I wonder why that stuff is red. And he asked, what would be red? Well, iron oxide would be red. Iron has strong magnetic properties.

So here he was doing these experiments on NMR, he’s painting iron stuff on top of it and he hadn’t thought about it. Purcell said “it never would have occurred to me to ask that question”, but Kusch asked the question because he’s interested in all these ways that experiments can go wrong. If you’re gonna do a precision experiment, you have to get all these little things right.

And this story shows that there are many different kinds of physicists. One can be really good at something, and another can be really good at something else, and they don’t have to be the same things. And there’s a role for each of them in physics.

And so Kusch was a great physicist doing precision measurements, Purcell was also great in doing a very different kind of physics. And neither of them could have done what the other one did.

Feedback and reading recommendations are invited at malhar.manek@gmail.com

Behind The Scenes

• When I read Jimmy Soni and Rob Goodman’s biography of Claude Shannon, A Mind At Play (my notes here), I was reminded of something I’d read in James Carse’s book Finite and Infinite Games.

  Here’s what I wrote in an email to the authors, quoting from Carse’s book: “…‘To be serious is to press for a specified conclusion. To be playful is to allow for possibility’. Given the formulation of information as the resolution of uncertainty (H = - p log p - q log q), it seems like being playful is being information-rich.”
  
  Rob Goodman very kindly wrote back: “That's a really insightful connection between the ideas of playfulness and information theory. It makes sense to me, and I wish we had thought to make that connection in the book!”

• I enjoyed A Mind At Play so much that I gifted a copy to the Managing Director of Morgan Stanley India.

• Fun fact: Jimmy Soni spent more years researching and writing the PayPal book than the time in which the PayPal story itself played out (from creation to the eBay sale)! It’s a testament to his meticulous research, which shines out when one reads his books.

MM: Let me start by asking you about chess: Claude Shannon played against Mikhail Botvinnik; Peter Thiel played 10 simultaneous chess games at the PayPal IPO (and is said to have been impressed by DeepMind founder Demis Hassabis’ chess skills); and your book The Founders segments the PayPal story as Sicilian Defense, Bad Bishop and Doubled Rooks. Can you talk me through the chess connection? [For interested readers, Raymond Smullyan’s book The Chess Mysteries of Sherlock Holmes is brilliant.]

JS: That's great. It's an observation that actually nobody has made before about the connection between the two books. You know, it's interesting, I hadn't even really thought about it until now. I don't think it's an accident that chess features in both books — one about Claude Shannon, who's an electrical engineer, a math professor and founder of information theory; he was an avid chess player, and was pretty talented. He wasn't competitive in the way that others were like at Bell Labs, but he definitely could hold his own. He wasn’t trying to be a chess virtuoso, but Shannon did build one of the world's first chess playing machines; I think it was actually the world's first chess playing computer. It's called Endgame. It was designed to play the last several moves in a game. He actually wrote an academic paper about chess; he also did some really breakthrough work in calculating the maximum possible combinations of moves in chess, things like that. So there's a real passion there.

And then on the PayPal side, it's a much bigger part of the story. Peter Thiel was exceptionally good at chess — at one point he was the top player under 13 in California, I think, something like that. I don't know the facts quite right there. But he was very talented.

And I will tell a couple stories. One is that actually one of the earliest investors in Confinity [which later merged with X.com to become PayPal], found Peter because of chess. Even while Peter was involved in the creation of Confinity, he was playing competitive chess on the weekends and going to tournaments. And there was a gentleman [Edward Bogas] who he had met through playing chess. When Confinity was looking to raise its friends and family round, he contacted this guy.

He's not someone that anybody would know, he's not anybody famous. He's not a venture capitalist. So I Googled his name and I found a phone number on one of those sites that lists four or five phone numbers, and hopefully one of them works. So I call this one number, that's a 650 area code. I call it and I said, “Hi, my name is Jimmy Soni, I'm working on this book about PayPal. I think you were an investor in PayPal very early when it was called Confinity. I don't know if you're the same person that's listed on this document. But I would love to talk to you if you are.”

And I'm thinking, this is gonna go nowhere, right? And I get a call back right away and this person said, I am the same person and I did invest. And I'll tell you a story. The story goes like this.

So, he had played Peter a few times. And Peter shows up at this guy's house and he says, “I'm building this company and I'd love to have you invest.” And the guy goes to the back, writes a check for a decent amount of money and hands it to Peter right away.

And I asked this guy, I said, “Wait a minute, this doesn't make any sense. You don't really know Peter — you know him as a chess player. How did you know that this was going to work out?” [Because the investment worked out very well for him.] And he said, “When I sat across from Peter on the chessboard, his chess style was merciless. And I knew that anything that Peter was going to do, he was going to be out to win.”

And so it's interesting because that quality does emerge throughout the story. The folks who were at PayPal weren't just competitive, they had a will to win. I think there's a difference between wanting to do something and wanting to win. It is the difference between players like Kobe Bryant and Michael Jordan and other players who play basketball, there is a desire to win that supersedes their competition. So chess is the place where that came out.

Now, the other interesting thing about PayPal is that it had a lot of talented chess players. Max Levchin (co-founder) and David Sacks (early employee) were both very talented chess players. So you had actually a number of people who were very good at chess. And in the book, one of the things I explore is this culture of game playing: chess, strategy games, poker.

The book has three parts, and I named each part after a different sequence in chess. So one is called Sicilian Defense. And there are two others [Bad Bishop and Doubled Rooks]. At the time, I thought it would be very clever if the sequence in chess was something of a metaphor to the the situation that they were in. And so one of them is called Doubled Rooks, and I think there's something around two power players in that section of the book.

Sicilian Defense, if I remember correctly, is something black can do to respond to a move that white makes, that will give it the advantage. But it's always the case that black is the underdog. And throughout much of its early history, the early iterations of PayPal, they felt like underdogs; they felt like they were never going to win, they felt like they were going up against much better opponents. And so I just had these ideas. And I felt like it was a little bit more clever than calling those sections part one, part two, part three. And I put in the chessboard diagrams because I thought they were visually appealing.

All of this is also tied into the fact that when I was a kid, I played chess. I was not competitive, I was nowhere near as good as the people who were a part of PayPal. But I loved chess growing up, I played with my dad. I have an eight-year-old daughter now, I've taught her how to play chess, we play chess regularly. And so chess is a part of my life. It was a part of the lives of the people that I studied.

And there's this joke that all biography is actually just autobiography. And so now that I'm thinking about it out loud, I'm like, of course, I ended up writing about this. I wasn't writing about baseball players, because I didn't play baseball growing up, I played chess. So it's sort of both very funny and then it's also very much a part of the substance of these cultures, and I don't think that's accidental.

One thing I'll add to this is that, in addition to chess, I’ve learnt that there's a high degree of overlap in video game culture and tech culture. And maybe that seems obvious to people. But I would go one step further. I heard this story, I think it was on the Tim Ferriss podcast about how Travis Kalanick, one of the co-founders of Uber, is actually one of the world's greatest Wii Tennis players or was at one point (“Travis Kalanick was ranked No. 2 in the world at Wii Tennis” — source).

One of the things that I found was that Elon Musk was an avid StarCraft player. And I would say that there's a difference between people who are casual gamers and people were very competitive video game players. And my understanding through both my own research and through other interviews and other things, is that Elon is a very competitive video game player.

And there was actually a culture at PayPal also of video games being played at the end of a work day. One of the things that happens in these cultures is there's a thin line between work and fun. So if you're at the office on a Saturday, you might be working, but you might also be playing video games for six or seven hours. And I think that's unique to this kind of culture.

Now granted, this is all again, with the caveat of, I played video games growing up — I grew up playing StarCraft, I grew up playing Doom and Duke Nukem and all these video games. So maybe I was just looking to write stories about this stuff.

MM: You gave a Ted Talk at Duke, where you spoke about learning history in terms of people and stories rather than facts to memorise — and your books exemplify this. Can you talk me through your own stories and writing processes?

JS: There's this curious thing in the lives of writers — people think writers’ lives are more glamorous than they actually are. And particularly if you write books — this is less true of just general writing on the internet or blogging — people have a tendency to think that if you write books, your life is very glamorous.

And maybe there was an era in which that was true, right? Maybe people think all writers’ lives are like F Scott Fitzgerald or Ernest Hemingway. Actually, a lot of writing is not glamorous at all. It is sitting in a room alone for many hours, just trying to take an idea or a thing that you're thinking about and bring it to life on the page.

Think about the scene: I'm in my apartment, working alone with big stacks of paper, it's not that interesting. It's not like I'm fighting in a war or leading a revolution or discovering some breakthrough in science. I am sitting by myself typing, trying to finesse words on a page.

And there's this line and I don't remember who said it, but someone once said that, the more interesting the life, the worse the writing. (For pure writer types.) And I think there's actually a lot of truth to that. Because what happens is good writing comes out of — at least for me, and for many of the writers I admire — it comes out of just daily improvement.

I noticed that in your background, you have Robert Caro’s book — and The Power Broker is a masterwork. But if you were to look at the scenes throughout his life, apart from the interviewing, most of Robert Caro’s life was just getting up every day, putting a suit on, going to work in his typewriter, and doing the same thing over and over and over again.

I think that my life has been interesting: I've had the opportunity to sit with some of the most prominent people in the world and get the chance to tell their stories. But I don't want people to think that the writing life itself is glamorous. And I would say that if you're looking for glamour, like, there's way better places to find it.

People think that what I do is go to cocktail parties — and it's actually very much the opposite. I generally don't socialize — especially when I'm in a true book-writing mode. I don't want to be distracted, I say no to a lot of invitations.

This is true of a lot of authors who are working on long form projects; the project just becomes a part of you, and it takes over every other part of your life. It's hard to do anything else. I suppose there's something interesting about the process of flying all over and meeting these people, but the truth is that a lot of writing is just boring, basic, “get up and do the work” kind of stuff.

MM: In your interviews with these subjects, of course you have done Caro-esque research but how do you go about formulating questions? Do you think about specific word choices and the phrasing and framing of the questions you ask? Are there examples of actual questions you asked that you could share?

JS: Yeah, these are great questions. The answer is yes. I think about all of that stuff a lot. One of the biggest things for me is that — and what works and why I think my books might be a little different than what's out there — is that I try to ask interesting questions. And I spend a lot of time on the front end, figuring out what a given subject has already said about a topic. And then I eliminate most of those questions from the list that I want to ask.

And so that seems very obvious. But here's why it's important. If you're someone important, or interesting, or famous or wealthy, you're often in the position of being asked the same questions over and over again. So if you're David Sacks, or Peter Thiel, or Elon Musk — you just get asked the same boring questions over and over again. So if I, as an interviewer, come in, and I ask the same boring questions, then they will be bored. Or they'll just give me the answers they're accustomed to giving — they become rehearsed.

If you're writing a book, that's the opposite of what you want as an author, because you actually want new material. I want Elon Musk to be answering questions that he has never answered before. I want to ask him something where he says to himself, “oh, never thought about that.”

I'll give you an example. One of the questions I asked Elon — this is at the beginning of one of our interviews — I said to him,

“I'm gonna go in a little bit of a different direction here. We've talked a lot about PayPal. One of the areas I'm writing about is the death of your best friend who died at the age of 54 or 55 (correction: 51) — around the age that you are now — and his name is Greg Kouri. My impression from what I've read and learned is that Greg was a really important figure in your life, in the transition from your life as a student to your life as a technologist. And I was wondering if you could say something about him and about the impact he had on your life.”

— A question that Jimmy Soni asked Elon Musk

And I don't think Elon had heard the name Greg Kouri in many years at that point, or at least a little while, because he's not a well known figure. And there was a notable pause. And Elon gave me this line that's in the PayPal book, he said, “Greg was a trickster with a heart of gold.” And then he went on to explain what he meant by that. And it was just one of the best moments in the book, because I had not seen that kind of description, or that kind of reflection anywhere else.

So I knew okay, this is new. I'm trying to add something that isn't there, but the only way to do that is for me to obsessively read everything I could about Greg Kouri. I found Greg Kouri’s widow, I interviewed her for two hours, I did all of this background research before I ever asked Elon that question, because I knew that this is going to be an area where he hasn't spoken about this.

If you were to go to Elon and say, “What's it like to build a startup?” He’d say, go watch my other 25 interviews on YouTube where I've answered that question a million times. But that's the trick. The trick is, ask the question where the subject says, “Hmm, that's interesting.” That's one trick.

The second trick I have found is, one of my principles is you want to look where other people aren't looking. And so if you are interviewing somebody — let's say they're a successful technology investor — if you ask them about technology, or about investment, you're actually likely to not get that interesting stuff. But if you ask them about college, and about their education, you might get more interesting stuff.

And so from my perspective, yes, it was useful to ask Max Levchin about the process of PayPal and how he made it more secure and how he built a technology team; we got to all that later in my interviews. But where I started was with his story of immigrating to the United States. And he's done some interviews on that, so there was a little bit of overlap, it wasn't completely new information. But I went deeper and I was able to ask him about his grandmother, his parents, physics and the University of Illinois Urbana-Champaign.

I try to find a way to tap into the thing that someone wants to reflect on, but they have never really been asked about it, or they don't get asked about it that often. It's a hard thing to pull off. And there isn’t an exact science to it. But I find that if you can do that, what you do then is also you build a rapport with a person.

One of the things I tried to do with all my books is I try to bring my characters to life. I don't want them to just be two-dimensional people giving the same answers. I want them to be figures. I can say that you're 5 feet 7 inches tall and 150 pounds — or I can tell a story about your family. The second is a better way of writing. And so for me, the way to do that is to look where other people are not looking.

But the other last thing I'll say — and I could go on this topic forever, interviewing is one of my favorite things to do and also one of the hardest things to do — one of the other things I do in an interview is I try to make it fun. I try to make it enjoyable. I'm not there interrogating you as though you were in a courtroom.

I try to make sure that I really listen to what people say. The most important people in my stories will just cancel the meeting if I'm boring to them, or if I'm uninteresting, or the interviews are not fun. And so you're trying to balance this. It can't just be all humour and good times, it has to be substantive, there have to be some questions that are hard. At the same time, if you are an interviewer with whom they don't actually have fun, then for the kinds of projects that I do, that's a death sentence: you're never gonna get called back, you're never going to be able to get them to respond to you again.

I always tried to make sure that I was asking about things that would be enjoyable to reflect on, in addition to asking about things that would be uncomfortable, or where I really needed substance. I heard from my subjects after the book came out, they said, “one of the reasons that we really liked talking to you or one of the reasons we kept talking to you is because it was actually fun.”

And I would say that I tried hard to make it fun. And that's not at the expense of being hard hitting; it's not at the expense of being serious. It's just that there's so many interviewers I've met, where they go so deep into the facts, they forget that they're talking to a person and not an AI Chatbot.

MM: You’ve spoken about volume builds quality — maybe there are parallels to this in how PayPal tracked their number of users in the ‘World Domination Index’? Could you elaborate on how that loop from volume to quality occurs?

JS: Yeah, it's a good question. This is especially true for writers. where the more you write, the better you get, the more you read, the more adept you are at reading. In those two disciplines, reading and writing, it is like, the more you do it, the more capable you will be. I've yet to meet the person who who violates that equation, or who can short circuit it.

I think that there are many domains where we have been fed the fiction that you can short circuit the outcome without putting in the reps; that you want to get somewhere and you can skip the 10 years, you can skip the 5 years.

But most of the things and people we admire are a by-product of compounding. It's rare that there are overnight successes. You look at any person that's in the headline today. And if you hear something about them you admire, it's highly likely that they did not just get lucky — they did something over a long period of time.

In the domain of music, if you look at somebody like Taylor Swift, it is easy to talk about how she is a billionaire today and how the eras tour is, I think, the highest grossing world tour of all time. People forget that she started doing music when she was like 14 years old or 13 years old. Her process, her desire started at a very young age. If you look at the most successful people in sports, the most successful coaches, the most successful technologists, they all started early and put in a lot of struggles and failures.

Max Levchin had three failed companies and one very modest exit before he started his fifth company, PayPal. Elon Musk started a college business, then had another business (Zip2) and then started PayPal.

Volume teaches you all of these lessons. It's not popular to say because everybody wants a quick way to get the thing they want. I get it. If you see somebody on Wall Street Bets making a million dollars overnight, or you see somebody making $10 million in crypto in a month, you think to yourself, “how do I do that?” I get it. But there are some games where patience, compounding and volume are the only things that matter.

And I would say that is true with writing in particular. The more words you write, the better you get. My third book is better than my first, my fifth book will be better than my third, and so on and so forth. I'm just becoming more talented and knowing what resonates with readers.

In terms of feedback loops, I build feedback loops in by either hiring editors, so I'll just pay people to read my work and then give me feedback. Or you have editors that you work with the publishing houses. And then the internet is the ultimate feedback loop. Especially with quick writing, you put something out, and you can get feedback right away. That's important.

But I think people tend to think that the feedback is more important than the volume. Because the volume is the not fun part. The not-fun part is when, for example, during the middle of The Founders project, I set myself a deadline. My goal was, I need to do 1000 words every day. And I did that for 100 days. I just had to do it. And I did it.

And the volume built a project. But I also thought that by the hundredth day, it was easier to get 1000 words out than it was on the first day. And it was horrible at times; there are times when I would wake up and it's the last thing I wanted to do. I mean, it was just truly horrible.

But I think that success leaves clues. And one of the clues of successful people is whatever they're successful at, they generally just do a lot of it. And the more you do of it, the better you get. It's not a sure thing; accidents and chance can happen, both good chances and bad chances — you can have two people who are identical, and one gets the lucky break and the other doesn't. There are all these other variables, but the truth is, part of the success of most things is volume — just doing more than other people do.

What makes Mr. Beast’s YouTube videos so successful? Well, he started when he was 16 years old, he was doing it for five or six hours a day, for 30 years. You do that and you'll be similarly successful. But I think it's hard for the human brain to operate on such a long time horizon. And so that's why people end up just shrinking the volume, and they go for the quick win.

Feedback and reading recommendations are invited at malhar.manek@gmail.com

Behind The Scenes

• We have a common passion for postal mail — he appreciates the costly signalling involved (both money and time) in sending a letter; I, on the other hand, have been a stamp-collecter (philatelist) and post-office-buff all my life!

• We both enjoyed Scott McCloud’s amazing book, Understanding Comics. In Rory’s words, “I picked it up thinking it was kind of silly, but by the end - and I read it on one sitting - I thought it was one of the best books ever written - or indeed drawn!”

• I first read Alchemy 2 or 3 years back, and it changed the way I think about perception v/s reality. If a courier service’s 3-day delivery rate is 95%, but customer surveys show that it is perceived to be 70%, greater value can be created by improving perception (taking 70 to 95) than by enhancing reality (taking 95 to 100). In this case, the bottleneck/constraint is not reality but perception. (Theory of Constraints is among the most elegant and important ideas ever, in my opinion.) The parallels with George Soros’ theory of reflexivity are intriguing: improve perception to boost valuations, use high prices to raise equity capital at low dilution, then use this capital to improve reality. Another gem of an insight: cyclists tend to be rude since they pay the price for losing momentum. I can go on and on endlessly, raving about Rory’s genius ideas on placebos, self-fulfilling prophecies and beyond (he is incredibly well-read, as you will find out), but I’d recommend his book Alchemy for more.

• He called Mumbai “a fantastic city, amazing city to be in” — music to my ears!

And now, dear friends, here’s Rory Sutherland.

RS: One of the really clever ideas that Eric [Yuan] who founded Zoom had, was he realised that most people attempting to develop video conferencing were focused on the video quality, but actually it’s the sound quality that matters. The human brain can cope with a large amount of pixelation and it can cope with people freezing, but it can’t cope with the voice breaking up. And there are really interesting reasons for this.

By the way, there’s a book I really recommend you read if you’re interested in this field, which is a book by a British guy called Andy Clark, called The Experience Machine. He has this theory that most of what we perceive is really a prediction — that the brain generates a prediction of what we expect to see, and then we use our eyes, nose, ears etc. to correct, effectively, for prediction error.

It’s very similar to how a JPEG works — so a JPEG achieves data compression by having an expected value for each pixel and it uses data to describe where the value differs from the expected value. And obviously it uses much less bandwidth and much less data to actually work that way.

So the optic nerve — the 6 megapixel or whatever it is that passes through the optic nerve — can concentrate on updating prediction error and alerting us to the unexpected, rather than being wasted effectively generating things that we were expecting to see anyway. It’s a really, really interesting mechanism. It’s an old theory actually, people like von Helmholtz in the 19th century came up with this theory, and what Andy Clark has done in The Experience Machine is he’s taken it forward. He’s taken it further with the benefits of modern neuroscience as well.

[MM: I had sent these ideas to RS before our call: “I remember you saying that the percentage of electric car owners who switch back to petrol cars, is very low. This metric seems to be key to analysing new, disruptive technologies. In reading Richard Dawkins, I found a compelling parallel. The Levinthal paradox in protein folding says that the number of possible paths of a protein is too vast and that sampling all of them would take longer than the age of the universe! Mr Dawkins resolves this using the following analogy: a monkey pounding away at a typewriter will probably take exceedingly long to print a sensible, meaningful English phrase like ‘The Eleven Rules of Alchemy’. But if a correctly-chosen letter/word is fixed in place (akin to an EV owner not switching back), this time reduces dramatically as the search tree is pruned.”]

RS: But I thought your Darwin point is very interesting, which is that it is impossible for monkeys to generate Shakespeare, except if there’s a mechanism where you stick with something that works, and you experiment with something you’re not sure about.

So, it’s very interesting that if you look at evolution, most animals are basically symmetrical, so it wouldn’t make sense to experiment heavily — although there are weird cases, so the exception to that would be the flounder, which is a flat fish that lives on the bottom of the ocean and so the eye that used to be on the bottom has migrated round to the top. So in that case there's this weird fish that would've started swimming vertically with an eye on each side, and then it effectively evolved its behaviour so that it lay flat on the bottom of the sea where it was very well camouflaged. But that's a really unusual case of an animal actually developing asymmetry.

I imagine Nassim Taleb’s point that we have 2 lungs, 2 kidneys, partly for the purposes of redundancy really. The odds of damaging your eye — I just had a retinal detachment actually, which until 1930 couldn't be treated, so if I'd been a one-eyed person before 1930, I would be completely blind for the rest of my life. But there are two happy things: one is they can now treat it, and the treatment is affordable which I imagine it wasn't in the 1930s; and secondly, because I had a second eye I was able to function fairly well. It strikes me there's a reason we have 2 of things.

The other reason is that symmetry is highly desirable for movement to the point where Robert Trivers actually investigates the correlation between joint symmetry and atheltic performance in Jamaican atheletes. There’s this correlation between how symmetrical your joints are and how fast you can run.

But your point’s very very good, it’s a pity that more people in marketing haven't read enough evolutionary science.

MM: Have you thought about how that relates to redundancy in language?

RS: So there’s a wonderful thing called the check digit, isn’t there? The way a check digit works in a credit card is that if you type 1 wrong number, or if you type 2 numbers the wrong way around, it immediately says there’s a fault here. There's always the potential that you could type 2 wrong digits and still get the check digit working but the most common errors which are transposition of two digits, or one mistyped digit, or one digit typed too often — the check digit is the last digit which is a product of the previous 15, and it effectively says “this one's okay”.

So you need to have a certain amount of redundancy. Obviously, if you had hugely efficient language, then you could have comical misunderstandings all the time. I always think it’s laughably stupid that in American English they produce “can” and “can’t” almost the same way.

In German, although it is “eins, zwei, drei” for “1, 2, 3”, if you’re reading your phone number over the phone, you’d actually say “zwuo”, not “zwei”, for “2”. There are a few other things in German where there is a very useful other word to avoid misunderstandings.

It’s interesting that there has to be redundancy in ordinary, oral communication; its different in instantaneous digital communication where you have a return channel. The other guy was Claude Shannon, wasn’t it, who did a lot of work on this?

MM: Exactly. His information theory equation involved an expression for information in bits as a function of logarithms of probabilities. [Claude Shannon is also one of my idols, along with Richard Feynman. The Shannon biography by Jimmy Soni and Rob Goodman, titled A Mind At Play, is brilliant.]

RS: I think one of the best things that has happened in the last 5 years is what we're doing now, which is that Zoom has replaced email, partly because email writing doesn't make sense for one-to-one communication. It makes sense for text messages like "are you there yet?" but for anything longer it doesn't make sense because we can speak much faster than we can type. But of course, if you're writing for an audience of 20,000 people, that obviously has time advantages.

The other thing is we can read a lot faster than we can listen. So there’s a benefit to the recipient in receiving email, but to the sender it’s a big pain.

RS: By the way, before I say anything else, I’ve got to say a big thank you to India. I was talking to my publisher about 2 weeks ago and India is one of the countries where it’s had really great success. It’s very difficult to get bookshop sales in the UK, it’s sold well in the US and continues to sell well on Amazon and in Kindle form, but India is the big bookshop success. I noticed when I was last in India, there was a copy on sale at Hyderabad’s airport bookshop.

There’s a problem in Britain, which is that the main high-street bookshops have very little storage space. And so all they’re really interested in is taking a Dan Brown blockbuster or a J. K. Rowling, filling the bookshop with 27 crates of  J. K. Rowling, and they’re not interested in anything that isn’t in the top 3% of popular sales, not necessarily in volume of sales over time — my book sells quite well over time — but B2B books don’t sell instantaneously in the way that novels do. So I’ve got a big thank you to the Indian bookselling market for doing such a great job of selling Alchemy.

RS: But you mentioned asymmetry, and if you look at the way emails breed, so you look at it from the point of view of, dare I say, eugenics. Unimportant emails tend to go to a lot of people, and important emails tend to go to just you. So if people hit “reply to all”, you have this fundamental problem that unimportant events fill your inbox and the important stuff gets buried.

There’s also a really interesting question about email, if you look at it from a systems thinking point of view — I’ve always wondered about this. A really interesting point of view is that actually, the reason we have a calendar and we have email, is they’re considered to be separate things, because they were historically: you had post, and you had a diary, which was a physical device. So computer designers, when they designed the interface, had a calendar and they had a communications vehicle. Actually, email and diary should be the same thing, by which I mean, it should say “go to this meeting OR read this email”. So really you should have a calendar which says these are the things you need to read, people you need to meet and places you need to go — they should be the same software. I’ve had crazy situations where I’m reading emails all about my flight and meanwhile there’s an email saying the meeting’s cancelled. 

The other thing is, notifications on the mobile phone have become useless because there’s no costly signal. So my phone goes “ding” whether my daughter’s in a car accident or there’s a special offer of “save 5%”. Now, okay, I could go through the various apps, but if there were a costly signalling mechanism, you could say “this is really important, I’m going to pay 10 pence and have the phone make a huge alarm noise instead of going ‘ding’”. To some extent, the physical post had that — you could tell from the stamp (first class, second class) and handwritten envelope.

When you don’t have a costly signalling mechanism in communication, it basically devolves to meaninglessness. What’s happened is that the burden of prioritisation in communication now falls on the recipient, not on the sender. [MM: Game theory provides another fascinating perspective on this. Consider the TCP backoff game, for example, where the Nash equilibrium results in self-interested agents flooding the channel with information, leading to a locally-optimal but globally-suboptimal outcome.]

The lack of 'skin in the game' for the sender is a problem we need to solve. You could solve it quite easily, though not infallibly, by simply requiring senders to mark their emails with a level of priority. Of course, someone could mark everything high priority, but people who did that regularly would get blocked and unsubscribed. And most people know that would be a bad thing to do and they wouldn't do it because there'd be social sanction.

Currently, email has neither a costly signalling mechanism (pay x price to indicate high priority) nor social sanction.

Shannon was looking at the transmission of information. But there is a secondary dimension of 'to what extent should you trust the information?', which isn't included in the Shannon model — you're assuming the sender is accurate in who they say they are.

Because email was designed by people who were Shannonites aiming for maximum efficiency and accuracy of transmission, they didn't spend enough time thinking about how you reliably signal importance and priority, and in what mood you should reply to something. I get tons of Christmas cards, you probably have Diwali cards, but it would be funny if you designed an envelope to look like a bill but inside was a Christmas card.

RS: By the way I was once studying how Amazon became so successful in India and I got into a rabbit hole about the Indian postal service, its history etc. The key insight Amazon had in India was, “we don't need to set up our own delivery infrastructure, it already exists” — and that worked like magic.

By the way, there were some countries that decolonised who said, “we'll get rid of all that [colonial] stuff, we'll start afresh”. And where India was absolutely brilliant was you said “we'll keep whatever works and we'll get rid of the other stuff”. What fascinates me is you're a democracy of 1.5 billion people that basically works. It's not top-down design, it's all bottom-up. For example, that wonderful system you have in Mumbai where your food is delivered to work at lunchtime: it's all bottom-up. That's why I've been saying for the past 10 to 20 years that the prospects in India for economic growth and innovation are much greater than China. The Chinese form is artificial and top-down, India's form is evolutionary and bottom-up.

I used to get really annoyed by British and American businessmen who went on and on about China. Of course, they liked China because you only had to talk to one person; you found the boss guy, you talked to him and that was it, everything was decided. Of course in India it's brilliantly messy and experimental and bottom-up.

There was this guy called Field Marshal Cariappa, who made the decisive decision that the Indian Army would never get involved with politics. In China, the Red Army is not the army of the country China, but the army of the Communist Party, which is a freakishly weird thing.

RS: In evolutionary science, there are a lot of butterfly effects (chaos theory, butterfly flapping its wings leads to a hurricane in another). I'm very interested in evolution; for example it's a random decision whether an animal evolved a convex eye or a concave eye. Now, it turns out that is really decisive, because you can evolve from a concave eye to a lens, and then the whole thing of the iris, light control, magnification and focus. But a convex eye, which insects have, you can't do it. If you wanted a fly to have the vision quality of a dog, its eye would have to be 3 feet across.

Physics is a science of how things are, a science of things that don't change. And evolutionary biology is a science of how things do change. And I think it's really, really important to have both.

In evolutionary biology circles, you can't mention Steven J. Gould because he started to believe in group selection, as did E. O. Wilson and Darwin, and as do I. But in current evolutionary science, that's a no-no, it's all happening at the level of the individual and the gene. But of course, evolution can happen at the level of a group.

[MM: Here’s an extract on this idea from Neal Stephenson’s Cryptonomicon — “From an evolution standpoint, what was the point of having people around who were not inclined to have offspring? There must be some good, and fairly subtle, reason for it. The only thing he could work out was that it was groups of people—societies—rather than individual creatures, who were now trying to out-reproduce and/or kill each other, and that, in a society, there was plenty of room for someone who didn’t have kids as long as he was up to something useful.”]

I would even go further to say that humans have been partly selected for by dogs. [MM: Kind of like “the wand chooses the wizard” from Harry Potter.] This is a really weird theory, but at the absolute low point of the Ice Age, the 30,000-or-so Homo sapiens who survived were probably those who knew how to domesticate dogs. Now that means that dogs would disproportionately domesticate humans who were kind to them…

[MM: We ended our call with my reading out an extract from Ashlee Vance's biography, Elon Musk: “The engineers were constantly baffled by what Musk would fund and what he wouldn’t. Back at headquarters, someone would ask to buy a $200,000 machine or a pricey part that they deemed essential to Falcon 1’s success, and Musk would deny the request. And yet he was totally comfortable paying a similar amount to put a shiny surface on the factory floor to make it look nice.” If this intrigues you, I can think of no better book to read than Rory Sutherland's Alchemy!]

Feedback and reading recommendations are invited at malhar.manek@gmail.com