Optimizing Your Inputs

https://www.joincolossus.com/episodes/85214124/srinivasan-optimizing-your-inputs?tab=transcript

Released July 6th, 2021

Balaji Srinivasan is a serial entrepreneur and angel investor. We cover ways to optimize your information diet, how technology is driving decentralisation, and what that could mean for countries, corporations, and individuals.

Introduction

Patrick: [00:02:49] My guess today is Balaji Srinivasan, a serial entrepreneur, and angel investor. Balaji is known to challenge conventional wisdom, and he lives up to his reputation in this conversation. We discuss a wide variety of topics, including advancements in health tracking, ways to evaluate your own information diet, and how technology is driving decentralization, and what that could mean for countries, corporations, and individuals. Before we transition to the episode, if you enjoy this conversation with Balaji, I'd recommend the Ethereum episode on our newest show Business Breakdowns. You can find that episode and more on your favorite podcast player, or at joincolossus.com. Now please enjoy my conversation with Balaji Srinivasan.

Evaluating and Optimizing Your Information Diet

Patrick: [00:03:30] So Balaji, I think the best place to start is with three interesting concepts that you've been thinking about more recently, the first of which is information diet. I love this concept. I devote my life to this concept as well. What goes in determines a lot of what comes out. Why did you get interested in this? What does information diet mean to you? What are you doing about it?

Balaji: [00:03:50] For a long time I've been interested in genomics, and metabolism, and things like that. I did my graduate work in the space, and I taught bioinformatics at Stanford and I started genomics company. And so I've been interested in health for a long time. And there's an interesting study many years ago and there's been follow-ups on it since, the saying, you are what you eat. I mean, you don't really think about this, but when you eat something it's literally incorporated into your body. And if you were smart about it, you might be able to track and say, okay, when you eat this piece of chicken, or this lettuce, or what have you, where are those lipids, those carbohydrates, those proteins, et cetera, where are those incorporated exactly? Does it go to your eye? Is it just all diffused? Does it gravitate towards particular things or what have you? And so there's interesting studies that have been done on exactly this where they're trying to track where it ends up deposit. Obviously easier to do in animals than it is in humans. You literally are what you eat in the sense of your body is reconstructed by this stuff. And that intersects obviously with genetics and intersects with what we know about biochemistry, and metabolism, and pharmacogenomics, and nutrigenomics, which is the response of body, you might be a caffeine fast or slow metabolizer, alcohol fast or slow metabolizer. Okay. All those things. So I'm interested in that for a long time. And more recently, we started to get better-quantified cell stuff. Some people have said, oh, IoT, that's a bust or whatever. It's not really, because it's something which it actually is working, it's just working in the domain of health. And I mean, actually a lot of home devices now, by the way, are IOT, ovens, and stuff like that. IoT coffee things that refill your coffee. There's a thing called endless or something like that-

Balaji: [00:05:27] Yeah. It's funny because it's just relatively few things that are purchased as much as that. And anything you can take off your plate is often worth it, just the attention aspect for at least some people. It seems like a gimmick, but then it's also like, okay, why do we have an LED on the screen of something? At a certain point, the conversation becomes basically free, ten percent becomes free. And so IoT stuff is happening. Okay. So recursing back up. So the fitness stuff, there's the WHOOP device, there's an Oura Ring, there's a Fitbit, there's Apple Watch. And these things have really become a big thing, especially in the wake of pandemic, because some of these things are actually monitoring devices that studies are showing that you can get early warning on things. We're starting to cook them with more and more stuff. In fact, this would have happened 5 or 10 years ago, the Apple Watch was actually nerfed by the FDA, it would have had things like blood glucose meters and so on. We've been able to do this non-invasively. I've seen demos of this, where back in 2013, 2014, there was a company, I think Sano Intelligence and one of their lead people was recruited by Apple, actually. And they had a device where I could wear it, and it basically felt like Velcro. And the thing is it wasn't pricking the skin, you weren't seeing blood spread out like in a blood test. Okay. But essentially what it was doing was it had something which essentially was getting into the skin, enough under the sort of lipid bi-layer there that you're able to actually do noninvasive monitoring of stuff. We start to know what's going on in our bodies. We know what's going on in Bangalore or Budapest, been on our bodies. And we have information on the other side of the world, but not something that is half a foot away. We don't know what's going on subcutaneously. It's actually this amazing blank spot on the map. It's like back in the 1990s, there's this concept of getting lost. You don't get lost anymore, you've got GPS. Now you never don't know your location. In the future, you will never not know your infection. This is one of the few good things about the whole COVID thing is enormous effort is being put into various kinds of non-invasive diagnostics, and monitoring, and all that stuff. So now what you can start doing is you can start seeing cause and effect for your diet. You can start eating something and you see it spiking your levels health, or your FreeStyle Libre or some other device like this that you're wearing that's monitoring your blood glucose.

And you can see that your resting heart rate spiked, you can see when you've got a good night of sleep or not. You kind of know this stuff kind of intuitively, but tracking it over time is actually very valuable. You hit what you measure. What are we optimizing today? Twitter likes. That sucks. It's funny, people say that we have nothing in common around the world. Well, every politician in the world, every celebrity is optimizing their Twitter likes, or their followers, or what have you. I think for those of us who have something, like math or something else like that that's a primary metric, this is not your all-encompassing metric, it's maybe top 10 or something like that, but it's certainly not number one. But for those people for whom it is number one, this is just this mind-melting kind of thing, because you're just optimizing popularity, which is benefiting Twitter, it's just rows in their database. It's not benefiting you that much. There's some benefit out of it, but not that much, especially on the anger and rage that comes out of getting attention sometimes. So just talked about diet, talked about the cause and effect of diet when you've got this IoT stuff. Now let's apply that stuff to information diet. Currently, there's that saying, if it bleeds, it leads. You can expand it. If it enrages, it engages. Legacy media and social media profit from fight club. Two people just go at it over some stupid polarizing article. And they might agree on 85, 90% of other things, they might live in the same community, they might know each other from work, or church, or whatever, but they just go at it over this thing, which frankly, neither of them would have even thought to bring up, but it's optimized for being polarizing. It slates our codex calls like a scissor statement.

So scissor statement, which is something which is obviously true to one party and obviously false to the other, which basically is like a natural selection process where media companies and social media companies are constantly sifting and selecting effectively for scissor statements because they're so enraging, and therefore engaging. And so what happens in this fight club, whatever, it's a thousand clicks, 10,000 clicks, everybody watches, fight, fight, fight, someone wins, someone loses, but most of the time both lose because both are harmed in this environment. But who gains? Twitter gains, the times' gains. They sold ads. It doesn't matter. Made money. The issue here is, I think of it as analogous to our diet, diet, because over the last 50, 60 years as restaurant culture has become more, and more and more mainstream, you've gone from having family cooked meals where the person who was your food provider was also your healthcare provider, to outsourcing all of that to capitalistic entities that are not actually economically aligned with you. And so, therefore, it's Coca-Cola, it's sugary foods, it's unhealthy foods that are delicious, that gets you to eat more, that's huge portion sizes, but you pay later with diabetes or disorders of metabolism and so on and so forth. And it's really something where it's an externality. People have talked about this, like Fiat food, like how the food pyramid is totally upside down and it tells you to eat all these carbohydrates and grains, when actually that should be the thing that you avoid to the maximum extent because we're showing it causes inflammation, causes type two diabetes. There's this gift of the obesity epidemic, Google that sometime, it's from Slate, it's almost 10 years ago.

And it's frightening because it actually looks like an epidemic, a true epidemic. Could it be, by the way, partially microbially caused? Sure. Could it be that this era ... You know how in previous areas, you ever seen the show Madmen? We kind of wince when you were supposed to wince, when like they're smoking all the time, pregnant women smoking a cigarette. I haven't seen the show, but I saw a few episodes. I don't remember all the episodes. The director knows enough about modern culture around smoking that they can kind of pan to a shot where the audience would go, they're smoking a lot. Yeah. That looks like a really smokey room. What are they doing to their health? But guess what? If the drugs at that time were nicotine and alcohol, maybe to a greater extent, I'll come back to that point, and it came sort of at the time, sure, nicotine has dropped off, but we're juicing ourselves with sugar, with caffeine and with whatever the dopamine stimuluses of social media slash rage stimulus, we're doing that. We're dosing ourselves to the absolute maximum. This is a sugar caffeine and social media society. Have ever seen that book, I think it's Apocryphal? Okay. I'm not sure if it's actually true, but it's interesting if it is true, you should do what I call independent confirmations. This book that claims that the Nazis were really on meth that was laced through a bunch of things in their society. They're like, that explains a lot why they're so angry. So I don't know if that's actually true. I thought it was a funny thing I saw in my peripheral vision worth diligencing and double checking. That's a general thing, by the way. I think lots of things on social media are consistent repetition rather than independent replication.

In cryptocurrency we have this concept of independent confirmations. You don't just approve a transaction right away. You wait for six independent confirmations, this cryptographic checks and so on and so forth. A lot of our mechanisms for information dissemination have advanced over our mechanisms for information verification. But fortunately they are on a screen, they're in a database, they're electronic. So maybe some of the truth can catch up to the lie, even though it's getting its boost on the cryptographic truth, curating your information diet more. Okay. So now pulling it all together, basically, just like you can see the cause and effect, you eat that cookie and you see your glucose spike or what have you, I think we're going to start to be able to see the cause and effect where you see this content, it's coming in through your eyes, it's coming in through your ears, and you see your heart rate spike. With the right analytics, you can see whether or not your biochemistry is moving in a positive or negative direction. You can actually see the effect on your body itself. Obviously, if people see a disturbing image or something like that, or a stressful text or something, you'll see it. You'll see it in the analytics if you've got enough analytics. What we also want to look at is the long-term effect. That is to say, what if you started to think about what would Men's Health look like in the era of Fitbit? What would Bloomberg look like if it was something which was measured by, did this stimulus of content improve your portfolio over time? What would it look like if you had learning content, then you saw whether or not you actually retain that with tests of mnemonic media? So this is a completely different model for media, because right now media is based on page views, prestige and profit.

So pages is obvious, just sheer popularity. 10 million clicks, whatever. Prestige is a little less obvious. That is popularity among the establishment. You're going for Pulitzers. That's a very small group of people that all clap each other on the back and are less and less representative, not just of America, but of the world. And then the third is profit. So short-term profit. Because COVID happens, a lot of newspaper's subscriptions are getting sold. Wow, you need to be glued to the news. Again, if it bleeds, it leads. There's an incentive for conflict. And given that legacy media corporations constantly interrogate everybody else's incentives, constantly paint them in the worst possible light, it's worth asking what their incentives are. There's actually an incentive. If it bleeds, it leads. Well, that means there's an incentive to make it bleed. More reporting. Is it causing the wars? Is it inflaming? Well, we know that that did happen with the Spanish American War. William Randolph Hearst, the yellow journalist, basically was like, something like, you give me the papers and I'll furnish the war, or something like that. I'm probably butchering the quote. But we know that there's at least one guy who basically started a war to sell papers, and we know that war reporting juices things, conflict juices things. So if these guys make more money in situations of conflict, that's actually bad. That's something that should be interrogated. We should be tracking that. Paul Graham actually had a plot from a few years ago, which showed that starting in 2013, the New York Times in particular started juicing their content with all these sort of bulk words. With the right analytics, we can just see that they just juice the sugar content of their articles. They suddenly just put in all these enraging words. One reason why may be that the Washington Post had just gotten bought in 2012 by Bezos, legacy media was down on one knee, Google and Facebook were going vertical. Have you seen the graph of print media disruption?

Balaji: [00:15:16] So from 67 billion down to 17. To me, it was really on the back foot, and whereas this paper going to get sold for scrap. Jack Shaffer actually had an article at the time saying that the Sulzberger family, which owns the New York Times. Isn't it interesting, by the way, we hear so much about Zuckerberg, and Kalanik and Andreessen, all the founders in tech, but most people cannot name the owners of media companies. I argue the reason for that is it's not so much that the owners of media companies get positive coverage, it's that they get no coverage. We are told that Zuckerberg, for example, exerts editorial control over Facebook. We're not really focused on Sulzberger, who exercises indirect control of the New York Times by appointing the editor in chief and the CEO, who fold up into the publisher. The thing is, look, scrutiny is good, but these are media corporations. They should be scrutinized just as much as tech companies are for their editorial decisions, for their personnel, for their conflicts of interest, for their incentive structure. Just to digress on this topic for a bit. Main things that tech companies are attacked, for example, there was an article a while back by Swisher in the New York Times attacking Zuck for having dual class shares. A few more years earlier, there's an article by Joe Nocera celebrating the Sulzbergers for what? Having dual class shares. It is something where it's wildly inconsistent. These folks will attack you for having ad trackers or whatever, while assuming their site in ad trackers. This is classic Russell Conjugation. Dual class is bad when Zuckerberg does it, but good when Sulzberger does it. It's like a true rip in the matrix where only somebody who just reads all this stuff and they catalog all of it-

Patrick: [00:16:46] I was going to say, it's crazy. I mean, just look at these two next to each other.

Balaji: [00:16:49] Look at these two next to each other. 2012 article, How Punch Protected the Times. 2019 article, Tech's Dual Class Stock. Let me just talk about this topic for a second, Russell Conjugation. I know I'm popping the stack and digressing on things, we'll come back up. Eric Weinstein has actually helped popularize this. And at first it seemed sort of like just a fun observation. So it comes from Bertrand Russell, and he observes that I sweat, you perspire, but she glows. The same concept can be gift wrapped in positive or negative connotation as one sees fit. Now, the thing about this is, and this is non-obvious thing, Russell Conjugation is actually a way to embed an org chart in language, a hidden org chart. Okay, let me explain what I mean. First, when you think about power, power is fundamentally a double standard that is to say, A can do something to B, but B cannot do it to A. That can be an explicit opt-in double standard. For example, you, as an employee, you join a company, you report to the CEO, you, as a private, you join the military, a general can tell you what to do, but not vice versa. In such a case, the right to exit, to leave the company, balance the control the CEO has. And in theory, you can also eventually leave the military, though you lose some rights upon signing that contract. But there is a tour of duty or what have you. The point is so that that double standard is consciously opted into by both sides for their mutual benefit, to win the market, to win the war, et cetera. Okay. One can critique these hierarchies, but they're explicit and they're opted into. There's a different kind of thing, which is an implicit hierarchy, where someone can do something to you and you can not do it to them, but it's not acknowledged explicitly.

For example, you cannot help dual class stock, but the New York Times company can, and it is gift wrapped in opposite sounding language. Now, once you kind of look for this, you'll see it a lot. You harass, but they hold you accountable or clap back. You are doxing, but they're investigating. He doxes, she leaks, they investigate. So the thing is the same action, documents were obtained versus hacked documents were put on the internet illegally. Those are two different ways of the knife went in, super passive voice description versus active voice putting a spotlight on it. These are all linguistic tricks that effectively embed an org chart in the conversation. Another example, the COVID stuff. There's so many aspects of this. But when you talk about the lab leak, you are a paranoid conspiracy theorist, et cetera. Okay. Once it's acknowledged by the establishment, well, scientists say, so on and so forth. And so this is like government by jargon. It is power by participle. It is basically org chart by adjective. It is essentially an org chart that's embedded in the language, and who is allowed to use what word and what is considered salient in the given situation is something there. Now what we're really doing here, by the way, is we're sort of deconstructing deconstruction. For co and these guys who are actually the roots of much of the woke stuff happening online, we can go deep into wokeness if you care or not, a lot of that actually does have an important kernel of truth to it, which is that one can find that there are power relationships in society that language can deconstruct. There are assumed defaults that really aren't universal default set by the laws of physics, but they're just imposed by some society at some point in the past that we've inherited today, and we don't think about too much, and that you can attack them with language and deconstruct them. And this is true, but you can also deconstruct deconstruction itself. And one way of doing that is to actually look at the double standard and the use of language, point it out, actually have a table of it. I'll give you another example, by the way, of this language double standard. So in 2019, there was an amazing one, which was YouTube, NYT back-to-back, The Making of a YouTube Radical. This was June 8th, 2019, The Making of YouTube Radical. And then it's like, want free speech in Russia? Try YouTube. Now, when do you think that article came out?

Balaji: [00:20:52] The next day. It was literally in the same paper. They screwed up the propaganda engine. And essentially, these are like rips in the matrix. Individually this argument can work, but once you see them side by side, you're like, wait a second. This is not a universalist argument, it's a tribalist argument. It is arguing that media tribes should have power. And that's really what it is. Media tribes should have power and tech tribes should not. Media tribe can have dual class stock, tech tribe cannot. Or in the case of YouTube, YouTube should not be broadcasting content that media tribe does not want in the United States, but overseas, YouTube and media tribe are aligned from now, so go ahead and broadcast content. Is absolutely nothing to do with free speech. Is absolutely nothing to do with corporate governance. Is everything to do with the Leninist who whom, and that's where the org chart is embedded in the language. It's the Russell Conjugation.

Patrick: [00:21:37] If you apply this back to the information diet concept, I think everyone listening will say, it makes a lot of sense that the information going into your brain is similar to the food going into your body. And with a tighter feedback cycle or measurables, we probably would change a lot of what we consume. I'm wearing the WHOOP, or the Levels, or whatever. It's amazing, very quick you're like, I'm not going to drink at night anymore because it blows up all my measureables. And just like the food pyramid, you don't spend that long investigating that to say, this should be inverted. And there's an 80/20 [predo?] thing where if you just stop with the grains and the sugar, you get a lot of the benefit. Is the same true here? If you just stopped reading the news entirely and just read source material, let's say, or research material, would that get the vast majority of the information diet benefit, in your opinion?

Balaji: [00:22:24] In general, I feel that the most constructive criticism is to build the alternative. Yes, deconstruct the establishment, use every verbal, technical, monetary, et cetera weapon. I think that's actually extremely important. But build something better. It's incumbent to build something better. See media records media, how do we build something better? So, first on the information diet, one very important concept, which I'm going to try to explain to your viewers, maybe they already know it, maybe some of them are very familiar with it and maybe some are unfamiliar, is a concept of on-chain proof. Are you familiar with this?

Balaji: [00:22:57] Okay. For example, a few weeks ago, Vitalik Buterin donated a huge amount of these so-called dog tokens to COVID support in India. And on-chain, you could actually see it. You know when someone would show you an unbelievable tweet and you say like, is that real? And you would go to twitter.com to that user and try to scroll to see if twitter.com loaded it for you. Why? Because anybody can kind of Photoshop a tweet. And if it's truly unbelievable, you're like, is that real? I have to actually go and check. And how you check? Well, the non-technical way of doing it is you go to twitter.com and you go to that, and it's basically, it's like you're looking for twitter.com to give a digital signature that says this is authentic. Yep, okay. Now with blockchains, it is sort of similar where one thing that's incredibly important that most people don't really think about that much, but it's very, very important for the future of the world, this multi hundred billion dollar sector. The hundreds of billions of dollars in Bitcoin stored on the Bitcoin blockchain, nobody really disputes who has how much BTC. I've mentioned this before, but whether Democrat or Republican, Israeli or Palestinian, Chinese or Japanese, groups that have had conflict in the past, in the present even, they're not disputing who has how much BTC. That is actually something where we've got global consensus on truth. And this is also the case for other sufficiently decentralized blockchains, like who has how much Ethereum? The actual amount, the raw numbers are not in dispute. Now, the thing is that that can actually be extended, not just to who has how much BTC, but everything in finance. Who has what stock? Who has what bond? Once you can gain consensus on a byte, you can gain consensus on a stream of bytes, and you can gain consensus on a data structure meta bytes.

And it's not a complete straight forward extension, but conceptually basically all the finance, everything that can be represented digitally can be put into this farm. And that's what DeFi is, but it's going to become. You may be familiar with this. But then you can go further and you can have other kinds of facts that start getting put on-chain. Some of these rely on some reporter, some oracle from off-chain putting the data on-chain. The Weather Channel is digitally signing and saying, this is the temperature in Poughkeepsie. Or Fitbit is usually signing and saying, this was your heart rate at this time. Why would you want to do something like that? Well, first you wouldn't necessarily want it all to be public on-chain. You might have it encrypted such that it's there, but it's basically nobody can view it other than you. Just like random bytes to others, only you or those with the so-called viewing keys can view it. But now here's what you can do. If you have, and you can start at the centralized, by the way, doesn't need to be on-chain. But if you have the next Men's Health and all your readers have, let's just say Fitbit or Apple Watch, you can see what the cause effect of your content is. You don't just put out some, here's great abs, or whatever. You put it out. Let's say it's like, here's how to improve your diet, and you actually see what delta there is on weight for these people. You actually start tracking something completely different, which is the reader's benefit. And of course you can just trust FitBit to report these metrics to you, and that might be fine as a V1 of this because it's non-trivial. Okay. Just the concept itself is a new concept for basically all health magazines, all diet magazines could immediately go over to this now that there's enough trackers out there. The V2 of it would be that it's put on-chain, so therefore it's not just Fitbit reporting it, but it's harder to falsify. Not impossible. The reporter, as I mentioned, could still falsify. But the metadata is hard to falsify, it's harder to mess with. And so the reason I say that is lots of media companies are very competitive about analytics. They don't trust this guy's metrics. How are they overstating it? Are your video views understated or overstated? Metrics are important and hygiene around them is important. And so, eventually I think these metrics kind of go on-chain for anything that you're basing big decisions on. Do I want to invest in this company or that company? Well, this company is showing a better improvement on-chain in the metrics, the health metrics, the finance metrics, et cetera, for its readers in this other company.

Informational Supply Chains

Patrick: [00:26:40] It's all making me think the bias should be on the information diet stuff towards information where the perspective of the source isn't a big part of the information. You're after basically almost raw data or unprocessed information.

Balaji: [00:26:59] So actually, there's a few different things. One is, we were able to interrogate the information supply chain all the way to the origin. Just like your physical supply chains, we can interrogate that all the way back to China. And actually that's very important nowadays. The information supply chain, you see it in the media, and that comes from a government study, that's based on an academic study that's based on a dataset. There's this concept in academia of reproducible research. This was big when I was a grad student. And in reproducible research the idea is, you don't just publish a PDF, you also publish the code and the data underpinning graph. Just in the same way a webpage is templated from a database. You're familiar with that? Hi, it's Patrick. That comes from a database. In the same way you have underlying data sets that underpin your scientific study, and the code generates the charts, graphs, and you also have the texts interspersed. There's various tools for doing this in R. There's something called Sweave. In Python there's notebooks. There's something called Distill.pub, which is a format for doing this. But the reason to do this, the reason it's called reproducible research is, at least for anything that's on the computer, this allows a grad student or another researcher to see exactly what you did. You don't need to explain to them that you chose this or that parameter to make this plot. And they can actually go and do independent diligence on it. They can say, "Hey, was your stuff really robust? Was it flat in this parameter range? Or is it actually hypersensitive and you cherry picked something." All of that type of stuff is transparent for the skilled practitioner when looking at the code.

You can wrap it in another function, drive it with new parameters. And most importantly, you can actually replicate the research by importing it and running it on your computer. Maybe you have some pseudorandom seeds that you set at the beginning, but basically you can replicate it. So anything on the computer you can replicate. This is reproducible research. There's another concept called open access, which says, basically every article that's being paid for by public funds should be public. And we've made progress towards this over the last 20 years. In the Public Library of Science, they actually had a video in 2003 talking about this. There's actually an article that I put up at [] .com called Crypto Sci-Hub, Talking about this. Your readers can take a look at that, but essentially open access says that if it's public research funded by public funds it should be public. I basically agree with that, but publishers have essentially gotten in the way, though there's been steady progress in open access. You combine these things and you get this constant, what I'm calling truly reproducible research. Where it's not just that some of the code and data is out there some of the time, or that the papers are open access some of the time. But rather that you publish the code, and the data and the PDF on-chain, such that you get a truly permalink. You get the ability to not just cite the paper, you get the timestamps. So you can prove you were first, which is important in academic culture. You have libraries on-chain that other people can import. And you also have a actual academic supply chain where you can trace the etiology, the origin of a fact or an assertion, all the way through the literature. And there's a famous paper that actually did track something like this all the way through the literature, and found it was just something that just got repeated. It was some medical nostrum, that actually didn't really have that base of evidentiary support. Where when you tracked all the way back, you couldn't find the broad table.

Patrick: [00:29:56] Is this is like the spinach thing, where they fat fingered the amount of iron in the spinach, and it got propagated forever.

Balaji: [00:30:02] It might be that. Iron in Spinach, it's something along those lines. But it's basically something where if you think about it, all the partial differential equations and so on a physics, Maxwell's equations or Newton's laws, which are not PDs, but ODs, typically. Those are continuous equations, which you really could only learn by effectively, scatterplots. That is to say, for example, the famous inclined plane experiment. You roll a ball down, you're just checking whether or not... What parameters affect the time that it gets to the bottom. The height of the inclined plane, how smooth you make it, and so on and so forth. You're just collecting data in a table, and you make a graph which is dot, dot, dot. And it's that interpolation, which is the continuous equation. So every equation that is written in a paper, underpinning it is a dataset. It might be a dataset from 50 years ago, but that curve is discrete at the root level. Somebody collected measurements and it led to this. Now, it could be that it's somebody like Feherty, who's this incredible observational genius, and Maxwell, who could put together these observations in this paper and that paper, without a full data set. He could do it from the verbal. But most of the time... Even that, by the way, is going to be effects then why. You could make it in a table. I observed this. This happened. The thing is, if we have this, if we have what I described, where the papers are on-chain, the code and the data, not just the PDF. And the citation graph is on-chain. And the import graph is on-chain, such that when you're using somebody's older paper, you're importing it. It's a big project, but it's a finite project. It's like Wikipedia. In theory, you could go and do this. You could take whatever million papers that have actually been important, and that have been cited. You could, one by one, put them on-chain. You could use Sci-Hub to do it. I'm not going to do that. But you could. Sci-Hub is basically the BitTorrent of science. Publishers didn't free the papers, so this bold person called Alexandra Elbakyan did, and she's running it out of Kazakhstan or some random place. But it's ridiculously popular in academia, because you can just paste in the URL and get back the PDF immediately. Very popular during COVID when the public needed access to this stuff, it starts to show what could happen if all this stuff was just burst free out of the paywall?

One thing I've thought a lot about is what does crypto Sci-Hub look like, and a crypto Sci-Hub might align incentives, such as the publishers also made money by putting all their papers online. I'm all about aligning incentives if at all possible. It's one thing to demonize a publisher as bad. They've got a business model. They need deed. Okay. How can... They're making a lot of money, but whatever. Can we align interests? If we can figure something like that out, that attacks the problem at base, because our entire society is based on "science". Why a government study? Why this regulation? Because science. Why do this? Because science. Well, what is that science? Well, some of it is "science". Where it's just some dumb study that appeared last week, and some of is truly fundamental science. And we're equating the two when they're not. And the way it assumes to them, is the number of independent replications, and if the stuff is on-chain, then we can start to do that. This relates to another concept I have, which is, at first it seems like a trivial or weird statement, but the only thing more prestigious than science is math. And they're not normally juxtaposed, but I would say math greater than science. There's a trivial sense in which that's true, which is theorems are theorems. Math is pure in a sense that science is essentially approximate. Math sewing science in another respect, which is when I talked about independent replications. It's not realistic to expect everybody to have inclined planes, and vats, and reactors and bubbling, alchemy type stuff, chemistry sets, or whatever at home for everything. Scientific equipment is expensive. So how are people supposed to independently replicate stuff? Well, yeah, they don't have all the scientific equipment, but you know what they do have? Computers. So they have all this mathematical equipment.

So you can run billions, and billions, and billions of calculations per second. So you might not be able to replicate the experiment, but you can certainly replicate all the calculations after the raw data set is put online. That's interesting. This is also, by the way, the reason that cryptocurrency exists, if you listen to the "economic scientists", the Nobel Prize, I'm going to get the exact term. Nobel Memorial Prize in Economic Sciences. There's a bunch of economics Nobels who have denounced Bitcoin. I think it's Krugman, Stiglitz, Myrddin, Shiller. And [foreign language] , does say, and yet it moves. Bitcoin is moving. Yeah, I know the prices down now, but who cares? Whatever-hundred billion from zero, it's clearly changed the world. Every potentate knows about it. It's legal tender now in El Salvador, it has transformed the tech industry. It has led to all these billion dollar companies. It's obviously a big deal. And yet these guys, their theory would say it doesn't work. It shouldn't work. It can't work. It should be banned. These are basically like the Soviet scientists and the late Soviet regime who had all gotten the Stalin Prize. They also had their economic pseudoscience. Some of the stuff like Kantorovich's work on some of the linear algebra was math, and it was actually portable outside of the regime. But most of it was just Marxist claptrap that was obviously false, once the regime collapsed. And actually a lot of, I think American macroeconomics, Western macroeconomics will be seen like that, because it's "economic sciences". It's not economic mathematics, which is what cryptocurrency is. Just to delve into that for a second, then we'll pop all the way back up.

In physics, you can do experiments. In chemistry you can do reproducible experiment. Even in biology, you can do reproducible experiments. There's human subjects regulations if it's humans. Medicine makes it harder, obviously. There's all these ethical constraints that we have to abide by, of course, but you can do experiments. You can say, "Case control. This person got this. This person got that." Even microeconomics, which is about the assemblies of humans, you can do experiments. The theory of the firm. You can vary prices. Every startup is in a sense an experiment, all the advice that Paul Graham, and Ben Horowitz and so on give out, is effectively stuff that you can abide by or not. And see what happens. So microeconomics is also subject to experiment. Macroeconomics, until very recently, was not subject to experiment. Basically for the most part, the only thing you could know about macroeconomics is communism didn't work. In a similar spirit, if the whole ramshackle structure of contemporary macroeconomics vanished into thin air, and the field had to be reconstructed from scratch, the sentence, which packs as much of the discipline into few as possible words might be, governments are not households. Not communism doesn't work, but governments are not households. Do you know what he means when he says governments aren't households. He means unlike in your households, they don't need to balance their budgets. They can print money in deficit and spend forever. There's this famous quote, actually from Paul Krugman, if you have money it's backed by men with guns. He says that to Joe Weisenthal on Business Insider several years ago.

That gives a different vantage point on the pronouncements of macroeconomists. It more priestly dictates than genuine science, priestly dictates, backed by men with guns. Because if you have currency, if you have money it's backed by men with guns. So now, rather than like the Nobel Prize in Economic Sciences it's like the Nobel prize in economic religion. Instead we have Nakamoto prize in economic mathematics, where rather than take the word of these economists on how well the economy is going. For example, in 2004, Bernanke gave a speech on the, so-called, Great Moderation at Jackson Hole, just to opening this up. He said... I want to just quote this because it's actually pretty amazing. So this is official federal reserve speech, February 2004, three years before the financial crisis started getting underway, before it really hit, "One of the most striking features of the economic landscape over the past 20 years or so has been a substantial decline in macroeconomic volatility. Several writers on the topic have dubbed this remarkable decline in the variability of both output and inflation, the Great Moderation." So essentially it's like this victory lap on something which is, now we know, completely and totally wrong. Actually the entire economy is going to gyrate in three or four years from then. We're still in this QE infinity, which we're not even calling QE anymore. We're calling MMT, but just printing all the money. And during the middle of this incredible, crazy thing, which is absolutely not the Great Moderation, some of the things that we are calling sciences are insufficiently rigorous. But now, as I mentioned, we have to have a better alternative. So what's the alternative? Well cryptoeconomics, mathematical economics based on mathematics says, "You don't need to trust Bernanke on the Great Moderation. You do not need to trust the CPI, the consumer price index on inflation stats. In fact, I'm going to probably fund a prize on inflation stats, actually just a decentralized shadow stats version of that. Totally on-chain, smart contract calculating it, prices scraped from all these websites like MIT's Billion Prices Project, but on-chain.

So it's fully reproducible, and it's more transparent and you can customize it to your own personal basket of goods. So now you can see whether the inflation is actually happening, or you're being gaslighted about it. This, I think is a major application of crypto. I can come back to it. The thing is, that rather than trying to trust that the Great Moderation's happened, or inflation isn't happening, or everything's under control and the stress tests have worked at the banks. Well, you can actually run the entire economy on your laptop. You can replay with Bitcoin. You can replay the history of an economy on your laptop, from T equals zero. And every satoshi is accounted for, every debit and every credit. It's on-chain. It is something that also every edge has metadata on it. Something like chain analysis, for example, which is mixed in the sense of it, it's basically surveillance on the blockchain. But in theory, you could know what those things, many of those edges were for. Was that a wire, or was that a purchase or an exchange? So the thing that previously was just this abstract thing that's captured in whatever, Solow's growth equation or something, is now... It's a data structure in Solow growth model. Oh, what I mean by that is, if you think about it, in the late '90s there was a concept of six degrees of separation. And that was this vague thing, fun thing, party trick, whatever. And then it became something very concrete with the social network, an actual data structure, a graph where you can do all pair shortest paths. That's an algorithm. It's a very expensive algorithm, but there's ways of bringing down the competition, to actually determine what the shortest path is between any two parties, and whether it's really six degrees, or whether it's 5.7 on average, or whether it varies based on time and space or all these other conditions.

So something that was just this abstract thing became a data structure. It's not a public data structure until decentralized social media where the graphs are actually open. I'll come back to that. But now the economy is similarly subject to inspection. The crypto economy is on-chain. It's transparent. We can all calculate our own inflation stats. We can determine whether volatility is increasing or decreasing, and not simply take the pronouncements of priests for this. We have decentralized it. This is a thing that goes really back in Western civilization for a long time. The establishment becomes ossified and incompetent, you break glass and decentralize. That's what Luther did, basically, you break glass, decentralize. The Catholic church doesn't have legitimacy, we're going to decentralize. And that's what Bitcoin is. It basically says, break glass, take the power away from these institutions that no longer have our trust, decentralized it, pull it back. Obviously, there's a lot of technical details associated with something like that. It's not a trivial thing to do. This starts to give a sketch of what economic mathematics looks like, rather than economic sciences or economic priesthood.

The Future of Cities and Network States

Patrick: [00:41:08] If I had to sum it up back like you said, we've gone in a lot of interesting directions, in the notion of information diet is as much about verification as it is about what actually goes in. And if you just blink or squint and look out, I don't know how long it's going to take to do all this, but let's say 10 years, or 20 years or 50 years, or something, that everything's going to get much more personalized, and much more based on relevant outcomes for the individual. Whether that's health, or mental wellbeing, or blood glucose, or whatever, and provable. So it begs this interesting question about, I don't know whether it's sovereignty, or what the right word is. If the promise of decentralization and provability of information is better outcomes for the individual, I'm really interested in how you think about collections of individuals and modern cities. I don't know how else to put that.

Balaji: [00:41:59] Okay. So first is, one thing I just observed the other day, is Twitter bios. They used to be something like, I'm Patrick. 32. I like the Steelers. That was the 2000s. That was it.

Balaji: [00:42:14] Yeah. Some stuff like that. That's right. And increasingly what they've become are emojis and hashtags signaling tribal affiliation. I'm with this gang, don't F with me, here's what I believe. Go if you don't like it. And so on and so forth. That's the most salient kind of thing. So first of all, why that arises, is because [Zach] and Dorsey and so on, they did connect the world. You essentially went from something where the geographic distance was primary. Similar to the geodesic distance was primary. That's to say the distance between two people on the surface of the earth is becoming less and less material than their distance in a social network. And that distance, by the way, is a distance metric. You can do cloud cartography on this. It's something that satisfies all the qualities of a metric. You can do graph lay outs and all of that stuff. And the point of this, is once you start thinking about distance in the social graph, well, nation states have borders. And those are formed over lots of wars and conflict over time, to get groups of people who mostly agree because they're physically close to each other. They share a culture. And Because they share a culture, they can have the same laws. And now that's all breaking apart, because you are closer intellectually, culturally, morally to somebody, thousands of miles away than you are to your next door neighbor, who you may not even know if you're in an apartment complex. That is the great fragmentation that's happening.

So what I think that's going to be resolved by is, at first, digital borders. So that's what these hashtags are. It's like, "I'm part of this group. You're a part of that group." It's like taking this graph and coloring it with ideologies. And then what's going to happen is there's going to be great migrations. This is already happening, that step to Miami is the 0.1 of this. Collective migrations. The collective bit is important. This is what I call the concept of Crowd Choice. Essentially, there's lots of things where you can take the 20th century version and just put it online. There's paper, and then you can scan it. There is university offline, you can put it into Coursera. There's cash and you can have PayPal. But really what we want to do is the full digital version. So PayPal goes to Bitcoin, or the scanner goes to a digital text file. So essentially when you're talking about collective decision-making, the obvious thing to do is, "Hey, we vote offline on paper, let's vote online districtly." That's like the scanner. Someplace to do that, Estonia does that, whatever. But a different way of thinking about it is, "Hey, the mobile phone is this absolutely incredible device, and enables all kinds of things. Can we think about collective decision making better.?" In particular, America's not just shaped by democracy, but by capitalism and by migration, all three forces. Voting with your ballot, with your feet and with your wallet. And all three forces. So what does that look like if we actually started thinking about it in a different way. And we say, "Okay. So in terms of collective decision-making, what if we optimize for the amount of consent?"

If you think about it, why is democracy considered legitimate? Because it's about the consent of the government. They have consented by putting in this vote, but we recurrently are in a 51% democracy, rather than a 100% democracy. Because that consent is the barest scraping of the bar. I mentioned this before, but it's like a Fosbury Flop. And you're just barely getting over the bar as opposed to clearing it healthily. So it's not a mandate. What happens is, basically if there isn't consent, you have to use coercion. So the guys who have 51% crack down on the 49%, they lose two points and it's just this seesaw back and forth. And that's what we've seen. So instead, how do we use the internet and the new tools we have, given that the old ways are falling apart and people are just trying to scramble, like, "Make America Great Again." Or for the east coast, they're trying unions or whatever, to try to prop up these failing media companies. And these are just echoes of a distant past, rather than leaning into what the future looks like and what the new technologies are. So what would that look like? Well, it would look like you have a group that is your true group of peers on social media. Not your physical group around you. Those are really just your co-specialists. Your co-ideologists are actually your true group. To say your neighbors often. You just don't know them. You don't know them anywhere near as well as you know the people online, who are your friends, or your enemies or whatever. But you know them. So you take your friends, who you know online. And you group together, and you basically say, "Okay, the 1000 of us, or 100 of us are going to move collectively." Even 100, by the way is a lot. And you appoint a leader, and you basically crowdfund a pool of funds to collectively fund the move. And at 100 people, maybe you can get the mayor of a small town to roll out the welcome wagon. A 1000 people, especially if it's a thousand software engineers, that's a hundred million dollars a year, let's say at 100K each. And that's remote revenue. You can do that anywhere now. Okay. Very big. And those 1000 software engineers can all collectively move to a city and actually make a big dent, a hundred million dollars of extra income being spent locally is a big deal. Then we start getting to 10,000, now you're talking a billion ARR. That's huge for most places.

So the point is, you can start actually, collective bargaining with governments. So this is a concept called the network union. So essentially 1729.com has an article on this. And the idea is, it's like a union, except you don't just represent somebody in a labor dispute. You can do that. But it's a network union, where it's got a union leader, and they do daily positive actions for their union members. Which could be something like, this person got deplatformed from a FinTech or media company, we're going to all tweet together, or lobby to get them back. Or this person was covered poorly, unfairly. We're all going to get that person's back. Or we all want to do something together, like we want to buy masks together. Or we want to have community goods such as a homeschooling teacher for us. Many of these things can be done online, but many of them, like the homeschooling thing, if you have a few people around Robin that are like the community school teacher, these are things that are easier if they're in person. And now that everything is remote. And now that Starlink has made the remote arbitrage possible, Starlink arbitrage is this huge thing, sell a city buy country. He tweeted that a year ago. I think that's still definitely accurate. Blue cities are seeing an exodus. They've just hit peak blue or whatever. Lots of red states are seeing an in in-xodus or influx. And they're also by the way, passing a lot of interesting tech and crypto laws like Wyoming, Texas, Florida, Colorado, Nevada, and so on and so forth. They're, again, doing interesting things there. Sharing this corrective, where folks are moving, and the next step, I think is for them to start moving together. This is the concept of Crowd Choice.

The idea here is optimizing for consent, because everybody exceeds to it. There's a leader, so they organize in a hierarchy. There are funds that are managed on-chain, because they're on-chain they're sovereign. You can decide how sovereign you want to be, but it's definitely yours. Say that mayor who you do the deal with can't just cease all the money out of your bank account when you land on their side. That will become a more and more important consideration. Just the fact that it can't be done, means that it's not contemplated. You just put it out of reach, and live and not get any fancy thoughts on this money by the community. And I think Crowd Choice starts to be an intranet native form of governance, where if you don't like it you can opt out. You leave that hierarchy doing those network union. So this is optimizing for consent. It's optimizing for choice. It's optimizing for collective goods. It's optimizing for competition, so that there's choice of government. I think the very first version of this is, we're starting to move from the two party system to the N city system. Do you know what I mean by that?

Balaji: [00:49:25] What we're now seeing, which is good, I think overall, is SF versus Miami, versus Austin, versus Wyoming, versus this versus that. And ideally what that does is it reduces Democrat versus Republican within the cities, and now pits cities against each other, competing in a more aligned way, which is way better. Because then you don't have people attacking each other within the city limits. It's just a war of words between cities. That's the ideal. Now of course, this kind of stuff can escalate to Athens versus Sparta or whatever. But basically, we've got a long way to run before then. Going from just a choice between two parties, to N different cities with genuinely distinct governance and genuine feature sets. Miami is really pro crypto and other places aren't. Some places are going to be pro drones. Mark Anderson and I talked about this several years ago. I have a bunch of tweets on this, but basically Mark has an article at that time on make Detroit drone valley. Essentially meaning don't try to compete on Silicon valley for what it was good at. This is like seven or eight years ago, which is software. Now Silicon valley itself is in the cloud, but anything physical, or anything that's regulatory, a jurisdiction can distinguish itself by having a regulatory regime that's favorable to cryptocurrency, or to drones, or to robotics, or to 3D printing, or what have you, that kind of stuff. So one aspect of the collective thing, which you mentioned is Crowd Choice. I think it's a critical concept. It's a very important concept. It's really the realization of Tiebout sorting. So Charles Tiebout everything you've ever come up with, some scientist-

Balaji: [00:50:54] Yeah. Nothing new under the sun. And in fact, actually this is one of my theorems, or premises, observations. Every ideology has been around since the beginning of time. What changes as the technological feasibility. In this sense, technology is the upstream driving force of history. Libertarian ideas have been around for a while, they're becoming more feasible, thanks to blockchain. Centralized ideas like communism are becoming less feasible, thanks to the internet and blockchain in some ways. You can argue they're becoming more feasible thanks to UBI, and seizures of funds and so on. But I think in the huge fight that's going to ensue this decade between these two forces. I do think decentralization ultimately wins, but it'll be a long multi-generational thing, potentially. And a huge drag out fight. Okay. So one concept of collective decision-making is Crowd Choice, where you're optimizing for consent, rather than desiccated, obsolete structures from 250 years ago that were devised when people had pen and paper. Some of the principles are there, but they're not contending with how the internet has pitted people against each other to figure out how to make the internet align people together. This new stimulus has pushed people. It's a new geography, fundamentally. The other aspect, the other thing I think a lot about, there's other things, but with the collective stuff, is imagine a basketball game where every single person, every player only saw their own score. There was no team score, and they saw how many shots they had, how many points. Nobody knew who was winning the game.

Now that is basically Twitter. People who think they're on the same team are really just Twitter maximalists. They're optimizing likes and follows. If they have some offline binding, like there's a company or there's a cryptocurrency or something, aligning them, that's different, and I'll come back to that. But other folks who are advancing in ideology often find themselves in fighting, because they don't have proper on-chain metrics. They're playing Twitter zero-sum status game, and consciously or not angling for influence within a movement or within a tribe, as opposed to delivering benefits to the community, or provable on-chain actions, that show they actually did something. They made some sacrifice. Proof of "actual work". So an alternative thing, which is interesting, is to think about what would a social network look like with collective dashboards. Where you had a group of people, who, the status gaining thing in the community, bands and metric that the entire community agree was valuable.

Now we have a model for this, it's enterprise sales. Within a startup within a tech company, you have a sales team, which is a bunch of individuals. It's true, but everybody's looking at the big board, where who's putting on the sales, which everybody knows benefits everybody, and pays peoples at least their base salary and comp. That's one model. Of course the sales team can be competitive. But there's other models as well, like looking at how many conversions were on the website. The entire company can of look at that and see, is that going up? Is that going down? Is it trending in the right direction? And so on and so forth. It's a collective metric where many people can drive towards it. One thing I think a lot about is how do we make team sports for social media. Video games are also like this, by the way, good video games, capture the flag. You know when your team has won and your team has lost. Social media is not like this, it's completely individual. But I think collective social media combined with crypto is very, very interesting. Because you have Onshape approval metrics. You start to do things.

The Recipe for a Successful Startup City

Patrick: [00:53:53] So I totally follow the granularity with which we can affiliate with other like-minded people, not just in terms of what we're interested in, but also in terms of the system settings, the rules of the game that we want to live within, and that cities are one really interesting unit where there are specific rules or standards or norms that are set. And that where we're probably heading is cities that are ever more specifically catered to one affinity group or collection of affinity groups. And I guess the question it brings to mind for me is like, let's say, I want to start, and I kind of do, let's say I wanted to make the town in which I live, the world's beacon for people that have a certain set of attributes that I care about. Like I want to attract as many people as possible. What is the playbook in your mind there for cities that will do that successfully? What are the ingredients or the dimensions that matter most for creating, I'll call it like a successful affinity town or city?

Balaji: [00:54:45] So this is a huge area, startup cities. I'm writing a book that has lots of stuff in this. Let me give a sketch. I'm going to read a tweet that I had last year, which I think has been influential. How to start a new city. Build a community in the cloud, organize the economy around remote work, enforce laws with smart contracts, practice in-person norms of civility, simulate architecture in VR, eventually crowdfund territory and materialize city into the real world. And the key is to go cloud first, land last. Build a community first. Don't worry about real estate. That can be solved once you have sufficient numbers of people with aligned values. Negotiate a deal with the state to buy land in the middle of nowhere with specified laws. So that's like the true de novo approach, where you have total route access to everything. It's like burning man. Just pick something in the middle of nowhere. And I think that's worth doing, the more middle of nowhere it is, the less people can complain about it. You should actually do drone overflight or something beforehand. Put that on chain and show that you really did build it from nothing, because when you do build it, they're going to be like, who would you take it from? Or you've got all these iPhones from nothing or whatever. So that's a de novo approach. If you don't want to do total de novo, then what I would say, here's one model, I think actually influencers will become mayors, will become governors, will become presidents, but first do that online. Build your community online, which you've already done, to some extent. Set up a network union where you can do that in a spreadsheet. It's just an org chart where you have, I don't know, a few lieutenants. It's like hiring. Do you want to make it a company? You can make it a company if you want. What you do is you conceptualize, okay, what benefits could we provide collectively to the PO shag subscribers? For example, maybe more analysis of this. Or where you try and figure out what the community service kind of thing is. You start providing that digitally. You include that with the subscription. Those five people get recognition for doing that. Maybe there's more dues that are paid that go towards supplements or income. And you start building the organization in the cloud this way. You build a hierarchy and then you start having people move out one by one. And the critical concept is you are basically like mayor for life of this town. And if they don't like it, which is totally reasonable, they can exit to another town and join another hierarchy of whatever level, TBD.

If they want to be the founder, they can do that. If they want to be one of the first whatever executives, they can do that. If they want to be the equivalent of employee number 5,000 and just go into a built environment, they can do that as well. And so they can dial up or down their risk tolerance and people can move back and forth between founder and employee and executive as they see fit. But critically what you're doing is you have a digital structure for this. It's based on something which we know works, nearly an org chart from a company where there's no one accountability. In a real way, by the way, a strict hierarchy minimizes bureaucracy. Because if you have, let's say an org chart where there's five people who fold into you and five people who folds to the next guy and five people who fold to that person, so then any person has maximum like three people to go through to get a decision made, because it's 555, like this. And so that scale is like five to 125 people. The minimum number of people who can stop you from doing something is degree of bureaucracy. The more people who can stop you have like governed by veto, even as your client as you actually noticed is governed by veto. Strict hierarchy minimizes bureaucracy. Now what's interesting is, by the way, if you think about what I'm saying, this strict hierarchy, this digital hierarchy that's being built is quite different from some other things. Two other things are out there. One is this idea that, oh, we have elections and there's a ceremonial mayor and whatever. And two that you have total decentralization where there's no leaders. Both those constants are in vogue, but I think this is the practical critique to both of them. It is a critique to the idea of just having a ceremonial mayor because got another article at 1729 called founding versus inheriting. The short version is that the East Coast, and by the way, you may have some East Coast listeners. I'm not actually-

Balaji: [00:58:42] Yeah, yeah. What I'm talking about is a certain institutional state of mind. There's insurgence on the East Coast and there's institutional people on the West Coast. And in fact, a lot of people in West Coast have actually left the West Coast now. And Silicon valley itself is not very institutional. So this is something where rather than East Coast versus West Coast, we could call it Fiat versus crypto. Because crypto is going to be synonymous for kind of the tech disruptor soon, if it isn't already. Basically let's call it this sort of Fiat mindset is actually one of inherited institutions. Inherited wealth is obvious. It's like Rockefeller's wealth. There's inheriting money that whatever great-grandfather earned for them. Inherited companies is something else. So like Sulzberger inherited New York Times company from his father who inherited from his father. It's six generations back. By the way, there's very few tech companies that are passed father to son that I can think of. Bezos built his company from scratch. All of this stuff is new money, as opposed to this generationally old money.

Again, an interesting point that in all the critiques of tech and so on, seems to have been missed somehow, in a thousand articles, right? This is why I say like, looking at their founders and owners as well is very important, because it's not going to be reported in their paper. That's what is not shown really often. It's like batias concert. It is what is unseen that one can infer and you can see on the side, but just in a different context. Talking about economics that I'm talking about informationally, but it's an analogous. Okay. But coming back up, it's not just that you inherit a fortune, it's not just that you inherited a newspaper or an institution. That's an obviously not just Sulzberger or Murdoch inheriting it from his father and so on and so forth. And I'm not saying by the way, inheriting is necessarily bad. Obviously there's people who inherited things who are totally fine. But where things do start to get bad is inheriting institutions. That one can never build from scratch. What I mean by this is the current mayor of whatever city in the U.S. could not have organized usually the police force from scratch. The current president or the last president or the last president before that, or whatever, and not picking on any particular person, could not have organized the U.S. military from scratch, or all of these departments. That's a massive beast.

That's like a George Washington level figure to organize the Federal Reserve for all its flaws, and I think we need to build an alternative to it. Alexander Hamilton built a Federal Reserve and it powered the U.S. through to present day. I think we need to replace it with something better. Those kinds of people did walk the earth. In fact, they still walk the earth. Vitalik and Satoshi who did Ethereum and Bitcoin, those are at the level, I think in the fullness of time will be looked at as comparable to what Hamilton did, if not even more important, like with Bitcoin. Bezos organized Amazon's logistics infrastructure, which if you've heard the saying about the military, amateurs talk tactics, professionals talk logistics. I think if Bezos was tasked with it, he could probably be a military commander since a lot of it is out of his strategic mind, all that stuff. So those people do walk the earth, but they're founders, not inheriting things. So the folks who are inheriting these institutions have inherited something they can never build in the first place. And the reason that's important, again, this article founding versus inheriting at 1729.com goes into this. This is kind of my place for essays. The reason this is important is, let's say you've got an heir and they inherit a factory that's cranking out widgets. And then the heir's heir and two or three or four generations down. They just cash the checks. They come in ceremonial, Hey, oh, it's founders. Hey, how's it going? And they've got some professional manager who's just cranking out the widgets. There's no immediate signal of anything being wrong. Widgets are coming out. Profit is coming in. It's all fine.

Then something like COVID hits. Some huge shock. The factory has to be turned from making widgets to making masks. Now this professional managers have no idea how to do this. That's the kind of edit that only the founder could make. By the way, a lot of Chinese factories did exactly this. They switched over to making medical equipment and so on. Totally different things. Making chairs to go to making medical supplies. They've figured out how to move the tools around. That change the system can only happen if one has intimate understanding of the technical business, financial realities of that entity. So the point is that this heir was repeating. They could not create. This is the same as somebody who grows up in a household and they may hear a language, but they can't speak it. There's cultural artifacts that surround them and they kind of copy them, but they can't create them from first principles. And if you cannot create, well, that's read only culture. You can't actually rebuild. You can't reinvent in a crisis. And that's what happened to America. All people could do is come up with slogans.

The only people who did anything in my view were the folks who crank out vaccines. That's basically what saved the bacon. The state was essentially incompetent at every level, from police failing and fire failing to the CDC and the FDA failing to state, local and federal government failing. It was just wild incompetence. It is not a first world versus third world anymore. It's like ascending world and descending world. It's like just descending world levels of competence. It's like decline this great thing that's now like basically a ruin. There's some rebirth there in Detroit, but that's basically what's happened. And San Francisco's incompetence and sickness has spread to all these other cities with homeless encampments and poop on the street and needles and mentally ill people who are not getting treatment attacking people in the streets and so on and so forth. That has spread, has metastasized. This expensive and wildly incompetence state is basically run by people who are ceremonial. They don't actually know how to run it. They're optimized crucially for legitimacy, but not for competence. Why legitimacy? They do have legitimacy because they were elected. That's a known process, but not for competence.

And so often people are caught between, oh, I want democracy. I want somebody who's legitimate who I feel like I voted for and so on. And someone else is like, oh, actually, a dictator is better. And by the way, the dictator will not necessarily be competent, but at least there's a chance of competence. They'll be able to make decisions cut through all this fluff, et cetera. And so you get a tug of war between people who think legitimacy and competence are necessarily opposed. Of course, the 49% didn't feel this person was legitimate. And that flips. The last four years there was one party that didn't think it was legitimate. Now next four years is the other party. And legitimacy and competence are thought of as being in opposition, but the founder combines both because why are they legitimate? Why is Zuckerberg the CEO of Facebook? Because he founded Facebook. Started in a dorm room. And every single person who joined Facebook, who signed up as an executive, who invested in it, there was a series of deals that he did, each individually beneficial with literally 3 billion people. Some of them obviously through intermediaries where it's like sign up at the website, but every trade, he made that person better off. And after billions and billions and billions and billions of trades, he is where he is.

So just like billions of bilateral trades like this. And that power was built up over time. It was not built up all at once. No one would have said, okay, give this 20 year old person the keys to a 3 billion person communications network that is beyond anything ever seen in history and beyond. By the way, David Cameron, while back, the premiere of the UK, appeared on a video chat with Zuck. And I was thinking at the time, I was like, but this is the leader of a tiny 60 million person social network talking to the leader of 3 billion personally. It's kind of funny. And yet there's something really true there. Like Zuck actually has power that he hasn't even flex, which is to say ... I mean, have you ever seen the movie, The Lawnmower Man?

Balaji: [01:05:54] This is from the 90s. And basically it's about like this guy who is given like super intelligence. It's like about an AI that occupies his body. He's just like a random lawnmower man, and he becomes a super intelligent creature. It's kind of like Terminator with Skynet, but there's a scene in it towards the end where basically he had said earlier in the movie that when the AI escapes from the lab, it'll make all the telephones in the world ring at the same time. So they think they've killed them or whatever. And at the end, what happens is, like closing credits, because back when there were landlines, it wasn't cell phones. Every telephone just starts ringing around the world. And the thing is, Zuck hasn't even flexed that pinky finger. He has lawnmower man level power. Like he could literally make all 3 billion phones in the world show a push notification. He could speak in tongues. He could put whatever he wanted in the feed. He has more reach than Xi Jinping and Biden and Cameron and every president of every country combined and more engagement. How many people tuned into the state of union versus coming to Facebook every day? It's got to be like a 100X, the number.

And yet he hasn't used that power, partly because, and this is important, his viewers kind of don't want him to use that power or they are different about it. Or they didn't really sign the social contract to let them use that power. They just sign up for something to communicate with their friends. They didn't sign up for Zuck as their leader. He's not really the leader leader. He's like the leader of the platform, he's not leader are the people, but somebody else could be. You could set up a new platform, which was premised on X is the leader. And people fold in a hierarchy. And if you don't like it, that's totally fine. But there's a thousand others like this, and you can go and found your own or what have you. There's a leader and they make hierarchal decisions. Think about the difference between Zuck with Facebook employees versus Zuck with Facebook users. Within the company, it's a strict hierarchy. You join and basically you're doing what Zuck wants either directly or indirectly. Orders are given, not like in a peremptory way, like do this or whatever. I mean, maybe if there's an emergency, but for the most part, if you're CEO, you learn very quickly that you can't use the I'm CEO card very much. You have to persuade. These are people with options. These are talented people. You have to say why it's important and you have to be willing to take feedback. All that stuff is very important for true leadership.

Basically it is a hierarchy within the company where people will essentially do what he wants them to do. It's completely different out on the platform where the people just know about and do anything. There's no direction. It's all entropic. All these particles moving in different directions, but it doesn't have to be. There's no upper limits to the scale of a network union. Your diamond, your triangle, your hierarchy could scale out. So this I think is a really interesting way to think about social networks, not just the social graph, but the social tree, not simply the social network, but the network union. And the network union we had defined it is, it's a social network with a hierarchical structure, a defined leader, an integrated cryptocurrency, and a sense of purpose. Why an integrated cryptocurrency? Because that gives wealth creation. That gives smart contracts to be signed by all the people. That gives rule of law. That gives wealth accumulation. It gives provable statements. It gives punching metrics. It gives lots and lots of things people don't even really fully realize. It gives encryption. So you can do encrypted messaging between yourselves, blah, blah, blah. It's basically like the stuff that actually turns a community into something real online. It gives you borders. It gives you everything.

And a sense of purpose is very important. Why are we doing this? How do we take all these entropic particles and focus them on something? Why Patrick's city? One way of thinking about this, by the way, just to linger on that entropic point for a minute, you know the term entropy? Entries where all the particles are scattering in every direction versus work in one direction. If you go to Hacker News or you go to Reddit or you go to Twitter with these new set of spectacles and you just look at it and you're like, wow, that's entropy. That's 30 different links. Every day I'm just like throwing random things in my head, 30 different things, because we have a novelty bias as humans. This is new. This is new. Twitter is new, et cetera. What there isn't is like a bias towards actually learning things. You have a higher order goal sometimes, which is, I want to learn, I don't know, elixir. To be clear, I'm not saying there's no role for serendipity, but I think we are over serendipitous, so to speak. We are over novelty seeking. We're not optimized for results seeking. Do you know what the Karen bandit is? The Karen bandit comes from decision theory where basically the idea is you've got a row of slot machines, K of them, and you've got to find that budget and you want to figure out what your strategy is for exploration versus exploitation.

This is something where the trade-off is, let's say, for example, there's just two machines. And the first one, when you put in a dollar, it gives you back 50 cents. But one out of a hundred times, it gives you $5,000. The second one, when you put in a dollar, it gives you back 75 cents 50% of the time, but $2 the other time. I'm just making up numbers. I have to actually work it out. But point is, that first one is something where if you explore it, it might just look like you're constantly losing money. You have to actually stick with that for a long time before you figure out, well, this is one of the jackpot payoff and it dominates the other one. So you have a finite budget also, because that exploration is itself costly. It's a really good mental model I use to think about business opportunities, about allocation of time and so on and so forth. So Twitter or Reddit or Hacker News is like this totally random jackpotty machine where ... I'm not saying you don't look at it some of the time, but I'd like to see what if we tried to allocate more of our time towards non entropic media, making it just as engaging and so on. It's engaging you on the upward arc of your life. You're gaining muscle, you're losing fat, you're learning things. You're earning money. You're accomplishing things that are of genuine value to you rather than just getting mad and clicking things. Just value to the legacy media or social media company operator. That brings us full circle back to the information piece.

Everyone is an Investor Now

Patrick: [01:11:35] Can you talk a little bit about what I view as sort of the last important piece of this sovereignty idea. So the whole thrust of our conversation is like evermore choice and tailored life information, everything for the individual. I think ownership, the joke like ownership is nine tenths of the law or whatever, applies here too. You mentioned this concept like everyone is an investor now. Behind that, I think is two things like, again, personalization of what you "own," but also the source of your income being more capital gains than W2 or cash or something. Flush out this concept a little bit for us of everyone being an investor and how that concept tucks back into the two major themes we've talked about.

Balaji: [01:12:13] So first is, everyone becomes an investor. Meaning, in the 1800s, everybody was a farmer. In the 1900s, everybody was in manufacturing. In the 2000s, everybody becomes an investor. The thing is that the transition from farming to manufacturing obviously led to communism. I mean, that was the industrial revolution, but it led to massive disruption in the 20th century, as people had to acclimatize to what that new world looked like. The other side of it was okay. It was a big deal. Ideologies were let loose by this thing. What's interesting by the late 20th century, the thing that people had denounced was in romanticized. Oh, all the good manufacturing jobs are now going overseas. This thing that a hundred years ago, people thought of as horrible relatives farming, now thought of it as great job. Look, I can just work a job and come back, whatever.

So now what is happening, and people, I think, with some legitimacy or some justification, bemoan the financialization of the economy. Everything is finance, finance, finance. I think what's going to happen is we're going to actually lean into that. And what I mean by that is there's many different trends, which when I say the statement, everybody becomes an investor, that can be interpreted on many levels. First is it's true in the sense of social media, everybody became a publisher. 3 billion people are now writing online, posting online each day. This is a massive increase obviously from even 25 years ago. So the Unabomber, you know why he sent all those bombs to everybody and like kill people at universities and elsewhere?

Patrick: [01:13:36] All I know about him is, because he's like weird philosophical writings, when you read them are bizarrely interesting and strange, but I don't know why he sent the bombs.

Balaji: [01:13:44] The thing is, this is a genius mathematician, also a lunatic. But basically he killed people for the distribution. At the time, getting an op-ed in the Washington Post was really hard. So this guy was such an attention seeking clickbait guy, he would kill people for the distribution. Remember, the guy's smart, his content parses to some extent, there's like a logical consistency there. By the way, he was a research mathematician at Berkeley that was proving theorems, like really actually off the charts smart in some ways. Where I disagree with him, by the way, is the idea that one can optimize for humanity while trying to reverse the Aero technology. I think just lead to masturbation and be bad. He idolizes this nature thing. I will say this, at least Kaczynski had skin in the game, in the sense he went and lived in the woods and many other people who idolized this primitive as sick thing, do not do that. But be that as it may, the point is that this guy who was smart, whose writings parse on the service level is willing to kill for the distribution. Now, with those lenses, think about Twitter. A lot of people there who are smart, but lunatics aren't necessarily willing to kill for distribution, though we haven't given them the opportunity. They're certainly willing to scream for it, certainly willing to attack you to kill people's reputations, to lie and to do whatever just for the attention. Might be a 100X, might be a thousand, 10,000 executions, but that also illustrates how scarce distribution was back then and how abundance is today and how we're still as society like learning to deal with this.

And that's why I think crowd choice and consent and the social smart contract become a big thing where people explicitly opt in to new communities with the constraints. And I say the social smart contract, it means you actually sign a social contract, just like you sign a click-through to get onto a site, but much more serious, more like a DocuSign. You actually read it then sign it. And it's on chain. It's a smart contract. So it's logged. It's clear that you consent it at that time for this duration, your funds are locked up. It's like a genuine commitment that you're making before you enter this startup city, before you entered this network state. That's where I think we land on the far side of it. But right now we're in this environment where everybody became a publisher. So that's one level in which you can understand everybody becomes an investor. Every social media platform integrates cryptocurrency. And therefore you're not just liking and tweeting. You're constantly investing on things, buying this, selling that. Bitcloud is actually like one of the first examples of this. There's pros and cons of Bitcloud. I won't get into all of that. I'm interested in it because it is a genuinely decentralized social media. Just from a technological perspective, I am an investor in it, but I'm an investor in like 500 other things and a small percentage of holding just a disclosure, whatever.

Actually there's also hive and steam and other things that are like this. And it's still early. This space is still early. But everything's on chain. So you can do things like print every single tweet ever, or the equivalent thereof. Print every post ever. You can print out the social graph, you can crawl the social graph. You can search the social graph. All these are things that Twitter and Facebook just don't let you do. That's one aspect of program building. But the other aspect is monetization is built in, so the cryptocurrency is built in at the edges. So you can pay people. You can invest in them, you can sell and buy. Essentially it makes the edges more functional. You can actually earn money on them and eventually do real work on them. Crowd something, sign smart contracts, form real businesses, offer SAS services for API keys, et cetera, et cetera. All of this stuff is happening on decentralized social media. It's in the early stages, but what's going to happen is we're going to move from social networks to truly digital economies. I think that's going to be the term for them.

So now what you've done is you've massively leveled up what is happening online. You've actually got people earning real money and they're constantly out thinking like an investor all the time. Do I put money into this? Do I put money into that? The number of investments soars. So does everybody's complexity of their personal finances. I think lots of people will personally incorporate. It's going to be all kinds of tension with existing government laws. The whole Uber thing was a big thing, independent contractors, everybody becoming an investor will put a similar strain on the system. It's just simply not meant for everybody to be basically like have a portfolio complexity, size of Soros, lots of stuff will be built to accommodate this, unless stuff is being built, but that's like one dimension of everybody becomes an investor. The second dimension is that influencers become more important because not everybody can diligence investments. So they'll Naval, they'll follow you, they'll follow people in angel list investments and back them. And this is sort of like choosing a boss in a traditional company. In a traditional company you're like, hey, you can go start a company on your own. Or you can find an executive or a CEO that you respect and join their company direction from them. But you also have a more stable income.

The same way you can go and just invest all the money yourself, be an independent investor, or you can join a fund where there's an influencer who has a good track record, hopefully an on chain track record where it's provable returns, again, on chain, on chain, on chain, because trust is vanishing in the world, but fortunately computation is rising. So we can compensate for the loss of trust for the increasing computation. Both of these are things that are catalyzed by widespread information networks. So you have on chain provable returns, people follow those influencers. And now they're giving up some of the returns, whatever percent, for hopefully a reduced risk, because this guy's an expert in managing the money. No one can say your company won't go bust, but let's say there's a buffet or somebody like that might be good for a long time. Also in this world if everybody becomes an investor, I think Bitcoin is the no op which everybody holds. I think we have a hundred million something people owning crypto today. In 10 years, I think It's very, very, very likely that it's going to be a billion plus, maybe even 3 billion. That might be even conservative.

If inflation hits in a big way, buy Bitcoin. Don't sell Bitcoin. That to me is very obvious, but it's not yet obvious, so I'm just saying it. In that way world, the world where everybody holds Bitcoin, everybody is on decentralized social media investing just like everybody is on Twitter and Facebook tweeting and publishing, there's also the world of Robinhood being there for a long time. That's a world of all these VCs and angels and founders and startups and exits. That's a world of essentially everybody becoming an investor, except it's a world where if in the 20th century, the 99% are labor and the 1% are capital, the flip happens this century where the 99% are capital and the 1% are labor. To say the 99% are investors, they're just retail investors trying to get in the deal. They don't want to get screwed, whatever. And the 1% of the founders who can actually turn money into things, because that's actually a very rare skill, extremely rare skill. One of the things that people don't understand until they're an investor, is this idea that money is abundant. Because people are looking at like, where's the money? I don't see the money. What do you mean money's abundant? Money is abundant for people who are hardworking with good ideas because they're rare. It is actually very non-trivial to convert money into a spaceship, into an electric car. A lot of people just think the CEO just puts their feet up on a desk orders around flunkies, but that's really not how it is.

Actually an interesting point is a lot of people who think that, think it's incredibly important who the political leader is, but unimportant who the corporate leader is. Let's say like Bernie is president of the U.S. or whatever, but it doesn't matter who the CEO is. They're completely dispensable and the worker could do the job just as good as the CEO. Interesting contradiction, interesting way of thinking about it. Another view is of course the political and the corporate leader both matter and leadership matters, which is my view. And then there's other view which is my view. Then this other view, which is, I think now obsolete, I shouldn't say obsolete, but the old school tech view, which is only the corporate leader matters, political leaders don't. There's like all the innovations happening there. The reason that has now become obsolete is we've all learned what happens when you have terrible political leadership. San Francisco literally just need to keep the lights on. They need to keep the poop out of the streets, and the lights on. They literally couldn't even manage that, like PG&E with power outages. They couldn't even do that because they inherited a system they couldn't build.

One way of thinking about that, by the way, is San Francisco city government, and the US government in general, imagine a board meeting where the board just basically assumes that after they all finished their stupid political knife fighting and they write down what should be done, that the company can just execute on it. Imagine if 80% of the company was devoted to the board meeting, 90%, 95%, the attention and energy. How well do you think the company does? That's all the political wrangling, not the actual execution itself.

Essentially, this model that I'm describing with the hierarchy where you have a CEO who appoints themselves digitally, just like you appoint yourself the founder of a company, you're the founder and people fold into you. That's a model that selects for both competency and literacy for the ability to actually execute, not just to inherit the thing. Everybody becomes an investor. We talked about the decentralized media aspect. We talked about the crypto aspect. We talked about the fact that everybody is doing VC and Robinhood and Angel, and this type of stuff. We talked about the sheer numbers going from roughly 325,000 people, I think with Bloomberg terminals, to easily 30 million plus on crypto, actually more like a hundred million on crypto alone. You add in Robinhood and so on, and you're talking probably hundreds of millions, also add in all the Chinese stuff that's happening. Add in the fact that governments are now going to be competing on money. Most people don't understand this yet, but we are entering the age of global monetary competition. All these constants are becoming variables. The dollar is facing serious competition from both the digital Yuan and Bitcoin, both on different flanks. I mean, people who run the dollar take this completely for granted. That's why they're printing all this money. They're trying to sanction various people. All these people, like thousands of different sanction programs around the world. They got pushed back though recently. They tried to sanction not Russia, China, and Iran, but Germany and India who are allies.

The US government tried to do that. Germany over Nordstream 2, for like a pipeline they're building with Russia, and India for buying some missiles from Russia to use versus China or something. I don't remember exactly what came of the Indian situation, but the German situation, the US at first talked a real tough game and then kind of backed off. One of the things I think we're seeing is the limits of American power, where empire is receding abroad because neither Democrat nor Republican really has the appetite for more forever wars. The little rips in the matrix like Crimea and ISIS and China and the South China Sea, Iran quasi-going nuclear or whatever's happening there, in North Korea kind of doing its brinksmanship, all of those kinds of rips in the space time continuum are sort of like on the borders of the empire, but it's like rumblings, presages the US sort of gravitationally contracting and troops coming home and less overseas intervention because it can't. Now, this is something which is maybe a long-term arc but there's huge possible disruptions to that. In particular, one thing I'm concerned about is that people will try to stoke a war with China as a cure for the internal division in America.

I know that there are some people, I think, misguided who are doing this. It's sort of like have a baby to save the relationship, but actually worse than that, which is to say it's not like the war on terror led to a more unified America. Everything gets politicized. It just led to this gigantic surveillance state where Republicans built this surveillance state in the early two thousands and the Batman movies at the time, there's like, should we use the power and so and so forth. 15 years later, that surveillance state is now being used to track Republicans and I think soon it's going to be used to track, unfortunately, everybody. The House Un-American Activities committee was actually started to go after Nazis and then post World War II it was used to go after a communist and McCarthy was running.

Lots of these kinds of weapons will be flipped. I expect, for example, that sometime in the 2020s, here's a prediction, okay, that the New York Times will have something on inflation or Bitcoin that Twitter says fact-check, because Jack actually recognizes that a lot of the stuff that they're writings is incorrect. I think that's, call it a prediction, and then you'll see the New York Times eyes bulge where they expect this fact-check stuff to only be used on the competitors of media corporations, not on themselves. How could they get something wrong? Oh my God. I think that's going to happen. Especially if it's on something technical, which tech people feel confident in like the article recently where they said something like Bitcoin was hacked and it actually wasn't, it was just a traditional subpoena process, but they made it seem like ECDSA was broken, which would be huge news.

Action Items: Practical Things You Can Do

Patrick: [01:25:49] Without really intending to do this, but it's cool that we have, thinking about the three big things in someone's life, what they consume, the context in which they live their lives day to day, what they own. Those are three big important things that determines the quality and direction of somebody's life. You've thought in deep detail about these things much more than I have and much more certainly than most people listening have. Just because it's hard to think about these things at a deep level. If you had to turn these into RX, like into a prescription for people listening, saying, "Wow, these are things I haven't really thought about. I want to take some next step." How would you turn all of what we've talked about into that RX, just at a high level, for those listening that want to do more?

Balaji: [01:26:31] Monetarily, I'd say buy Bitcoin. If you're more adventurous also buy ETH, like 50% BTC and 50% ETH. Put into it what you can afford to just never think about, literally if it goes to zero for years, drops 80%, you don't even care. You're not looking at the price. If you want to, like Coinbase has pretty good security, but there's cold storage services also. Do that, number one, because that alone, I think is the dumbest, but most obvious thing to do to prepare yourself for 5 or 10 years or more of significant financial and monetary disruption, number one.

Number two, if you're interested in health, there's a few things that I think are producing pretty darn good results now among like the tech crew, whether it's a Fitbit or an Apple Watch or a Whoop or an Aura or something like that. You should probably get something like that. Optimize your sleep. There's also cheaper ways of doing it; stuff like Eight Sleep, for example, it's like a really good mattress, there's Casper. Getting a really good mattress given that you're spending 33% of your time sleeping, very important, don't skimp on that. Even if it's whatever thousands of dollars, it's probably worth it if it gives you even 10% better sleep. If you have noisy things around children or construction or something, figure out how you can get either whether it's headphones or move or something to truly limit noise when sleeping, because you'll track your sleeping and you'll start to see those things that improve it. For example, cutting out drinking at night and so on. Okay.

The intermittent fasting stuff is talked about, but it's not very precise often. When you fast, how do you fast. OMAD is something that a lot of tech people are doing now that is, I think very effective where it's what it sounds, like one meal a day. Basically the simplest version is you have a one hour or two hour window where you can eat in that window. You can basically eat whatever you want and you should eat healthy food, but you can basically eat what you want and you just don't eat outside that. Some people do dinner 6:00 to 8:00 PM so they can eat with their family and that's it. Basically, it's very easy to remember. It's essentially like a significant calorie restrictions because as long as you just don't eat at other times you can get in the habit of not eating at other times, it's kind of hard to eat a full day in one meal. The satiety kicks in. It's like 50% or more drop in calories. You start just dropping weight. It's also simpler because you just have fewer dishes to do and all that type of stuff. Right? The other way you can do it is like lunch 12:00 to 1:00 PM, whatever you want. Then you drink coffee or water, or you can have flavorless water or Vitamin Zero or something like that over the day, that kind of stuff. You can track your blood glucose and make sure that things are getting back to normal. That's like intermittent fasting and diet. Home gym, if your folks are into that, that's maybe an obvious one, but basically investing in health. If you want to get more serious about it, David Sinclair's podcasts and stuff are good to listen to. A Harvard professor who is into longevity and he has kind of a thing of resveratrol, vitamin K, vitamin D, NMN, and Metformin as sort of like a combination regime and the guy looks pretty young. He's got the science and so on behind it, and he's got the dosages or whatever there. If you want to listen to him, you can do that. None of that stuff is expensive. Metformin, you might need to get a prescription for, you can get like a pill box and start kind of doing that. That's like a V1 longevity stack. Longevity stuff is going to get bigger and bigger.

These are individual things on wealth. On wealth, also, basically we built actually a search engine for this called teleport.org several years ago. I think had we done it during COVID times, it would have been a Hundred X bigger. It was fine. It was a fine result or outcome. Nomad List is also good. I think search engines for digital nomads are going to be a huge thing where you can punch in and find the best place in the world to live and work. There's actually multiple map overlays. If you're a single person, you might care where the bars are. Drinking is bad, but whatever, going to the bar if you want. If you're with kids, you might care about where the schools are. You might care where the airport is. You might care where Starlink is. Gas tax, property tax. All of these are layers on the map and superimposing all of them and doing the calculations is actually very non-trivial. You can do a first cut on that and try to find, okay, if you could live anywhere, where would you live? Getting a place where you can be outside and there's sun is probably good if you like the sun. Finding a place that's congenial, and one of the key concepts by the way is you can do launch through arbitrage. Once you're outside the office, you just need to be within whatever number of hours of the other people, so plus minus three hours is usually fine. If you're the founder, it's your decision. Ask your CEO or whatever, your exec, but plus minus three hours probably is manageable. Which means if you're in the US, South America opens up, Canada opens up.

There's an interesting hack, by the way, that I've been thinking a lot about. It's a great website called Time.is. There's another thing on Wikipedia with the time zones of the world, which actually are more complicated than you might think, because there's like weird fractal things with the small islands and stuff. One interesting hack that I've been thinking a lot about is Guam. It's something where any US resident or US citizen can just go there. You have basically a permanent visa, but it's within two hours of Asia by time zone. You can be an American citizen working remote for Asian companies by living in Guam. I think that's going to be a huge thing. You should probably tweet about this, the Guam arbitrage. I think Guam gets offices built there. Guam, there's also like Northern Mariana Islands and so-and-so and so forth. They've actually got pretty good internet and it's like an island paradise. You can be a US citizen and participate in the Asian economy without getting a passport or anything like that or like a permanent rez. All that stuff is hard, it's becoming easier. Taiwan has a gold visa and so on. I tweeted a bunch of stuff, Taiwan's gold visa, Estonia's e-citizenship, Singapore has stuff. Monaco has stuff, Dubai has stuff, Chile has some stuff. Everybody's trying to attract entrepreneurs, France is trying to do this. You've got a lot of choices, but if you just want to remain a US citizen, then Guam is kind of interesting, potentially. I haven't beta tested it myself, but I think it's conceptually interesting as also an example of staying in the US territorially, it's a US territory, but work in Asia so that opens up a set of companies that you can potentially work for.

Thinking a little bit outside the box. That's wealth. Wealth is buy Bitcoin and probably ETH, work remote, minimize consumption if you can. It's much easier to reduce your expenditure of five x than to increase your top-line five x. Going remote will do a lot for you on that, just getting out of big cities. Health we already talked about obviously the home gym is kind of obvious. The OMAD, like the intermittent fasting, the longevity stack, sleep stuff. I'm not a huge person on this, but I've dropped something like 30 pounds doing this. Let's see where it goes. I used to be like as jacked as South Asian physionomy would permit prior to my first startup, like an obsessive and doing anything if I'm doing it. Startups have a way of prioritizing other things.

This is an interesting thing by the way is like as CEO or founder or whatever, you're basically always putting everybody else in front of yourself. You're like, oh, I could work out or I could finish this email. I could work out or I could respond to this deal. Oh, people need me to be there for them, therefore I can't be selfish like this. Who cares about my workout if this, I don't, I can't eat healthy. Eventually, I realized that was actually a false economy, like going to credit card debt, maybe an obvious point, but the younger you are the easier it is to do that without realizing you're putting miles on something, putting in credit card debt. Actually, what you want to do is do something where it's communicated to people like, "Yeah, we're all going to invest, not just me as founder or me as a CEO, but you as an individual, we all want you to invest in your health."

Don't be ridiculous about it. In the sense of sometimes we do need to like crank on something because the site is down or something, but in general emergencies not withstanding, you want to budget for your health and time for health because just in a pure enlightened self-interest manner, that means the CEO is healthy all the way through, the founder's healthy all the way through, the executives and employees are healthy all the way through, health insurance cost is more alertness, more fitness, all the type stuff. Enlightened self-interest for your group and focus on health. The metrics let you do that. You can actually do group health.

I think curating your information diet is important. Just going from first principles, like every once in a while, what I'll do is I'll just draw a pie chart of how I'm spending my time and how I want to spend my time on a whiteboard. It's always disaligned. Then I think about, okay, how do I kill all these things on my calendar unapologetically and boost these other things. At least I start vectoring in that direction. Other things I do, whiteboards everywhere just to have extremely fast boot up, just take notes on something. Some people have paintings. Some people have photos. I just have whiteboards everywhere with just little notes and equations and thoughts and stuff like that. It's also useful. There's two books I think are quite good. Nir Eyal's Indistractable and James Clear's Atomic Habits, where they're complimentary, James Clear basically says how to teach yourself to do good things and Nir says how to stop yourself from doing bad things. The basic idea behind Atomic Habits is you have a stimulus, you have a response, and you have a reward.

It is not simply enough to say, "I must workout." It's like you put your shoes in a spot, like your running shoes, in a spot that you'll always see, and you have your socks and your shorts or whatever there so you can change right into them. When you see them, you're like, "Oh yeah, that reminds me. I'm supposed to work out." You run, and then your reward is something like you post your Fitbit thing somewhere, your proof of workout, you log something. There's something which is like, "Yeah, I've got an accomplishment." This is the way to kind of pack your brain circuitry so it's stimulus response reward. This is why people are on Twitter. Hit somebody, get a like. Right? Maybe we can use that for good things. Nir has the inverse of this, like Indistractable, how to gain more focus and whatnot.

On the collective stuffs, that's all individual stuff. I think you need to achieve individual financial independence to achieve ideological independence, to achieve collective independence. Once you've got a bunch of people who are doing that, who are financially independent, who are healthy, who are mobile, who are conscious about their information diet, then they can work with someone like you or with an influencer more generally, or become an influencer or a founder by themselves. By the way, that's a new thing, founding influencer. Like a concept just like founding engineer, founding influencer, I think is on par with founding engineer where the influencer brings a community and their knowledge of the community, the engineer brings the technology that's a great co-founder pair. The community is more forgiving of the bugs, basically the product, efficiency of the product doesn't matter, the product does matter, but the influencer knows what product the community wants.

Now, you start basically having software that starts ... The key is a hierarchy with a sense of purpose. There's something you're doing for the community. You might need to survey them. You may to talk to hundreds or thousands of people via direct message or what have you. But if you decide to do this, you can probably find something that a bunch of people complain about that you've probably talked about that's actually something that they would pay for, because that shows some commitment. You just set that up effectively as a SaaS service and you start providing it. You organize this online, start the digital stuff. Maybe it is just a summary of content. Maybe it is reviews of things. Maybe it is crowd sourcing of information this week. Maybe it is helping somebody with something. Maybe somebody launched something you want to signal boost them. It's something. You've got some daily action. You've got some common purpose that starts putting this networking together. Then, that becomes something that can do more and more things over time.

Patrick: [01:37:33] This has been so much fun. I wish we had more time. Maybe we even find a way to get more time. I ask the same closing question of everybody, which is to ask, what is the kindest thing that anyone's ever done for you?

Balaji: [01:37:44] The interesting thing about the question is that kindness almost feels like it was sort of undeserved in a way, or is like a bounty of some kind that kind of arrived from you didn't need to do it or whatever. I'm not sure if this is the kindest thing anybody's ever done for me, but it was surprising to see some of my former students have made really touching images or videos. Like people who I got into Bitcoin in 2013, I taught this class that had 250,000 students and had Bitcoin tutorials in there. Bitcoin was like at 60 bucks. Some people had their life just totally changed by that. They've done very kind things over the years, sent me really nice messages. They're like, "I was at X and now I'm at Y. Thank you." Just very unexpected gratitude because all I just did was type some stuff online that I knew. I didn't realize it had this positive impact on people. That was touching to me because you have a very left brain view of the world or actually right brain is the logical side, right, left brain is the emotional? I always forget that.

Balaji: [01:38:48] Right. Yeah. If you have a very left brain view of the world, it's like, "Wow, that was unexpected. That was really nice. Thank you." That's kind of one thing I think former students or people online sometimes say just don't expect really kind things. One thing many, many, many, many years ago, I was kind of a smart ass at times in high school. There's a physics teacher and he was talking about centrifugal force. He was explaining it and was saying, "You know when you wash clothes in the drying machine ... Yes. What is it, Mr. Srinivasan? How could you possibly have a question now?" Just at the beginning of that, I was like, "Don't you dry clothes in the drying machine?" That got me kicked out of that class, but there was a kind administrator who allowed me to take physics C electricity and ... Not a C grade, but it's called AP physics, electricity and magnetism. I'm not sure if it's still called that. Allowed me to take that just independent study, so I wouldn't fail the course. I would just be able to do it myself. I could do that for also mechanics, which was the first half of that. That was very kind. He didn't need to do that. I was like, "This is awesome. I don't have to go to class and I can just learn it myself out of the book at my own pace when I want to." That got me into more physics and math and whatnot, and taught me how to learn, how to self bootstrap, because that same kind of thing of just sort of being a little bit disobedient, like I was trolling him. I was. I was a kid, but it was also technically correct, which is the best kind of correct. Showing me that small mercy actually put me on a good course where I would, would I say it was a life changing event? No, maybe not a life changing event, but I wouldn't have gotten a good grade had I not been able to do was able to get like a five on everything that I learned that I actually learned faster out of school and helped shape my thinking on things. I don't know if that's a good version of it, but that's something.

Patrick: [01:40:39] I like it. Very consistent with what you've put out there in the world of self-bootstrapping and self-sovereignty and all these things. I've learned a lot today. I really appreciate your time. I'm sure we'll connect again sometime soon. This has been a blast. Thank you.