The Genesis Episode: Reteaching Math as a Liberal Art
1 hr 19 minAudio
Magic Internet Math Hub: https://mathacademy-cyan.vercel.app/index.html Brian Hirschfield X: @fundamentals21m npub12eml5kmtrjmdt0h8shgg32gye5yqsf2jha6a70jrqt82q9d960sspky99g Website: https://zeuspay.com/btc-for-institutions Rob Hamilton X: @Rob1Ham npub1emdtsxly9m68m00x206t574jttp65vk0c2m89ms038q047yz7ylqcac9aw Website: https://www.anchorwatch.com/ We’re launching Magic Internet Math. In this kickoff, Rob Hamilton and I set out our mission: to reteach math as a liberal art for Bitcoiners, builders, and anyone who wants conviction in their own reasoning. We share our origin story (a December 18th road‑trip call that should’ve been a podcast), define the show’s “bar‑level” approach to concepts like groups, fields, vector spaces, and probability, and connect them to real systems - Bitcoin, elliptic‑curve crypto, libsecp256k1, and why linear algebra underpins LLMs. We also talk survivorship bias in Bitcoin, the perils of rhetoric over first‑principles thinking, and how to stay sovereign in a world of shiny tools and FUD. We outline what’s coming next: deep‑dives on linear algebra and graph‑based knowledge for AI, approachable cryptography basics (binary/hex, finite fields), conversations with maintainers and researchers (e.g., libsecp256k1, FROST), and how to learn publicly - mistakes and all. If you’ve ever felt “the dumbest person at BitDevs,” this feed is your anonymous on‑ramp to build math legs without the priestly gatekeeping. 'Motivate the Math' podcast (Apple Podcasts): https://podcasts.apple.com/us/podcast/motivate-the-math/id1790329241 'libsecp256k1' (Bitcoin Core): https://github.com/bitcoin-core/secp256k1 'Bitcoin Core' (repo): https://github.com/bitcoin/bitcoin 'BIP‑32: Hierarchical Deterministic Wallets': https://github.com/bitcoin/bips/blob/master/bip-0032.mediawiki 'BIP‑340: Schnorr Signatures for secp256k1': https://github.com/bitcoin/bips/blob/master/bip-0340.mediawiki 'FROST: Flexible Round‑Optimized Schnorr Threshold Signatures' (RFC 9591): https://www.rfc-editor.org/rfc/rfc9591 'Programming Bitcoin' by Jimmy Song: https://programmingbitcoin.com/ 'Understanding Cryptography' (Paar, Pelzl, Güneysu) 2nd ed.: https://link.springer.com/book/10.1007/978-3-662-69007-9 'An Introduction to Statistical Learning' (official site): https://www.statlearning.com/home 'Base58' - Bitcoin protocol school: https://www.base58.info/ 'BitDevs' (NYC hub and resources): https://bitdevs.org/ 'Rock Paper Bitcoin' podcast (site): https://rockpaperbitcoin.fm/ 'Obsidian' (official site): https://obsidian.md 'Model Context Protocol' (official site): https://model-context-protocol.com/ 'Claude Code' by Anthropic: https://www.anthropic.com/claude-code/ 'Claude Opus 4.5' model page: https://www.anthropic.com/claude/opus 'AnchorWatch' - Bitcoin custody + insurance: https://www.anchorwatch.com/ 'BIP‑340 overview (alt mirror)' - bips.dev reference: https://bips.dev/340
Audio and video play straight from the show's own hosting. PodSubs stores and serves no media.
Transcript
From the show. Tap a time to play from there.
Right. We are here. Did it. Hello. We did it. It's finally happening. This is happening. I'm really excited, man. What are we doing? Let's see. Well, do we wanna start at the beginning where you, you know, there's a special moment in every man's life when another man asks him if he wants to do a podcast. And it's kind of one of those things that you kind of cherish till, you know, the end of your days and your twilight. As I'm sitting on my deathbed, I'm gonna have that regret of I wish I was doing more podcasting.
Yes I think that's probably gonna be the number one regret people have. Yeah. When they hit the end. That's right. You should have been doing more podcasts. That's why I do five. And then Yeah. That's why when this one this particular one was on its was on the ropes Mhmm. For a second. Yeah. So you I will today's 01/11/2026, and it was December 17 or eighteenth or something. Eighteenth. Eighteenth. I'll never forget this day and for a variety of reasons. There you go.
So we were oh, that's right. Yeah. So we were I was on a road trip. Was driving basically myself, my dog and my family possessions for Christmas holidays while everyone else took the plane. And you said, hey. Do you have a minute? And I said, I have twelve more hours of driving today. I have nothing but time. And at this point, I'd already kind of gone through a bunch of podcasts just, like driving along the road. And then you and I ended up talking for like two hours, which should have just been its own podcast. We should have just recorded that whole conversation somehow.
That's a normal phone conversation with me. People are starting to realize. Well, I guess this is what's the name of this podcast now? The Magic Internet Math? What was I believe we went we're going with Magic Internet Math. Excellent. So do you you wanna set the stage as the torch bearer and the one carrying over the previous feed? Yes. I would very much like to do that. So first of all, for coming on. Rob Hamilton,
we are magic Internet math. And what the heck are we doing? Okay. So I think we have a very noble and yet psychotic ambition. Those are the two best qualities of an ambition. And it is, I will just say it's to, I think reteach math, really re bring math to our civilization, newly,
with newly formed eyes and also as a liberal art. And that's, I feel like we live So I feel like part of like all of our Bitcoin journeys are, there's certain skills that we were taught not to get. And even if we, you know, we were demoralized out of getting them and they're important. So like when you think of like, what's the attack surface, everybody like is really on top of their upset. Most people are pretty on top of what they eat.
You know, they're pretty, you know, they their physical fitness, they're fairly on top of that. You know, they can think to themselves, if I'm in a jam, I might need to outrun somebody or beat them up or shoot them. All of these things are like things that people are on top of when they think, boy, when we go through this fourth turning, what's really the skill set I'm going to need? And, you know, it's not like you're gonna have to do a math problem to get through the fourth turning.
You might. You might have to prevent someone else from doing a math problem. But it's more, I think where this is really important is thinking through, know the fourth turning isn't going to last forever, and most of us are really, most of us, let's just assume we're going to survive it because we've been doing the work, right? And I can't think of a sadder thing than doing the work to survive the fourth turning only to get completely bulldozed in the next era because you really didn't necessarily think through the complete set of skills you might need.
It's kind of like playing basketball and locking down on defense, but you don't get the rebound and they get another possession and you have to go defend yourself all over again. So the way I view mathematics is like a core skill set in a self determined human being. A self determined human being needs conviction in their own reasoning. And this isn't something we think of all the time, right? But
you have to know when you're right. And that's like a big question is how do you know when you're right? When it comes to fighting, like I think you know, right? You're like, I got this, you just have this feeling like I fucking got this, right? Then every once in a while somebody really surprises you and you're like, oh, I guess I didn't have it, But I understood the conviction. Understood what it took to know that I was capable of this match.
With regards to your own reasoning, that gets really difficult. And we see it in the world in like some of these FUD attacks, like that's where I think I really see it on the battlefield. Like where does this get people? I think something we discussed is something like 98% of people who have ever owned Bitcoin have already outlived it. Right? That's right. Yeah, and that was I forgot the exact context of the setup of the conversation,
but I made the observation that most people, especially if you view it by amount of Bitcoin, have already outlived their Bitcoin. If we did an actuarial analysis of like the Bitcoin as a cohort, it would have already Most people, their Bitcoin survives them because they have spent it or lost it all already. Especially when you count it by the numbers when you have was it Half of the Bitcoin distribution happened in the first four years from 02/2013,
half the Bitcoin ever in that will ever exist in circulation were issued. Right. I think most of those people don't have their Bitcoin anymore. There's a couple who still do and you the culture and the like the kind of like the the shared oral history of Bitcoin. We always focus on the people who still have the Bitcoin. Yes. We call that survivorship bias, and we'll probably talk a lot about that on this podcast. It's funny. The first level, like click you always get with people with like Laszlo and the pizza stories like that idiot sold 10,000 Bitcoin for two pizzas. He must be regretting it. And everyone forgets that Laslow is the guy who came up with GPU mining as a concept.
Yep. And Satoshi actually tutted him into trying to share some of it because my personal theory is that Satoshi always knew that you were going to be able to use graphics cards to do more optimized like operations. Right. And that was going to outperform the CPU for high end computation, and so he knew that was always eventually going to happen. That's actually a funny side story. The amount of work the Genesis block has way overshoots and indexes the amount of required work to make it. Like, it it surpass I forgot the exact math, but you can go look up the Genesis block header. It has more lead way more leading zeros than you would ever possibly need to make a Genesis block, a consensus valid one. Yeah. So my theory is that like, I think I got this theory from John Seth that
Satoshi was tinkering with physical hardware and understood how it worked because it was almost kind of like a little artifact for you to realize later, like, wait a second. This guy did way more than one CPU's worth of work. Right, right. I guess to pull it back in, though, on math, so when I went to college, I actually got a double major in math and economics. But I also almost failed seventh grade math. So I've always, especially later, I've always liked numbers and math. I had a difficulty when I was younger trying to fully grok them and I like getting like my bachelor's degree in math
and appreciation for kind of like it's the foundational rules of the system, whatever system that you have, whether it's how computers work, it's how really any sort of concept like right now LLMs being the hot new thing, it's linear algebra. It's a bunch of large matrices doing multiplication, right? And I've always found that even if you don't have like the full like PhD doctoral level analysis, understanding the rules of the system and being able to have enough tools for self sufficiency to explore
that system is a valuable skill that should be encouraged among everyone and not everyone has to get a bachelor's level of understanding of math. They should be able to at least bread crumb out their reasoning and understanding so they can figure out where they want to tap out and stop. Right. And when we talk about this with Bitcoin, how many people actually understand elliptic private key cryptography and like finite fields and like doing the multiplication with to be able to actually do all this stuff and how it all works, right? The asymmetry of public private key cryptography.
People understand my seed phrase is special and it should be kept safe, but they don't understand what actually is going on under the hood. And I think part of this podcast journey is kind of being able to, these are often things that are kind of like glanced over like, hey, go to a BitDevs or read this book if you wanna learn more, but there aren't actual conversations around that. And most of the technical podcasts, the first thing they usually do is wave over the theory and talk about like how something technically works, but never getting to the math level. It's like here is the level of technical understanding you need to use this product or service or technology, but it's never like the one level deeper of why is this good technology or why does it work the way it does in a secure way. So I think that's kind of part of our journey here. I think the way when we were on that two hour signal conversation was we want to be able to break it down for like if you're
at a bit devs and you go to the bar afterwards and you're just talking about stuff, right? This isn't going to be super deep in the weeds, but bringing a bar level comp, like standing at a bar level conversation of these math concepts to give people some sort of extension for them to help them. So like an olive branch or some sort of like reaching out of hand to get get you at least into the first level of being curious about these concepts and thinking about them for yourself. Yeah. And you might participate in the conversation even if you don't know what the hell you're talking about because you recognize at least a couple of the terms because we
like, I feel like if you so this this is the second generation of what I started last year, which was called the motivate the math podcast. And I'm gonna leave that RSS up. This is gonna be a new one, but I definitely think people should go check that out because that was the really that was like the that was the Genesis block that had way more work than it needed. Like, the the amount of work it needed to basically start from a complete noob level and explain what groups are, which was sort of psychotic.
But like, I think that people who went through that have some like kind of sports ball like understanding now of concepts like closure, inverses, identities. And these things actually matter when we talk about, say, you know, what's the difference between elliptic curves and Schnorr signatures and like those kinds of things like that. So like just having sort of the sports ball type bar level conversation, familiarity with a lot of the jargon, it's like knowing notation
in a certain way. Right? And there's a great quote, I use it all the time by Von Neumann, the mathematician who said, you never really learn math, you just get used to it. Being exposed to a recurring conversation, it's sort of like going to BitDevs all the time. You go to bit devs all the time, you're not so intimidated after like three or four times, you just might not be that intimidated talking to a guy like Rob, where he's just like going, he's going a mile a minute, but you're like, wait, I understand some of the things he's actually saying, you know? Yeah. And
I think like related to that, my own journey with Bitcoin was I got into Bitcoin in 2013. I had just graduated college and the actual gateway for me was I had buddies that were like, hey, you're a Ron Paul and the Fed guy. You should check out this Bitcoin thing. I was holding gold and silver and I was kind of like I was technical in that I was always like I was the token IT guy and I understand computers pretty well, but I wasn't a programmer And I kind of rolled my eyes. Was like, like, like, I think my eyes just glazed over, like too complicated, probably a scam, like baseline heuristics. Right. But then a couple of months later,
Dogecoin came out and a bunch of my buddies started mining Dogecoin And I download the we're all gamers. We have our graphics cards, right? And I'm setting up and the I could see it was all in Comic Sans. I understood Internet culture to be like, Okay, this is satire, right? This is sat But you had to download the client just like you would with Bitcoin Core. You had to run your miners, point your hash rate. I was mining at a Dogecoin pool and it had me realize I was like sending money around to my friends. Was like, okay, understand this has been a way to thaw the ice and get myself to start grappling with these concepts of what Bitcoin teaches.
And then I started going to bit devs and there were bit devs New York City bit devs in 2013 was like a dozen, maybe two dozen guys in a room going through C plus plus poll requests. And I walked in, I was like, damn, I am the dumbest person here. Right. But that gave me the opportunity to sit there and think and listen and start my own exploration and journey that with Bitcoin. And I think maybe our goal a little bit is for those that are in Bitcoin and kind of always kind of pushed off on the technical understanding of things to be able to
take that step for themselves right have a friendly environment for them to hear people who are grasping with the concepts themselves and being able to start their own questions and start building their own foundational understanding through their own reasoning, right? Like you and I are not experts that should be taking directly at our word for any of this, but we're trying to bootstrap and start up people kind of thinking about these more technical mathematical concepts for themselves and start trying to connect concepts and ideas and how it relates to Bitcoin or their life and being able to kind of have fun with it.
Yeah. And to be clear, anyone who listened to the first iteration, Motivate the Math knows I'm not a mathematician. I am not even I don't even have a college degree in math. I was an actuary, which is what mathematicians called morons. You know, oh, you can use a five function calculator and you just make a bunch of money working for insurance companies. That's great. So like an actuary is typically if you like, you know, if you're inside and like the math scene, right, at companies, they're considered like too dumb to do the real math. I ended up quitting being an actuary and becoming a quant. I had to self teach myself basically, to get to calculus, differential equations, all of those things. And I think that's a big part of how I ended up here.
That's a long journey that I think is worth sharing at some point. But, you know, college degree is in something that is like very simplified. I never would have made it through like a lot of the hard math classes. My oldest daughter is a math major. She's already took a couple of real analysis classes that I never like, I would have got like, I realized like I would have learned everything, but I would have gotten Fs on all the tests. And that's sort of how and that's like, I accept that that's how I am, right? I can learn things and I can explain them, I probably won't do well on an exam.
But I am the strongest at the moment person applying for the job of teaching people this. So I'd say if you're a better mathematician than I, jump in the ring and start teaching and when that happens, find that guy. I will also say like the contextualization. So for me, the big thing is the context. The context is like decisive
And I believe I probably have a very unique world view here in wanting to teach math as a liberal art. And I think that is super important in having children. So I have, again, my oldest daughter, she's 20. She's a junior in college, pure math major. And, you know, I didn't know I didn't know this is how it would happen, although I wanted it to happen. She had a liberal arts education her whole life, and that was really important to me. It was I actually thought I thought there was a pretty good chance she would study math, and it was really important to me that she had humanities,
that she like understood people, she had art. So like they go to Waldorf schools, which is really focused on art. And like why do they focus on art is because art actually builds will. So when you like struggle with an art medium, like copper or blacksmithing or a canvas, you are at the mercy of that medium. And you have to figure out how to dominate that medium into and into what you're trying to get it to do. And you you know, that is like that that builds will in a child and in a person. And when you talked about math earlier,
that's what I was hearing. I was like, is another medium. And it's appropriate in some places. It's inappropriate in other places. Meaning like, LLMs are dominated by linear algebra. And like one of the reasons I have very one of the reasons I'm pretty confident in the world of LLMs is because I have a lot of conviction that it's really hard to fuck up when your entire world is a vector space. Now we can talk eventually about why I would say something like that. Right? But like linear combinate the world of linear combinations is really hard to fuck up, meaning it's hard to find
like, I feel like if there's aliens on a planet somewhere else that their linear combinations work the same as ours. Like that is like one of the base atomic units of how we think. So like, I have a lot of confidence that that is probably gonna work. Right? Whereas I don't know that the physics that people have mastered, sorry, the math people have mastered in physics, I don't know that that applies in other mediums necessarily. And I think when I think about Bitcoin and cryptography, a lot of the danger we run into is
just these notions of equivalence. Like we, I feel like as human beings, we project equivalence onto things all the time. And it's not always correct. In fact, most of the time it's not correct. And we are, you know, we're subject to a knowledge system that actually wants us to make incorrect projections of equivalence, you know, and this is why we have stupid things like, name you name the stupid thing in the media, right? Like, woke shit, whatever it is.
Slop. We have a lot of slop and a lot of it is just because we're being told that A equals B when it doesn't or A projects to B when it necessarily doesn't necessarily. So like the when you when you start studying like abstract algebra, you get a rigorous, it's like a really a rigorous approach to like, when do you get to say something equals something else? And sometimes two things that don't look alike can sort of equal each other because they preserve the same mathematical operations. And we got a lot into this in the first iteration of the podcast.
So like that's kind of thing like I feel like in you, so why that's why it matters like humanities, like understanding math as an art is like, is this medium appropriate for the kind of math you're trying to do? And does it makes like, this is something that human beings, there's a certain math that human beings will try to do without knowing it's math, which is, you know, I think all of these types of people are the same. Something like that, right? Where we do these projections that are totally, and then we take it for granted, right? We, they're like, we treat them like theorems in our brains and we build off of them. And
I'm kind of like the messenger who you might want to, you might not like very much to say, hold on a sec, right? We need a rigorous framework and you don't have to study math, you don't have to have like, you don't have to have been this gold star student and you know, through this. That system was designed to make people not learn it. That system was designed to just sit to actually keep you from having a reasoning base that you have conviction in because they want to be you wanna they want you to have conviction in their reasoning.
Right. But, like, I've seen this meme going around that, like, people oftentimes aren't actually engaging in rhetoric. Like, well, they're engaging purely in rhetoric, and they kind of view words as spells, and they're just trying to use their spells on you to persuade you because they're not grounded in any sort of, like, deeper truth. They just have, like, a spellbook of catchphrases and things that they say in rhetorical tricks and framings and how to almost short circuit your reasoning. And the idea here is to kind of
understand things from a deeper principle. I think we should talk about kind of like our larger ambitions for the podcast. But a perfect example, while you were just talking, I was battling with chat GPTs and this is a perfect example. I made an offhand comment earlier about how the Genesis block had all of this extra work in it. And I threw in a chat to see like, can you verify for me that the Genesis block had extra work in it? And I've been wrestling with it for about five minutes now. And it's like, well, the difficulty of the Genesis block was one. So there really wasn't any extra work. And I was like, that's not how this works. I was like, the solution to the proof of work, look at the block header.
There are extra So Genesis block has, if everyone's familiar at a high level with Bitcoin mining, you prove about like how many leading zeros you have in your number to show how much work you've done because you've succeeded the target, right? So if I pull up the current, for context here, if I pull up the current Bitcoin block, the amount of leading zeros in the block hash is one, two, three, four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, fifteen, sixteen, seventeen, eighteen, nineteen, twenty, twenty leading zeros in hexadecimal, which is base 16.
Right? So you have 16 leading zeros when you have 16 options. Right? So you have you're taking 16 possible like options and you're saying that at 16 times, you're always picking your one out of sixteen, sixteen times in a row. Like that's a very like difficult thing to be able to achieve. The Genesis block has one, two, three, four, five, six, seven, eight, nine, ten. It has 10 leading zeros. Block number one, on January 9, the first block that Satoshi made. My personal theory is that Satoshi put out the code and the Genesis block and the client and was waiting for someone else to do something with it. So there's a provable observable gap of basically almost a week, like six days,
where Satoshi was kind of like taking a step back and see if someone else would take it. And eventually he's like, okay, I'm going to start mining blocks. This is like because everyone knows Jan three was when he hit the button, but the block wasn't mined for So the Genesis block was mined. That's where you get the times in the Coinbase output. He says, you know, chancellor on the brink of second bailout bank. So he wanted to prove like, hey, the network started on this date because I took the morning paper. But the next block that built wasn't on top of for six days. Right, that's right. So the whole network didn't do anything for six days.
And the thing known in Bitcoin is you actually cannot spend the money in the Genesis block. Like the actual initial 50 Bitcoin. You can't spend that. But just to go back here, there's 12345678910 in the Genesis block. There's one, two, three, four, five, six, seven, eight in block number one. So you have two less leading zeros, which means a lot less work was done to do that. And ChatGPT is adamant. It's like, no, the difficulty
was one for block zero and block one is identical. I was like, you're not I'm not asking what the network difficulty was. I'm asking how much effort was put into making the Genesis block versus. And so this is all just like a little bit of microcosm to say, especially in a world with LLMs where it's like very eager to very quickly give you an answer. You're going to take this stuff as baseline fact. Yep. If you don't understand and reason this stuff more. And I think our mission and like what we're doing this for is to provide
a safe entry level point for someone who is not technical to talk about how math intersects with Bitcoin, life, technology in general. Like we could probably do multiple episodes on just linear algebra to explain like what a vector database is and like how context works with LLMs, right, to understand how to get more and more insight out of them. They may not always be Bitcoin related, but I think the goal here is to provide a casual environment for people to start exploring and playing with these concepts for themselves.
Yeah, and I think what, so now the resources that we have to offer people listening to this are in a just a different stratosphere from what they were in the first iteration of the podcast. So if you want, so I have a website now. So this is why December 18 was such an important date. December 18 is the Jan three of magic Internet man. Because that's when Rob came to my meetup and decided to interrupt his long long drive.
That's right. He introduced he introduced me to a not to vibe coding in general, but a particular the particular skill that we've all been obsessed with. I wonder how this episode is going to age just because of the time it's being released, right? Things you know so quickly that we're gonna laugh at ourselves. It's gonna be seem very antiquated a year or two from now. He basically reached his hand out and brought me into the wave of Opus 4.5 and Claw Code. Yes. And with that, I stood up this magic Internet math website, which I'll put in the show notes. I wish I could tell you it's magicinternetmath.com,
but it's not and we'll hope to get there one day. But I'll put the website in the show notes, but there are there's like 30 courses now on there. And so one of them is this basic cryptography class. So the first thing I put in there is a total drill down. If you want to get good at converting from binary to decimal to hex and all that, It's the first page in that in the cryptography class. So you can basically just go in there and you can click a button and pop up a little calculator and there's a bunch of quizzes. And the easy quizzes are multiple choice, but the hard ones you actually got to put the right answer
in. It's really good thing for like, if you wanna know like, if you're an adult and you kind of like fucked off in math your whole life, you're like, what could I do? Well, like learning your times tables is probably a good thing to do. And then learning how to convert from just natively from binary to hex and decimal, being able to do that is a really good idea. And I've gone through like I've spent hours on that one page just doing the quizzes and I think it's worth doing and I'm probably going to create a game off of it
that know, just these certain things that I really want to drill down. So anyway, so like the ability to do that, there's also there's three linear algebra classes on the site now of varying, you know, of varying difficulty. So there's a I don't know that there's like a super true blue beginner linear algebra class, but there's a basic algebra class anyone can start with. And they're like, it literally starts with like what is a field which is just like the very properties of like when you're adding two numbers, it's you have commutativity
meaning the order doesn't matter like that kind of stuff, associativity like that very basic, very, very basic stuff that gets you into eventually what we call linear algebra. I like to just call that vector space, the math of vector spaces where it's linear combinations of items. Strictly like ax plus by equals something, where a and b are like integers. And so we have this website that definitely
is you should check out. Got a map URL, put it in the show notes. URL will be in the show notes. It's all public. It's all free. Everything's free at the moment and it probably will stay that way for a while. So yeah, that's so like we have some we have some tools now also that like, so we want people to be comfortable. I think when you brought, when you said you showed up, you've told me the story before too. And I've had the same reaction when you talked about going to the bit devs and feeling like you're the dumbest person there. Like,
are the things I seek all the time in life. Like, I seek to find myself in those situations. Now a lot of people are very uncomfortable in those situations, but this podcast could be such a situation where you don't have to look anybody in the face and see them later and be like, hey. I remember when you were dumb. I remember when you were the dumbest guy here. Right? You can just listen to the podcast in anonymity, right? And I would definitely say stick with it because this is worth doing,
right? But it's not it might not feel good. You might not feel not feel like you're making a lot of progress necessarily. But again, the website, I have a beginner section on the website. And like, you probably don't realize this, but you can learn like calculus as you know, as a basically as a noob. You can start learning it you know, the only reason they teach it so hard and stupid is to make you hate it. I know that sounds like slop, but I do have a strong belief that the powers that be don't want people to don't want people good at math. They don't want people building a bomb
against them. Right? That's they don't, you know, they don't want another Manhattan project of a bunch of people. So, like, even so like they rigged the system to demoralize people. And you know, some people were really good anyway and survived this system. Some of them got PhDs in physics and then they basically just put them on things like strength things that don't matter, like strength theory so that they would never do anything significant. Like Satoshi was a fucking mistake in their system. He was lab leak. He's a lab leak. That's right. Sort of like and I like to think of us as lab leaks. Like, are errors in the system trying to and then we're trying to essentially
select and find people who wanna be errors. Right? Because it's gonna take an error to like really win on the frontier. Like if you really want to be out there with what is going to be bombarded at you, like the FUD attacks, like I talked about the FUD attacks before, like maybe like the first FUD attack, which is parabolic price of Bitcoin. That's like what unlocked, like that's what's what got most people to sell their Bitcoin. They're like, oh, look how much it look how much it's gone up. It's two x, three x. You know, I bought it at $2 and now it's like $6 Fuck, dude. I'm a genius. Right? That's the first understanding the concept of market cap and that when it went from 2 to $6 it went from like a 20,000,000 to a 60,000,000 market cap and you think that's the total addressable size of this market
is 60,000,000 right? So the first like kind of, I call it FUD attack, it's very passive. It's just like let people, let people get rid of it naturally because they don't understand how valuable it is. Then they started to get these flood attacks, they started hitting things like socially, right, that you're a psychopath, you're a loser, you're a power consumer, you're all this shit. They try to make you feel bad about it. And a lot of people I think said it wasn't worth it and said said, you know, I don't know. And it's again, it all comes down to not knowing really the value.
And then, you know, like this past year, we had the fucking node stupidity that really I I feel like really, I don't wanna get you going and get act people activated, but like the reality is I think, oh, this leads I think the reality is a lot of people had a problem with the realization that they were second class citizens, that they this realization that they had they didn't have the power to really affect this thing. And they got it you know, they got really worked up and scared about it. So, like, know, this is where, like, I saw very clearly
that first, I saw that technical people sidestep this thing, like, pretty beautifully, and it was really the nontechnical people that got really caught up in this. But, you know, but, like, I don't wanna talk about that issue so much. I just wanna say that, like, these things these these attacks are gonna get better and better and better. Meaning, like, they're you know, it's not gonna be so easy. The bar, I think, for your own conviction in yourself is going to be is gonna rise more and more and more, and it's almost like your ability to hold your Bitcoin is going to be like a big fucking tournament
that you just keep going through every round, hope you survive every round. Yeah. No. I agree. To be I'll work backwards and be concise and say, one, this is where the biggest like, in that whole discourse and debate was this isn't a technical debate. This is a philosophical debate. And that was a basically Jedi mind trick to have you be ungrounded from reality. We can just leave it there but saying like oh this is philosophical, this is not technical when Bitcoin exists as software in the world and it must operate as software. You have to contend with the reality of how the world works in a distributed system. So I'll just say that, like, that was like the first, like, obvious, like, slight of hand of being like, Ah, see, you don't have to learn technical things. I can just
do rhetoric and you should listen to me because I have rhetoric. It's an instant it's trying to remove your sovereignty. But it's an instant like just switch of like, don't think for yourself, don't do the hard work of understanding how things technically work, just listen to the rhetoric. But I think the messaging that's important to take away from this, this is what I took away at least and I think this is what I would want people to know is that bitcoin is a protocol driven, it's a protocol and if you're all in,
like, on your life and your life savings and you're, you know, you're even close to all in, you think you've got really got skin in the game. And I know a lot of those guys had skin in the game, and this was, I think, part of why they had such a problem with it. Hold on. If you're all in, you had skin in the game, You are not a you're just not a first class citizen if you don't really understand the thing. And I don't really understand the thing. I don't even know if you really do. Like there's only a few people that do. The question, right? So what do we do about that? So this is what we were talking about on that car trip, right? Is that like
the libsecp library, the actual library that secures the Bitcoin. It started when Satoshi initially wrote it. It was 500 lines of code and now it's like 50,000. And the amount of people that could walk through and understand that entire code base maybe gets to two hands, but it doesn't go much beyond two hands. At the upper bound, we're talking 20 people, right? Deeply, intricately, like understand all of those mechanics. I want to say that this episode
is special because we're just kind of doing this cold to get the momentum going and not over plan and just get something out there to start the conversation. These future episodes, I think our goal is to either have those people on to be able to ask questions and just have a good conversation, or you and I are just going to pick something like linear algebra and how it intersects with LLMs, and we're just gonna spend two hours talking about the math around it. Right? Like, we're gonna this is like a unique one in the capsule because this is just us getting the momentum going, getting started, and kind of at a high level talking about the thesis of what we're trying to do here.
So like that, our goal is to get the people who write the Lipsch P library on to actually talk about what they're working on and like the work that they did. Like how does it go from a 500 lines of code that Satoshi wrote to the 50,000 lines today? Like what does that journey look like? How does that all work? And you and I doing a little bit of like poking around and research ahead of time so we can ask some good questions and getting into all of that and all of the different things they have to think and account for. You were talking about like being able to listen to this podcast is like an anonymous person and being able to
being able to like without like putting yourself out there like learn clumsily along the way. What I'm actually looking forward to is people calling me out and checking my math and improving my understanding. I'd like to actually learn publicly and have like an open dialogue, have conversations because the way my brain works is I I just have a bunch of axiomatic principles of like how things work. And then whenever I'm thinking it's improv jazz assembling concepts and building on top of those axioms. So whenever someone like says, actually, this thing you thought was true was not true. I'm like, amazing. Thank you. I can now iterate and improve all my understanding of how math or Bitcoin or life or anything else works, right?
I think And they will. They They will. Gonna mutually benefit from this. It's not just we are we are far from a priestly class explaining to you how math works. We're going be clumsily stumbling along with you guys, and it'll be a great opportunity for everyone to learn, whether it's listeners calling in or getting someone who's like a cryptographer on and correcting our misunderstandings of how things work. Right? I think that's the missed opportunity is that whenever you get these people on podcasts, it's from a marketing angle of like, here's this new product thing and whatever,
but it's never like the math. It's never the actual like logic of it all. Right. We're like, oh, Frost is here. So the cool thing about Frost is you can do off chain key rotation and you can do a two of three and make like a single stick. It's always like the end product marketing bumbling. And it's not why is that safe, how is that secured, right? And that's I think what we want to do is we want to kind of zig where everyone else is zagging and we want to kind of go deeper in that side and not run away from those technicalities because we think there's a lot of beauty
in being able to find things like that and how it enriches our lives and I think it just gives us all a better understanding of how things work. Yeah it's so funny man. So you mentioned libsecp library. That's the whole that's the reason I'm here and I exist I think to anybody listening because my story is, you know, I got into Bitcoin pretty much almost four years ago to the day. And, know, the first thing I did when I came in here was
start verifying, which is what they tell you to do. So I spent the entire year reading books. You know, I found base 58 and that was amazing. So transactions. I got really good at decimal hex and all that stuff going through all that too. Didn't know what endian meant and all that stuff. That was really cool. Got into Jimmy Song's programming Bitcoin book which I that was like the first time I think I even ever got introduced to a finite field and I almost put the book down after a day and then at some It's chapter one. Finite fields is chapter one of that book. Yeah. Chapter one and it didn't click until it did and then I was like, oh my god, this is wild.
The other like, I always shout that book out because it's one of the few programming books where if you go on the GitHub, the code still works. It's so many books you'd think it just doesn't. So shout out programming Bitcoin, great book, people should get it. You know, I go to the Bitcoin GitHub once I start knowing what I'm doing, and then I go I look at all the formulas. It all makes sense. Things are nice. And then I said, okay. Now now it's time to check out LibSec,
the libsec library and I go on there and there's no formulas. There's just nothing but like what looked to me like millions of hard coded numbers separated by commas. And that was truly like I was already so I already like spent a year getting myself like all in here with Bitcoin. And and I'm like, I'm starting to I'm like, oh, I'm gonna validate everything. This is gonna be great. You can validate it. Everybody says you can validate it and I'm doing it. And so I'm building a big foundation and then I hit this libsec library and I just panicked. Was like, oh my god, this I'm never gonna validate. How the fuck am I gonna validate that? What what is this?
What even is this? Okay? Am I really dependent on understanding what this is? And, you know, I bought a cryptography book because, like, I guess I'm gonna have to do that. I'm gonna have to learn what cryptography is to figure out the hell is in this thing. And that was the beginning. So in the cryptography book, I've I'll attach it here. I think it was the Par Impelzl, which is a pretty good mathematical cryptography book. Parm Parmpazil, p a a r and p e l z l. We'll put it in the show notes.
That showed me all of, like, the abstract algebra and the number theory that goes into cryptography. And I was like, oh my god. I never learned any of this before. This is crazy. And that began this rabbit hole that I'm in with fully with math because it's like, you don't just you don't just get a book on abstract algebra. I mean, I'm I'm I am on my eighth book now and I'm still like, just still barely, still barely, it's like mixing glue, but like with moments of like, oh my God, algebra, abstract algebra, number theory,
all of these things, it's a beautiful thing to be engaged in. But I'm not sure how much closer it has gotten me to validating this LibSecP library. So maybe also one of the subconscious things I did was start this podcast with Gary only to bring it forward a year to them to now be here to actually start having these guys on the podcast so we can say, hey, what the fuck is going on in that library? How do we begin? How can you help us? Because here's the thing. Okay? If you can't validate everything,
you are gonna have to trust somebody. And, you know, there's going to be I really do believe there's gonna be some some potential threat attack where we have nothing to go on except like Andrew Polster tells us it's okay. Right. And, know, I I've I've met Andrew. I actually I believe he's sincere, and I believe he's I I'm very impressed with his caliber, but I don't wanna be I really don't like the idea of not being self determined fully. And
I don't know if that's an attainable goal for my life, but I'm going to I am like literally going to spend my life building myself to have a chance at being that. And maybe it makes a difference for like a thousand people in the world who look to me for the answer and don't want to trust Andrew Bolster and maybe 50 more people like me start podcasts and start decentralizing this effort. But like you got to start somewhere and this is where we are. And I think so I think we're in a great spot
to start getting some answers and to start really moving the needle on this. It took three years. Yeah. This was the this was one of the key load bearing ideas in that conversation we had that is like the podcast, like why to do the podcast is that genuinely there are many parts. Anyone who is involved in Bitcoin, I'll say this, anyone who has exposure to Bitcoin, even if you're buying the ETFs, you're you're believing this stuff, right? That the idea that you can keep money safe in a certain way. And for most people, they just say
someone told me about it. They're good with computers, the end. That's my, know, little Tommy always helps me fix my Windows computer when it bugs up and he said Bitcoin's cool. And because of that, I think Bitcoin's cool and the tech works on a distribution scale. That is over 90% of people who do anything with Bitcoin is for sure. They have one person and that's their that's who they are delegating and outsourcing all of their thought process to.
Yep. It doesn't scale to have every single person go super deep on all this stuff, but there needs to be a path for people who want to be self determined to be able to walk down it and understand it. And I think tools like LLMs and the ability to have like a custom tutor at a moment's notice explain things to you is a great way to flatten that curve of difficulty. But we want to be able to have more people. And I experienced a lot of this in building AnchorWatch just in, Okay, I understand.
Yeah, you know, you have BIP 32 and you have your private key and your public key and you have your seed phrase, which kind of encodes your master secret and kind of walking through this whole thing and okay, like I'm going to now build a wallet. All right, so if I'm building a wallet, I need to understand. And that was a whole process of I understood these things abstractly, but writing software to actually do that thing brought a new level of evaluation and understanding. And not every single person's going to build a wallet and get super deep into the details. But the idea though is that you shouldn't have to take that level of a swing to have a path to understanding this stuff better. And I think these conversations and
the concepts we're going to dig into and the people we're going to have on is going to provide a nice way for people who wish to learn more and have a approachable rung up the ladder to be able to start climbing it for themselves. Yeah. I think less and less people that run Bitcoin wallets in the future are gonna have any any idea about the math at all. You know? And I say that just because of how good the vibe coding is now. It's only gonna get better and only, you know, the more if you look at a pie chart of today of people building
in Bitcoin, they're probably pretty mathematical. They're not in the priest class and that's why we're hoping to at least try to get us all closer to not fully trusting the two or three people that really understand it. However, that pie chart is gonna go much, much more into the you know, non mathematical, having less and less mathematical ability. And when I and like, even like people who do like object oriented programming understand like abstract algebra a little bit. They have to, because
you're just understanding that you're preserving certain operations, you're preserving certain qualities in different objects. That's just going to go it's just going go by the wayside and the thinking. And so what I really so there's the podcast, which is a support beam for the overall Math Academy, which just aims to be a structure for people to continue to learn math, even though their avocation may not naturally be building that anymore. And you're going to
probably teach yourself how to code or teach yourself how to build things and without going to college and having to even go through basic any real basic math classes. We're going to create a structure here that really helps not only like it's not just like, oh, I need to understand certain math to implement this thing or to apply it this way. It's a way of life. Right? It's like caring caring about being right, but not just like not caring about being right like as an identity or identity politics or anything. Caring that you know you're right, like caring about that and being curious
is a lifestyle. And dude, I know people want this. Was at the Bitcoin John last week and which you presented at and Matt let me gave me ten minutes to present the math website And the response I got was so it was the most enthusiastic response of anything I've ever gotten. Last year, wrote a book and put it out and I got to go speak at a bunch of conferences about it. But nobody's
nobody really gives a shit. You know, like nobody like, you know, even like things that people think will help like Bitcoin adoption, people like like that. But I didn't get I got a different sense of enthusiasm from the math from this math effort. My off the cuff theory on this is that we've had five years of bitcoin content creation of people writing books and putting their own spin touch and flavor on it. And I think the introduction of these tools like LLMs
raise the bar of the level of effort that's required to get people's attention now. Like when your book came out, it was like the it was probably like the thirtieth like institutional Bitcoin narrative or maybe like 20. Right. But 20 or 30. But people haven't actually done the effort of building out a website yet, and that's way more approachable now. Yeah. And you want to find new modes and mediums for being able to interact and reach people and something reading a book, like an audiobook, maybe it's passive.
If you're sitting down reading, it's a little bit more of your active, your full attention. But it's a very democratizing thing. I think many people who did not write software are now going to write software and people who are already writing software are going to level up their own. Like everyone's kind of going one step up the curve in their own ability, the accessibility to kind of step up one level in their sovereignty and their understanding and first principles thinking for everyone, everyone's taking one big leap up. So people who probably may have been bashful
a couple of years ago about going through a coding website now realizing, wait a second, I could take this and a little bit of something else and I can kind of keep leveling up. I think it's an interesting development. Think building out the website is what you're doing now is going to be like every single person, instead of releasing a book, going to release an interactive website and they're going to start leveling up like the consumer experience for all of this stuff. I've been thinking about that myself for starting a blog. And the one thing that helped me off was like, yeah, even if I get the words down, I don't want to half ass it on a medium blog. I want it to be interactive. I want it to kind of like tell a story or using it. I'm realizing, wait a second, now I can write the blog and I can make it own interactive website pretty trivially
and kind of kick something off. So I think that's you're sensing where the trend is going be going in general for people that are putting out content. It's going be putting a higher level of effort and making it bespoke to your voice and getting your mission across, which in your case is, you know, in leveling people up in their understanding in math. For sure, yeah, and you know to that end, I don't know if I've told you this, but one of the things I've built now for the website and in general, I built a NostrilBot of me. I have I uploaded
180 podcast episodes that I've done. My book, all of my Nostril Twitter and LinkedIn posts. Yeah. And now I'm starting to test it out. And it's like, feel like, because again, like I just made a website of all these courses and I'm like, yeah, you know what, in six months, everyone's going to have a website with a bunch of courses on it. What's I feel like for this like it's us, it's the context. That's why when I say the context is decisive, I still think I'm probably the only person who cares about teaching math as a liberal art.
And I think that's super important and I think people want that And that's so I'm going to continue to find ways to do that. And, you know, like this podcast is such a it's such a key support beam. Because you can't talk, you can't have humanity. I don't think you can really get humanity across in any other way than conversation between two people. Well, think that just goes to like our sapien OS. I heard someone use that term recently like our monkey brains like oral history long before our ability to
even write. We've always had conversation and I'm blanking on the exact context. It was something I was listening to. It's going to kill me now, but I don't remember the exact reference, but basically that there is a higher people place higher trust on people who they listen to rather than read words from. And it's you kind of it's why podcasting is such a powerful medium is that you kind of pass like the monkey sphere barrier and you seem like a real person if it's audio, whereas if it's written word, you understand like your brain's able to associate that it's like something distant. Whereas if you hear two people talking and having a conversation, it's an acceleration for
building trust in relationships, even if it's a one way thing because you're putting the content out there and there's always an asymmetry from listeners versus actually producing the content. I think for what you're working on with generating all this math content, it's your way of getting like this big force multiplier and the podcast is part of that distribution to be able to get that idea out there further because people aren't going to organically stumble across a coding website unless they follow your feed. The podcast is a different angle of getting that distribution out there.
I also think it's really dangerous to train to give people the skill in math without the humanity. It's sort of like training psychopaths, know, it's like training giving somebody military training without any, you know, without Well, any it's so or everyone who works for all of the tech companies who just implement their code to who implement algos that are kind of destroying humanity, right? Right. Like who would willingly do that? I don't think anybody would willingly do that, I just think that you have peep that you know, this goes back to people train themselves in math and they need jobs and this is who hires them.
Yep. And what you get, right, what you get is a world that is devoid of, you know, is devoid of humanity and you get really psychopathic implement and really good targeted implementation of things that are probably bad for civilization. The perfect example of this is Chamath when he was leading a group at Facebook, right, when they're privately held it was a small massively growing company. He doesn't let his kids use Facebook, right? He understood the neural pathway and like the brain hacks for the growth and the engagement and the activity but knew it was corrosive so his own children don't interact with it. And I think we're going to probably see our own version of this with LLMs and bots. I think
I think the idea of an AI tutor is a really powerful concept. Like, if I was back in seventh grade failing math and I was able to have my own conversations and work my way through this, I probably would be able to accelerate my learning. And there's definitely utilization of this, but just unrestricted, ungated LLM image generation access like in general is going to be pretty corrosive for developmental age, right? Yeah. You don't want to be like learning your own because it goes back to this conversation I'm having with GPT about the difficulty. It's finally copped up to me and said, well, probabilistically for Satoshi to find a block with this level of hash is is a zero point four percent chance that he randomly at first glance found that. And to me, I say zero point four chance technically possible. Yeah, one out of two fifty. Sure. Like it could have technically happened, but I'd like to like I'm able to construct my own understanding of this knowledge and kind of piece these things together. Whereas if I took the first answer, I would have shut down an entire line of inquiry. So you need to have your own metacognition
and your own first principles thinking as you wrestle with these tools. And when you're really little, that's what you're doing as a kid is you're learning like it's the scientist in the crib. Like when they're playing, they're understanding how the world works either socially through play or physically with I take my bucket of sand and I make a sand castle and I add war and it falls apart. They are learning by grappling directly from a first principles basis their sense organs, how the world actually works. And if you just curate it, here's all the answers immediately, don't think for yourself, you're going to short circuit and fry brains. So I think we're just like with Facebook, people like we're going to have to wrestle this technology and find a way. And that's what we're trying to do here is have people actually grapple with the difficult first principle things for themselves.
It's challenging and taxing, but trying to give them a pathway to explore these concepts with their own kind of metacognition intact. Yeah, it's funny. In Waldorf education, when I started my kids in it, this was 2015 when I started sending my kids to Waldorf schools, and actually the reason I did it was I was I was reading a lot about Silicon Valley executives not wanting their kids to use their devices. And I was like, that's interesting. And then they're like, and then they're like, they send their kids to these Waldorf schools where they don't allow these devices. Like, that's really interesting. I started looking into,
know, there's a strong, first of all, there's the literature in their pedagogy, if you go back to like Rudolf Steiner and this is an extraordinary, he's an extraordinary thinker and he left a lot to, for us to learn from, but like even just the idea of teaching something to a child, like teaching them how to read is an injury and there's an age appropriate time to do it And it's like, I never I never thought my my older sister taught me to read when I was three because she wanted me to be dominating
this other girl across the street. And, you know, it was like, that's just how that happened. But like, you know, according to, you know, Steiner's world, it's an injury to actually teach somebody the word for something. It's been like, so there's a certain idea what that you just let them live and experience life for themself up until certain ages where you then say, you know, this is a chair.
Right? This is and you know, like that it's like a break. It breaks. There's something there's a there's a certain self that actually gets injured. It's not your physical self. You don't bleed. You don't break any bones. But there's a certain aspect to a self that does get injured when they when you learn something. And that's I think that the powers that run our knowledge system are well aware of this, right? And there's certain you know, I think we all are all fairly broken people
that you know, we're taught a bunch of things. And I say this all the time, like my parents parents lived in an eighty year cycle where if they just trusted everything the government told them, for the most part, they they won. They won and they didn't have to really think that hard about it. Now it turns out my dad was a PhD in electrical engineering and you know, taught me a lot about doing hard things, just the value, you know, a lot of a lot about it. Didn't teach me much about, like, necessarily the humanity required, but
instilled strong value still and that, you know, like, he would he would really tell me a lot about what it was like to have to get a PhD in electrical engineering. Like, why, you know, like, it's difficult, like, how hard it was and how much you know, how beat down they get and, you know, how humble that you have to be. You know, one of the, I think, things that I think is interesting is if you look at, like, math students at Harvard versus MIT, you know, at MIT, like they all think they're shit. At Harvard, they all think they're the greatest thing in the world, but like who actually who actually does things? Who actually,
you know, who adds to the body of knowledge in mathematics, it's the MIT guy, right? And they all, they're just told they're absolute shit, and you know, there's a reason for that, right? Like for us too, it's like there's no high priest class, you have to be hungry to and desperate. I would consider myself hungry and desperate and that's what I want to get across to people, like if you're in Bitcoin and you're not hungry and desperate to understand some of the things that really your your livelihood is going to depend on, like I want you to kind of think about that.
Right, I'm hungry and desperate because I've based my I've literally put my life savings beneath the protocol that I don't fully understand and can't verify. I would echo and agree with that compounding with not only my life savings but also building a company around all of this stuff, right? Like I am fully in. Like there is no Yeah. It's right when I started the company it was the idea of burning the ships. Like there was no going back and to make this actually work, it had to be done with full force of effort and concentration
and being desperate to make it work. Because I think that going back to MIT example, I think being privileged in the sense of being told that you're special and you're great isn't going to derive greatness. Like as with any art form, the constraining of the canvas is the domain in which you make the art. You need to be pushing yourself with a survival instinct of sprinting forward for a chance to achieve greatness.
That is ultimately what is required of anyone in any domain or field. They're not going to just like passively roll out of bed one day and be like, oh, I'm awesome, I'm great, and I found 10,000 new innovations, right? Like you have to have this constant panic sense of driving you forward of if this doesn't work, I'm dead. Yep. And, you know, if if you haven't, like, delved at all into the cryptography, it is like it's as ruthless of a medium as, you know, whether or not when you sit on a stool, it holds you up. Like Right.
It's either, you know, the door opens or the door doesn't. And I like to think of cryptography as digital physics. Interesting. So this is something we haven't discussed yet. Yeah. So it's like, in other words, what is the force required to open the door? And it's now it's measured really is measured in force, sort of brute force CPU cycles. And we live in an age where analytics can like there's certain
things other than total brute force that can lower the threshold for opening that door. And people think quantum might be one of those things. If you go back knowing that, you know, it's like you go to your first cryptography book, like knowing the letter E is the most common letter is one of those things that like, okay, maybe I can make some inferences that lower the requirement a little bit for, you know, to get this door open. But the reality is you still ain't getting that door open unless you got the kick.
Totally. No, I think that's an interesting way of like physics also just being the rules that govern our physical reality. Cryptography is what regulates the rules and constraints of operations for the digital world. There have been examples of this over time where entropy is a perfect example. If you're using cryptography you need true random numbers and what does it mean to get a true random number? This guy, Alex, his Twitter handle is
raw underscore avocado, Alex Walz. Oh, yeah. Has this great he hand wrote out all of the ways Bitcoin Core derives entropy for making your initial secret. And that came through a disciplined yeah. Like, there's four different parts. Low level processor instructions, entropy from the operating system, dynamic events and static events. And he wrote out this like color coded entire tree.
And the reason I'm talking about entropy is that entropy, if you it's garbage in garbage. If you have shit entropy, you're not actually securing anything. Right. And that's kind of one of these foundational rules of how the world works in when it relates to cryptography and digital security. And the reason why quantum is this like topic du jour right now is says, what if we can reinvent physics? Is basically what that question is. It's like, wait, if we can take all of these fundamental assumptions and we can discard them, could we do something better? And I think it's
interesting within Bitcoin right now because all of the existing people that are in Bitcoin working on development don't think it's a nothing burger, but they think it's something that like, one, we have time to respond. The compute isn't there yet, and they're working on contingencies now. But like a lot of big institutional investors just walking down the street are just like, what about quantum? Right. Quantum going back to this mental model I had earlier that some people aren't thinking from first principles. They have a spell book of sayings
and that allows you to kind of shortcut independent reasoning because you just say these words like rhetorical words like spells like you're a magician. Quantum is one of those things because no one who no one who is saying that quantum is a threat to Bitcoin understands how the digital elliptic curve cryptography works today, nor do they most of them really even understand how the quantum stuff works. They just get to have this like magic spell and say, ah, what I've been able to do by saying quantum is a threat, I get to take the entire cumulative knowledge base and understanding of how web system security works,
and that will become irrelevant. I don't have to understand it. It's now irrelevant. And you're able to just say the word quantum and you're able to kind of like wave your hand and just throw out any of the independent reasoning and thought process that built this cathedral of how cryptography works and say, but I can break it hypothetically in the future. And it's its own version of like, I think quantum is like a legitimate thing to consider and think about in the development of Bitcoin, but it is slop. For You just say the word quantum and say, shut up, nerd. I don't have to understand how all of this like math and entropy and public private key cryptography works. I just say the word quantum and I say this magic word quantum as a spell and everything you've said is irrelevant because quantum.
I don't understand physical qubits. I don't understand logical qubits. I don't understand Shor's algorithm. I don't understand the asymmetry in this of PNP equal NP. Like, I don't worry about any of that. I just say the word quantum. Right. So I think we'll probably do an episode or two over the coming year on just quantum stuff and probably starting with libsec p and how the thing works today. Then we could talk about the emerging tech and kind of where these assumptions start to break down. Yeah. It's a good example. Because it comes down to how strong are your legs when you are faced with a conversation. And so we're not gonna we're not skipping leg day here.
That's all. And we're we think we can really start doing some deadlifts here to show you have a leg to stand on. That's really what this is all about, right? We're too invested in this thing to have a lot of smart people run away over a conversation that honestly, they just listened to a podcast for a couple for a few weeks would hang in there a little longer. Yeah, absolutely.
Yeah. And I'm looking forward to getting into this stuff because I think it's going to be the first time in a public domain I'm going to be using large language models to like draft out like the concepts and the script and peer reviewing it too because just like I just before with the interrogating GPT about the Genesis block, I'm going to have to use it to maybe accelerate the template, but I'm still have to interrogate it and get in there to get like a nice tight script for a conversation and making sure I understand things and I'm not taking things for granted. I think it's going to be a really good I think we're going have a lot of fun with this. I think between us picking random like things that we're interested in and going off and building conversations around things that maybe aren't directly related to Bitcoin.
And we were just talking about graph theory the other day. Yep. And this is like maybe a small tangent. This is a small little nugget of a future episode. But the way large language models work is you have this large matrix. We'll go to this in more detail later. But when you like upload a document to chat GPT and what it'll do, it'll do what's called vectorizing. It'll take the text and make it into numbers so it can be put into this large matrix multiplication. You're dealing with high dimensional space, but once you start adding
over, I think the number is like once you get past 10,000 documents that you want to document within your like LLM AI work, the model starts to break down because they all start to compress and seem very similar to each other. So at that point, the large language model is just guessing what context is relevant. And what they're finding to be really impactful at the moment is graph based knowledge. And think of graph based knowledge, not quite like a graphing, like a like a a like a a equation, but the idea that I've I have a node, it's called Bitcoin. And then on that node Bitcoin, maybe I have keys and then maybe I have like network operation. Right? And then in the key section, have public private keys and you can kind of I know I'm visually doing this and this is an audio only podcast,
but the idea is that if you have concepts and those concepts directly are linked to other concepts, you can kind of view each concept as like a circle and then each link is aligned and you can get this like web of different ideas. LLMs are becoming way better at this graph based knowledge using things like Obsidian and the Obsidian Model Context Protocol, MCP, to get more complicated concepts because rather than it making a random like dice roll guess because it's seeing this cluster of numbers in like a vector database, it's organically
navigating a knowledge system the way humans think as they think about related concepts. And so it's a very funny thing because it's the way the human brain works. Also, it's something that has really high impact if you're leveraging these tools, and it's all because of understanding math. So by understanding math, you can understand how to better leverage these tools and systems, whether it's Bitcoin, LMs or anything else. And I think like we could do we easily could do a two hour podcast just on graph theory and the vector database and like this little thing I just did. We usually could do a two hour thing and everyone could walk away with a better understanding
of why the magic box that you type into and gives you answers works and how you can better leverage it for your own sovereignty and your own thought process to empower your own metacognition and not just being one shot by the LLM. There's a lot of people who say I've had conversations where people are like, well, per chat GPTX, and I'm like, that means nothing to me. Like, you might as well just say per TI-eighty four calculator this. Like, I don't like, you need to get you need to show your work. It's just like in math class. You need to show your work. I'm not just gonna give you full credit for showing me the answer, you need to show me your work, and I think that's kind of, this whole podcast is an effort of us showing our work.
Yeah. And navigating the concept for yourself. I'm so tempted to not say anything and let this hour and twelve minute thing end. But I do want to say one thing. Go for it. I have to. Okay. Because I think a lot of so I think it's important to understand learning in addition to the machine. Sure. And so I would the first thing I was going to say was the this like first generation of LLMs, when we say it was vector based and linear algebra, it's also probability. And so I think like we got to like we we want to be able to understand
just sort of, and that's like, you know, I know I sold actuarial science short, but this is probably really where that education sort of shines and being able to understand probability based outcomes. But when it comes to learning, right, what you were describing with the big vectorized databases was one, are at least two big dimensions in learning. First one is pattern recognition and that's pretty much all these things ever did for a long time. Pattern recognition,
that's linear algebra and probability and that's really where your LLM is just a more advanced version of your TI-eighty eight calculator in terms of it does, it's thinking a little bit more about what's the likely answer, but nonetheless, it's still doing pattern recognition. So the other big piece of it is what I call intelligence, but it's more of concept connect, like connecting disparate
fields of pattern recognition together to get some sort of orthogonal, orthogonal meaning unrelated, completely unrelated new information. And that's so like, there's a back to the website, there's a class called the introduction to statistical learning, and that is based on probably the book that anyone who learns machine learning uses. It's actually the intro book. There's a second book that's really the seminal book that people use, but this is an intro book.
I made a class out of it, and I only say this because you can go to that first page and there's a sort of canonical illustration called the bias variance trade off. And what it tells you is that no matter how you tune a model, right, even in the most optimal sense, you're basically at best going to be 40%, 50% signal at best, right. If you really tune it between like what do you want to trade up, you want to trade off bias or variance, best still you're going to have a lot of noise. And that's why when you say it gets to a certain size, the noise is overwhelming and it just can't it hits a limit of what it can
of how much new information it can actually allow into the system. And that's really where then this jump. And that's when people talk about like intelligence, I believe they're describing this jump where now because they've this jump where they're allowing new information into the system has now, you know, sort of reset the base camp for what they can now pattern recognize again on, and it just expands it. This is like, you know, if anybody listens to my other podcast, Rock Paper Bitcoin, and I said that I wanted to 10x myself as a podcaster last year, but I knew it wasn't gonna be by doing 10 x more episodes.
It was I I created this term called orthogonal divergence. And I said, well, if I do a bunch of disparate podcasts, like a math podcast, that's 2x. I do a fish podcast, that's another 2x. And then a coffee podcast, say that's another 2x, right? And this is like how, like this is how you, you know, it's not just from overwhelming 10 x ing the amount of thing that you're already doing. Right. It's creating a new dimension, right? It's creating a new axis, right? And that's how, and that's so to me when you're thinking of learning, right, the pattern recognition is typically the thing that is always what people consider learning.
And that's what you're taught all through school, you're taught, this is what you're taught to do, recognize patterns. And then it's these, but it's like the ability then to bring in these, to create new dimensions, right? Yeah. That is really exponentially expands what your learning capacity is. There's a quote and I'm making sure isn't there I'm
gonna just attribute it to I think it's an Einstein quote that basically like showing mastery in something is being able to explain it to a six year old And that's actually deep knowledge because you can't just straight pattern recognition, all these complicated concepts and explain it to like a child. Actual intelligence is transposing all of that pattern recognition into a new field domain and being able to explain it to someone who who doesn't have a Ph. D. In physics. Right. And I think that is actually like I used to be a math tutor and that was always one of my strengths was taking a complicated concept and recontextualizing
it in an entirely new domain. And that is not like that is a form of intelligence to your point. Like there's pattern recognition, understanding that piece. And I agree with you that most of the education system is trying to fill out the worksheets and understand the patterns and get to the next checkbox so you could do the next thing. But the actual intelligence is being able to like orthogonal divergence sounds like a version of that, right? Being able to take Newtonian physics or like Einstein relativity and transposing it into an entirely different domain and being able to explain it from those principles.
And I think that's a sign of a good educator. That's a sign of someone who actually deeply understands the concepts because they understand where the rules are and when to bend and break them and where it still makes sense. So I would I would concur with that. And I think that's part of our journey here is we're trying to transpose all of these abstract mathematical concepts into a new field of domain so that people can have a conversation, hear a conversation over a podcast and maybe act as a way for them to be as a stepping stone to building out their own knowledge base for themselves, to think for themselves. That's ultimately what we're all looking for here is people just thinking through things for themselves
and not being intimidated by the math and understanding that it is a foundational part. I think I agree with the liberal arts experience and being being human is to understand math, and we're just trying to provide a new form factor for people to be able to level up on that. Love it. Got it here. Awesome. Peace.
In this episode, we picked back up on our red-team tour of Bitcoin cryptography and tried to make sense of the last few weeks of fast-moving math news. We unpacked why the recent excitement around the exponential time...
OpenAI just dropped roughly 300 math papers with Lean proofs, mathematicians are crashing out, and the vague-posting has started: "math is breaking, so cryptography is next." Brian and Rob Hamilton refuse to hand-wave....
Magic Internet Math: https://www.magicinternetmath.com/ Shout out to https://x.com/uncleJim21 for the Alice in Wonderland story and the following rebuke of Lewis Carroll: https://www.youtube.com/shorts/33FYXpEWgyA This...
We’re back because math keeps refusing to stay in the classroom. In this episode, I unpack two very different kinds of “math pop culture.” First, we revisit entropy, seed generation, and wallet security through a...
Magic Internet Math is back. After the summer that set the Bitcoin world on fire - the Cold Card entropy disaster - Brian and Rob Hamilton (AnchorWatch) reunite for a two-hour deep dive on the event Rob spent three...