Skip to content
PodSubs
ShowsPeopleGuestsJobsNewsEvents
More
BlogPodCred
Menu
ShowsPeopleGuestsJobsNewsEventsBlogPodCredNetworksSign in
Join nowSign in
  • Shows
  • People
  • Jobs
  • News
  • More
    • Guests
    • Events
    • Blog
    • PodCred
    • Networks
    Join nowSign in

Magic Internet Math

Magic Internet Math artwork

Vegas Recap and Elliptic Curve Point Operations

May 7, 20261 hr 27 minAudio
ShareXLinkedInFacebookWhatsAppEmail

The Study guide: https://ecc-study-guide.magicinternetmath.com/guide.pdf In this episode, I (Brian) finally introduce myself properly and Rob and I kick off with a Vegas + Bitcoin++ recap before diving back into Chapter 6 of our elliptic curve study guide. We talk small, dev‑focused conferences versus mega‑cons, the Hoover Dam power-plant tour, and the standout conversations around quantum risk, BitVM, and Binohash. We also share plans to bring a math‑track BitDevs to Philly and to run a hands‑on Codex32 workshop soon. Then we slow things down to the math. Using the curve over the reals as a warmup, we revisit the group axioms that make elliptic curves useful for Bitcoin: identity (the point at infinity), inverses (reflect across the x‑axis), closure, and associativity. We sketch point addition and doubling, why doubling accelerates scalar multiplication, and how this geometry-algebra fusion underpins private→public key derivation. Along the way we touch BIP‑68 relative timelocks, why secp256k1’s simple y^2=x^3+7 form is performant, and where post‑quantum work like FROST and research at ZeroSync/Localhost is heading. 'Magic Internet Math (site)': https://magicinternetmath.com/ 'bitcoin++ (developer conference series)': https://www.btcplusplus.dev/ 'Hoover Dam (Bureau of Reclamation)': https://www.usbr.gov/lc/hooverdam/aboutus.html 'BIP 68: Relative lock-time': https://bips.dev/68/ 'BitVM whitepaper (Robin Linus)': https://bitvm.org/bitvm.pdf 'Binohash: Transaction Introspection Without Softforks (paper)': https://robinlinus.com/binohash.pdf 'ZeroSync Association': https://zerosync.org/ 'StarkWare (STARK-based scaling)': https://starkware.co/ 'libsecp256k1 (Bitcoin Core library)': https://github.com/bitcoin-core/secp256k1 'Codex32 overview (Blockstream/Andrew Poelstra)': https://blog.blockstream.com/codex32-a-shamir-secret-sharing-scheme/ 'FROST threshold Schnorr (IETF RFC 9591)': https://www.ietf.org/rfc/rfc9591.html 'BIP 32: Hierarchical Deterministic Wallets': https://bips.dev/32/ 'An Introduction to Statistical Learning (ISLR)': https://www.statlearning.com/home 'Localhost Research (Bitcoin research org)': https://lclhost.org/ 'OpenSSL (project site)': https://www.openssl.org/ 'NIST note on removal of Dual_EC_DRBG': https://www.nist.gov/news-events/news/2014/04/nist-removes-cryptography-algorithm-random-number-generator-recommendations

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.

UnknownAlright. Welcome back to the Magic Internet Math Podcast. Hello. Hello. Hello. What's up, Rob? What's up, Rob? I would say about, like, 65, 75% recovered from Vegas. Just kind of bouncing back. The brain starting to, you know, get get back to normal, a little less stimulated, a little under oxygenated. You know, they pump those rooms full of oxygen to keep everyone bouncing and going. It's brutal. You know what I noted? Know You what I've something I noticed is that

I've never actually introduced myself in a podcast where I'm like, oh, this is my cohost, Rob. I am Brian. Never done that. And, like, the AI stuff never the AI stuff attributes my ideas to, like, my cohost. It drives me crazy. So that was for the AI.

UnknownThere you go. As as I think we're gonna start with a little bit of Vegas Bitcoin plus plus recap, and then we'll, go through chapter six of the study guide. Tragically, we've lost an episode, but that just gives us another chance to do it more concisely the second time around. So I feel pretty good about that. I just wanted I mean, we're here now. We're all here now. Everyone who's listening is here now, but I just I do need to eulogize the last episode.

Yes. Had a a two hour episode where we struggled and fumbled through trying to explain, like, the elliptic curve operations in chapter six. And so while while it's always disappointing to lose material and maybe and the reason why we lost it just for the listener's benefit, it was was that there was a technical issue, and Rob had echo. So he was coming through his mic and my mic, but, like, one full second

apart, and it was impossible to listen to.

UnknownYeah. It was it was rough. I I got the raw audio, and it was definitely It's my fault for having you coming through my is my I I know the setting and everything is my fault, but your voice was coming in twice, and it wasn't like a millisecond difference where you just you could ignore it. Yeah. It was like overriding what you were you know? It was echoing over myself. It was awful, unfortunately. We also had a good banter with Alan Farrington as the board as the, the pseudo board of the, Magic Internet Math Academy

podcast.

UnknownOnly three people have heard this episode. Yeah. That's right. That's right. It's nice to have Alan back in the fold, and we're definitely gonna signal. He so he'll he it's likely that he'll be our first guest. Yeah. He's a good he's a good warm up guest. Low stakes. We'll we'll find a fun thing to banter and chat about. If we can get it for the before we get deeper. If we can get it done before he just he jumps into his next rabbit hole, we will get this done in next maybe the next episode. Maybe. Yeah. The

other thing to call out is if you're hearing a study guide and you're new to the show and you're like, what the heck are you talking about? Magicinternetmath.com.

UnknownThe study guide specifically is this the elliptic curve one on the Magic Internet Math site? Yes. It's def and it's in all of the show notes. So you wanna go to the show notes, the first link. It's like magicinternetmath.com and then slash something like ECC study guide or something like that. But just click on that link, and it'll pop up a big PDF. Yep. And I'm assuming you I'm assuming anybody's here. You may or may not have listened to the first four episodes we did on this subject, but we've been going through this study guide for the last four or five episodes. That's right.

UnknownAnd so from there, we're going to do a little bit of there's a formula. There's a podcast I listen to that does movie reviews, and the first half is always they call dudes talking and it's just bantering. So today, instead of going right into the map, we'll do a little dudes talking about how we've spent our past week, and then we'll get into chapter six. I was in Vegas for, well, I was in Nevada for basically four days, three days before the conference in Vegas even took off because I went over to the Hoover Dam, which I intentionally set it up that way because I wanted to have a fun conference hanging out with my friends before I wore a suit for three days. Mhmm. And, I would say it was

it was a good time at the Hoover Dam, Nifty's Bitcoin plus plus was very informative, was always a good time. You get to kind of hear people working on projects that, you know, I would say Bitcoin plus plus is a very math heavy conference. There there's not a lot of there's no price talk. There's no econ, macro, global, whatever. It is just the hard facts of the protocol and advancements and what people are working on.

UnknownYou know, just for people's benefit, if you've never been to a conference or maybe the only conference you've ever been to is the big Bitcoin conference. If you ever go to one of these small focus conference conferences, I mean, it's like you can't even imagine the experience relative to the big one. It's it's like you're around the same people with the same focus for a couple of days. And to me, that's where I feel like you you end up with the longest lasting connections. Whereas, like, at the big Bitcoin conference, you run into people, you'll say, hey. Can I DM you? And it's like, yeah. Yeah. Yeah. Yeah. Yeah. But, like, at the

at the smaller conferences, you're getting, like, a lot of reps and doing, like, a really accelerated selection process, getting to know people, and making kinda, I think, longer much longer lasting connections.

UnknownI'd agree. I'm now looking through what we have going on here. Just really quickly at Bitcoin plus plus coming up, they're gonna have one in Vienna in May, Nairobi in June, and July is gonna be Canada, Toronto. And it's called Bitcoin plus plus finds consensus. So Really? The Toronto one might be might be doable. I have the most milk toast consensus change in that it it actually is just above the line of being, like, not doing anything and actually being impactful. As a small aside is I wanna be able to take

relative time locks Mhmm. Which is BIP 68. And when they encoded it, there's a 32 byte field to to be able to to flag these time locks, but they only use the first 16 bytes. And what that ultimately means is that you can only do a relative time lock for up to fifteen months, 65,000 two to the sixteen, sixty five thousand once every ten minutes is, like, roughly fifteen months. Right. So that's that's that's, like, the the the

When they did it, they were like, this is for payment channels. Who would ever need more than a year of like inactivity for a payment channel? And I'm like, I want to use it for a vault. I want to have a dead man. I want to have Bitcoin secured in a way where if the UTXO hasn't moved in three, five years, I can do something as a recovery mechanism. You can't really do that today with that. And so as I'm looking through the schedule this morning, I may look into researching, getting a panel talk for that. I think it's like the most slam dunk, straightforward thing that would like who's going to who's going to object and be like having a three year relative time lock destroys the network or changes Bitcoin as much? You know what I mean? It's like, it's just

UnknownI have people in my mind, but I'm not gonna say them now. Well, so the Hoover Dam was great. Did a tour of the Hoover Dam, the power plant tour, which was really, like, awe in the true sense, like, awe inspiring. Just like it was awesome. Taking that elevator down and and, you know, it only takes, like, maybe forty five seconds, but you're just dropping the whole length of the Hoover Dam, getting to the basement, seeing all the power generators. Was it like the whole like,

Unknownthere, or was it just you and a couple buddies? Two days. So was split across two days for the power I mean, the trip Hoover Dam, was that, like, a side trip that you and a couple buddies went on, or was that everyone went Like, that was the conference venue was overlooking the Hoover Dam. Like, it was at the conference venue was in was in the Hoover Dam behind, like, everything. Like, it Yeah. The pictures Pictures were cool as hell. The pictures are sick. Yeah. Yeah. Yeah. It was very awesome. And I think what there was a good, there was a debate with Reardon and Alex Pruden talking about quantum stuff

and if it was legitimate FUD. It was good discourse. There was something I really need to start wrapping my head around because I've seen this talk now three times was Misha's talk on pipes and, like, encrypted witnesses, basically a way that you could build these, like, off chain, like, verifier circuit proofs to be able to encumber a bit a Bitcoin UTXO with a covenant. It takes like a terabyte of RAM to run, though. And they're working on it, though. So like, I'm I'm very eager to learn more about it.

I'm. Yes, that conference was great. Was there for a couple of days. Got to link up with someone who was also at the conference. You who's also a big math guy. I think you got to meet, Sam Parker. Right? Absolutely.

UnknownYeah. It was awesome. Shout out. Very cool. Shout out. Sam Sam's a good guy. I met him on Clubhouse back in, like, 2022. And right he he was always the guy who would walk around the Bitcoin conferences and says, we need to find a way to be able to encode a zero knowledge proof circuit into Bitcoin script. And he he would always follow-up with, I have no I have no proof or thesis as to why this can be done today. I just know it is. We should start working on this. Like, he was just, a true believer. Like, a priori, like, was already at the end being like, this is something we can do. We need to work our way backwards to doing it. And then Robert Linus came out with the BitVM paper, and he was just, like, off to the races. He was, like, right there. Sam Sam is, like, for Matt.

UnknownRight? Yeah. He's just like, I know these things are self evident. I mean, I could prove them, but but you could you could do it just as easily as I can. It's I I don't have time for that. Right. You know, have other things I have going on. And so here's all of my ideas. Go ahead. Do your stuff with them. And then yeah. And then and then the next thing you know, an actual academic actually does pick it up. Yeah. Right? Exactly. You got to meet Robin Linus. To call sorry. Is it fair to call Robin Linus an academic? It probably is. I mean, but it we need to make it's like I I

feel like in this process of magic Internet math, we're I'm redefining what an educator is and redefining what a math teacher is. Maybe we're redefining what an academic is in this space because he's Right. You know, a different kind of academic.

UnknownAbsolutely. Yeah. I I I think it's academic in the sense his understanding of of Bitcoin is of a, you know, like a PhD level. Like, he knows exactly every single wart and bump and, like, little, like, like, a little quirk of how the whole system works when it comes to Bitcoin scripting. He is a true elite Bitcoin wizard when it comes to script wizard. He Bino Hash has been my funniest, like, little, like, sometimes, like, in the day, I'm, like, going for a walk and I'm just trying to like, I think about Bino Hash.

UnknownI There's a lot on my website now about Bino Hash. And I mean, part of it is we're trying to glaze them enough to come on. Well, I think he'll come on now. We we had a good time, like, chatting with him in Vegas. I think I think he he's seen our face a couple of times now. He knows we're real people who just love geeking out about all this stuff. I mean, hitting him at Opnix and already having binohash.magicinternetmath.com Yeah. Loaded into my site was, like, I think, rung his bell a little bit, you know, and then get a follow-up with him at the conference, you know, saying beef beef steak and

just just continuing to hammer the nail. Yeah. There's a two

Unknownand a half hour Stanford lecture he did. It, like, got a whiteboard talking about the intricacies of Bino Hash that I still have queued up in my, like, watch later list to to get through. My very selfishly, I want it in there. I want it I want it live and running, like, across the network with, like, good tooling because, someone from Starkware took the bino hash model and made a post quantum signature scheme. And it is like the ugliest script you've ever seen, but I don't care. Like, if we technically can move the Overton window from if we don't do anything, we're dead to we have an arc. We have we have a thing and it sucks, but it

but we could ride out the storm in this. That changes the entire Overton window on what is the next incremental step. If you have something that changes the entire threat model.

UnknownCan I oh, you know what? I'm gonna I'm gonna introduce something. We're not we didn't say we talk about this. Okay. But I'm connecting it. This is a this is a math thing, and it's a Picadillo I have. I'm I'm not gonna get into the whole Alex Pruden thing, except I have one I wanna just hit one of his talking points, and this is for the math podcast. K? This isn't for, like, our friends. K? So he has this one argument, which is if there's even a 1% chance

of this happening, we should really, you know, we should really turn on the money faucets and research this and make a plan. Right? And it's this argument I hear all the time. If there's only a 1% chance, if there's just a 1% chance. And I really think this is an attack on people's ability to figure out what a 1% chance is because, like, he's literally sitting next to Reardon on stage who's like, forget what James was was doing. I mean, James had his Duplexing. Was he was stealing the conference. Right? James stole the conference. But like, again, I want to just stay focused here and just say that objectively,

somebody like Reardon, everything he everything he says suggests that a 1% chance is a wild exaggeration of probability. Right? And so what is it about us where, you know, what is it about us and humans that we, for some reason, that if there's a 1% chance argument is actually effective with people? Mhmm. Because that's the one thing he's still clinging to. Well, it's also presupposing that 1% is the right number. Sorry. And just to set up, like, this is a you know, if every if anyone doesn't really know, this is this is this is the post quantum type conversation.

And, you know, there are there are entities inside the Bitcoin space that are, I think, you know, some kind of fear mongering their way into research funding. And I'm be giving a very biased take. But that's but there was a you know, there's a debate now amongst sort of people who you would say are balls and strikes types mathematicians and these people and the clash it was like Clash of the Titans on stage at both BTC plus plus and at the Bitcoin conference. But, like, there's this little struggle.

And what to me, what it where the whole thing really breaks down is this one point where, you know, the fear monger says, well, if there's only 1% chance that this thing can happen, and then we're talking about something that there's no evidence that'll ever happen. And it's like the if he's really effective, the most he's most effective with somebody like Reardon, I think, who's so intellectually honest, who, you know, is like, no, you're there is a problem. I just don't know why we're talking about this. Like, I just don't know why we're talking about. I don't know why you're talking about your little thing. There's a bigger problem.

And like, it's hard if you're intellectually honest, it's hard to just say just totally dismiss this guy. Right? And you're gonna just argue every point. And I think it would just be this like, there's not a 1% chance of this happening. Like, there's no way he would even say so.

UnknownYeah. My my my main discomfort and the thing that I tried to balance by my priors and my my gut here is I frame it up as I don't like being short the innovation of mankind. Right. Naturally. Like that is the reason why like that's what they're preying on, though. Of course, that's what this argument is preying on. Yeah. And you see the advancements in AI. Where are we going to start seeing breakthroughs in physics? Because, actually, another huge shout out, to, I would say, not a listener yet, but a kindred spirit of the,

magic Internet math extended universe. Talking with, Jesse Posner in Vegas.

UnknownAnd Welcome to the Kindred Spirit Club. Yeah. He, was I was talking to him about, like, large language models, because that's a lot of what he works on now at Vora. And he was talking about from, like, a general information theory world. The way in the words we communicate, like in these mathematical embeddings that there's and how we're starting to like pull apart from these like large language models, breakthroughs in other fields in that, like the domain of our language is like a hyperdimensional space with all of like these nuances and contexts and maybe even tone of language and like how you structure the syntax of your sentences

communicate like there's almost like a a universal weights and biases of how the worlds work, of how concepts interrelate and, like, this is a new domain and how we're able to link these things. So maybe advancements in large language models will be able to allow us to make breakthroughs in things outside of language because the harnessing of the language allows us to anchor concepts and to and build on top of that. So that's like the that's why it's a nonzero. The question is is what level of zero for a post quantum cryptography

cryptographically relevant quantum computer to be able to actually do something on that? Yep. From nonzero to a 1% chance is quite a leap.

UnknownAnd by the way by the way, I I just interlude, go if you wanna learn about the statistics and math behind large language models, you you can go to magicinternetmath.com. And I think it's might be islr.magicinternetmath.com, but you can go on there and find the class called introduction to Statistical Learning, and that's what ISLR stands for. It's and it's a reference to this book series. It's a seminal book series on the math behind all of this. Lovely. So check that out if you really wanna deep dive deep into that. Yeah. Yeah. So, like, I I just think exploiting our sense, and I don't remember if this was part of the lost episode or we actually talked about this on air, but the people have lost their ability to assess risk for themselves. And I was attributing it to the lack of to the

removal of coins, like removing coins from the world. Like, we used just flip coins and sort of get a sense of, what's the what's the probability of getting 10 heads in a row? Like, you know, it's like Yeah. And you start looking at the math and it compounds pretty fast. Right? And, you know, what like, we just don't have an we don't have an innate sense. I think we're very vulnerable to this argument. That's all. And I think I wanna I wanna point this out and hammer it so that the more of us are like, well, so where are you getting the 1% probability, sir? Yeah. Where are you getting that?

UnknownMy my split the difference on this is if you, you know, like Jonas Nick is working on shrinks and shrimps and all this post quantum stuff. A really cool thing I learned in Bitcoin plus plus was there is a pretty strong belief now. There's a research organization called Localhost. They're trying to think of most similar to a brink or chain code of, like, working on just big Bitcoin research. They're setting up an actual,

a a quantum, like, arm of their research working with, like, leading, like, physicists. But they're pointing them instead of just like saying we need to do something, they're saying, well, Jonas has this shrink and trims method. We may be able to actually make these kinds of signatures threshold signatures, meaning you could do a two of three, which would be massive because like all the post quantum schemes right now, to my understanding, they don't use BIP 32. So you can't have a master seed phrase that like links to many, many, many keys. You have to do each individual key generated as a one off and you can't do any sort of multisig in these things, which is agonizing.

So they're trying to point this in a productive direction of this is the domain of things that we care about with Bitcoin in a post quantum world. How do we make it better? Then we just basically have Jonas hanging out, hammering away on the arc, building it. And there was a it was a fun conference, though. Like, Bitcoin plus plus was good. Going to Vegas and seeing everyone was amazing. You wanna talk a little bit about your time in Sin City?

UnknownYeah. God. I I I don't like to bore people with my little experiences. I'm just gonna keep I'm gonna go high level here and just say, you know, it's the first So it's the fourth year that I did the Bitcoin conference. It's the first year I had this purpose. Like, it's the first year I really had this magic Internet math focus. Last year, I had just finished, like, writing my book, and I was talking a lot about that. And so I have a good contrast of how people respond

to, say, my focus on the book Bitcoin for institutions versus magic Internet math. And it's almost like it's almost like last year, I had a you know, I don't know. Like, I I was in a fire, and I had a burnt face and skin. And people are like, oh, you have a book. That's nice. And this year, was like, oh, people were genuinely interested in what I was what what I it's an amazing contrast that like, it's not that nobody cares about the book because that that came you know, it does come up, but it's one of those things where it's not in that that is not an important thing in this time to people I care about, apparently. Whereas

this is something that got me you know, it was like a conversation extender. And even people who I talked to about this when they see me later on, they're like, hey. You know, have more I have a couple of things, more math ideas I just wanted to bounce off you. And it's just one of these things that's really engaging. It's engaging people in a way I've never really experienced before. And so like for me to get this validation of, you know, validation of signal that I'm on the right track.

You know, it's like a bummer last year to be like, well, I don't know if anybody really cares about this book, you know? And then now it's like, it's very validating to, you know, the response that people have to the podcast, to the book, to all of that. And so then the other, so when you combine it with the show, the live show we did so I did we did a live show at Grassroots, but, like, I just wanna, like, dig in for a second because we were with all the meetup organizers, and that conference

also was me sitting for two days thinking myself as a meetup organizer. Like, what am I really doing? You know, I'm with I'm with people that are launching and running really big meetups, and I don't do that. I just run my little meetups, little hangout. And so you take that experience and all of the people I met in Vegas, and a lot of them happened to be bit BitDev organizers. It just happened to be that way because that seems to be who is gonna be gravitationally attracted to this conversation.

Right? It's just one of those lucky weird things. And I basically made a made a decision, and I've been starting to tell people I'm starting to bid devs in Philly. And, you know, I'm not like the technical guy really on transactions. Like, I'm not like Rob on that. You know? I'm not I have I have a ways to go to become like that, but it's like I think having a math track for BitDevs is actually it was a light bulb moment for me. Like what we've been doing, right? Why not connect to not connect it to bit devs? Yeah. We talked about this offline, but having been to

Unknowndozens of bit devs over the course of my life, Each city has its own kind of flavor and style as to what the BitDevs covers based on the strengths of the hosts. So I think it's very natural for you to do a math track for sure. Yeah. I'm excited. And the I've really been this motivated to, you know, start something. I'm like, when I started my own meetup, I was like, no sponsors. No one's gonna get you know, I'm like, dude, I need a perfect classroom. I don't care who has to sponsor this thing. You know? It's it's a totally different mindset now going in. I'm really ready to ready to build something big that's gonna matter.

UnknownI love it, man. Yeah. That's gonna be good. I'll have to find an excuse to swing over by Philly sometime as you're getting stood up and we can do one together. Hell yeah. Hell yeah. And so like that, I'd say overall, that experience for me was it was all the people I met. Obviously, I mean, it it really rung my bell. I have to say just on I it really rung my bell. The only panel I attended other than the ones that I did was the the main stage

quantum panel. Yep. And something about watching James operate there did something to me.

UnknownYeah. James, he's a true treasure in gem and Bitcoin. There's a longer conversation about him that I don't wanna take over the rest of the podcast, but I think he is a kind of voice that we need in Bitcoin coming from a totally different it's almost like his entire his, like, epistemology, like, of how he views Bitcoin that is just sorely needed, just sorely, sorely needed. So

Unknownthe term the term I have, and I've talked about this a lot in my other podcasts. So this is what he really what brought up for me was the term I have is standard bearer. Okay? And, like, how we we have two types of people in the world. We have diplomats who meet people where they're at, who we very much need, and we have standard bearers. And it's really alienating and hard, and you gotta be a 100 x the person to be a standard bearer than to be a diplomat.

And Yeah. That that's and you don't see it. You just don't see it every day, and you don't see it on stage. You just don't see it. You know, most people in polite society don't show don't you know, usually, it's like behind the scenes. Hey, dude. I'm gonna pull you aside. I'm a give you my standard bearer. I'm a chisel you right now. But, like, there's something about that moment where he decided he was gonna do it in public on the main stage. And it was obviously it obviously resonated

because, you you know, there are metrics in the world now that just can just validate what a big moment that was. But for me to watch somebody operate that way with the like, it was it was cinematic. You know? It showed me something. It showed me something about what what's possible for us. You know? It gave me a glimmer. Like, you know, we might we might we might win one day. You know? We're build on the backs of people with, like, that have the integrity to carry us. And I just saw I saw something truly extraordinary. I'm not trying to glaze him too hard either right now, but, like, it really like, that was that that's a takeaway I feel like. Like, I feel like I'm gonna remember that for a long time. For sure. For sure. Just a dumb carpenter, but he's, inspiring us all. Like another carpenter in history's past.

Yes. I was saying I was referring to Karen Carpenter who was very skinny, but, like, you know? Okay.

UnknownOh, man. And then I had your She could really play the drums, man. Yeah. And then you had your, ideal, crossover event with fish. Right? Oh, of course. For for a different podcast to do the full debrief. I don't wanna take too much off back off. No. I understand. But I will say, like, those are like, big time FISH moments are usually moments where big things, big transitions, big changes, big moment. You know, like you know what I mean? Like, big moves get made in my life

that when I look back, there's, like, five big moves that have ever been made in my life, and they've all happened, like, after big fish shows. There you go. But, like, maybe this one this one, it's interesting because the what's my memory of this weekend will much will be much more the conference than the fish show. There you go. There you go. The first time I could say that. You know, it's the first time I could say that. And so maybe there's there's a little bit of a changing of the guard in my maturity level. I am in my fifties. You know? People have been asking, like, when are you still doing this fish thing? Like, this is still, like, a big deal for you? And I don't know. Maybe it's becoming less of a big deal because

what we're doing here is just mattering a lot more.

UnknownI think so. I think it's there's a whole there's a lot going on with where Bitcoin is in the global ecosystem, the attacks that are it's kind of the posture and have to deal with now are way more larger scope and existential. There's big important things. The I'm trying to think. The there'll be I I did one panel on insurance at the conference. I I poked around and listened to a couple of them. There was, four different quantum panels. I'm gonna have go back and listen to most of them. I was running the news desk during that big one with James O'Byrne and Alex Pruden and

Hunter Quick, Hunter Beast and Reardon and Alex Thorne. That was a I was doing live desk commentary with Aaron Redwing. Did you get a chance to meet Aaron?

UnknownNo. You know, I met her last year. We were Oh, did you? Well, because yeah. We were we actually were doing a podcast together, a live podcast at Inscribing Bitcoin. No way. Oh, it's Yeah. I let Rod and Dick drag me to Inscribing Bitcoin, that that was another eye opening moment. Probably my lasting moment from last year's conference because I was, like, I was so you know, I might have been on, like, the I might have been on the line of, like, you know, BIP one tenor. Like, I could like, there's an alternate universe where I became a BIP one tenor.

And going to Inscribing Vegas and finding out those people are actually kinda cool, and they actually build on Bitcoin. They actually spend Bitcoin. They're super enthusiastic. I don't know. That did some that also did something to me. Seeing them as real people. Yeah. Totally. Like, know, maybe I don't wanna ruin my life to make sure I never have to interact with these people. No. No. No. No. No. Like, if you ever need a good ambassador, Erin Erin's

Unknownone of my favorite people in Bitcoin by far. Yeah. So we were sitting down doing a podcast at the event, and, she sat down. And that that's that's, like, the probably the most I got to meet her. Got it. Got it. One funny observation I learned in Vegas is that smoking sections are fake and gay, and that if you just have really good ventilated systems, you can just you'll be fine. Like, it's actually not that bad. But, like, I I I'm sorry I wasn't able to find time to partake in the cigar lounge with you. But,

it's very like, you you can walk through a casino floor and people have been smoking in for, you know, hours and hours on end, and you walk by twenty minutes later, you don't even smell it.

UnknownYeah. Well, I'm gonna shout out Keith Gardner, Brandon Carpalis, and Avi Burra, and the guys that were in the cigar lounge. The guys we we made at this the Venetian cigar. It down. We held Yeah. It At some point, they tried basically like, they were letting they were letting us sit there, and they were letting us sit in the sports book, smoke cigars. And at some point, they started enforcing, like, no. It's like $50 a person to, like, sit here. Hang up. Yeah. You have to spend $50 in drinks and food

to actually be here. So at some point, they started enforcing that on us. So we had

Unknownto had to adjust. No worries. Yeah. Migrate. Oh, man. I'm trying to think if I had anything else big to say about Vegas. It was good. I think maybe as a little peek behind the curtain, we're gonna work on some Codex 32 stuff. You had dinner at my house last month and you took a Codex 32 book. I swear I have a couple more of them around here, so I'm gonna have to go find them. But wanna do we'll do a Codex 32

preparatory, like, run through you and I will each go through. For those who don't know, Codex 32 is a project made by, Andrew Polstra that he basically made a series of paper vowels, like paper computers to be able to do Shamir secret sharing. So you could take an individual private key or like an XPRIV, like a seed phrase, and chunk it out across many shares. And with only a pen and paper, you can actually reconstitute the shares, which is an incredible ambitious pursuit and pure love of the game mathematics. That is someone who is for the love of the game,

understands this stuff in the 0.01% understanding in the world and wants to be able to give back in a fun, meaningful way. So we're gonna do that. I think we're we'll probably be able to swing on, Jesse Posner. We have plenty of you know, there's there's you know, we could talk Frost and Bitcoin stuff, but he's also super deep now into AI modeling. It was very interesting. He was basically looking at the latest academic papers of how many bytes a prompt injection has happened ever

in large language models, And he is trying to conservatively scope how can you have a secure large language model by piece by piece chunking in information in such a way that it would be smaller than any publicly observed ever prompt injection and using that as a way to help secure, like a private large language model. So he's he's also a not a classically trained mathematician. He's a lawyer who basically when he went and got his math education, it was all primary sources. So he spent, you know, years going through. He never even got to set theory. So he's just gone from, like, the beginning. So he's like Fermat too. Fermat was a lawyer. Fermat was not a professional

Unknownmathematician. I don't think they actually had professional mathematicians in the time of Fermat. There were, like, some academics. But it's not it's it's something we have to go into at some point in the history when we talk about math history. But, like, it's not like it is today where all the academics are piled into one centralized area where they can be sent on wild goose chases so that we don't you know, so they don't threaten the power structure. Right? But we didn't have it was much more decentralized in Vermont's day where you just had lawyers obsessed with or Pascal. Guys like that existed that were not, you know, in the ivory tower.

Let me just say something about Codex 32 real quick because so I was at your house. You gave me you showed you know, you go you, like, showed me this workbook, and it was, like, showing me, like I felt like I was in a movie, and I was being shown, like, the freaking holy grail. Okay? This coolest thing. It's like a first of all, it's a beautifully illustrated workbook with, like, a wizard on the cover and this kinda, like, lion, the witch, and the wardrobe, like, art theme to it. I'm like, okay. This is fucking cool. I didn't know I I did not know this was, a Blockstream product or anything like that. I just knew it just looked so cool. And you open it up, and it's like a pop up book. Right? But even like the

it's like, the even the text has, like you know, it's illustrated in an interesting way. It's a thoughtful book. And, anyway, it's all super cool. It's got things you cut out, and I took it home, and I was like, I'm afraid this thing's a collector's item. That's how it occurred to me. So I went on I I went on and bought, like, six more just so it can really, like, tear this thing apart. Right. Still keep one as my collector's item. Like, this is gonna like, this is like, it commemorates

the The moment. Commemorates the dinner we had at your house. That's

Unknownright. That's right. But I didn't wanna I I needed one to stay, like, I needed one to stay in good condition. But I'm really psyched to dig into that. Because those Books By Design are supposed to tear tear them all up and start using them. Yeah. Right. Exactly. They are workbooks. They are not textbooks. They are workbooks. You're supposed to take scissors and rip out pages and and make some paper computers. So, like like, this is definitely something I would enjoy doing in the bit devs, put perhaps. It's like one it's a codex thirty two workshop day.

Yeah. Right?

UnknownSick. Yeah. Very cool. Yeah. So I just have to buy like a 100 more. You know? So this is where yeah. I just I I need to find a sponsor who's gonna like really wants this thing to go to go well. Amazing. I mean, I think I I would just get started and then you just put in repetition, get compounding. You have the John, which is one of Tomorrow. The point community events. I have the John tomorrow. John tomorrow. And I'll be announcing the BitDevs tomorrow at the John. Sick. We we will we got How about you start the interest?

Yeah. Thomas and Marty will be at the John's. It's gonna be a big one. That my point is, yeah, it's gonna be a big John. It'll probably be sixty, seventy people there.

UnknownLove it. Yeah. Absolutely love it. Having been to a John, it's a special place. But you were at the John because what so the John does a bit devs like once a year. And so that's when I was there. Yes. And it was almost exactly a year ago to the day that you were there. You and S It was May. You're absolutely right. Yep. And that was a great time. We sat on the sidewalk smoking cigars, drinking scotch, and then we were in the pool hall for a while. It was just a great time.

UnknownYeah. But that was, like, the last BitDevs that we did. And we do it once a year. And I just I remember when I first when I first started with the John I have, like, a near perfect attendance. The only Johns I've ever missed were because of car accidents. That happened that happened on the way. Okay? Plural. It happened twice. But, like drivers. Like, the initial idea about a bit devs at the john was, like, Philly's too retarded

for for it. Like, Philly's not they they can't handle it. And then then we had one at the old Fergie's that Jay came and ran. Wow. Incredible. Yeah. So, you know, Matt was not messing around for the first one. Brought Jay.

UnknownAnd then It's all downhill from there. Well It's all downhill from a Jay bit Jay bit devs is what got me into Bitcoin in 2013. It is all downhill from a J Bit Debs. Well, this so then Jay and Evan did the second one, and hilariously, we were in that we now were in the Chinese restaurant, and it was like trying to compete with the karaoke was this is like a pro like Jay. Like, only a pro like Jay could navigate this

and still do a bit.

UnknownThat that mean for for those who haven't been to the Bitcoin, John, it is in a, like, probably what? I would say 100 plus table Chinese restaurant. Yeah. Table. 100 plus table. Yes. Tables. Yes. Tables. And there's a whole wing that they could basically block off with a stage that they use for bit devs and and for the John. And it's amazing because the meeting gets started and then wait waiter waitresses just start walking around taking orders for dinner. And so everyone just, like, gets to have catered Chinese food to their to their plate while the John happens. So

UnknownAnd my experience There's like a 20 foot screen. Yeah. Right? That's normally there for karaoke. Yeah. Right? And it plays like these hilarious Chinese videos That are just wild concha you know, if we're not presenting It's in a strip mall. It's in a strip mall. Like, it is the most Philly thing ever. Like, in a strip mall, the karaoke wing of a Chinese restaurant is where it happens. It is the most Philly thing, and I love it. So we're actually really excited about tomorrow.

Anybody in the Philly area I I will probably put this out assuming we don't have sound issues again. I'll probably put this out, you know, today. So if you wanna come down to Philly, come join us. Alright. On that note On that note. Chapter six. Let's do it.

UnknownForty forty two minutes in for those if you wanna time stamp it for for the if you wanna time stamp it for the the meeting notes for when the maths discussion starts. I mean, for people For the true die diehards that don't care about life. They just wanna be in the math. Well, for people like Rob, we're only twenty one minutes in. I'm busted for, two x speeding most of my podcasts. Yeah. So, yeah, kudos to you guys. This is not a podcast I could two x speed. I could two x speed the banter, but once you get into, like,

deeper concepts that actually require a lot of rigor to process the thoughts,

Unknownthat you gotta slow that down. Yeah. Right now, if you guys it's time to come out of hyperspace. Yeah. Right. Right. De engage the warp drive. Yeah. And you get the white lines and, like, now we're gonna go back into we're gonna go back to real time and cover cover elliptic. So we're at we're in chapter six of the study guide, which is now, like, what, like, what we had been doing was building the group theory. Right? We've been building the group theory, and we've been doing it under this context, how to validate your Bitcoin, how to validate that you're like, if you're given a public key, how do you know your private key is a point also a point on the curve? Well, what does it mean to be a point on the curve? So and then it's

probably worth repeating that, you know, a group has the four properties. Number one is closure, which means any operation you do produces a point inside the number system of your group. In our number system, it's the points on the elliptic curve which is a finite field modulo p. P being two to the two fifty six minus Two to two. Yeah. Minus nine seventy seven. Right. Two to the two we call it p so that we don't have to say that all out. Right? Yes. Okay. So the we have a finite field modulo p, and we can go back you can go back and kinda listen to what all that means, but this is a new number system created for this purpose. It was created

I don't know why it was created, but it was utilized and leveraged for this purpose.

UnknownMhmm. And so, you know, the points on an elliptic curve so this is what we got to where we I think we mentioned that there was a the you lift the curve as an equation. We'll get into that. It has an equation, which makes it sort of a continuous shape, but then we do it over this instead of doing it over the reels, we do it over what's called the finite field modulo p. And so the curve is a collection of p points that you can imagine in your mind's eye take the shape

of the thing you see on postcards and podcast illustrations and, you know, T shirts, this elliptic curve. Right? Mhmm. It takes that form in aggregate of all these individual points, but it's being all these individual points that gives it that makes it part of group. Okay. So that's that was closure. So closure all closure means is that there's an opera group has closure under an operation means that when you run the operation, meaning, say, in our world, it's point addition. Right?

Point addition or doubling of points on elliptic curve yields another point on the elliptic curve. And so if your public key is a bunch of repeated additions of your private key And you do have closure. And you have closure, then you know that you know your private key is a point on the curve. Right. Right. So that's that was, like, where and then so then we had the three other properties. One was associativity, which we don't really spend a lot of time on, but it just means that when you do the addition, it's like if you have you're adding you can be adding

an indeterminate amount of points, and it doesn't matter which one you add first. You're gonna get to the same answer. Right? That's it's a conversion of closure where it's like, the closure still holds no matter what, no matter which two points you operate on first. Right. Within the group. Right. And then we have the idea of an inverse. Sorry, we have the idea of an identity which is like, is there a point where you do the addition and you don't change the answer? Right? Like, in in our world in the world of, say, integers and under addition, that number is zero.

Right? In the world of Multiplication,

Unknownit's one. Right. Is there a equals x. Which seems like a kind of a pedantic thing to worry about, but the reason we worry about it is so that we can talk about the next the last property called inverse. Most important for when you think about Bitcoin stuff. Like, obviously, all of these things are required, but the inverse is, like, the magic of the cryptography and elliptic curve. Yeah. And so the inverse is assuming you have an identity. Right? The inverse is the number or the point on the curve

or the member of the group that when you apply the operation to another element, you get back to the identity. Now that oh, I hate myself for just having said that. I feel so concrete example. Yeah. In addition, say with the integers under addition, we know the identity is zero. Right? So if my number is if I'm dealing with a point called five or an element called five, my inverse is negative five. Right. It's the thing that gets me to the identity. But I need to know what the identity is in order to know how to in order to navigate this, in order to identify the inverse.

Yes. So in in the in the if I'm dealing with the integers and multiplication, there is no inverse because the my identity is one under multiplication. And so in order to get to one from five, I'd have to go to one fifth, which is there's no integer for that. So, like, that you have to say, well, under the rationals that exists, rationals being right p over any any two integers divided by each other. Under the rationals, I have an inverse under multiplication, but I don't have an inverse under addition. And so we get to this

It turn it we you know, it ends up with an elliptic curve over the number system called the finite field modulo p. Yep. We actually end up with all the group properties.

UnknownRight. Right. And as the one last, like, little nugget I'll throw in there for a little context for Bitcoin, the entire, like, cryptography, the security of the system is it is trivially way easier to multiply numbers together than it is to factor them out after the fact. If I give you one massive really big number at the end and you try to break apart the components that make up that number, that is way, way harder than it is to say, just take a bunch of numbers and multiply them, which in this case,

in this group like in this group world, the cyclic group world, it is actually just adding over the generator point within the cyclic group continually over and over again. And if you add something together X time, like N number of times you're adding something, you're also just multiplying by N. So that is where the just for why do we even care and think and talk about this? This entire group over the elliptic curves of reels is we're talking about the tran like, the transformation of going from a private key to a public key and that that you reveal and how can you reveal a public key that has a concrete, defined mathematical link to a private key and not actually give up what that private key is?

UnknownSo if you can somehow validate this idea that the points in the elliptic curve form a group under this number system. Right? And, you know, you understand what it means to be a cyclic group that you you're really in a I think you're really in a privileged position in terms of your ability to have conviction in Bitcoin, in my opinion, and it's relative to those who can't. And for me, I I've been starting to say, you know, it takes, like, maybe you could do this in, like, twenty hours. Yeah. You can you can go from kind of having some math background, but not, like, you know, not

feeling like you really not not feeling like you really grok the math. You can go from sort of that to this point in, like, twenty hours, something like that Yes. In my opinion. Yeah. Yeah. The

Unknownthis is, I would say, one of the two big concepts. This year, what we're talking about now is understanding how can I have property in Bitcoin? Like, how can I own Bitcoin, which is really not really owning in the traditional sense? It's just having a very large number that no one else knows what it is. And it is of a is of a depth in space such that someone will not be able to trivially find your number because it's a cosmic needle in a haystack at two to the 256

that, like, more atoms in the observable there's more possible Bitcoin keys than atoms in the observable universe, which, like, no one's going be able to traditionally grind through. We're going to park the quantum stuff because that's what breaks down those heuristics. The other thing, though, the other big idea is proof of work. And the reason why is proof of work is how you maintain global consensus amongst these things. And what miners do with proof of work besides just mining and like finding a new block, what they're doing in that new block, though, is they're updating the ledger of transactions.

And so if you can understand hash functions and proof of work, you don't have to go every single line detail of like a miner's block template and all of those pieces. You can if you want to. But once you understand how a private key can map to a public key in a secure way, in a provably known secure way. And, the mining piece, you then have the two big concepts of individual property, global consensus. And with those two things together, you get Bitcoin.

UnknownAnd it's might be helpful and useful to know from a group theory perspective that a hash function is not a it's act it's considered a semi group. Meaning, it has some of the properties of a group, but you don't have it actually It doesn't have closure. But it doesn't have an inverse either. Right. So would be undoing the hash function. Yeah. It's just useful to have things in this context. It helps under it helps us understand. And it's like, if you don't wanna get one shotted and taken by something like post quantum,

you know, it's useful to know that hash functions and elliptic curves elliptic curve private keys are two different entities.

UnknownIt it also helps scope out when people talk about post quantum stuff. Hash based signatures work because those quantum computers can't do that same trivializing over finding the inverse because there isn't an inverse in a hash function by design. So there and there here was a fun fact that got lost in last episode, but and you know the answer now, but it was so when Satoshi launched Bitcoin, he did it with p two p k addresses, and that's everybody knows about p two p k addresses now because of thank you to the quantum people.

Yeah. But I'd asked you, do you know how long he had how long it was before he introduced p two p k h into the protocol?

UnknownIt was, like, under a thousand blocks because the scripting language is already there. Yeah. Thirteen days. And people I was asking around the conference just as a fun fact, you know, and it was that that was a shock to many people. Yep. That was so fast. Right? You can even make the case it technically existed since the beginning because he didn't, like, fork Bitcoin to add it. It just was a natural consequence of Bitcoin scripting that you could, instead of committing to a public key, commit to a hash of a public key. It was discovered not invented, this p two p k h.

That's right. Right. That's right. But it's just in you know, like, it's good to have this history, and it's good to understand these things with, like, with the knowledge of group theory as your ally. It's like, know, it gives you you know, and you're not you're not so impressed when you hear somebody say you know, when you hear somebody arguing for quantum talk about you use group theory, right, as we did. Right? We were like almost fanboys when we heard Alex talking about

cyclic groups on Danny's podcast. We were like, oh my god. What's going on there? And then it's like, dude. Yeah. Yeah. Well, you know, so but the whole point of what we're doing, right, is we don't want you, you know, don't don't you wanna be loaded with this knowledge and not so impressed with others that are. Right?

UnknownRight. And it really helps. So here we go. So now having said all that, we jump in. Ready to jump into elliptic the elliptic curve? We call it elliptic curve over the reels. We start with we start with the reels. Do you guys know so I was just I was just about to ask the Do you guys know what the reels are? Gotta keep it real. Gotta keep it real. The real numbers are like, it's a name for some like, it's a name for something we all like, most people listening to this just think our numbers.

All the numbers. Right? Right. We think about all the numbers. The real what we call the real numbers. It's basically,

Unknownlike, how do you even describe it? You have, like, the way that all positive and negative numbers? Like, it's just the entire universe of the number line? Yeah. So it's it's anything you can yeah. It's the number line is the best way to describe it. The like, it's all positive. Negative infinity to positive infinity. Anything that sits between those two is all points in between. So, like, you have if we start building with, like, the natural numbers or, like, all the positive numbers. Right? But, like, we needed more inverses. We needed an inverse. So we decided, well, let's invent negative numbers too. Right? So we have positive numbers, negative number. These are whole we call whole numbers. Right? But then we decided let's we want rational numbers because we want inverses on multiplication.

So we have rash we have a rational number system, I. E. Any fractions, right? P over Q. Maybe another story for another day, but the Pythagoreans had a kind of a cult like take on rational numbers. I don't know if you're familiar with that story. Okay. Well, let's do it real quick. There is a Pythagoras. Everyone knows who the heck Pythagoras was. He had a he had a group of, like, followers called the Pythagoreans.

And, like, it was they were like a gang. They were like a math gang. You know? In March, we had math gangs. Well, you know, top the top people were in in these clubs and societies. And one of the things that Pythagoras held dear to himself was a worldview that there were no such thing as irrational numbers. Meaning, he believed any number can be achieved by dividing an integer over another integer.

Right? Now what's interesting about that as one of the Pythagoreans decided to note, right, his name was Hepasus or Hepasus. Not really sure. As Marge Simpson might say, I only saw it. I only read about it, you know, before she said run like the wind. Hepassus, he basically said, hey, dude. You know, if I take your Pythagorean theorem and I take a triangle with one you know, with the two adjacent ends,

one, the hypotenuse is something called the square root of two. Right? And I got some I got some earth shattering news for you about the square root of two. Like, I actually did a very like, a three line proof that it's there's no way that's a that that can be a rational number. There's zero chance. Right? And here's now that here's where the the the story and the legend happens is the Pythagoreans lost their mind. The Pythagoreans basically said, dude, we hear you, but Pythagoras could never see this proof because it'll destroy him. It'll just it'll, like, literally undo him. Okay?

And supposedly, they took this guy out to sea and threw him out threw him out to sea and drowned him. Incredible. Well, what's incredible about this it's cool you didn't know about this because what's so cool about this story is that in math circles, like, the highest act of subversion you could do is to prove the square root of two is irrational. And there's like, there are books that just do nothing but show all of the proofs. And it's almost like you're it's almost like mathematician

doing it doing their graffiti on the Berlin Wall. Like, you you you know, it's your one act of sticking it to the man is that to do another proof that the square root of two And another thing.

UnknownJust do another one. It's does this make Pythagoras the first pleb slot map mathematician? Yeah. Well, that deal with the reality. He just he's like, no. It is. People. It's pretty relevant. I just well, I'm gonna leave it at this. It's relevant to where we are today in Bitcoin. For sure. For sure. The Pythagoreans, there's a definite analog of the Pythagoreans, right? Like, you know, Pythagoras might have been a feature, a person who you might

have said in Bitcoin has contributed some great things, but maybe his maybe has gone a little off the rails with his worldview. And he has followers want to protect him.

UnknownOkay. That's why some stress equations. That's all to say. But so that's all to say. If you take all of the numbers that we just mentioned, like the ones that are integers, rational numbers, plus all the irrational numbers, like the square root of two and pi and e. You take all of the numbers that make up the number line, you have what's called the real numbers. Right? And that's not the elliptic curve we're gonna use in Bitcoin, but that's the basis for that that's the basis for the elliptic. In other words, the domain for the

sort of prototypical elliptic curve is gonna be over the reels. And once we establish all of that, then we're gonna do a little switcheroo on the domain and say, well, we're gonna look for cryptography cryptographic purposes. We're actually not gonna do it over all the reels because it's not that doesn't that's not a group. Yeah. So right now, that was a long way to describe the reels. But, you know? So now we have this equation that everybody knows.

Right? This equation that y y squared equals x cubed plus seven. But there's a general we want us to say there's a general equation called y y squared equals x cubed plus a x plus b. And a happens to be zero, and b happens to be seven. This is called the wire stress equation. And we were fumbling a little bit before fumbling around a little bit before the show looking in. Weierstrass is one of the, like, all time elite legends

of mathematics. His name is everywhere. If you are an undergraduate, you learn very quickly the Bolzano Weierstrass theorem, which is about converge so, yeah, it's about convergent sequences. It's one of those seminal theorems about how how you do real analysis. Okay? And Wirestrasse himself, really, this was this was a footnote on his career when he decided to look into elliptic curves. But you guys should know elliptic curves were not invented by Satoshi. They weren't even really invented by cobblets or people, you know, people of that

They were they were around, you know, they were around for a long time. This thing called an elliptic curve, which not to be confused on ellipse. This thing called an elliptic curve is something that mathematicians have been interested in for also say hundreds of years. I'd say hundreds. Yeah. Right? So then so so then the question is, right? We have this like, special version, the wire stress version that Satoshi chose, right, which has a equals zero and b equal to seven.

And I think it's worth it to just talk a little bit about or speculate based on maybe what we know, speculate as to why Satoshi chose that version.

UnknownYeah. I actually I you know, for people who are longtime listeners of the Magic Internet Math podcast, you already actually had this alpha months ago. But in the recent discourse around if we upgrade Bitcoin, quantum stuff has been well, if you look at NIST, you know, the National Institute of Standards Technology, like, basically, it's a federal government organization that also has a lot to do with, like, accreditations,

and they also do accreditations for things like elliptic curve cryptography. And famously, Satoshi did not use a missed approved curve. Now someone correctly kind of called out in the comments, longtime person in Bitcoin, Dan, does a little bit of work in the sorry now, but his case bitcoin.com called out that SHA two is technically an it it is a NIST standard, like cryptographic hashing operation.

The the interesting piece, though, is that Satoshi picked a curve that was not NIST standard. And the best way I would summarize why this is important is right at the end of the talk, someone bumps the thread a year and a half later and talking about like how they feel really uneasy about this curve. The person says, I'm not at ease knowing that every signature in Bitcoin is implemented using a very particular curve and unusual elliptic curve that has been selected for some unknown reason.

And that reason, the chooser, Satoshi, is unwilling to elaborate on. And Greg Maxwell with one sentence Sorry. This was, like, 2010. This is circa '20 Well, so this is so this is when the thread gets bumped later. So the original post was at, like, January, like, 2011. Like, it's actually, like, Satoshi is still around, but he's on his way out. No one knows that yet. And, Greg, someone bumps it, like two years later, like, hey, I just discovered this and this is concerning to me. And Greg Maxwell says, you mean you're uneasy that he chose the only standardized curve that was available at that time that did not have unexplained parameters. The reason why we like this is important is like when

when people choose these curves, the constants and the values mean a lot because you can spend a lot of time iterating over every permutation of the Y cubed equals b x squared plus a x plus y squared equals sorry. Sorry. Yes. Not to confuse them, like not to mess with the study guide people. But if you just spend a lot of time rapid brute forcing, you can find curves that if you you basically can inverse the entire thing. And that's kind of a

thing you don't want to be able to do in Bitcoin. Right. You don't you don't want to be able to that. And there's actually a dual electric curve deterministic random bit generator that NIST put forward as like, hey, we should all use this standard. And this was in twenty two thousand and six. So this is pre Satoshi. So Satoshi picked this. It's really important. It had the backdoor that the NSA had so they could actually reverse all of the cryptography at a moment's notice. And this is why any and this was just like a little cultural comment from the conference.

The amount of people, easily a half dozen people who would come up to me, knew I was technical, would ask my opinion of quantum with a straight face would say, well, NIST wants to deprecate elliptic curves in the next couple of years, and they have done a lot of really great work. NIST has several approved post quantum secure signature schemes. That's the federal government saying you should do this. And in no other domain ever in Bitcoin would be like, well, the federal government says we should do this. We should probably do it. And for some reason, because it's like most people don't understand the math, they're willing to, like, delegate to an authority. And rather than thinking about like rather than kind of delegating

that authority to the cryptographers who work just in Bitcoin, who are like ride or die multiple like coming on multiple decades of working in Bitcoin and just true believers that have put in massive proofs of work. They would just be like, well, like the government said, it's cool, so why don't we do it? And it's like it's a total blind spot. I think it's a blind spot because people don't take the time and effort to drill down into these, like, core principles of, like, how Bitcoin is secured.

UnknownIt's amazing, and it really makes me realize and double down how important it is to continue to do this. If all if all we do is like a one notch upgrade on people, like if you were a two in math and you got yourself to a three, at the very least, you'll you may not trust the government who are who knows they lost on this elliptic curve perhaps. Consider this scenario. Say they just know that Bitcoin's elliptic curve is was a very good choice. Right?

Well, it kind of makes sense that they would want to put out standards that eliminate it, and then that they can rebuild on their on their stuff again. Basically, not allow the Satoshi mistake to happen again.

UnknownRight. Right. It's you know, so that's important. Like like like, these curves are not trivial. And I think we did this in either the first or second episode. We went into the exact mechanics of a Koblitz curve and, like, we went through every single letter in sec p two fifty six k one and talked a bit about the libsec p library and the history of that in Bitcoin. I think that's what we'll eventually when we when we build our way up, we'll we'll get Jonas Nick on just to talk about how he started from an open SSL import of the libsec p two fifty six. The the the sec p two fifty six k one being just an open SSL import for signatures and how they turned it into like 50,000 lines of code.

Like that's just that's the pod. That's the whole pod. You could do a second one, I guess, on shrinks and shrimps and all the post quantum stuff. But like there's a cathedral that he's already built and everyone's like, okay, yeah, but what have you done for me lately? Like, where's my Hey, think, boy, where's my post quantum scheme? And like, he just finished like, there's a cathedral behind him of, you know, fifteen years of effort and no one cares about it anymore because that's vulnerable to quantum computers. And this is like you talk about changing the stuff in Bitcoin, man. Like, this stuff is Lindy. It's been securing Bitcoin for seventeen,

sixteen years. Like, we don't just trivially throw this away.

UnknownTotally. So I'm thinking back to there are certain things we tried to do last episode that I don't think we'll try to do this time. Sure. You know, we were getting we've we've got really got into the discriminate. That was really difficult. The thing, I guess you know, the reason first of all, the reason why we start over the reels is because we need to know things. We there's certain there's certain things we need to know that require a continuous curve. We need to know what's called the slope Mhmm. Slope of the curve. So this is a clunky one that I feel like we should just do real quick. Is this what what is the slope of the curve

actually mean? So if you imagine you guys can understand a parabola, like y equals x squared is a parabola. And you imagine any point. Right? And if you try to draw a line, what we call tangent to that curve. So, like, there's a line. There's one line on earth that touches that curve. It's singular, and we know there's that line has a slope. Right? Whatever that line is that touches the curve once, right, grazes it.

Think about at the point one one if you were to draw a line tangent to that curve. Now we happen to know from calculus that the slope of that curve is two. Mhmm. And just think rise over run. Right? It goes up two, and it goes over one. Right? It's just you you just imagine this in your mind's eye. I just wanna get across what does it mean? What does the slope of the curve mean? It's the slope of the tangent line at a point. Okay? Right. And so we need to know that.

And the reason couple of reasons we need to know that. One is it's sort of the fundamental of point addition. It's also we need to define the identity of this entity. There's an identity point that happens to be what's called the point of infinity. Now Mhmm. It's the elliptic curve is interesting because it's it's symmetric on it's symmetric over the x axis. Right. And so we don't have function. We we normally deal with functions that only operate, like, the top half. Right? So we're like a like, think of, like, the square root of x, which is like a parabola tilted on its side, but we eliminate the symmetric part. Right?

Elliptic curve is like two halves. Right? Like a South Park Canadian face. It's like two two halves. Right? But but they're mirrors of each other. They're reflections of each other. So there's this point, the leftmost point where, you know, the two halves come together. If you try to imagine the slope of the leftmost point, it would be a vertical line. Yes. Which we would say has slope infinity. Vertical lines have a slope of infinity.

But what matters here is just that this point is the identity point. And it's be meaning if you take this point and add it to any other point, it it remains unchanged. That's right. Okay. That's one that's the first thing. So you we needed an identity to for the group properties. Right? And then we and we needed the identity to know how to determine what the inverse is. And it turns out this is the one of my favorite properties of an elliptic curve. It turns out the inverse of any point on the elliptic curve. So you take any x y point on the elliptic curve. The inverse is x negative y. It's literally just the it's the it's the bottom half partner

of the top half. And if you add those points together and you get to that leftmost point, you get to that I we they call it the point at infinity. Okay? That's just the name pointed infinity, but it's that's the identity, or they called big o. So we're starting to we we're using the reels to establish the mechanics here, right, before we then and when I say we, I mean, like, the the, you know, the The royalty. The greats are doing this. Right. Bringing it into the knowledge system. We're just explaining it. But, like, this is we're using the reels to establish

these basic primitives here. So then so that we can then we can, you know, so we can sort of see it and visualize it and say, yeah, this makes sense. A computer can do this. Right? It's another reason we use the reels. Right? It's like, oh, can can we at least have a computer do this? Right? So, you know, so we collect we're sort of collect the only thing we have it now established in terms of the group properties. Right? We've sort of established the we the identity and the inverse.

There's this idea of closure. We haven't established what the operation is. This thing called point edition. And I don't think it's like that's just something that you probably need to go to the videos to really see drawn out. In the first time I saw a GM point edition of an Elliptic curve, the first time I saw it was in the par puzzle puzzle cryptography book, which I talk about a lot. That's the first cryptography book I went to. It blew my freaking head off. I couldn't believe how beautiful this point edition was. I couldn't wait to go open Python and start coding it and figuring out if I can actually do this. It's all so great.

So do you have this point addition and point doubling that I would suggest going to the study guide and working through because it's not it's actually not hard, but it's very, very eye opening. It'll open something up in your brain space. I don't know, Rob. Do you wanna, like yeah. Is there anything you wanna add on that one? Yeah. I would just say the point addition with the graphs provides a visual cue of how

Unknowncomputers do these operations of just taking the two points on the curve, and that is, like, adding them together, and you can get to a third point on the curve. We don't think we have to get too, crazy with it. I know last time we were trying to, like, verbally explain the visual of Yeah. Of taking the two points and adding them. I would I rather than do that, I'm gonna say if you look at the study guide, it is trivial to look at. In section 6.3, you can see the additions happening. And rather than trying to stumble for ten minutes explaining it,

I'll leave that there. The I I and remember, just to maybe take a half step back and talk about this in the context of Bitcoin, what you're doing is you're taking a point on the curve, your private key, and you're adding it over this elliptic curve as, like, the group of modulo p, and you're doing that again and again and again, you're actually taking the the value

of your private key, and you're doing it that many times into the generator group to get a proper shuffle to get to whatever your final public key is that you share with everyone. Just to kind of explain a little bit of, like, touching grass of reality of why do we care and talk about this stuff. This is what we're talking about right now is this transposition between a private key and a public key. And so you you do the addition, the amount of times the the number of your your private key is, which is just multiplication. That's why we say it's, you know, your private key times

the generator point equals your your public key.

UnknownWe're take we're going back a step and saying we're going we're we're taking a step back with multiplication with this first principle that it's really just repeated addition. Exactly. Yeah. Yeah. Do you wanna take a stab at point doubling? Well, yeah. But Real quick. There. Yeah. Just so when we talk back to the wire stress version that we use in Bitcoin, You know, one of the things that gets brought out in a study guide was that

speculating as to why Satoshi chose it or one of the other reasons Satoshi chose it was that the zero value for a makes makes it so that it's a much faster calculation. Yeah. The I think 30% faster if having no a term, meaning this this x q plus x cube plus seven, y squared equals x cube plus seven. If you think about it for just a real quick second, if you for anyone who's done, like, high school calculus,

okay, I don't wanna scare you if you haven't, but, like, if you have, you can imagine taking derivatives of both sides. Right? Y squared, you have two what? You have two y times slope. Okay. Two y times the slope. And then on the right side, you have three x squared. You've execute you know, the derivative of x cubed is three x squared, and then you have the derivative of ax is a. Right? So it's a little clunky to have this constant a value in there. If you can if a goes to zero, it's a much if you've you now wipe this derivative clean, the derivative being the slope.

Right? So the the slope no longer no longer has this other a term and it just operates faster. So in addition to whatever other reasons Satoshi was looking into for this equation, it was another very convenient property that to choose what's called the cobb you know, basically, it was the cobblets curve that set happened to set these parameters a equals zero. So it's just to be clear just to be clear, somebody will correct me on this probably. But the wire stress determined

the sort of this x square x cubed plus a x plus b. I'm not sure if the wire stress version is the cobbles version is the one that said a equals zero and b equals seven. Right? Yeah. So now why why is b equals seven? Not sure. Maybe seven's the smallest prime number or something like that. I'm not that this we'll have to Oh, why is it plus seven? Yeah. Why why what we understand why a equals zero is a powerful is a powerful parameter.

Right? Not really sure why b equals seven matters that much. Right?

UnknownRight. Right. Right. It's the smallest positive integer that gives the curve a prime group order. Boom. Boom. There you go. There you go. Thank you, Claude, for the quick, research assistant role there. The committee, tried one two three four five six, and seven was the first time it produced a valid cobblets curve with the right properties. Yeah. Like, how drug brain could you get? Right? There's like, we just tried every value until

we got one that works.

UnknownIt's beautiful. Now the simplicity of that, the the brute simplicity of that is beautiful. Now to pull things back a little bit, like the associative property, and this is taking it to 6.4 with doubling. Why why would you why do you need to double a point? Like, why would you ever get to the same point? Now remember, you take your private key and you multiply it, which is adding a set number of times the generator point that many times. So what you can actually do is you can take that generator point and send and just add it to itself a bunch of times, which is doubling.

And it's way faster to do that. So that's why we're about to talk about the doubling piece. Point doubling, if that point, since you're just taking the same generator point over and over again, you can just double it and double it and double it and get the generator value to the correct size that you need for your private key. And it's it's something that is just a very strong performance shortcut, and it's what every single wallet does. Anything that's touching a private key, it's doing this point doubling over and over again the generator point to be able to securely run it.

And I'm looking through the study guide here.

UnknownI think the curve is I need to refine the curve. I think for whatever reason The visuals? The visual didn't put the result on the curve. No, did not. Yeah. It did not. You can always do another pass at this. If you're listening in a couple of the visuals. Listen for this whole set, this whole chapters, the visuals are slightly off. You'll just imagine the dotted line making it all the way to the point because, you know, it's not true because if a point existed outside the curve, then it wouldn't be a close group. That's right.

UnknownBut don't worry. I will fix this. Hopefully, by the time you all are hopefully, the time you all are looking at it. Yep. And 6.5 inverses, we already talked about inverses in that if you look at the visual shape of the curve, you could just take any given point and you can actually flip it around to get the reflection across the x axis. And I think that kind of lays out the core foundational pieces. This is now theorem 6.1,

and it does a full it's a proof of of this basically being a generalizable approach of how you can use the elliptic curve for the point doubling and the addition to be able to kind of map and traverse everything. Yes. Sorry. It's not approved truly. It's actually just kind of like a walkthrough example of, like, the different pieces to abstractly reason about how this would work, you know, in in in real life.

UnknownYeah. And now if maybe we'll close the episode on I'm gonna there's the sort of this thing that the study guide does, which is it sort of pulls from thought just like thought leadership of Rudolf Steiner who lived in Weierstrass's time or maybe a little after Weierstrass's time. He lived in the early he lived in the late eighteen hundreds, early nineteen hundreds. And I I use I use them a lot to help me think through, you know, what is

how do our brains really, really grok a lot of this? And in this chapter, we have this concept called the curve as a whole. I'm gonna read I'm just gonna read out of the study guide this piece, and maybe we'll we'll either talk about it or just close on it. Okay. So in one of his lectures, Steiner describes how the thinking mind grasps a mathematical concept, not as a collection of parts, but as a whole. A single self determined reality.

A circle is not many points equidistant from a center. It is a unified form perceived as such by thinking, and the individual points are consequences of the whole rather than constituents of it. An elliptic curve exemplifies this. We defined it by an equation, derived addition formulas point by point, and verified the group axioms one at a time. But the curve is not the sum of these parts. It is a single mathematical

organism whose algebraic structure, geometric shape, and group law are three phases of one reality. The addition law is not imposed on the curve from outside. It arises from the curve's nature as naturally as the area of a circle arises from its roundness. So this is why elliptic curves are so powerful in cryptography. They're not an arbitrary choice of group structure. They are a natural mathematical object whose internal coherence,

geometry, the algebra, number theory, it's all fused into one. This internal coherence produces exactly the properties that cryptography requires, which is a large cyclic group with a hard discrete logarithm problem.

UnknownRight. Yeah. And all I'll say to that is we're we're we're constantly talking about this in the context of you managing your Bitcoin. You actually your Bitcoin node is doing all of this math for every single Bitcoin transaction on the network. And the fact that is lightning fast trivial to verify these signatures is a core part of how Bitcoin can remain decentralized. If these algorithms were not efficient and let's say it took an hour to validate a Bitcoin block, then you wouldn't be able to actually be trustless with the network because you would you'd always you'd have to either trust someone

that the math is correct or you'd always be perpetually behind in the blockchain history.

UnknownThese are great, great points. Yeah. You know, you just reminded me of, like, how I thought we were gonna close on that one. But, like, it just literally reminded me of how, like, AT and T was created and started. It was, like, it was essentially front running. It's like front running horse racing results. This was a problem that the results weren't getting to the betting to the betting desks in time. And people were quote, you know, with the Bitcoin version would be they were double spending these

Yeah. These gambling results. And so the AT and T came in and said, can do this over wire and everyone will get it at the same time. And we will eliminate this problem of bad information that people have an incentive to exploit. Right? Very similar, very similar trajectory. And maybe we're this is the new, you know, yeah. This is like the new version of AT and T, whatever we're building.

UnknownYeah. Absolutely. I think that was a good summary of chapter six. Yep. Next time we have collective curves over finite fields. Yeah. And we'll, we'll we'll go carry through that. Alright. Magic Internet math in the books. See you guys.

Open the transcript file

Hosts and Guests

Nobody is credited on this episode yet.

Credit the people on this episode

Be the first to rate this episode

More episodes

All episodes
  1. Exponential Got Less Exponential

    Nov 2, 20261 hr 10 minAudio

    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...

    Credit the people on this episode

  2. Red Teaming Elliptic Curve Cryptography

    Oct 11, 20261 hr 27 minAudio

    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....

    Credit the people on this episode

  3. Hash Functions, Linear Algebra, and Alice in Wonderland

    Sep 28, 202650 minAudio

    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...

    Credit the people on this episode

  4. Kayfabe Math is Now Mainstream

    Sep 13, 20261 hr 17 minAudio

    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...

    Credit the people on this episode

  5. Legalize MATH

    Sep 2, 20262 hr 2 minAudio

    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...

    Credit the people on this episode