On The Brink with Castle Island - James Prestwich (Summa) (EP.04)

Episode Date: September 30, 2019

James Prestwich, the founder of Summa joins the show to discuss blockchain interoperability, developer incentives, and how economic interests can influence the protocol development process. For more information please visit our website at www.castleisland.vc and follow us on Twitter @CastleIslandVC.

Hello and welcome to another episode of On the Brink with Castle Island. I'm your host, Nick Carter, and today I have the privilege of having James Pruswitch on the show. Many of you will already know James. For those of you who don't, he's the founder of Summa, which is a startup building tools to connect blockchains like Bitcoin and Ethereum. One of his best-known contributions is helping to design the T-BTC system, along with Keep, which is a way of making Bitcoin usable on the Ethereum network. So far, cryptocurrencies have been relatively siloed.

So whether or not these cross-chain swaps that James works on are successful is a critical question how the industry will develop and how these blockchains will come to relate to each other. Are we going to end up with one dominant chain that individuals store their wealth in and occasionally move in and out of when they want to transact on some other platform? Or will we have a plurality of blockchains ending in some relative equilibrium? Or maybe cross-chain movement will end up being taken. too cumbersome, then the industry will consolidate to just one or two blockchains. So we brought James

on because he's perfectly situated to answer these questions. The other reason we want to have him on is because he's a very unique perspective on the industry in that he actively participates in technical debates in Bitcoin, Ethereum, Zcash, and a handful of other blockchains. James has a great reputation for being relatively candid and forthright about some of the flaws with these technologies. And that's another reason we were excited to have them on. One of the objectives of this show, aside from technical discussions about cryptocurrency, is to explore the political implications of the rise of non-state money. And it's our belief at Castle Island that public blockchains are a new form of institution,

granting individuals around the world access to a transparent, fair system of property rights, which are not dependent on the authority of any one state. but these new institutions are not perfect by any means, and in some cases they fall prey to the flaws of our legacy, monetary institutions. Chiefly among these is the opportunity for insiders to exploit their status, and that's what I was really excited to talk about today. As monetary systems, cryptocurrencies like Bitcoin aim for predictability and issuance, and they aim to distribute new units fairly by making them costly to produce. The leadership, in theory, has no advantage when it comes to printing money. Many would claim that this is an improvement on existing monetary policy, which could be argued has led to a massive run-up in financial assets over the last decade. Since these are mostly owned by the wealthy, this is increased inequality in society.

So cryptocurrencies claim to offer an alternative to this, but discretion still exists in those systems. Numerous examples exist of insiders exploiting their status. It's not always done maliciously or in a self-serving way, but it certainly poses the risk of delegitimizing these systems. James says really well-defined views on the topic and plenty of anecdotes to boot, which is why we're so glad to have them on today. I really enjoyed this conversation and hope you do too. Brought down by bad mortgage investments, Lehman, which has 25,000 employees will be liquidated. The federal government loans American International Group, AIG, $85 billion. This is a different kind of market, and the Fed is asleep.

The federal government is stepping it to stabilize Fannie Mae and Freddie Mac, the two mortgage giants that have been threatened by the housing crisis. The Bank of England has pumped 75 billion pounds more into Britain's ailing economy with a new round of quantitative easing. You print a couple trillion dollars and all of a sudden people start to worry. So out of this worry, we have something called a Bitcoin. Bitcoin. Welcome to the On the Brink podcast with Castle Island. I'm your host Nick Carter and I have before me James Presswitch. Welcome James. How's it going, Nick?

Thank you for being on the show. It is a privilege to have you on. It's good to be here. This office has a nice view. It does. We're looking at Cambridge right now. So James is the founder of Sama, which is a company that builds custom-built interoperability solutions. And if you don't know what that is, you will learn in the following hour. He is one of the creators of the cross-chain group, formerly of storage. And he is just generally speaking, speaking highly interdisciplinary in crypto, very lucid, very sanguine about cryptocurrency and not afraid to voice his opinions, which is part of the reason I'm so thrilled to have him on the show today. Now you're making me a little afraid to voice my opinions. You're talking me up too much. Now you have to. So James, thank you for being on one of the very first episodes of On the Brink with Castle Island. We're very excited. Let's start off by talking about summer, your main thing. How would you describe summer to an outside? side or that doesn't know a lot about crypto. So we work on making the existing coins interoperable

with each other. So you have Bitcoin, which most people have heard of. You have Ethereum, which very few people outside crypto have heard of. We're trying to take these things and make them play nicely, remove the walls between the gardens, and generally, like, improve the state of the crypto ecosystem. Is the main chains you connect are Bitcoin and Ethereum? Right now, we have current projects with Zcash and Cosmos. So we're actually working with the companies and foundations behind those chains to bring Bitcoin to Cosmos, to hopefully bring Zcash to Cosmos and Ethereum. What is the importance of connecting chains to each other? We're trying to build a financial ecosystem and we've just put all of our money in different

boxes and spread those boxes out. We can't actually like transact from Bitcoin to Ether. If you have Ether and I have Bitcoins, there's really nothing we can do together today. That doesn't sound like a very good payment system or financial system or really any kind of system. So you're proposing a better world where we don't just yell at each other all day and try and destroy the other chain. And we perhaps one day even cooperate. You can still yell at each other. That's pretty fun. But crucially, we'll be able to trans-

We'll be able to cooperate if we want to. Across chains. And so my understanding of cross-chain swaps is that there's a very question. are still somewhat niche and in the R&D stage. In five years' time, do you believe that these will be a very popular product and a way to realistically convey value between chains? We're really optimistic about it. The existing cross-chain swaps have all been based on the atomic swap,

which is discovered around like 2012 or 2013, if I recall correctly, by a guy named Tier Nolan, or a guy who calls himself Tier Nolan on the Bitcoin Talk forums. So the Tiernolan swap has a lot of downsides, and we think like a lot of the new relayed and SPV-based constructions remove the downsides and make this actually worth using. The main swap that you propose is not atomic. It is a SPV swap. Is that the correct nomenclature? Yes.

We call it SPV swaps or relayed swaps. SPV is an old Bitcoin acronym, and don't worry about what it means. but basically it's running a light client for Bitcoin in Ethereum. Okay, maybe for the benefit of the non-experts, could you describe it in not the simplest possible terms, but the intermediate terms? Okay, so a blockchain is supposed to create this history of what has happened. And for proof of work, the rule is that you follow the heaviest valid chain.

Okay, and so full nodes will follow the heavy, chain, check all of the validity rules on everything. So every transaction in that chain must be valid. A light client follows the heaviest chain, but does not check validity for everything. It only looks at small portions of the chain. And so it's much, much cheaper to run a light client, but you're kind of making this assumption that the chain is valid. So you're kind of assuming that miners will not mine invalid blocks. which is often a safe assumption, but it's definitely weaker security than a full node. So what would be the way to interfere if you were a minor and you're really intent on interfering with the system?

If 51% of the miners decided to extend an invalid chain, all light clients would break. This includes almost every wallet out there, like your Electrum. So if miners were to start doing that, like it would break a lot of the ecosystem at the same time. Because despite the cautionary warnings of Bitcoin developers, SPV, is still fairly popular. Yeah. Most people I know use light wallets or just fully custodial solutions. And a wallet where the transaction is actually broadcast by the servers of whoever is running it, would you call that light or is that sort of trusted? There's a lot of different tradeoffs here.

It really depends on are you sending this to one server, multiple servers. there's room for good light wallets that have little trust for the servers, but there's not room for light wallets that have no trust for the servers. For that, you need to run a full node. One of the things I like about your SPV swap model is that you can specify your desired level of security. Is that right? Yeah, you get to choose. That's how proof of work works, right?

Is once something goes into a block, you wait usually six confirmations. Six is just a rule of thumb. Really, you're waiting for an amount of security that you feel comfortable with. And so because we're running a light client in Ethereum for Bitcoin, you get the same choice within the stateless swap construction. So you can select an arbitrary number of confirmations worth of security and attain peace of mind that way. Yep.

Right now, I think we have like hard limits on something like 30 or 40 confirmations. I'm thinking about bringing that up to a hundred. 100. And ultimately it depends on how large the swaps end up being whether people want those monster 100 plus confirmation conditions. Yep. It does get more expensive, the more confirmations you're waiting. But we're working on relays right now that bring the costs down an awful lot. I think our current relay design saves something like 90% of the gas off of the original BTC relay. So this is an interesting topic because, as you may know, we had a billion dollar Bitcoin transaction yesterday, and it kicked off a firestorm, similar

to when Benance was speculating about incentivizing miners to roll back a transaction, which was a theft from Benance, which kicked off a firestorm. I remember that incident very well. Yeah, those were some good times. People got very angry at me. They did. And yesterday was no exception. People were angry again, not at you this time, but just in general.

because as it turns out, we still don't have a good model in Bitcoin for when a transaction is final. And in fact, the whole Bitcoin model is predicated on the fact that there is no absolute finality, because that could be relatively brittle, so to speak. And so Bitcoin prides this malleability over certainty, which people do not like. It's really interesting to me because I really believe in that is that, hypothetically, we could reorg Bitcoin from Genesis. We have now hard-coded assume block, right? Well, the assume valid only like disable signature checking. We don't have any more hard-coded checkpoints that I'm aware of.

So we have places where we assume all of the signatures before this block are valid, because validating signatures is the expensive part of sinking a chain. so we can save some cost in sinking, but as far as I know, we don't have checkpoints. I may be wrong about that. Ethereum definitely has checkpoints. Yeah, there's many things that are called checkpoints, and it tends to cause rank war whenever they're brought up. Yeah. But as I was saying, like, I really like that aspect of proof of work is that we follow this

very high level, very general, like very simple set of rules, and it creates consensus. this. In practice, I think there's really like two things here is one, we get security from proof of work as confirmations build and we wait until we have a subjective sense of security. You know, like how comfortable do I feel with this transaction? But two, when we were talking about the finance thing, people came out of the woodwork to say that Bitcoin should never allow reorg of longer than, well, people had different opinions, but longer than 10 or 50, blocks. And I think personally that it is extremely unlikely that a reorg of 100 blocks would be allowed by the community. There's a technical reason for 100 blocks. Because that's when the

coin base outputs become spendable. Right. That's when collateral damage from the reorg grows. Because those coin base transactions cannot be replayed by the attacker or reorder, right? Right. And there is actually like a really fun rounder. about way of replaying those coin-based transactions by incentivizing miners on the new fork to replay them, by giving them coins whose transactions are dependent on them. Nobody's written that code and nobody ever will. And I think it's extremely unlikely that the community would allow a hundred-block re-org. I think so too. I mean, the chaos it would cause by stripping, most likely stripping miners of their hard-earned coin-based rewards would be catastrophic. And the collateral damage

from those Coinbase rewards having been spent already, because each of the Coinbase reorgs that are wiped from existence would have child transactions that would also be wiped from existence. And so it kind of fans out and affects more and more people. And it happens not because of a double spend or a purposeful attack, just because the long reorg caused accidental collateral damage. But in your mind, is it a virtuous feature of Bitcoin that we have? have this convergence, even if it can cause pain in kind of the short term? I think that Bitcoin's security is rooted in its community and in social consensus around what the rules are, and that this is an interesting example of like a breaking point in

the social consensus, where on the one side we have this idea of proof of work and following the heaviest valid chain. And on the other side, in practice, it seems like people would ignore the heaviest chain if there was a reorg longer than a certain amount. And we have precedent for this. This happened. Yeah. These long reorgs have been engineered in the past around the 0.7 to 0.8 consensus failure and around the inflation bug before that.

So we have this interesting fallback to social consensus, but it's not clear that the community has rules for how to handle this. So it seems like friction point for Bitcoin's social consensus, and things probably get a little unpredictable if it ever happens. Yeah, that's something that makes me nervous. Not that I think it's likely for a long reorg to occur, but in the case that it did that the community has not, maybe even it's a little contentious to say there is a community,

some people would deny that as well. There is definitely a group of people who run Bitcoin full nodes. Yes. And so the thing that makes me nervous is that there is no really agreed upon contingency plan one way or the other. As you say, it is likely that there would be some resistance to a very, very deep reorg, which would be manufactured in the social layer. But that is not codified in any way. Right.

And a lot of the actual outcome would depend on what the major custodial systems do. What does Coinbase do? What does Cracken do? What do all of these exchanges do? What's interesting is that the last time the exchanges had a very firm opinion on something, they were overruled, I would say, which is during two acts. Most of the exchanges and big custodians were in favor of two acts. Yeah.

I hope that we never have to test this, because we, we can't make like firm predictions about what will happen because it depends on social conditions, marketing, and human behavior. I think what some of the contrary case on the we will just default to some social consensus is that if a social contingency plan agreed upon plan existed, this couldn't there would be exploited by some entity who, for instance, made a very large ODC trade and then they wanted to get their money back and they appealed to the community. I was defrauded, would you all coordinate to help me get these coins back or whatever,

but under false pretenses? I think that in some ways keeping the behavior undefined makes that course of action more risky and therefore less likely for a financially motivated attacker. Right, because there is no game plan that they can design against. Yeah. So obviously I find all of this stuff really fascinating. I don't know how much applicability it has to life right now. Like, Bitcoin seems to be working fairly smoothly,

despite the Twitter arguments about whose billion dollars, that is. To me, an ideal digital cash would not be one where there was any way of really knowing. It would be more of a chomian digital cash. It would be nice if we had no way of knowing that a billion dollars moved. Yeah, precisely. But it seems that the auditability trade-off of privacy may be, too costly to incur. Yeah, we have reasons to believe that there's no private system with an auditable money supply,

that you just can't have both of those in the same system. And empirically, we have now a small body of evidence showing this to be the case. I would say the zero coin protocol had a typo, I think, leading to a potential bug, which I don't think was exploited, but could have been Crypto Note had a bug. Again, I don't think, oh, the inflation bug was exploited in Bitcoin, but not Monero. The bike coin community did not seem to mind.

I'm not sure there is one. And then Pivics, I think, had issues as well. Yep. There is the Zcash inflation bug. So virtually all of the privacy coins have had these issues. Virtually all major coins have had inflation issues. Oh, right.

Even the latest bug on. Bitcoin was potentially an inflation exploit. So the difference is that you would be able to tell. In Bitcoin, when there was an exploited inflation bug, we rolled back the chain socially and fixed it. That's right. What was the supply 80 billion BTC? There's something like trillions of Bitcoin. Yeah. That's the issue with digital cash. There's also the lesser-known inflation bug that was fixed last year that was never exploited and the recurring block reward bug that was fixed by BIP 42.

the initial Satoshi client would restart the inflation cycle every 256 years, give or take. So that was a problem that we were bound to face at some point? Yeah, we've already fixed it. So I guess the point is that bugs are sort of inevitable, and the important thing is that you can recover from them and discover that something bad is happening. While blockchains are in their infancy, we kind of need auditability so that we can have some checks on their operation.

Having very few ways to interfere with them is a feature, but having no way to know that the chain has failed is unacceptable right now. And then in theory, once we're very comfortable with the nature of the protocol, and it's ossified to some degree we layer on privacy features. Now tell me about TBTC. TBTC. So TBTC is an Ethereum token that is redeemable on demand for Bitcoin. And you, through Sama, have created this with Keep, kind of a collaborative effort. Right. So TBTC is a project of Cross Chain Group. So Summa and Keep are the members of Cross Chain Group right now.

We've worked on this very closely with the Keep team for the last six, eight months. The design has been kicking around for a little longer than that. And it's nice to finally get things public and at the door. And so the design is public, but the system is not live today, right? The code is running on a private test net and has not been published yet. But we are live in private. And the objective is to create high-powered Bitcoin-flavored collateral, which can bounce around Ethereum and the various Ethereum applications. Yeah.

The defy narrative is very strong in Ethereum right now. Lots of people talking about collateralized. lending, Maker Dow, compound. There's a lot going on right now. And I think one of the core theses of TBTC is that all of that stuff's just better with Bitcoin. Bitcoin is a larger, more liquid market. There's more value in Bitcoin than Ethereum by a factor of more than two to one. And so if you want good collateral, you might want slightly less volatile collateral. Yeah, Bitcoin and Ether are strongly correlated, but that correlation has kind of been breaking, and Bitcoin has been a little less volatile than ETH recently.

Historically, it has been, although it's still pretty volatile. Yeah, Bitcoin's always going to be volatile. It falls out of being a fixed supply system, right? Do you think that you'll be able to tempt Bitcoiners into using Defi applications, or is that you're just designing the system and then? Well, this is kind of one of the nice features of the system, right? is we have to convince people with Bitcoin to put it into TBTC, but they don't have to be the end user of the defy systems.

We can have this separation between the people who have put up Bitcoin and the people are using TBTC in applications. And you as a holder of TBTC can be completely passive, can have no association with the Bitcoin chain at all, and just be holding the bearer asset, using it in defy, putting it in uniswap, doing whatever you want with it. So TBDC is essentially a full reserve Ethereum Bank with Bitcoin as the depository collateral. Right, exactly. You can think of this as like the Bitcoin standard.

You have a TBTC, you have a note, and you can present it at any time to get the underlying Bitcoin. And unlike the gold standard, the collateral is auditable. It's in a segregated account. You can verify that it's there. Yeah. Really, the whole meat of the system is verifying that that Bitcoin reserve exists and is available. I've made a lot of noise about agitating for exchanges to implement proofs of reserve schemes, although none have done it really, or not recently at least, because it's pretty hard, and the cryptography actually isn't really settled yet, so there's no share. standard. Also, traditional audit firms, this is kind of toxic to them. So I guess the alternative

potentially, if we do believe in this vision of high-powered auditable collateral, would be a crypto-native solution like this, as opposed to trying to bolt proofs of reserves onto an exchange, but they're using a whole bunch of other processes than just blockchains. Yeah, I think that there's a lot of like interesting traditional processes that can be automated this way. TBTC is, is essentially an automated proof of reserve system. That's music to my ears. So going back, well, I guess still on the topic of interoperability, has been a very popular buzzword over the last couple of years,

and there's a few different approaches to interop. So you have the, I call them interop within a walled garden, which may be a little bit uncharitable. So you have the cosmos and the Pocodot model where they're sort of fit for purpose, but they require a lot of work up front to get people into the ecosystem and then the interrop is less frictional.

And it's only within the ecosystem. And then you have interledger which from what I understand tries to be more packetized value mostly underwritten by ripple so far and as I understand it requires these connectors to make the system work and then cross-chain swaps

which are more I don't want to say hacky but they don't require modifying the base chains but they just build a system on top to make them communicate. Is that the taxonomy or is there something else that I've missed in there? When we're trying to make a taxonomy here, we typically split it into four categories. We have atomic swaps, which are based on like pre-image revelation for a digest or revealing some key. Atomic swaps are not actually cross-chain communication. The chains never learn anything about the other chain.

We just set up something where humans can interact on multiple chains. After atomic swaps, we have what's called stateless SPV. This only works for reading proof of work chains, and it has some interesting security caveats. So we've built a lot of stuff around that. It's difficult to parameterize the system and have confidence in its security. Then the next step from there is what's called relayed SPV. So this is having a smart contract that tracks the remote chain. If you want to get pretty technical about it, this is when,

the local state is dependent on the remote chain's fork choice rule. So my contract state is dependent on Bitcoin's consensus process. And then the inverse of that is merged consensus. And that's when my fork choice rule is dependent on remote state. So Cosmos is an example of relayed SPV. All of these zones are watching each other's consensus protocol. And I, IBC is a standard way of doing that. Pocodot is more of an example of like the merge consensus where the local parochain is supposed to be following the central chain. The jargon is getting decidedly heavy at this point.

Yeah, I'm sorry. Has there been no effort to harmonize the jargon among the different intro protocols? No, no, no, no, no. And then side chains are an example of merged consensus as well, is the side chain is following the parent chain. Yeah, I guess I did omit side chains initially. Why would you say they appear to have really declined in popularity in the last couple of years? Well, you know, we haven't really had strong use cases for them. And as Blockstream and others found out, they're a lot harder to make than they appeared at first. Is the reason because you just fundamentally cannot trust minors and you have to place an unacceptable level of trust in minors?

Initially, it looked like we could do side chains with economic guarantees, which is what we want. If we can't get cryptographic guarantees, economic guarantees are the next best thing. But I think while building out side chains like elements and liquid and rootstock, we figured out that we couldn't actually do that. And the best we could get was federated side chains, which is what liquid is. which is essentially a consortium of entities that all have a shared interest in the side chain working. Right. Hopefully they all have a shared interest in the side chain working. But a lot of the security model reduces to an 11 of 15 multi-sig.

And so side chains like the use cases never really materialized. The tech ended up being harder to build and have worse security guarantees than we thought. I think in the future we'll see more development. and innovation here. You have like Paul Stork pushing for drive chains, which would move us more towards economic guarantees than where we are now. I've been waiting. I think many of us have been waiting with baited breath. We will probably continue waiting with baited breath. Although it's actually under active development right now. Yeah. And if you look at rootstock, they were able to get essentially a soft fork. It's a little complex to explain what it is, but they were able to get a

commitment to the rootstock chain into the coin-based transaction of many miners. Same as name coin, right? Back in the day. Did they get a commitment into the coin base or did they get, anyways, that aside, rootstock was able to convince miners to embed information reliably into Bitcoin. And so we might actually see more like minor driven soft forks to add these things if there is a compelling reason to do so. So let's say that some form of cross-chain swaps or genuine interop solution becomes ubiquitous, successful, widely used. I guess there's two theories of how this might actually affect the balance of power among these protocols.

One, call it monetary pluralism, whereby network effects are actually obliterated because it's so frictionless to move into whatever protocol you want, that you have a good number of competing chains. And then the other is that, well, actually, monetary, monism, whereby you have one super dominant protocol and people use the swaps to move for a short period of time into alternate protocols, but then hurry back into the preferred SOV protocol. So both of those are pretty cartoonish, but which would you say is more likely? I lean towards monism here.

This is assuming that we have a well-developed and ubiquitous swaps function. Yeah, we're working on it. The reason I think that is it is not possible. to remove friction from financial transactions. If you look at financial transactions, whether it's payments, markets, whatever, the less friction there is, the more slippage there is. Like, always. It may be hidden from you like it is in credit card transactions. It may be brokerage fees that you pay. But the easier it is to make a transaction, the more you are going to pay for it. So you think what's likely is that we end up with,

one larger chain. I just kind of reject the idea that we can have completely frictionless transfer of value between two assets. So the premise was unsound. Yeah, I think the premise there is unsound, and that leads me to reject the conclusion. Well, that's some solid reasoning. I can't argue with that. So for everybody that's listening, James' is this great blog called prestwa.c.ch. So it's technically a Swiss blog, Confederation Helvetica, I think. And yes, that is just my last name with the dot. So it just so happens that there's this great top-level domain that you can get, which is dot-C-H, and it gives James just a great domain for this blog.

So the blog, there's a lot of different topics in there, but one thing that James does is he finds all this incredibly arcane esoterica from Bitcoin and other cryptocurrencies and meticulously documents it. And one of the most interesting blogs is all about standardness in Bitcoin and some of the really strange and arbitrary rules that are in there, which were hitherto undocumented from what I understand. Yeah, I did an okay job of documenting them, but there's really a lot there. I guess if we want to talk about this, we should start by explaining what standardness is and why it's important. standardness is the network relay rules for transactions. If I'm running a full node, what transactions will I send on to other full nodes?

This is not the same thing as what transactions I will put in a block or I will accept when I see them in a block. It's actually a stricter set of rules. So there are many transactions like the Coinbase transaction, for example, that you will accept in a block, but you will never relay to other nodes. So they're valid from someone who's validating the whole chain. Right. They're consensus valid, but they're non-standard. And so as a consequence, they don't make this hop

through the gossip protocol when nodes are communicating. Right. So wallets are all programmed to only make standard transactions. Because if your transaction is non-standard, you can't broadcast it to the network because no one will relay it and no one will accept it from you. Although you could ask a minor to, on a bespoke basis, include your non-standard transaction. Yes.

And that actually happens quite often, surprisingly. I've done it a number of times in the past. Why would you need such a thing? Well, back in the day, everybody used this thing called counterparty. It was a competitor of MasterCoyne. MasterCoyne is now better known as Omni, the home of Tether. Former Home of Tether. Still partial home of Tether.

You know, last time I... I looked, about one in eight Bitcoin transactions were Omni. That's correct. It's in the tens of thousands of Bitcoin transactions a day are actually Omni. Yeah, and all of those are tethered because nobody uses any other Omni assets. So back in the day, we had Counterparty, and lots of people issued tokens on Counterparty, including my company, storage. And so we had to do counterparty transactions. And what these are normal Bitcoin transactions with extra data appended to them.

So we put this in the op return, but counterparty had this interesting requirement for multi-sigs, where you could not use PetaScript hash multi-sig, which is what everyone's multi-sig is for a long time. You had to use the older bare multi-sig. So this is all a really roundabout way of saying it was really difficult and annoying to use counterparty multi-sig. So back in 2015, 2016, one day I noticed that not only was it difficult and annoying to use, the network stopped accepting transactions once they were done. We were trying to send like a two of three multi-sig transaction because we're a responsible company and we keep things with multiple signers anyway.

And I noticed that it started going through less and less often. So at first it was just, you know, one out of ten transactions. would fail and I'd have to resubmit it somewhere else. And then it was five out of ten. And then pretty soon it was all of them. And I would have to spend an hour trying to submit these transactions just to move tokens around. So what had happened there? There was a change as part of a normal Bitcoin note upgrade to the standardness rules. And I won't even try explaining what happened because it's really weird and arcane, and it has to do with how many signatures are in a transaction.

But there was a change to the standardness rules. And now these counterparty multi-sig transactions that we had been doing for years were non-standard. So you were essentially disenfranchised from the chain, the protocol sort of orphaned you. Right. We had these funds that were locked up in a previously working thing, and now they were inaccessible.

And so we had to make these transactions, get them the raw like blob hex transaction and email a miner each time we wanted to make a transaction. So you still had recourse, but it became much, much more cumbersome. Right. We were in this lucky position of having a social relationship with a minor because we were an established company in the space. if you were a normal everyday user, you would not have that recourse or it would be much harder

to get in touch with a minor. So this is such an interesting story to me because it illustrates, I think, a very overlooked issue with money transfer protocols, which are malleable and change a lot. Even in Bitcoin, which we think of as something that's pretty stable, it has pretty frequent updates. And sometimes unintentionally or intentionally, those updates might disfavor or favor some sort of class of transactional activity, even though from my interactions with Bitcoin developers, they seem to do a lot of outreach to sort of groups that might be making

arcane kinds of transactions. At least the debates often focus on that. Like, let's not get rid of this obcode or something. Yeah, this is something that's really interesting to me, because there are some, like, just hits of technical debt in Bitcoin, like something called Find and Replace, that are left in untouched because we don't know who might be relying on that feature. And a lot of work is taken at consensus layer upgrades to ensure that we're not breaking existing people's funds. But very little work is done for the network layer, for actual standardness and transaction relayed.

And that's where you were excluded at the relaying layer, not the consensus layer. And we were excluded by, you know, it seemed like a trivial change. I've went back and, like, looked at all the Git issues and the PRs, and most of the devs had no idea that this would happen and were unhappy about it in retrospect. Some of the devs were happy about it because it went along with their political views. And so you can kind of think of this standardist change, whether it was on purpose, to serve a political agenda or by accident as an exercise of power in the ecosystem. Just the ability to change the default software in an extremely minor way that people thought was

trivial, changed the political landscape of the chain, resulted in essentially the abandonment of counterparty by many different projects over the next year and a half, which of course contributed to the rise in Ethereum tokens because there was no viable token platform on Bitcoin. So it had long lasting and interesting impact, even though it seemed like such a minor change. And I think this is so critical because you see, as you say, what appear to be minor changes have these wide-ranging economic impacts. And this case study, I think, is a really interesting one because it shows us that this can occur even in Bitcoin, which we think is this very scrutinized thing and where decisions are really deliberated over. And so then, of course, it happens, I would say, a whole bunch of

different protocols for sure. And I can think of a few examples. Probably the most famous example of developers being accused of using their discretion to an advance, an economic agenda is, of course, the block size debate, right? Where the consequences are very, evident. If you had a business which was dependent on low fees, you have an interest in black size being greater. And if you didn't, you didn't have that interest. And this isn't just a future of Bitcoin. I think if you have a monetary protocol, which is conveying billions of dollars a day, there are going to be so many competing interests from startups and incumbents that all would seek to shape the protocol in a certain way that benefits them.

And we have a word for this, which is just lobbying. And it happens in government, of course. It's considered a bug, but kind of a necessary one. Something that people don't realize is that lots of congressmen actually rely on lobbyists to help them write bills because the lobbyists are the subject matter experts a lot of the time. So to me, I would say that many blockchains are actually really vulnerable to capture and to lobbying. Most obviously the ones that have a pool of capital. which is protocol derived, probably.

That's a very pure way in which economic interests can be installed. Right. And it's really interesting to observe this in practice and to be involved in the governance process of multiple chains. We're involved in multiple EIPs that are going into the next Ethereum hard fork. We're involved in a ZIP that we hope to get into either this Zcash hard fork or the next one. And, you know, like we keep in touch with the development. of many different chains. It's interesting just how much lobbying, how much capture exists

and governance for these software projects. And I would say outsiders do not realize the extent to which this occurs. Definitely. Even people like me are not aware of the politics that happen at the core development level. And I think there's a big incentive to actually keep these things quiet because you don't want there to be a common understanding that this is a technocracy much of the time. Yeah, we talk about tyranny of structurelessness is chains like Ethereum have not put in a rigid governance structure, and that means that they get captured by whoever shows up and has an interest and works on things. With arguably Prague-Pao is a good example. I'm not saying of capture, but just the people that care the most about it are the ones that show up,

right? Yeah, and capture is a really strong word. Generally speaking, everyone that I've interacted with in Ethereum, in Zcash, in other chains, is working for what they genuinely believe to be the betterment of the protocol. And ProgPow definitely falls under that umbrella. But if you make the protocol very sensitive to the kind of short and medium term desires of the economic actors, you may well be optimizing for entirely the wrong things. Right. Definitely. And especially in like complex chains like Ethereum, we're adding pre-compiles, we're adjusting gas costs, and those things have direct effects on what applications are viable to run on Ethereum. And so the EIP process is, in a way, exercising editorial control

over what you're strongly encouraged and strongly discouraged to do with the chain. Right now, we're seeing a decrease in the gas cost of data, so large transactions in Ethereum are getting cheaper. We expect to see increases in storage access and storage reads and storage rights in Ethereum. So if certain applications are going to get more expensive, it's interesting to like see the process in action and know that everyone is trying to work to improve the chain and make it more useful. But the changes that are being made serve specific use cases rather than necessarily the public good. So I think this is really the most critical question here.

Does catering to these individual use cases mean that developers are over-optimizing early on, or is it nevertheless useful market feedback in kind of evolutionary sense? This is information that the protocol then slips commonly encodes. I think because there are so few users, it's probably over-optimizing right now. The governance process is sensitive to the people who show up and have a voice, and right now that means engineers and not end users. As an engineer, we have very few end users, and so we're catering the protocol to the needs of the early engineers, and that might negatively impact actual adoption and actual future users. We don't know yet. Right, but it's interesting building for a future.

where the subsequent usage is unknown at present. Yes, we don't know what Ethereum is going to be used for in six months. So not the easiest problem. Right. And a lot of the times having more rigid governance structures is not any better. It just gives more rules for the game. And I suppose I definitely agree there. And the interesting thing with Bitcoin is that there is this assertion that there is no governance in Bitcoin,

which I think is false. To me, governance is just a fancy word for how decisions are collectively made, how power is asserted in the system. And so clearly, given the Bitcoin changes, there is power. Different individuals have different ability to force through various changes,

although there is this good separation of powers, I would say, between miners, developers, the various economic nodes, and so on. And this is something where Bitcoin being a larger market it means that there are more stakeholders and that there are more people whose voices must be heard and listened to. But within Bitcoin technical changes, those technical changes aren't just like the manifest will of all Bitcoiners.

It's like individuals wanted those and did the work of navigating the governance process to get them. Yeah, and it always kind of cracks me up when some developers will talk about the will of the people in Rousseauian terms, you know. as if they had consulted this mass populace, whereas in fact it often comes down to what a handful of individuals want and have the ability to do. Yeah. Bitcoin is great because it has a lot of checks on the process. There are many, many entities who can functionally veto Bitcoin changes.

And so if you can pass enough of them to get a change into the protocol, that means that it is more likely to serve many stakeholders rather than the needs of just a few. It also means that the changes are very general. We expect them to benefit almost all users of the chain rather than the needs of an application. Which in theory is good. Yes, in theory is good.

At the same time, we had no Bitcoin protocol changes in 2018 and we expect none in 2019. We might see some in 2020. It's preparation for full ossification. Yeah, I hope that, We get in a few more things before full ossification. That was actually going to be my next question. In your ideal world, when would we ossify and what would you jam in there before?

I don't think full ossification is possible. We talk about quantum computers breaking ECDSA eventually. We need a hard fork within the next hundred years to address the timestamp issue. I guess we'll get around to that eventually. Eventually someday. I like that it operates on a nice Bitcoin. geological scale. Right. I don't think we're trying to, like, ossify the protocol. I think we're trying to extend the time frame of changes. Okay, so we will never ossify. That's very sad, because I really

believed in that. We're going to get to the point where it will be functionally ossified for probably, I don't expect changes to Bitcoin protocol between me at age 50 and me at age 80. But the important thing is that you think it'll be around when you're 80. I expect so. Well, provided there are people around when I'm 80. At least one copy of the UTXO set scrawled on a clay tablet somewhere. That will still count as Bitcoin. That's still Bitcoin.

Carvedin Stone on the Moon. So you occupy this very interesting position because you are highly competent in a number of different blockchains, which is rare because people tend to specialize for ideological reasons or just because technically they're somewhat different. And so you have a bit of reputation

as being an informed critic of both Bitcoin, Ethereum, Zcash, and so on. So how would you say your critiques are received? Are they received with grace? And does this vary across the various constituencies? So Bitcoin and Ethereum are large enough that, you know, it varies within the community, how things are received. At this point, you know, I assume a lot of people who don't like me have blocked me. But within Bitcoin, there's definitely people.

agree with and people I strongly disagree with. And some of them are very mean to me, like around the finance thing. In Ethereum, it's the same way. Like, the community is large enough that I have a lot of friends in the Ethereum community and I have a lot of people who listen to me and, like, care about what I think. And at the same time, there's just a bunch of people who have written me off as some, like, Bitcoin Maxi Troll. Which is funny because your business employs Ethereum, a great deal. Yeah, I've written a lot of solidity. I've written a lot of code for working with Bitcoin. transactions and like working with Bitcoin and most of my solidity is Bitcoin oriented. So do you feel that to have a critique of some blockchain system, one needs to be intermediate

or expert level acquainted with it? Because I see this a lot. It's this attempt to raise the barrier to critique, if you know what I mean. Yeah, I definitely kind of agree with you is we should be open to honest critique from any skill level. And typically, I've found communities to be willing to educate people as long as they think that those people are intellectually honest. It's not always like a polite education. There are a few bitcoins who have given me a talking to an IRC, but like I took my licks and learned from it. So I think we should be willing to accept critiques from new entrance to the space and hopefully educate them and learn from them if necessary. That's a very diplomatic take, James.

So the reason that I kept discussion to Bitcoin and Ethereum here is that those are the only two, like, large communities in the space, where there are multiple factions with different opinions rather than a single monolithic entity that is in control. So criticizing other chains is often a complete social game. Am I insulting the individuals who are in control of? of that project? Or will they feel like I am insulting them as individuals? I made this big mistake, which was on January 1st, 2018. I made this hashed prediction of all these blockchains I thought would perish in that subsequent year. And three or four of them, I think, really did actually die. So I was pleased with that. But then when I revealed the predictions a year on, it was like this ultimate shitstorm. Some people got very upset.

They don't like it when you predict that their coin is going to die, even in hindsight. time commit to each one individually and open the ones where you're right. Yeah, selectively reveal. Can't go wrong. This work is very personal for a lot of us. It's something that we've put our time and our emotional effort into for years. And like I empathize with that, but at the same time, if you are in control of an ecosystem and you are not receptive to criticism and you are not willing to change your views as a result of outside interaction, then your project is doomed to fail. Yeah, I would say it's highly dysgenic. What you require is a free market of ideas which can be fought over and the best ones submerge. And when you discount outside opinions like that or you only listen

to ones that confirm your viewpoint, what you're doing is you are disincentivizing outside contributions to your ecosystem. I as an engineer, I want to be out there working on many chains and on anything that interests me and I want to contribute my time and expertise to these things. But there are very few projects like I think are worth contributing to, and some of those are not receptive to outside contribution. And so when we talk about a rigid governance structure or a dev fund or something like that, those things have a chilling effect on outside contributions to your ecosystem. You're limiting your contributions to people who understand your governance structure

and are willing to navigate it. They force you to engage in a very regimented way, which itself is enforcing some sort of behavior codes. Right. And I don't want to learn what Politea is or how to write a proposal and submit it in order to work on decred. And like the existence of that structure

makes me less likely as an outside contributor to put something meaningful. And it tends to attract people who are willing to play the game more than they are willing to contribute. Which seems contrary to the open source culture to some degree. installation of this bureaucracy.

Yeah. What we've done is we've put up a pot of money that is available to people who are willing to play the game, who are willing to play the governance game and play it well and navigate these social situations. The pot of money is not available to people who are contributing unless they're also playing the game. An extension of that is if this pot of money is going to like a company instead of a fund, a governance structure, it's even worse.

If protocol-funded financing goes to a company? Yeah, because I can go work on that protocol, and I know that no matter how much work I do or how well I do or how much I benefit it, someone else will be getting the payments for it. So do you think the Bitcoin model, which is essentially patronage now for large companies which rely on the protocol, do you think that is the best model? I think it is the worst model, except for all of the other ones that people are trying.

Very good. Very good. I've not seen anything else that produces positive results in practice. One thing that's very encouraging to me is the progressive distribution of the power centers in Bitcoin. The emergence of square crypto on the scene is highly eugenic in the dictionary sense of the word for me. In that Bitcoin developers are now distributed over a number of organizations, there are some unaffiliated, but the ones that are salaried are present at at least four places, five, I believe Zapo had a salary to engineer, although they just got acquired. But you have the DCI. You have Square Crypto now with actually a good cohort. Yeah, Matt's a great guy. Blockstream, chain code,

of course. Who knows if other Bitcoin companies will hire a dev. Outside of like core protocol development, the tooling for Bitcoin is split up among many companies. We've worked on a lot of tooling, but there's lightning working on BTCD. You've got purse working on the Node. ecosystem. There are many different economic interests that are making Bitcoin easier to use. So I think the way we're going to wrap this up is with some social and political commentary, but looking outside of the crypto industry. This podcast is on the brink. Matt and I believe that we are on the brink of a potentially new financial system, which is fairer maybe in some ways, although having listened to this podcast, you may not think it will be any fairer. And what do you

think the most significant effect of crypto on society will be in the next decade. You talked about this being a fairer system. And really one of the main benefits here is that individuals can still participate meaningfully in this system. Whoever you are listening can go out and you can involve yourself in the Bitcoin governance process and you can have an effect on it. And you can run a full node and you can participate directly in verifying the rules of this system. We are still young enough that anyone who wants to can go and do that. And that I think is inherently a better system and has an impact on society. Some of my interest is in keeping it that way as long as possible. I totally agree. Compare the cost of a gold full node, so to speak,

which is a really fun exercise. What does it take to verify some inbound gold payment you receive? I don't know that much chemistry. So I actually looked into it and you can, can do it with just like a jug of water and some math to find out the density. Okay, okay. The old, like... The Greeks had it figured out. But actually, you can trick someone by having a gold bar, which is full of tungsten, which is the best metal, by the way.

Shout out tungsten. Tungsten maximism. Because tungsten has essentially the same density as gold. So what you really need is a penetrating machine. There's this thing called an XRF spectrometer, which you have to buy, which can actually, I don't know, it's like a radio wave or something. thing, which can tell you if it's gold or not. So that's the equivalent of a full node, but it's pretty hard to use. I think it costs about five grand. Compare that to a Bitcoin full node,

which tells you not only the integrity of the Bitcoin you're receiving, but also the integrity of all the Bitcoin in the world. That's pretty good. That's doing okay. I don't think that we've had the option to participate in systems like this before. It seems new. I don't have $5,000 dollar spare to spend on a XRF spectrometer. Maybe we should get one. So do you believe that there are non-monetary applications of blockchain? So all we've talked today about is monetary applications. Do you think there is scope?

I mean, you worked at storage, so maybe the answer is clear, but do you think there's scope for non-monetary blockchains? I'm open to that possibility. I'm still tracking like a number of storage projects and other things. We are 10 years into this, and we have not seen benefits materialized from a lot of these things. So if anyone pitches you a non-monetary application, you should be more skeptical of it than of monetary applications. Yeah, that's probably my stance as well. And lastly, I know we've discussed a few overlooked issues, but what would you say is the most significant issue in crypto today, which is not really being discussed?

We've talked about this a little bit, but there are very, very few users. of crypto. And we're all throwing like money around on companies and startups and there was the token bubble and none of that has like panned out. The significant issue is that, you know, we're all kind of in denial about how like mediocre things are going. James, this is meant to be uplifting. You asked me about a significant issue. It's not like an uplifting topic here, Nick. So the issue is that there are no users and no applications and things aren't going as good as we thought. We have a lot of like propaganda about how we're changing the world and we're building better financial systems and we have not yet provided benefits to many people.

See, this is why I called James Very Sanguine. Because you're a truth teller. This has been James Presswich. Thank you so much for coming on the podcast. Oh, thanks for having me, Nick. It's been great. This has been another episode of On the Brank with Castle Island Ventures. To subscribe to our newsletter and find out about our portfolio, visit castle island.vc. A big thank you to all of our listeners, except those of you that are busy putting strawberries on the blockchain. See you next week.