Ethereum has just successfully completed The Merge AKA its long awaited transition to proof-of-stake. With the change, the network is no longer dependent on power-hungry miners, a la Bitcoin. But what is the broader significance of this change? How does it work? And how did it come about? On this episode, we speak with Christine Kim of Galaxy Digital, who breaks it all down. This transcript has been lightly edited for clarity.
Points of interest in the pod:
Why The Merge is such a big deal (5:25)
Why it’s called The Merge (8:01)
Proof-of-Work vs. Proof-of-Stake (9:42)
The benefits of Proof-of-Stake (14:02)
Is Ethereum really trustless? (17:42)
Risks of concentration (22:46)
Have censorship risks gone up? (29:30)
What constitutes a successful Merge (41:36)
Scaling Ethereum through sharding (48:51)
Joe Weisenthal: (00:09)
Hello, and welcome to another episode of the Odd Lots podcast. I'm Joe Weisenthal
Tracy Alloway: (00:15)
And I'm Tracy Alloway.
Joe: (00:16)
Tracy, something very big is happening in the crypto world. And it's not about lines going up or down, because so much of what excites me about crypto is that the lines move so much. That's fun, but there's something else and it's not strictly price-related.
Tracy: (00:31)
The Merge, The Merge. I feel like someone needs to make, like, a movie poster with ‘MERGE’ written in creepy letters and make it look like an old horror movie kind of thing. But yeah, The Merge is happening in Ethereum.
Joe: (00:43)
People have been telling us, ‘oh, you’ve got to have a Merge episode.’ You know, there's so much going on. We can't always get around to every topic. But Ethereum is switching its consensus mechanism from proof-of-work, which is like kind of like Bitcoin mining, very energy intensive mining operation that most people associate with crypto, to proof-of-stake. And so it'll be a different approach to creating and validating blocks. And they've been working on it for years and years and years and it's finally happening.
Tracy: (01:11)
So I'm going to caveat this discussion with the fact that I haven't been following this very intently. However, I do find it interesting because it gets to something that you wrote, I guess it was a year ago, about the crypto world kind of splitting into these two different camps and on the one side you have people who are very into the technology and, you know, one aspect of technology is that it evolves, it changes. And so in that respect, this is Ethereum evolving and changing in a very big way. But the other camp in crypto are the sort of Bitcoin maximalists, the fundamentalists who don't want to see anything change about Bitcoin and the technology at all.
Joe: (01:54)
I'm really glad you brought this up. I think that is spot on because there's this question of like, ‘Well, what is crypto? Is it software? Or is it money?’ And I think there's these two camps and if it's software, if software is a really big component, then a part of software is upgrades, right? Upgrade cycles. And you know, it's like Microsoft might update its browser or something every year. And then at some point they're like ‘the old browser will not be supported. You can't use it anymore.’ Whereas the sort of money stance, the sort of hardcore Bitcoin view is no, if you're going to have money, you don't ever want to be told this money isn't good anymore. Or there's been a hard fork or there's been some sort of change to the network and you have to do something, you have to change something. You have to update your software to use it. And I think these are fundamentally very big things. And I think this Merge, this huge switch from sort of traditional mining to proof of stake is really important in sort of culturally setting Ethereum apart from Bitcoin in its willingness to change the rules, from time to time, of the network.
Tracy: (02:57)
A touchstone moment for crypto. But it also gets to the idea of pros and cons of different types of blockchains, right? So Ethereum is trying to solve one problem here, which, you know, might, for instance, be energy use, which a lot of people have focused on. A lot of crypto critics have focused on saying that mining wastes a ton of energy. So why don't we try to fix that? But on the other hand, does the new design come with its own set of problems? And we've already seen some noises around the idea of, well, maybe you're making it less censorship-resistant and we're going to get into that.
Joe: (03:31)
Yes. And so one of the developments over the last several weeks is the Treasury’s sanctioning of TornadoCash, which is a way for people to obfuscate their Ethereum transactions through a mixer. And it's the first time that Treasury has ever sanctioned a piece of software, which is pretty interesting. But then it also raised the question, well, if they can sanction a piece of software, why can't they sanction Ethereum itself? Or can they tell Ethereum holders or Ethereum stakers in these new proof of stake mechanisms ‘You're going to get in trouble if you process blocks from entities that are trying to launder money.’ That could open a whole new can of worms. It's always evolving. We're always trying to keep up here on the podcast. So, let's try to learn some more about it.
Tracy: (04:15)
Let's do it. I don't even know what a validator is versus a miner.
Joe: (04:19)
A staker, a validator, a miner. I don't know any of these words. So we're gonna be talking to someone who does. I'm very excited to welcome to Odd Lots Christine Kim. She is a research associate at Galaxy Digital. And she's been writing about these topics for Galaxy clients for a long time. So Christine, thank you so much for coming on. How did we do there in the intro? You know, this is like foreign territory for us. How did we do there?
Christine Kim: (04:43)
That was excellent. I think we're already done, you guys have condensed this entire podcast and everything I wanted to talk about.
Tracy: (04:50)
Oh, we can stop now.
Christine: (04:51)
Yes we can stop now.
Joe: (04:53)
Tracy and I, we had a busy day. We started prepping about 15 minutes ago and I said, ‘Tracy we're gonna be talking about proof of stake.’ And Tracy turns to me like three minutes later. She was like, ‘I think this is gonna, lik,e create some centralization risks for Ethereum” Basically got it right away..
Tracy: (05:08)
Thank you Joe!
Joe: (05:09)
I was really impressed! Ok, so from your perspective, Christine, let's just start with how significant is this moment for Ethereum? And what's the real goal? We know reducing energy consumption is part of it, but why don't you put this in context on the Ethereum roadmap for us?
Christine: (05:25)
For sure. I think it's hard to overstate how big this upgrade is. The transition to proof of stake has been part of Ethereum's original development roadmap back when the blockchain first launched in 2015. Developers had thought that this upgrade would be ready earlier in 2016, but due to the technical challenges of actually swapping out the consensus mechanism of Ethereum while it's live, that brought forth delays.
And so people have been asking for this upgrade, developers have been working on this upgrade for around seven years now and it's truly, it was almost to the point where people had thought that it would never transition — that this transition to proof of stake was just a pipe dream. And so the fact that this is the week in which Ethereum will finally fulfill one of the promises that it had made to its users, its investors back when it first launched, I think is pretty monumental.
And it's really not just about the changes to its energy consumption, right? I think another big change that people are really looking forward to is Ethereum's monetary policy switching from being an inflationary currency to potentially a deflationary currency. The annual network issuance of the network is expected to drop from around 5% to less than 0.5%. And if you add in coin burns, which is like a new mechanism that was introduced back in October, or I should say last August, with EIP 1559, there is a lot of excitement around Ethereum's issuance of ETH and the supply of ETH actually contracting over time with more activity on the network, you're gonna see more ETH being burned and that impacting total supply. So I think that's another big part of it. It's like the economics around ETH that's going to be changing after the merge.
Joe: (07:22)
So I just want to ask one short question and make one point, we are recording this on September 13th, The Merge, the event is expected to happen in about a day. By the time you're hearing this, it should have happened. If it totally blows up or something, we might have to rerecord the entire episode, or maybe we'll put this out as an artifact of history that might have been, but just a real quick technical question. This is called The Merge. Does that imply that the proof of stake Ethereum already exists? And now the two networks are merging together? Is this other chain that has a different consensus mechanism already operating?
Christine: (08:01)
Yeah, that's exactly right. So the proof of stake blockchain of Ethereum has existed since, I think it was December 2020, it's called the Beacon Chain. And right now a very small portion of ETH is issued on that parallel chain and there's individuals and stakeholders that have already invested their money into that chain, the Beacon Chain, and for The Merge what's going to happen is that chain is going to become fused together with Ethereum main net today. But in the process of that change, all of the issuance that happens on Ethereum currently, which goes to miners will disappear, will go to zero. And so the only issuance of it that you have left is to the validators that are on the Beacon Chain now. And that's really just a fraction of the total issuance that's that's being generated today.
Tracy: (08:52)
So I have a ton of questions already. I'm also kind of hoping that our producer is able to put in sound effects every time we say ‘The Merge.’ Okay. On a serious note, can we back up for a second? And can you maybe describe the difference between proof of work versus proof of stake and also how Ethereum got into a position where they have two different types of chains? So Beacon versus the normal Ethereum chain, and then also on top of that, could you, maybe to describe the difference between proof of work and proof of stake, could you walk us through, like, how will a new Ethereum be created under this new regime and what happens to the miners, in this case.
Joe: (09:38)
I started by asking two questions in a row, so that now Tracy is asking a three part questions.
Christine: (09:42)
You're going to have to remind me if I forget to answer one of those. But let me start with just giving a broad overview of the difference between proof of work and proof of stake. And then we can go from there. I totally agree with you guys that like The Merge is a really big event and educating people around how this is actually happening and even the technicals beneath it, which sometimes can sound boring, is like really what's exciting. So I'm glad we're talking about this, but anyways, so I think it's useful to start off with what is a consensus mechanism or a consensus protocol, because that is what a proof of work and a proof of stake blockchain is.
It's basically, this is the mechanism that defines how nodes in a blockchain come to agreement about the state of the network. So what are the account balances? The transactions, the updated transaction history of the blockchain. There needs to be a way for all the computers that are connecting to the network, also called nodes, to be aligned about what the canonical history is.
And so it's really about how do you process blocks on a blockchain? How do you finalize those transactions? And with proof of work, you do that in a very energy intensive way. You have these actors that are called miners that are solving a very computationally-intensive math problem. These are called hash functions and every miner is competing to be the first person to find the correct solution, because that means that they get to build a block, include transactions in it, get the reward from the block.
But for proof of stake, these actors, these miners are replaced by validators. And instead of solving that very computationally intensive puzzle validators are voting on blocks and they're attesting to blocks and they just get randomly selected by the network, according to an algorithm of who gets to be a block proposer. So on Ethereum, it's not that when Ethereum’s transitioned to proof of stake, I should caveat these validators are not competing. They're just randomly selected to propose a block and they'll get rewards from that. They'll also get rewards from voting on blocks and attesting to the validity of those blocks. But the question is, why are these validators on Ethereum? How do we keep them honest? Because with miners, you've already expended so much computational energy. You've kind of input in a very high cost. You're not going to lose the chance to earn those rewards.
After you've sunk in a particular amount of cost with validators, you haven't really sunk in anything, you haven't expended any energy to vote or attest to blocks or to propose blocks if you did. It's very negligible compared to what miners do. So what validators need is, we need a different way to keep these validators honest, and that way is through a stake. So sometimes people call validators ‘stakers,’ they try and use that term interchangeably.
But at the core of that is validators at least on Ethereum are staking a large amount of ETH. They're staking 32 ETH which I haven't checked the prices as of late, but it's a significant amount. And if they do try and cheat the network, if they do propose a block that goes against the rules of the network, that they're trying to confuse the network, trying to attack the network or change the validity of the chain. There is a potential that that amount of stake that they've put into the network gets slashed. So it's a very different way of keeping actors, honest for miners. You are expending a lot of cost up front, and that cost kind of keeps you honest. But for validators, what you're doing is you're locking up your capital and you're letting the network kind of hold onto it. And the fear of having that stake slashed is what keeps you honest.
Joe: (13:41)
So yeah, you mentioned it. So you have to put up right now at around $1,600 an ETH, you're putting up a minimum to be as a validator, $51,000 US dollars. And I guess it's kind of like a security deposit, huh? The risk is if you do something untoward to the network, if you vote bad, if you try to approve invalid blocks, you lose part of your security deposit.
Christine: (14:03)
Exactly. And so this mechanism for proof of stake is seen, one of the benefits of it is that it is more ecologically friendly. It has better for the ESG narrative because you're not putting in so much energy just to build a block, you're being randomly selected by the network to build a block. And the reason why you can be trusted to make sure that that block is correct and is valid, is because you've already invested a certain amount of stake to the network, right? And for Ethereum, I think the original idea was, okay, we have this mining consensus protocol that is already finalizing transactions, progressing blocks. We should move all the applications and the users to a new consensus protocol, to a proof of stake network and blockchain. But the concern, there's a significant amount of complexity around that because what is that moving process going to look like? Ethereum over the past couple years has just skyrocketed in terms of active addresses, total value locked.
I think the amount of value that has grown on Ethereum, the amount of activity that has grown on Ethereum, has made the vision to just simply move users and applications and value to another new chain infeasible, very difficult to do. So instead of moving users on the current Ethereum chain to a new proof of stake, blockchain developers had thought of this alternative idea, where you launch the proof of stake version of Ethereum, and you simply fuse that version of Ethereum to the existing Ethereum. So one of the cool things about The Merge is that it doesn't impact the application layer of Ethereum. It really only impacts how blocks are finalized. So as the current Ethereum blockchain is progressing blocks, it'll communicate those blocks back to the consensus layer of Ethereum, which is the Beacon Chain. And the Beacon Chain will start to take over the responsibility of finalization.
So that question of, you know, why is it that Ethereum has a parallel proof of state blockchain? And why is it that we're going down this roadmap of merging together the blockchains rather than simply upgrading the existing proof of work chain? I think it really comes down to simplicity and it comes down to how do we do this upgrade in a way that doesn't result in downtime and doesn't result in disruptions to a network that has just grown so much quicker than I think core developers had anticipated when they first launched Ethereum. And I think that's also why we've seen a lot of delays to this upgrade because the value of this chain and the amount of user activity on it has made it so that, you know, this upgrade when it happens, has to be done in a way that's very airtight, that poses the least minimal amount of damage and of risk to the users and to the applications. So I think I missed some questions?
Tracy: (16:59)
I think you did all three, actually. That was great. That was really good. So one thing that I find odd about crypto in general is that the problem that they're trying to solve is the problem of like, how do you do trustless transactions? Two parties don't trust each other, how can you use technology to get in there and make it so that people can transact with one another in a protected way. But at the same time, it feels like so much of it is like, or at least, especially in the Ethereum case, so much of it is built on consensus. It's like two parties can't trust each other, but we trust the system as a whole to reach a consensus. And that's basically how, you know, proof of stake is working. How do they actually get to that consensus? And what happens if like one validator in a transaction rejects a block?
Christine: (17:54)
Consensus really is at the core of these technologies. It's a really good question because these systems are meant to be trustless. It's meant to cut out the middlemen. Like you said, and for proof of work, consensus protocols, that trustless interaction between miners validating and earning rewards from block production is actually much simpler than proof of stake consensus protocols, because proof of stake consensus protocols don't rely on an external good. It doesn't rely on energy. It just relies on an internally created asset. Like ETH, you have to assume that ETH is a worthwhile asset for proof of stake to work. So that question of, you know, when a validator rejects a transaction or when a validator creates a block that all the other validators think is false, or goes against the rules of the network, that's something that the protocol level of Ethereum as a proof of stake blockchain is automatically checking for.
So there are certain rules around how you can propose blocks. So one of the ways that you prevent against double spends, basically like somebody saying that I spent $5 and I can spend another $10 from the same address and not change like the account balance, that is prevented by the network basically checking for double block proposals. Like if a validator, were to propose two blocks at the same time, that's like a slash event. That's something where the stake that they've put into the network gets reduced. And there's also other ways in which validators can keep each other honest, even if those automatic rules aren't able to catch all the activities. So this kind of goes into the censorship question of like, let's just say, we've noticed that a certain validator continues to reject transactions from an address that's on the OFAC sanctions list.
Validators can coordinate to basically blacklist those malicious validators, because when you've put your stake into the network, you've also told the entire network that, ‘Hey, this is my validator ID. This is my address.’ You are no longer an anonymous stakeholder. Whereas for, I think miners, when you're dedicating hash rate or hash power to basically computational energy to the network, that kind of labeling system is harder to do. But for validators, once you've locked in your 32 ETH to the network, it's held by the network and it's also identified by the network. So another way in which validators keep each other honest outside of these like automatic rules, is this ability to kind of put bad behaving validators out of the network. Now this requires social consensus. This would require some sort of an upgrade, some sort of way for everybody to coordinate against those validators.
But that's kind of like another final resort where I think it helps to understand, like, how is this network coming to consensus? Initially it's like these rules, these pro pre-programmed rules that are part of the protocol, but sometimes rules can't always catch all of the malicious behavior. And in the case where you're not able to catch the malicious behavior, there's also this additional step that you can take. Sometimes it's called social slashing, where validators can basically remove certain bad acting validators from the network and slash their stake, even though they haven't necessarily gone against the technicalities of the rules, but they can just kind of coordinate to do that.
Joe: (21:39)
So I think this really gets to what was gonna be my next question. And it's one of the criticisms of proof of stake, which is, okay, most people probably, I assume most people who own some ETH don't have 32 of them, or don't have $51,000 worth, but it doesn't mean they can't participate in staking. And my understanding is like, okay, take some random person [who] buys a few ETH, leaves it on Coinbase. Coinbase can then be itself a huge validator But talk to us about the risk of a few mega validators, because the risk is I think that you are essentially undermining decentralization such that, I think there's something called Lido. There's Coinbase, probably a few others, but the risk of everyone just putting their money with a couple, and then you just have a couple public well-known entities who are in theory dealing with the laws of their land, the law enforcement of the countries they operate in, and the executives of these companies, a handful of entities with an incredible amount of staked ETH [and] therefore network power under their control.
Christine: (22:46)
It's a big concern. I definitely have to say that it's always been known — this potential for a lot of stake to become controlled by centralized entities like Coinbase, like Lido. But I think the recent sanctions against TornadoCash were just like a wake up call for the community. And because of that, there has been a lot of conversation around what would happen if, you know, these entities started to censor transactions. I think first it's not totally clear that these exchanges and these centralized taking providers will need to, but in the event that they do, we shouldn't understate the role of independent validators in the system.
So if by chance there's a transaction that Coinbase or Lido starts to censor. They're not gonna include it in a block. Eventually the network will pick an independent validator to propose a block, and that independent validator will not be judging transactions by, ‘Are they on the OFAC sanctions list or are they not?’ They'll just be picking it from the public mempool, but let's just say, you know, for the sake of argument that a hundred percent of validators, that not even 5%, not even, you know, 20% or 15% of validators are even independent. In that case, there can be changes to the protocol made so that a certain amount of transactions are kind of enforced by validators to include into their block. This comes back down to another kind of area of discussion, which is around OFAC compliant relays. I don't wanna get too technical to this but...
Joe: (24:21)
That’s alright. I think we said at the beginning, I think ‘relay’ was one of the words that I don't know what it means, so let's go for it.
Christine: (24:27)
Well, great. So relays are basically a third party software that validators will connect to in order to earn additional rewards on the blocks that they create. So you get a certain amount of reward for just producing the block. You also get rewards through transaction fees. These are additional amounts of ETH that people can add to their transactions for greater priority. They're sometimes called priority fees, and there's also MEV — maximal extractable value — which is what happens when transactions are ordered in a certain way that allows for arbitrage, allows for sandwiching, basically profits from positioning trades, usually decentralized finance trades in a very specific way. So validators are not super savvy in identifying opportunities for MEV. Validators are really, you know, operators that we wanna assume are just running a piece of software.
They're just running the consensus protocol of Ethereum as is, and they're just sitting back like earning the interest on their 32 ETH. But if they wanted to earn additional ETH, they can connect to a relay, which is this third party software that connects block builders to validators and block builders are the ones that are interacting with searchers, which are very highly specialized users that are able to look at the mempool and bundle transactions in a profitable way. And these block builders, they construct a block. They construct a very profitable block that gives more in terms of rewards than just a regular block that validators would create on their own. So some of these relays are operated by entities like Flashbots and Flashbots has publicly and has been for a very long time compliant with regulatory laws and has said that, you know, we are an entity that will be censoring transactions that are on the OFAC list, or transactions from addresses that are on the OFAC list.
And they're kind of a major, they're going to be one of the major relay operators, but there are other relay operators like BloxRoute that have said that we'll operate relays that validators can connect to that. They won't be censored. And going back to the hypothetical, that like all these relays are suddenly censored, let's just say, there's not even one. Then validators can enforce something called the CR list, like censorship resistant list. This is a technology that's still in the works, but is something that developers could potentially roll out if they see that, you know, all validators or all relays and MEV extraction is just kind of going to a specific relay, like Flashbots, that's, you know, censoring transactions. And it's very difficult for validators to include, even independent validators, to stay competitive to earn ETH, and to do so in a way that's censorship resistant, what developers can release and what they're considering as a potential solution is implementing CR lists, which are a portion of the block, that validators stuff with transactions directly from the public meme pool.
So instead of receiving from a relay, an entirely pre-built block that's already censored, that validators themselves can only accept or reject. They're able to enforce, you know, a portion of that block. You must include these transactions. So it takes away the power of block building and of censoring transactions away from the block builder, because there is this assumption that because the validators, the validator base of Ethereum, won't be completely controlled by these like centralized entities. We want to keep the power and keep the ability to, like, include transactions more in the hands of validators than in the hands of these other, potentially more centralized entities.
Joe: (28:18)
By the way, Tracy, you know, in addition to THE MERGE itself, I know for at least like a couple years or year, I've been getting tweets about how we had to do an episode on MEV and like this, which I think might be like the crypto version of a payment for order flow, or things like that. Christine's answer, that was a reminder that that's we’re probably going to have to do a whole separate episode at some point.
Tracy: (28:42)
Maybe we should just do an Odd Lots series where we go through, like, every term. Some of these terms are kind of like weird, like I'm going to ask you, I have to be careful how I pronounce this, at some point, but Christine, I'm going to ask you about ‘sharding,’ later in the conversation for sure. But okay. Before we move on, just on the censorship centralization issue, I'm curious what the Ethereum people have said about this, because one of the unusual things about Ethereum versus a network like Bitcoin is that you actually have a figurehead in the form of Vitalik Buterin. And I'm curious what he's said on this issue, because I think most crypto people are ideologically opposed to centralization and middlemen and aligning themselves with government requirements like sanctions and things like that. But at the same time, the more we talk about this, the more you could kind of argue that Ethereum is sort of going mainstream and maybe refining itself so that it better fits into the existing financial and legal system. And that could also be a strength. So I'm wondering what they've said on this issue.
Christine: (29:52)
That's a great question. I think that the community, especially in the aftermath of what happened with the sanctions against TornadoCash, we've seen a lot more segmentation, a lot more disagreement, I think in the Ethereum community about what is the best way forward. There are people in the Ethereum community that are a lot more cypherpunk, a lot more [aligned with] the vision of what the Bitcoin community is that even the slightest amount of censorship on Ethereum should be condemned.
Even if it's only Coinbase, even if it's only Lido, those entities should be punished and should be, you know, removed from the network. I think that's a very extreme view and most Ethereum core developers — And I don't speak for Vitalik — but I would assume that him too, most have landed in this middle ground of even if there is centralization, even if there is censorship happening by these exchanges, so long as there's even a small amount of independent validators that are processing transactions, these non-compliant transactions will get into the Ethereum blockchain eventually.
And that means that Ethereum is kosher, that Ethereum can still be considered a censorship resistant network. And then I think you can go to the other extreme where, you know, very big entities, major entities in the sense of like they’re huge figureheads, I guess, in the Ethereum community and that's entities like Flashbots that are very open about the way that they are compliant and about the way in which they don't see a future in which they're operating, you know, in like North Korea or like places in which US sanctions don't matter. So just very pragmatic and realistic, I think, but not trying to fight the powers, like not trying to really rock the boat by choosing a different path.
And so I think there's this tension, this tension between even the middle ground of developers and of individuals that want to preserve the censorship resistance of the network, but have to face the realities of like these big players that are core to the infrastructure of Ethereum, like core to what Ethereum is today. You can't necessarily just cut off all the exchanges. You can't necessarily just cut off Flashbots because they literally built the software for how validators are gonna earn MEV. So in that future, you know, you have to negotiate, you have to think about other third ways. And there's actually a really great talk by Vitalik, who recently went to the Stanford blockchain conference, about how he foresees different ways to decentralize the block building community. And so really, I think developers have landed on like, how do we improve the situation? How do we decentralize Ethereum further? But just recognizing that in the short term and in the medium term, there's a high potential that transactions will be censored. And that Ethereum, as a staking community, as like a validator community ,will be controlled majority by these regulated entities, which is a pretty, I think, alarming fact.
Joe: (33:19)
So I'm thinking back to something you said earlier about this idea of like social slashing, such that a theoretical validator, even if they didn't technically break consensus mechanism rules, could potentially lose some of their coins if the other validators voted in such a way. And you know, going back again, it's like Vitalik himself could say anything he wants, but Vitalik doesn't have to deal with like Gary Gensler and Vitalik doesn't have to deal with Treasury. Brian Armstrong on the other hand does.
And so Brian Armstrong … are there any attacks? And I use attack liberally. I don't mean a hack, but I mean, are there any like attacks that essentially work through the social slashing mechanism, such that a government, some government somewhere, or the US government or the Treasury specifically, can do something damaging to the Ethereum network through these entities? And, you know, is there a form of social slashing that Brian Armstrong might have to do or might have to push for potentially in some theoretical future where it's not about penalizing a entity that broke consensus mechanism rules, but something that they have to do, sort of at the behest of a government?
Christine: (34:30)
Like, would the government potentially call Coinbase and say ‘Could you censor these transactions?’
Joe: (34:37)
Yeah, ebcause attacks on any blockchain are difficult. It would be difficult for a government to attack the Bitcoin blockchain in part, just because it would be hard for a government to acquire the hashpower potentially to acquire enough chips, such that it could take control of the network or execute a 51% attack or something like that. But again if, you know, Coinbase ends up as the dominant validator and Coinbase has to go by all these rules, is there something that the government in a future scenario could pressure Coinbase to do from a sort of like social consensus standpoint that other members of the community might view as being damaging to the integrity of the chain?
Christine: (35:23)
Yeah. I think in that case, where Coinbase does enforce these regulations from a government authority that the entire community doesn't also agree with, it would cause a split. It would cause a split in the chain of Ethereum, versions of Ethereum that are compliant and non-compliant, but I think that would also undermine the very value of Ethereum. So it's almost like thinking through like doing this thought process of what a social slashing event could look like and the split that it would cause should deter any proponent or anybody who's thinking of doing this because it might like irreparably damage the value of the chain.
Joe: (36:06)
A good example, perhaps, I'm thinking of is like, what if the government said ‘Uniswap and other DeFi exchanges are illegal stock markets that are unregulated by the SEC. Coinbase, can you make sure that you don't validate any transactions that interact with these DeFiexchanges?’which would be like a massive rupture because DeFi of course is crucial to how Ethereum works. But one could imagine, say like you can't be processing transactions for a rogue stock exchange or something like that, which would be an attempt at imposing a very severe kinds of censorship. What happens then? And how does the network heal or find a way to route around such a big entity being told by the government that it could no longer process DeFi transactions?
Christine: (36:52)
I think ideally, I mean, this is very like an ideal sentiment. I don't know if this what actually happened, but ideally users of Ethereum recognize that this is not appropriate or is not behavior that they should support. And they take away their stake from Coinbase. They don't stake through Coinbase, like as a staking provider, Coinbase falls and, you know, other decentralized staking providers like Rocketpool and potentially Lido down the road if they do fully decentralized, that these are the staking providers that step up. But of course this requires a great deal of cohesion among the community, and like a shared belief in a shared value of, ‘Hey Ethereum only makes sense if it's censorship resistant. Ethereum only makes sense if staking providers can't actually censor transactions.’ And I think there's technologies that are being looked at, like zero knowledge proofs, to try and obfuscate even the contents of a transaction so that the power of validators to even know what kind of transactions they're validating is completely out of their control.
But of course, I recognize that that's not the reality today and that Coinbase does have the ability to build their own blocks and include whatever transactions they want as validators, validator node operators. And in this case, I think it really is up to the Ethereum community to choose staking through providers that they are confident uphold like the values and the ethos of the community. But what complicates even this is that right now, you're not able to withdraw stake from staking providers. That functionality is not enabled yet on Ethereum. It will probably be enabled, at a minimum six to 12 months after The Merge happens. And so that interim where, you know, we've already seen a lot of stake go to Coinbase and a lot of state go to Lido, I think the question remains of how users can coordinate.
And one of the ways is, you know, again, like we talked about — social slashing, but I think that kind of possibility is more deterrent. Like I don't think that it ever really comes down to it. I think it comes down to Coinbase censoring and users not being able to withdraw their stake and basically more independent validators being spun up to try and ensure that other trans that all these transactions that are, non-compliant still get included in the blockchain and arguing to regulators that, Hey, even if I censor transactions, it doesn't mean that Ethereum as a blockchain is any more like regulatory compliant, that these transactions are still gonna get included one way or the other. And that, you know, from a profit point of view, it doesn't make sense for us to even continue as a staking provider.
So like kind of argue that as a staking provider, it doesn't make sense for us to continue to censor transactions because they're gonna get included into the chain one way or another. And I think a very similar issue we saw with certain Bitcoin mining pools back in the day censoring transactions and those mining pools quickly being condemned by the Bitcoin community and kind of like social of social pressure, changing how their policies work. But I think hopefully, you know, a similar thing could happen in Ethereum. But again, as I mentioned, there's like those degrees and those schisms that are being created where certain big players in Ethereum, don't actually like fully subscribe to the decipher punk vision. Right. And I think in that case, it's not a hundred percent clear how cohesively the Ethereum community will act
Joe: (40:39)
Tracy, by the way, zero knowledge cryptography. That’s another episode we have to do.
Tracy: (40:45)
I feel like every answer you give Christine, and they're very good answers, but they just throw up a billion more questions. So I'm wondering, you know, what happens if a bunch of validators decide to kick out Coinbase for censoring transactions that like the CFTC is asking for, then are they immediately in violation of the CFTC or US law or something like that? But okay, maybe a slightly less thorny topic, how do you judge the success of The Merge? Is it price of Ethereum goes up? number of transactions goes up? Gas fees go down? I don't even know if this has any impact on gas fees. That would be interesting to hear from you about. But what are you looking at when we're deciding whether or not not this was a successful exercise?
Christine: (41:36)
I actually take the very minimalistic point of view. I only want the chain to finalize. That's it. I don't care about the price. I'm not looking at Ethereum addresses. I'm not looking at transaction activity. Really for me, what I deem as a successful merge is that after it, that the proof of stake blockchain fuses together with Ethereum main net and that version of Ethereum finalizes and that it is able to progress through epochs like be able to verifiably create new blocks, come to consensus. There's a really, this is a shameless plug, but there's this report that I've written on how to watch The Merge, that you can find on galaxy.com. But it illustrates that what you're looking for is basically the progression of two epochs which are, they’re intervals of time.
And in order for an epoch to finalize, you need at minimum like two thirds of active validators attesting to that epoch, saying that the transactions and the blocks that were completed in that epoch are all kosher and are all good. Once you've have, have had two epochs of that, you consider the network finalized. Because after that finalization point, it's very hard to revert the transactions or the blocks that had been created before that finalization point. So really what I'm looking for is just to see the chain finalized because it means, and the reason why is because it means that The Merge and the technical shift from just swapping out your consensus protocol, has worked. It doesn't say anything about how, you know, popular that upgrade was. How traders and the users view this change of proof of stake. It just says that, ‘Hey, this swapping of this transition from proof of work to proof of stake, worked. This very risky upgrade that requires two different hard forks worked. That transactions and blocks are continuing to be processed.’ And that, like, if you were to send a transaction on Uniswap, if you were to send ETH to another person, you now don't have miners processing those transactions. You actually have validators doing it behind the scenes. And that functionality is good to go. And that functionality, we don't have to worry about it breaking anytime soon.
Joe: (44:14)
So I just have one more short question. And it's a question that I'm thinking about a particular Twitter user. So Dan at CMS Holdings. He always DMs me and he says, ‘I love Odd Lots, but you guys are so negative all the time. It's always gloom.’ And we've spent a lot of time talking about like risks of The Merge and, you know, centralization and censorship. Other than the decrease in electricity consumption, talk to us about what's the exciting thing here? What's the good thing besides that, that this is going to open up at your view and the long term ethereum roadmap. What's the positive here?
Christine: (44:49)
This is going to sound very bearish, but there's not actually too much.
Tracy: (44:53)
Well, we tried Joe.
Joe: (44:55)
I tried Dan, sorry, if you're listening because I asked a question just for you. I'm sorry. I tried.
Christine: (45:02)
I think one thing is, you know, the validators that have been so faithfully on the Beacon Chain earning this issuance, it's a very small issuance in comparison to what miners get. But then again, validators aren't expending a lot of energy. So of course you're not gonna get that much. Validators, one thing they can look forward to is, they're going to start earning transaction fees, priority fees, and they're also going to start earning MEV. So that kind of reward, which, you know, compared to the network issuance they get, which is locked, they can't actually move that around. They can start moving around, and realizing the fees from transactions and MEV, they can start, you know, sending that over to exchanges. They'll actually start earning that. So I think that's kind of a positive for validators and that it just becomes more profitable to run one.
And then the other thing I think about Ethereum price is that you've got a massive supply drop, you know. All of the supply that's going to Ethereum is going to drop from around 5% to 0.5%. And in addition to that, you've still got coin burns happening. So that 0.5% in times of of high network activity will very likely drop to a negative number where the total supply is actually contracting. And so you've got, you know, a bunch of users that are locking up 32 ETH and then you've got, you know, issuance of the network dropping significantly. I think the liquidity of ETH, I'm not really a trader, but like that the supply going down I think will have a positive impact on ETH price over time. And I think that's something that people really look forward to and that ETH will become, I don't like this term, but quote unquote, ‘ultrasound money,’ you know, instead of having like a supply limit, you know, you've got this supply that's shrinking over time.
Obviously the ESG narrative of Ethereum will continue to thrive. And in comparison to Bitcoin, I think there's going to be a lot more narrative around, you know, the way that you mint NFTs, the way that you do all these things are no longer as energy intensive as they used to be. But I think one of the reasons why I say like all of this, isn't all that positive, which it is, it is very positive, is that I've been really waiting for a long time around Ethereum scalability. And The Merge really doesn't do very much for Ethereum scalability at all.
So I'm really looking forward to the fact that after The Merge developers will really focus on scalability. And I think that's one of the things, developers have just been so focused on pulling off this upgrade. After this is done nd out the door, I'm really looking forward to developers tackling some of the other big issues on Ethereum, like count abstraction and scalability and staked ETH withdrawals, etc.
Tracy: (47:50)
What is sharding?
Christine: (47:53)
We've come full circle...
Tracy: (47:55)
I honestly have zero idea.
Joe: (47:57):
It's a ‘D’ in sharding.
Tracy: (48:00)
There’s a ‘D,’ yes. [Sighs] Podcast pitfalls. Sharding. So I see people on Reddit talk about this a lot. They're like, ‘oh, who cares about The Merge? What I'm really excited about is sharding,’ which again, sounds terrible, but could you just explain what that is?
Christine: (48:16)
For sure. I'm going give a high level overview and then I'm gonna give a shout out to a really great report around sharding. So sharding originally [laughs] Corry. Sharding originally.... I don't know why, but when you say it, I've never thought of the term sharding as weird or as like strange, but now that you say it...
Tracy: (48:35)
You know why? Because everyone in crypto, they're not talking to each other. They're just writing. And if you write ‘shard,’ it's fine. But as soon as you start saying it out loud, everyone's gonna laugh
Joe: (48:47)
This is such an insightful point about language...
Christine: (48:51)
So sharding originally the idea of it was look, the Ethereum blockchain has this limited transaction throughput. The block space of Ethereum, which is, you know, there's a certain number of transactions that can fit into a block. And these blocks are what get processed and built on top of one another. And you can't stuff a block more than it's limit. You can increase the size of a block so that it can include more transactions. But if you do that, then it becomes more computationally intensive for miners or validators to propagate that block throughout the network. And so you have a higher chance of chain splits occurring. You basically increase the load on validators and miners. When they're running a node, you have to have very sophisticated software to be able to continue to propagate these very, very heavy blocks throughout the network.
So there's a good rationale for why you want to keep the size of blocks manageable for an ordinary node. It helps with the decentralization of the network, but anyways, so you've got a limited amount of transactions that you can include in a block. And if there's very, very high amount of transactions waiting to get included, then you know, you, you have very high fees. You've got long wait times. What if we were able to partition the blockchain so that instead of blocks being confirmed by this single Ethereum blockchain, you have mini blockchains also called shards that are all processing the transaction load of Ethereum in parallel together. So you've got like, let's just say for hypothetically like 64 mini blockchains that are all looking at the transaction memepool of Ethereum, which is this public space where everybody sends their unconfirmed transactions. And these miners in these validators on these shards are picking out, you know, transactions from there.
And they're all working together to process Ethereum blockchain as a whole. So that greatly greatly improves the transaction throughput of Ethereum and the scalability of Ethereum. However, it's an extremely complex design. 64 mini blockchains and even like, thinking about how like transaction atomicity — I think I'm saying that wrong — but basically like how would you be able to communicate like the finalization of one transaction on a specific shard to another shard? And is there latency between that communication? So basically that was the original idea for sharding, but again, like the complexities around sharding, the many unanswered questions around how transaction execution would work atomically throughout the whole network, those questions started to change how Ethereum developers think about sharding. And so now that roadmap and that vision is scrap — Ethereum developers, as a side note, has gone through many, many, many different iterations of how they think they're gonna scale the blockchain. And now they've landed on this other idea, which is very much focused on modularity. So instead of having transactions all execute and all finalized on the same chain, what if we abstracted away the burden of transaction execution to a layer two and with a technology — technologies like zero knowledge, technology like optimism, which are, you know, different types of rollups — I know I'm using a lot of technical types.
Joe: (52:20)
No, these are all future episodes.
Tracy: (52:23)
We're writing them all down for the inevitable series.
Christine: (52:26)
These are good for deep dives, but as at a high level, what if you abstracted away some of the responsibility of executing the transactions to a different network and that network can batch together and compress those transactions, and then only verify like the proof of that batch transaction to Ethereum, right? So that like greatly frees up the transaction and the block space of Ethereum, because now not all transactions are finalizing on Ethereum. You've got batches of transactions that are finalizing on a layer two, and you're just PR sending down the proofs of those compressed transactions to Ethereum. And so danksharding is a new iteration of sharding that really focuses on making the cost of rollups the cost of these batch transactions cheaper and introducing a lot more modularity to the Ethereum blockchain and achieving scalability.
Tracy: (53:22)
I’m sorry, did you say danksharding?
Christine: (53:23)
Yes. So that's actually the sharding roadmap for Ethereum now, and it doesn't really have anything to do with sharding, like the original idea for sharding. And this is where I plug Jon Charbonneau Hitchhikers Guide to Ethereum where he talks a lot about this, but yes, you're right. Danksharding is the weirdest term, I know, but that is the real version of sharding that is more likely to be implemented today than the version of sharding that I explained before.
Joe: (53:50)
All right. So we have danksharding, sharding, layer 2s, optimistic rollups vs zero knowledge proofs, the Ethereum ESG narrative versus Bitcoin, MEV. You've given so many future episodes for us to now build on. Christine Kim, thank you so much. You're the perfect guest for coming on. We say that a lot, but that was so clear and so good. And I know people have told us we need to do a Merge episode, and I'm glad we didn't just like rush it. We got a great guest. So thank you so much, Christine, for coming on the podcast.
Christine: (54:20)
Thank you so much for having me, you guys. This was lovely.
Joe: (54:23)
This was a lot of fun. Thank you so much.
Tracy: (54:24)
Yeah. Thanks, Christine. That was really interesting. And I don't say that about every crypto episode we do.
Joe: (54:44)
Danksharding. The future of all of finance is going to be whether these coders can make danksharding work. It's like a real word. It's one word.
Tracy: (54:52)
No, I know. But this gets to a real point, which is like, if you're, portraying yourself as the future of finance or the future of money, like can't we get some different terms? Things that people could say aloud in a meeting?
Joe: (55:05)
The one that has to go is ultrasound money. That has to go.
Tracy: (55:10)
Well, it also begs the question of like, if you were gonna create ‘ultra-ultrasound money,’ could you just like evaporate it, just burn it into oblivion? Like is that the soundest money there is… ?
Joe: (55:21)
Just the one coin left.
Tracy: (55:22)
Yeah the one coin. Okay. On a serious note, I thought that conversation was really interesting, mostly because it gets to that fundamental tension, which we kind of alluded to in the intro, which is if this is technology, if this is software, software is supposed to adapt to the needs of the people using it, or the needs of the market using it. And so it throws up these questions of how best to adapt. What are you sacrificing as you try to reduce energy usage and all those kinds of thorny questions?
Joe: (55:52)
Yeah, I think Ethereum is in an interesting position straddling the sort of two worlds, right? Because it was sort of one of the earlier chains. And as Christine mentioned, there is still a significant faction that has that sort of OG cypher punk anti-censorship impulse. On the other hand, it is a more corporate chain and VCs, that's what they they've put a lot of money in, and financial institutions experiment on it. Then there is like the pure software and, you know, some of these newer chains like Solana etc. That's just like a company launched that. I mean technically the company maybe doesn't control it, they're faster. They could probably upgrade even quicker than Ethereum. They already started on proof of stake, etc. So the question is, can Ethereum sort of navigate the two tensions, the sort of like community decentralized cypher punk tension with the software world? And this is a big moment in terms of I guess navigating those two worlds.
Tracy: (56:53)
It really seems like that's what they're trying to do. So it'll be fascinating to see what happens. And then of course, if, you know, if someone like this CFTC — to use your example — were to come in and say something and you were to get a validator, like Coinbase, kicked off the network, like it would just be fascinating to see how that consensus mechanism actually worked. And then what happened to all the other validators?
Joe: (57:16)
It really is interesting that The Merge is happening so soon after the TornadoCash sanctioning, because that's like the first time, right? Government's like, ‘no, we're going after a piece of software.’ And so, you know, it does raise the stakes potentially for the government has done all kinds of things with crypto, but it's usually like at the fiat onramp level, right? They're like, okay, you need to apply KYC, AML to the money you're bringing onto the exchange. But then once you're on, once you have the coins, then the government has basically been pretty hands free. And this is potentially a change right at a moment in which some of these big centralized entities are gonna have a lot of power over the network.
Tracy: (57:57)
Yeah. I'm also just interested in the PR aspect of all of it. If, you know, if the government says like, ‘we don't want you to deal with North Korea or don't let a north Korean entity like mine blocks or make transactions on your blockchain, and then you have a bunch of people going ‘well, no, actually we're censorship resistant. And you know, this is about making censorship-free money and transactions and all of that.’ That seems like a difficult position to take, or at least a, a difficult one when it comes to like broadcasting that message.
Joe: (58:30)
Speaking of PR, I suspect that the crypto industry is going to really turn on Bitcoin fast. And they're going to say, ‘look at this electricity guzzling blockchain, we have something that doesn't guzzle electricity anymore.’ Penalize those proof of work people. I think that that battle is coming — the sort of the ESGification of crypto and the vilification of change that don't move to proof of stake. I think will be a big story.
Tracy: (58:56)
Oh, I totally agree. But it's also really interesting to see Bitcoin kind of embrace that position in the system. And I think I wrote about this at one point, but Bitcoin proponents are positioning themselves basically as the anti-crypto now. That's the foil off of which they are playing. And I don't know., it's just been, it's been fascinating to see that narrative be created. All right. Well, okay, we could talk about this forever. Shall we leave it there.
Joe: (59:22)
Let's leave it there.
You can follow Christine on Twitter at @christine_dkim.