S15 E70: Jeff Garzik on Hemi & Bitcoin Development
Search this conversation
0:48Vlad CosteaHello there and welcome to the Bitcoin Takeover Podcast. This is season 15, episode 70, and I can't believe that I'm finally seeing this because I've been chasing you with emails for months. Since I saw that YouTube video that you posted about the origins of Satoshi Nakamoto and your opinions that it must have been just one person, I really wanted to interview you, Mr. Garzik. You are a legend. You are the first big blocker. You proposed an increase in the Bitcoin block size to seven megabytes back in October of 2010.
1:18Satoshi responded to that, that it's a bit too early. I actually met someone at a conference last month who participated in the conversation in that BitcoinTalk thread. He kept coming back every year asking, is this still too early? His name is ArticMine. You can actually find him under that thread. You're also one of the top donors on BitGive. You have been a Bitcoin Core developer for like five years. And you're one of the first people to figure out that bridges are the way to the future, and that's going to be maybe the first topic that we're going to approach during this discussion.
1:58And what else? You actually got responses from Satoshi, and you're so OG. You did TED Talks and you spoke about Bitcoin as a living organism way before people were writing books about it. So thank you very much for all of your work, and it's good to have you here.
2:14Jeff GarzikOh, well, thanks for having me. I definitely am glad to finally connect after many emails and tweets and chases. I know that the audience has a bunch of questions, so let's dive in.
2:32Vlad CosteaRight. So I understand that these days you're working on something that's called Hemi and is a bridge between Bitcoin and Ethereum. So my first question for you is, why did you choose to work on this?
2:45Jeff GarzikWell, a little bit of what Hemi is and a quick soundbite. Hemi is a Bitcoin and Ethereum layer two blockchain that bridges both the king and queen of crypto. Bitcoin, the king of crypto, the root of the crypto tree, the trunk upon which all other cryptos hang, flow through, etc. And Ethereum, which I think is kind of the Grand Central Station of crypto, if you're going anywhere but Bitcoin, you're probably going to go through some Ethereum based technology. And it was making that observation and then looking five, 10, 15 chess moves ahead as to what comes next.
3:37What comes next for Bitcoin? How does Bitcoin scale? And I argue that just as Ethereum and the rest of the blockchain universe are scaling through these rollups and layer two blockchains, that's how Bitcoin is going to scale up as well. It's fundamentally you can't put all the coffees on one blockchain, any one blockchain, nor would you want to. And this is organic. This is decentralized. It's the way to scale in an open market, free market way.
4:16Vlad CosteaI guess in hindsight it was so obvious because if Bitcoin did not scale, you would have other chains fulfilling the demand for transactions and for smart contracts that were not getting built on Bitcoin. And maybe for you it was super obvious. But for me, I spent so many years waiting for Blockstream to deliver that promised sidechain. And then I hoped that the Lightning Network would eventually turn out to be the scaling solution, the one that's going to make every other altcoin obsolete. And that's pretty disappointing here. In the meantime, I guess you started working on bridges with your company Bloq, with a Q and then you started Hemi.
4:59And I guess I got to give it to you. You're sort of right.
5:04Jeff GarzikWell, thanks. I think it's inevitable based on macroeconomics. That's the direction that we all had to go. I think Lightning is very useful for some really interesting use cases like holding your crypto and participating in centralized exchange order books. For example, that's a bit radical. The current CEXs, Binance, Coinbase, et cetera. They're looking at Lightning as a deposit method. And again, what is the next chess move ahead? And that's keep your crypto until the microsecond that a trade is ready to go through on Coinbase, Binance.
5:50And then the Lightning Network could very easily facilitate something like that. So I think it's great for exchanges. I think it's great for payments. But it's got also macroeconomic issues as well. If you have to pay a $20, $30, $50 fee just to onboard onto the network, again, that becomes a bottleneck for onboarding what I want to do, which is freedom for the billions of people on this planet. I want to metaphorically, if not literally, bring Bitcoin to everyone on the planet. How do you do that?
6:30You've got to scale up with these Bitcoin connected, Bitcoin secured networks. And obviously, plug for Hemi, we use a protocol called Proof-of-Proof that secures our blockchain into the Bitcoin chain itself, as well as we embed a Bitcoin node into our EVM node, the tightest integration, I believe, of any of these layer two blockchains. It really extends the EVM to something that I've wanted for almost 14 years as a Bitcoin developer, extending the EVM to directly talk to the Bitcoin blockchain. You can program your smart contracts to read the Bitcoin chain.
7:19You can program your smart contracts to create new Bitcoin transactions. So it's really kind of super-powered Bitcoin script from a certain lens, from a certain point of view. And like I just said, I've been waiting a decade for that level of power in scripting and smart contracting on Bitcoin. And I argue that now it's here with Hemi.
7:46Vlad CosteaBut why did you choose to go for Ethereum for smart contracts and scripting? Because you find a bunch of nerds everywhere who are going to tell you, you know, this other network is much better. It's more expressive. It enables more advanced smart contracts. It scales better and blah, blah, blah. There's even a proposal by Blockstream, which is called Simplicity, which is a programming language that they created for contracting that they say can fit on a T-shirt. But I've heard a lot of jokes that it's not really that simple.
8:20Jeff GarzikI'll take you in the audience back in history, 20 or 30 years. A key lesson that I learned in Linux and working on the Linux kernel, which is, of course, open source. I love open source software. Open source software is the most robust software on the planet. Open source is, as we say, with many eyes, all bugs are shallow, meaning that if it's open, you have many more reviewers. There's just a mathematically higher chance of finding any problems and fixing them. And among many other attributes of open source on the freedom side, the permissionless innovation side.
9:09And one thing, one key lesson that we learned was that software is with us for a long, long time. And specifically, that bleeds over into not only just, let's say, the EVM network or the EVM itself in Ethereum and other L2s, but tooling, auditing, as in humans who learn smart contracts. They learn where the sharp edges are and advise others. That takes years to grow these sedimentary layers of software and people on top of a core technology. And so that's why you see in the general technology space, the C programming language has been with us for 50 years.
9:59And it's not going away anytime soon. And C++ and Java and many of these other programming languages, there are brand new programming languages that are absolutely better. But at the same time, you don't have those decades of tooling, debugging. And again, people that know where all the hidden traps are in that particular language. And so you build up this set of field experience. This this tooling and this general knowledge among the community of what to do and more specifically, what not to do in order to build secure software.
10:42And so it's really about importing what Bitcoin has and Bitcoin strengths and all of the tooling and infrastructure, both economic as well as technical around Bitcoin. And then marrying it with that experience of Ethereum and EVM security auditors, Ethereum and EVM debuggers, tools, et cetera. We have a term in the office, wet cement, that refers to new technologies, new script languages. They might be interesting. They might give you a few more transactions per second. But you don't have that strength of tooling. You don't have that what I call a... I love evolutionary biology and I make a lot of biological references to Bitcoin and Ethereum.
11:40You don't have that Darwinian trial by fire that a new script language doesn't have. And the EVM and Bitcoin script do have. There's a lot more trust in something that's been deployed for years. That's been beaten up by hackers for years. Bitcoin's market cap, let me look at CoinMarketCap real quick, because we're hitting some all-time highs here, is $2.1 trillion as we record this. And as a result, that means a $2.1 trillion hack bounty for anyone smart enough who thinks they can hack Bitcoin.
12:24You know, you can instantly make yourself a trillionaire if you're that talented, that smart. You think that you can figure out what others haven't figured out. And that's how we know that these systems are ultra secure. That's how we know that Bitcoin is secure and Ethereum is secure. It has years, miles, seasoning behind it that tell you that it's incredibly improbable that there's a major security bug in Bitcoin at this point.
12:57Vlad CosteaThat's a good way to put it. And I guess Satoshi Nakamoto, you also mentioned this in your presentations that he used off the shelf cryptography. He didn't create anything. He just used and pieced together elements that were already there.
13:15Jeff GarzikThat's exactly right. That was incredibly important for myself and other outsiders coming in, discovering Bitcoin for the first time is he didn't invent a new hash function. He didn't invent a new digital signature algorithm. He took algorithms that were well studied, well trusted, matured, tested, cryptanalyzed to the nth degree and then put them together in a brand new way. But he was ultimately taking existing technologies off a shelf.
13:54Vlad CosteaYeah, and going back to Hemi, it's actually curious to me that you chose to use something that's called Proof-of-Proof, where, you know, the entire industry is sort of divided between proof of work and proof of stake. But what is Proof-of-Proof?
14:11Jeff GarzikProof-of-Proof, in a nutshell, is the protocol that we use embedded proofs, which are then put into a normal Bitcoin transaction and secured with Bitcoin's proof of work. So we're really building on top of and leaning on Bitcoin security, which, again, Bitcoin is the most secure, the most bombproof chain on the planet. The most likely to be here in 10 years, 20 years, 50 years, 100 years blockchain. That's Bitcoin. And so reversing or walking the next step, the next chess move of that logic, how do we inherit Bitcoin security?
14:59There's obviously a lot of different techniques, merge mining, several different types of sidechains. But if you want to use a censorship resistant network, which Bitcoin is, that's one of Bitcoin's fundamental features, censorship resistance, you want to use that for finality of our blockchain. So we lean on, we use keystone blocks. I'll have to double check my numbers here. My CTO, Max Sanchez, will probably dial in and correct me, but I believe it's 12 Bitcoin blocks: we emit a keystone hash, which is embedded in every 12th Bitcoin block.
15:46And that gives you ultimate what we call super finality on the Hemi layer two network. So Proof-of-Proof is an alternative to merge mining that relies on Bitcoin's censorship resistant properties.
16:03Vlad CosteaSo how are you inserting this keystone element? Is it a timestamp or is it an inscription? Or what kind of protocol are you using?
16:15Jeff GarzikMax Sanchez designed a custom protocol a few years ago. He live tested it in a blockchain called VeriBlock on the Bitcoin network. And so again, leaning on the Darwinian trial by fire, or another way to say that one of my favorite quotes from Linus Torvalds, the inventor of Linux, is that the internet is the greatest test lab in the world. You can test as a software developer on your local machines, on the servers that you have access to. You can test in a test lab, but you never really can trust software until it's survived the greatest test lab in the world, the internet.
17:02Out in the wild, out in the field. That's the best testing you're going to get. And Proof-of-Proof that VeriBlock blockchain was our live test. It worked great. And I have full confidence in that protocol. And that was one of several inspirations for Hemi specifically was the Proof-of-Proof protocol. And a little bit of a plug: docs.hemi.xyz. You can read all about Proof-of-Proof and the other Hemi protocol details.
17:38Vlad CosteaYeah, we just got a shout out from Amir Taaki on Twitter. I guess he is listening. So hi, Amir. It's really good to have you here.
17:47Jeff GarzikYeah, he's another OG. He did dark markets in the early days. Oh, shoot. What was that company that got some VC funding? It was like open market or dark market or but he was one of the kernels of that. So definitely a lot of old and familiar faces.
18:09Vlad CosteaYeah, I think he worked on Dark Wallet first and then he participated in a hackathon where he presented DarkMarket, which later got rebranded as OpenBazaar because, there you go, maybe not too friendly sounding, especially in the aftermath of the Silk Road.
18:30Jeff GarzikThat's right. That's ultimately where just to riff on that, where I think looking forward, the space needs to go is we need more privacy. You know, there's a lot of great work on homomorphic encryption, a lot of great work on zk-SNARKs, zk-STARKs, variations of those. We're fans of that entire category of work because, you know, I have very strong privacy advocate blockchains, Bitcoin in particular, Ethereum, even more so are they're not anti privacy, but they don't default to privacy. And I think, you know, a version of the unicorn world in my head kind of has a Zcash shielded-transactions-only type of privacy view.
19:31That's if we're going to if we're going to enable, you know, freedom, self sovereignty through crypto for, you know, the billions of people on planet Earth, they need privacy by default.
19:46Vlad CosteaOh, I agree with that. And it used to be the case that this was part of the conversation that Bitcoin needs to become a fungible currency. But now, and I guess you know much better than me, you are a person who participated in the creation of BitPay and was one of the main people behind BitPay for a while. And you saw how the adoption of Bitcoin as a currency diminished over time instead of increasing. And then the narrative shifted from a currency that is supposed to bank the unbanked and whatever was being said at the time to a store of value.
20:25And it's a bit disappointing because there was a lot of potential there. And I guess you were there to see it. And what I want to say to remark based on your statement is that I'm happy that you're still excited about Bitcoin because I've messaged other developers from your class. I think a couple of days ago I had some interactions on Nostr with Wladimir van der Laan and he seems jaded. And also Mike Hearn, he actually responded to my email and he said he's not interested in Bitcoin anymore.
20:59And Gavin doesn't respond to any emails. And here you are launching something being optimistic about Bitcoin. What's the secret behind it?
21:09Jeff GarzikWell, I'm optimistic about humanity. And I'm both optimistic and pragmatic, I guess I could say. Fundamentally, I try to reason from first principles. I think the word for people younger than myself is based. And it's ultimately pragmatism about freedom. And what I mean by that is that if you want freedom, you need at a minimum three things. You need free speech, you need end-to-end encryption, and you need cryptocurrency. And focusing on that last category, again, what is the cryptocurrency most likely to succeed? I try to take the long now.
22:04That's longnow.org if you want to visit that. The long now view of multi decades. If we want to secure freedom for the billions on this planet and looking at cryptocurrency specifically, again, Bitcoin is going to be the most likely to be here in 50 years. The most likely to be here in 100 years. And so naturally, if we're going to build something that enables freedom, enables decentralized finance, I'm a big fan of DeFi, enables the ability to onboard people for pennies rather than $10 or $20 per transaction fees.
22:45You're going to have to build on Bitcoin because that's, again, pragmatism says what's the most bombproof, most secure, most likely to be there blockchain. And you're going to use Ethereum smart contract technology because, again, that's passed through the Darwinian trial by fire. It has layers of tooling and layers of people behind its security as well. And so that led to Hemi. Some people say, you know, kind of all roads lead to Bitcoin or whatever their favorite topic is. I like to say all roads lead from freedom if you're reasoning from first principles.
23:33And so if you start with that, you end up with Hemi, 10 chess moves ahead.
23:40Vlad CosteaOh, that's very beautiful. I actually have a question from one of my sponsors, which is SideShift.ai. They have this beautiful exchange with lots of integrations, which is they would not call it a decentralized exchange, but at the same time, it's something that is more permissionless than the average exchange where you sign up with your ID card and driver's license and submit your bank statement and whatever. And it's useful for swapping between different types of cryptocurrencies. And their question for you is if Hemi has a token and if you guys would like to work with SideShift to get integrated.
24:18Jeff GarzikWell, the lawyers will tell me not to answer that question directly.
24:25Hemi will definitely have a token. I'm not going to give the specific release date for it, but we're launching first without that token. We've got an incentivized testnet right now. Again, test, test, test open to the Internet. That's the best way to test. We're launching mainnet very soon within probably the next 60 days, if not sooner. And then the TGE, the acronym everyone's looking for, will come after the mainnet launch. So it's not about a token. It's about the network. And ultimately, you know, a token does come with that, absolutely.
25:15But I don't want to get into specific details, specific listing requests, et cetera.
25:21Vlad CosteaThere is a question that's very relevant right now in the chat from Wayne George, who says, what is a practical example of why someone would use Hemi? Sorry if I missed it in the intro.
25:35Jeff GarzikWell, I'll focus on categories of people that use some of these products. We're all familiar with the Bitcoin maximalist, the Bitcoin maxi. I argue that there's a new category of user, the Bitcoin moderate, is they're interested in Bitcoin. They want to hold Bitcoin. Maybe they're a newcomer in the past year, in the past two years. You know, the gray hairs on my chin say everyone's a newcomer, in my opinion, versus what I call my blockchain age, which is 14. So very few people have a blockchain age of 14.
26:20But no matter what stage of use you're at, the Bitcoin moderate, I argue, wants to do something with their Bitcoin, whether it's collateralize that, earn a little bit of yield, kind of like a savings account by putting it into a lending market or a DEX LP pair or some of these other DeFi vehicles so that you're not just holding, but you're holding and compounding your Bitcoin. So there's an obvious economic incentive to do that. And you can do it pretty passively in a lot of different ways.
27:04So that's one key user and use for Hemi is decentralized finance. And there are DEXs, lending markets, over 30 different dApps on the Hemi L2 right now as we speak, loading their deployments, loading digital assets onto what's going to be Hemi mainnet inside of 60 days. So users can use the standard range of decentralized finance tools plus, and this is pretty unique to Hemi, it's a little bit of a developers view. But what we've done is we embedded a Bitcoin node inside an Ethereum node.
27:57And that's super unique. Again, I mentioned or hinted a little bit earlier, you can use from the advanced EVM smart contract language, the ability to look at the Bitcoin chain, send and receive payments from the Bitcoin chain, Ordinals, Runes, BRC-20, all of those types of protocols as well. And you can do it in a way that interacts with the rest of the ecosystem. On the Ethereum and EVM side, we talk a lot about composability. And to simplify the analogy a little bit, I love Legos. My son loves Legos.
28:42I love Lego Mindstorms. Building robots out of Legos is just the funnest thing in the world. And metaphorically, a lot of these decentralized finance applications, they plug into each other, they're compatible with each other, they can talk to each other. That's composability, where you have a single token, for example, you bridge Bitcoin over to Hemi, or I should say we use the phrase tunnel a lot. It's the same as a bridge, but we like to call it a tunnel. And once you tunnel Bitcoin onto Hemi, you have a token on the Hemi blockchain that's a claim or receipt for that Bitcoin.
29:30You can put that in a lending market and then borrow US dollars against the Bitcoin that was on the Bitcoin chain. And you can use that for day-to-day spending money, or you can use that to further earn yield or do some other trading or decentralized finance type tools. But ultimately, it all comes back to that Bitcoin node embedded inside the Ethereum node, giving a massive expansion on programmability. So I think Bitcoin developers will be very interested in building Bitcoin apps on Hemi because of that newfound flexibility and that newfound composability as well.
30:16Vlad CosteaIt sounds to me a bit like trying to build BlockFi or Celsius, but without the company, in a way that's only a protocol as opposed to having a political entity in the middle that must be trusted.
30:31Jeff GarzikThat's a very good summary, in fact. Again, decentralization is about making the blockchain a neutral referee and not having a centralized counterparty that can run off with your money at the end of the day. And that's part and parcel with that. Also, if you don't have a centralized company in the way, there's far less or none of the KYC AML practices that other centralized exchanges unfortunately are forced by regulation to engage in. And this winds up with the end result that those centralized companies collecting KYC AML data, very, very private data on users, they get hacked and a lot of those databases get leaked.
31:27And so KYC AML laws in practice wind up damaging the privacy and security of the individuals using those companies. The decentralized side is non-custodial, it's composable, and you're not sharing all this private data about individuals all over the Internet, which is bad data practice. It's bad for users and has led in several cases to users being victimized, the literal antithesis of the purpose of KYC AML laws. So decentralization also means eliminating single points of failure or SPOFs. Single points of failure are, speaking very, very abstractly in engineering, if there's a single point of failure, you can almost guarantee that eventually a long-running engineering system will hit that single point of failure and will fail.
32:34And decentralization is about automatically, by creating systems that automatically route around failures, automatically continuing in operation if any one piece fails. So from a technical level, from an economic level, from a human level, from a legal perspective, you want decentralization across all those planes.
33:03Vlad CosteaLet me backtrack a bit and maybe disagree with you. Because what you describe as, okay, it started from the fact that you mentioned that you're trying to minimize trusted third parties or single points of failure. But the so-called maximalist ideology is telling you that it's better to use KYC services than to appeal to something which is using another token that is not Bitcoin. Because obviously that's shitcoiny and that's shitcoin tech, and for some reason this ideology was co-opted by the bankers. But I don't want to get there, I just want to say that what you describe as a moderate Bitcoiner is actually what used to be the maximalist of a few years ago.
33:47Because the maximalist used to be someone who's trying to maximize what Bitcoin does, as opposed to saying, you know, Bitcoin is perfect the way that it is and we should not change it, we should not do anything except for watching these price charts and praising everyone who makes the number go up. It used to not be this way, and the fact that you were starting a new chain or you were starting a protocol parallel to Bitcoin like Namecoin is a maximalist project. Satoshi contributed to the making of Namecoin because it was a new idea to use a blockchain and he said, you know, it's not a good idea to put this on Bitcoin, so why don't you start a merge-mined blockchain?
34:28And from this point forward, every idea, I guess there was a major shift around the time when Counterparty and Omni came out, Omni being formerly Mastercoin, because they were trying to be this Bitcoin 2.0 protocols that were being built on Bitcoin, and then the developers, and I guess you can share some experiences about this, were very much against OP_RETURN and stuff that was being presented as arbitrary data as opposed to transaction data. So that's how we got Ethereum, which was, I guess, it comes from colored coins, Vitalik worked on that previously, but we could have had everything on Bitcoin and it could have been super maximalist.
35:14And the fact that we don't have it is maybe another consequence of maximalism, trying to make stuff happen for human freedom, as opposed to just giving up and saying, you know, Bitcoin cannot do this very well, so let's not do it at all.
35:31Jeff GarzikYeah, I think you hit on it at the end, and I'll give a spicy take for you in the audience. I think that Bitcoin just fundamentally lacks self-sovereign contract addresses, self-actuating self-sovereign contract addresses. You need someone in a multisig or with a hardware wallet, software wallet, et cetera, to actuate a Bitcoin script, which moves around UTXOs, and that's all it does. It doesn't actually literally hold a balance for anyone who can present arbitrary smart contract code, whether that's me or another wallet, and that is the fundamental aspect which allows Ethereum to have on-chain DEXs.
36:30Beyond a certain point, Bitcoin doesn't validate what's in an OP_RETURN. It's just opaque data, and Bitcoin miners can't and aren't in the business of validating opaque data that's stuffed into Bitcoin transactions. So Omni, Counterparty, colored coins, all of those, the fundamental, you know, there were two fundamental limitations. The first was, again, lack of self-sovereign contract addresses. I love, you know, just a sidebar, I love all things robotic, all things autonomous, I love the current AI era that we're in, and I love to think about, and I wrote in 2013, a decade ago, about Bitcoin holding autonomous agents and how they could evolve in the Darwinian biological sense of evolution and not be controlled by any one person.
37:30You know, so pretty radical robotic AI agent type of stuff, and again, Bitcoin was limited in these two key ways. That's, again, that's not knocking Bitcoin's script. Bitcoin's script is very secure. It's very easy to evaluate, but if your validation is not 100% on-chain, then it's not decentralized. There's going to be ultimately either a counterparty in the case of centralized type businesses or a secondary set of validators for this second protocol. And then that becomes, again, with our referencing our recent discussion of single points of failure, that becomes the weak link.
38:21Is this new validator network or wallet validating protocol that is required on top of Bitcoin, but not validated by Bitcoin miners that must be used in order to achieve whatever the goal is? So it's always been a half and half of Bitcoin script only goes so far in validating, you know, 10%, 20%, 50% of a BRC-20 or Ordinals or inscription type of scenario. And you have to have other software sitting on top to make sure that's not bad data. In contrast, Ethereum, everything is on-chain.
39:04It's 100% validated on-chain. And you can prove to yourself that it is decentralized from that particular perspective. So Hemi was about mirroring those two perspectives. And so I very much believe in kind of building multiple fast lanes that run alongside of the Bitcoin chain. But the spicy take is whenever you hear, you know, it's either on Bitcoin or bust, I think that's just Bitcoin maximalist cope.
39:41Vlad CosteaYeah. Speaking of Bitcoin maximalism and riffing on this topic, another sponsor of the show is LayerTwo Labs, which is this company that's trying to build two-way pegged sidechains with help from miners. Basically, it requires a majority of hash rates to enforce these sidechains and they're called drivechains. And the CEO Paul Sztorc is apparently a big fan of yours and has been following your work for a while. And he's curious to know what you think about this way to scale Bitcoin with sidechains which are enforced by miners.
40:16Jeff GarzikI've always been a fan of Paul's specifically, so shout out to him as well. And the drivechains effort. I follow that on Twitter and I should say X and occasionally retweet some of his posts. I guess I can't say retweet either now because it's just fully X. But drivechains are one of the few sidechain-type proposals that I take seriously. There are a lot of other kind of wish fulfillment or one-way peg proposals. The challenge is in getting miners to agree to these new protocols.
41:07And I always have a, you know, maybe I'm a decentralization maximalist. I try to look at engineer design, think about protocols that don't require negotiating with people just to engage my network from a security perspective. And so that's always been the mountain to hurdle is negotiating with individual miners and getting a certain percentage of the hash rate to secure your network. That's it's a valid, it's a secure, it's a promising way to do things. But it's also, you know, if you're if it's a video game, the level of difficulty is set to high.
41:59It's set to hardcore.
42:02Vlad CosteaYeah, it's pretty difficult to get consensus for anything. And I guess, you know, from the days of the block size wars, which I don't think was necessarily about block size because that was increased with SegWit. It was mostly about which team of developers is going to be in charge of Bitcoin. But that's just my take on the events looking back. And it is a bit disappointing because the only conservative way to scale Bitcoin, which has very known consequences and very well researched results, is to increase the block size.
42:37Everything else is just a hacky workaround. And it would be nice if we just agreed that, sure, we're going to scale with these layer twos. But it would also be nice if more people could onboard these layer twos in a non-custodial way at a reasonable cost. Because we spoke earlier about the lightning network and I guess we can also speak about Hemi and these drivechains. The ideal way to get onboarded is to make your own on chain transaction to peg into them. Of course, you can get onboarded directly on them, but there are some tradeoffs involved.
43:11And I was going to ask you, what are the tradeoffs for Hemi?
43:17Jeff GarzikWell, I would say that the primary tradeoffs are that if you're purely in the Bitcoin world, you are moving to a different blockchain. And fundamentally, there are more pieces involved. As an engineer speaking abstractly, whenever there are more pieces involved, there's more complexity. Whenever there's more complexity, there's always additional risk. So the purist view is if there's a single chain and you're just using that chain and no other chain, then there's going to be less risk. But you're also higher cost. And again, from a humanity freedom perspective, you can't get everybody on one chain.
44:05You just can't. And so I follow that logical next step. So I think that ultimately, there are more risks than just Bitcoin and nothing but Bitcoin, the Bitcoin maximalist perspective. But I choose to take those risks such that we can open the door for billions of people. We won't even be able to onboard my dream billions of people onto a single Hemi chain. But using Proof-of-Proof and some of the other technologies that Hemi uses, ultimately, you can scale out very organically. This is referencing the block size wars and looking at both Bitcoin and Ethereum scaling.
45:01Over on the Ethereum side, there's multiple proposals for sharding. It's kind of an L2, but really tightly coupled with an L1 Ethereum. There's been other Bitcoin scaling proposals. But I come back to my biological analogies. What is natural? What is organic? What do the economic incentives push people to do? And then I look at that and I say, respect the market. Don't try to tell the market where to go. Listen to the market. What is the market telling us? The market is telling us whether it's Bitcoin or Ethereum, we don't want to pay $5 a transaction fee for every transaction we do, $10, $20.
45:50You want to scale out and the incentive is to scale out organically as those networks grow, but remain economically and technically tied to those home chains. And that's what Hemi does, both from a technical and economic perspective. And even with other L2s, our competition, that's a natural Darwinian outgrowth, a natural evolutionary outgrowth of where Bitcoin and Ethereum are today. So I think it's positive for the ecosystem. People are coming back to Bitcoin because of L2s, and that trend is just going to continue. It keeps people in the Bitcoin and Ethereum camps.
46:47There's obviously a forest of blockchains, a forest of tokens out there, but the king and queen of crypto have staying power for a reason. And that's technical, that's economic, that's human networks, that's network effect.
47:06Vlad CosteaYeah, there are at least a couple of other projects that try to build bridges between Bitcoin and Ethereum. One of them is BOB, Build on Bitcoin, or BOB. The other one is Citrea. And there's another one, I think, by Alpen Labs that's trying to accomplish something similar. And I was going to ask you, how is Hemi different from these? Because I guess these are relying on ZK rollups. And as far as I could see in the documentation, I didn't see this particular buzzword, because it seems it's very popular these days to use ZK rollups.
47:38And also, your assistant set this interview for one hour, and I guess that ends now. And I just want to ask if you're fine to keep going.
47:46Jeff GarzikWe can keep going for another 15 minutes or so, but I do need to have a hard stop after that.
47:54Vlad CosteaOkay, that's good.
47:56Jeff GarzikBut yeah, on the rollup side, I'm very supportive of all this innovation. Just to talk about innovation very abstractly for a second. One of the things that made me sad about Bitcoin maximalism is you take it to an extreme, you're just really dumping on developers who just want to try something new and interesting, who want to build something that potentially, to use the cliche, makes the world a better place somehow, some way large or small. That's innovation. And the process of innovation is the process of trying and failing.
48:42You have to try and fail and try and fail. The famous quote about Thomas Edison, the inventor of the light bulb, is that he tried well over 100 times to invent the light bulb and then finally succeeded. And he only, you know, I'm probably butchering the quote, but it's something like Thomas Edison said, you know, someone asked him, you know, why did you try 120 times to invent the light bulb? Why didn't you just give up? And my answer is that's the natural process of iterating and innovating is you have to see what works, what doesn't work, to range bound the problem, to narrow in on the problem and the solution.
49:28And his quote is something like, I didn't fail 126 times, I succeeded once. And, you know, that was what was, you know, necessary for the world. And that's the process of innovation, which rollups enable, which the Proof-of-Proof protocol, POP, that Hemi uses enables is everyone, you know, worldwide is trying to figure out how to scale Bitcoin, how to scale Ethereum, how to scale Solana and some of these other blockchains, how to test new features. And as a developer, as an innovator, most of those are going to fail.
50:15And that's not dumping on developers. That's just economic pragmatism. Most experiments in that nature, they're either not going to see technical success, or they're not going to see market success. They're not going to pass through the Darwinian trial by fire. And that's OK. That's, again, the natural process of innovation. But if you have a group of chirping canaries in the corner saying, you know, don't do that. Don't develop. There's no reason for innovation. Bitcoin's already perfect. That's, to me, a non-optimistic, non-innovative and kind of, you know, shrouding attitude.
50:57So I disliked that kind of attitude. I'm far more optimistic. But I do see that innovation in general, it requires iterating. It requires failure. Then bringing it back to your question, like BOB and some of these other rollups, I'm aware of at least 70, 70 Bitcoin L2s. I'm aware of at least 50, 50 ETH L2s. But as far as I'm aware, Hemi is the only one that's Bitcoin and Ethereum brought together using Bitcoin as security. And again, we didn't, like our competitors, just take an EVM, airdrop it onto a Bitcoin anchoring method and call it a day.
51:49We've been building on this for two years. We've been working on Hemi for two years. And it's technically complex not only for the Proof-of-Proof protocol, but embedding that Bitcoin node into the EVM, extending in a backwards compatible way the EVM such that smart contracts on the Ethereum EVM side can talk directly to the Bitcoin main chain. That's new, novel, different. And from a Bitcoin developer perspective, that's extending Bitcoin script by 100x or more.
52:29Vlad CosteaI have a question from another one of the show's sponsors, which is Hodling.ch. And they're a self-custody service. They're basically advising you how to best set up your wallet. And they say, when it comes to self-custody of large amounts of Bitcoin, what do you see as the biggest risks and challenges for individuals or institutions today? And in your view, what best practices or innovations could make self-custody both more secure and easier to manage?
52:57Jeff GarzikThe phrase I usually go to is multi-multisig. Or another way to describe it is hierarchical multisig. Multisig has been with us again for almost 15 years. And here's one area where I do feel that Bitcoin lags a bit versus some of the security tools that have evolved around the Bitcoin ecosystem. And that is not a technical limitation of Bitcoin, but I argue Bitcoin businesses, Bitcoin projects just have not pushed the state of the art as they should. And so hopefully your sponsor and folks like that can push the edge in a particular security direction.
53:48Now, what do I mean specifically? A multi-multisig is a reflection of the fact that you need security not only in a multi-party or sharded type of approach, but you have groups of shards which themselves, that's where the hierarchical part of the multisig comes in, which themselves feed up to a permission to move a large amount of Bitcoin from A to B. Some large institutions, they'll have red, green, blue teams that are geographically separated and each of those have a multisig. And then you need, say, the red and blue team or the green and blue team to sign on the second level of the hierarchical multisig in order to move a large amount of Bitcoin.
54:46There's also an aspect that's very little explored social recovery. What happens if my boating accident is not a wink, wink, nod, nod, I lost all my crypto in a boating accident, but I literally am de-digital. My phone is gone, my ledgers are gone, everything is gone. Are you just up the creek without a paddle or is there a recovery scenario that's valid and secure? And social recovery, getting a quorum of your friends to commit over time that this guy swears to me he's really lost his crypto, and that's another hierarchical multisig.
55:38That's another approach that adds layers of security. And finally, you need to go beyond hardware wallets. You should be putting paper, like physical paper, in safes that are geographically distributed. So the full large amount crypto security type plan, it's quite involved and I think wallets make progress every day with increasing self-custody security. But there's a long way to go as well. There's still room for improvement.
56:17Vlad CosteaYeah, I have another question from Ray Youssef, who is the CEO of NoOnes. He used to work on Paxful previously, but now he is trying with this peer-to-peer marketplace to get Africans in general to join this economy. And do you still believe that Bitcoin will be used as this medium of exchange that should onboard more people who are unbanked? Or do you think that we should move on from this vision and use other currencies that are more optimized for payments?
56:50Jeff GarzikWell, it's more about the current behavior of any currency, rather than specifically Bitcoin, specifically the US dollar or Ethereum or other cryptos. It's difficult to use something as a medium of exchange if the price is very volatile. It's just a user difficulty. And if we fast forward 10 years, 20 years, 50 years and Bitcoin volatility is quite low, then the ability to use it as a medium of exchange increases. And so I wouldn't give up on it. I would just evaluate the present state of things and do a forward looking vision of what will probably be the volatility in two years.
57:46Three years, five years, 10 years and derive the answer about medium of exchange from currencies behavior.
57:56Vlad CosteaOkay, that's a good answer. Right now, I only have five minutes and two questions that I really want to ask you. One of them is more cheeky, to put it lightly, because it's about the fact that Greg Maxwell has been, at least according to historians who tell the story, he's so right about everything in Bitcoin and he has had this grand vision in which he was always ahead of his time. And you actually worked with him. And the question about him is if you can point out to any of his flaws or fallacies or moments when he was not necessarily right.
58:29And the other one is about SegWit2x and if you believe that the world would be better with Bitcoin if that one actually happened, as opposed to getting canceled in November or whenever that happened, or I think it was December.
58:47Jeff GarzikWell, I'll give two spicy and egotistical answers as well. I just, a little bit triggered by prep for this show, I tweeted out a link to a 2015 Medium post that I wrote, 'Bitcoin is being hot-wired for settlement.' And I'd like to think that I've been right about many of my predictions about outcomes. I think that if you take a look at economic incentives and again be very pragmatic about where those incentives point to, it was always inevitable that if you restrict the block size to a certain amount, then just like a, to make an analogy, the champagne waterfall that you see in hotels, the champagne comes down to that first champagne glass.
59:45Once it fills up, it spills over to the next level of glasses and spills over and so on. Well, if you squeeze bandwidth in a certain constraint, it's inevitably going to spill over. Whether it's spilling over developers to Ethereum in 2012, 2013 timeframe, or literally spilling over transactions because transaction fees are high, it's incentivizing the creation of layer two networks. And so that, you know, tooting my own horn a bit, I was, I think, completely right about that. Lightning, again, you analyze the economics of what it takes to onboard not millions, but billions of people onto the Lightning Network and using that for normal payments, is the incentives just weren't there.
1:00:44They incentivized against that for a number of different reasons. You know, I have to pay money in order to receive money. It's just ultimately macroeconomics are a real challenge on the Lightning Network. And so I'd like to think that a number of my predictions have borne out in that particular avenue. On the SegWit, the SegWit2x side as well, again, my egotistical answer is Bitcoin would be at $250K and not $100K if we had done SegWit2x. So, you know, wherever we are, we would have been farther down the road had SegWit2x upgrade occurred, but it didn't.
1:01:36And again, being, you know, very, very based, pragmatic, where are the economic incentives taking us? That spills over to today, the demand for rollups and layer twos like Hemi.
1:01:51Vlad CosteaOkay, I have like 30 seconds. What's your favorite scaling solution for Bitcoin and what's your favorite privacy solution for Bitcoin that was maybe not activated?
1:02:02Jeff GarzikWell, the scaling solution, kind of like what I was just saying, is layer twos and obviously Hemi. And privacy is, again, near and dear to my heart. I think we have a long way to go on the blockchain side. I think we have a long way to go on the wallet side on privacy to get to where I think users need to be. You know, you're not going to feature, you know, you're not going to onboard, say, Apple and Samsung's balance sheets, corporate balance sheets in their entirety in, you know, one particular dream scenario until there's contract privacy, payment privacy that is 100% secure, 100% shielded.
1:02:54Kind of the Zcash 100% shielded transaction type model. So I think that's where we need to go. And that's speaking to Bitcoin, that's speaking to Ethereum, that's speaking to Hemi. ZK has a good track record and probability of delivering that, but we're not there yet. And that's kind of the last piece that we all need to complete that freedom triumvirate I mentioned earlier of freedom of speech, end-to-end encryption. So freedom of communication and freedom of payment.
1:03:38Vlad CosteaMr. Garzik, thank you very much for your time. Thank you for teaching me about Hemi. And I wish you best of luck with the project or whatever else that you're going to do in the future. And you're always welcome on the show to talk more.
1:03:52Jeff GarzikOh, thanks so much. Your audience can find us at hemi.xyz or @hemi_xyz on X.
1:04:01Vlad CosteaOkay. And you are @jgarzik, with a K at the end. But I guess most people know this already.
1:04:07Jeff GarzikYeah, @jgarzik on X as well, though, be prepared for geopolitics and drones and all sorts of gadgetry far beyond your usual crypto Twitter fare.
1:04:22Vlad CosteaThat's fair game. It's a lot of fun.
1:04:26Jeff GarzikIt is. It is. Life is fun. We live in interesting times.
1:04:32Vlad CosteaThank you very much for your time.
