Hi everybody, this is Charles Hoskinson broadcasting live from warm, sunny Colorado. Always warm, always sunny, sometimes Colorado. Today is February 17th, 2023, and I'm going to post a little link here with more information on contingent staking. We're going to have the great Twitter space debate, so I'm giving you guys a little update and also showing you a link.
I reached out to Rick and Bullish Dumpling; they both agreed to jointly co-host. Now, JJ is working with them to get some dates set up for when we'll do the Twitter space debate over contingent staking and what options are available. I wanted to show you guys this quickly because it will help with the Twitter space. One of the community members created a board where you can either do a pro or con for contingent staking. You can see some arguments for the pro side and some arguments for the con side. This gives you an opportunity to anonymously put in support for one side or the other. The point is just to aggregate a lot of the information in one place, making it easier to deal with whatever debate comes.
For example, one argument against contingent staking is that it goes against the core protocol idea of security and establishes a contradiction in its design. Classifying staking as good or bad is a power that should be left out of the primary system. An argument for the pro side is that contingent staking gives stake pool operators an additional tool to express their freedom of association.
What’s really happening here is that we’re not necessarily having an argument on facts; it seems like we’re having an argument over values or potentially opinions that are starting to step beyond what’s actually occurring. One of the things I’d like to see in the con category, because I’m still not getting it, is a clear explanation of how an additional capability somehow reduces your choice as a consumer or compromises and censors the protocol.
For example, there’s this contradiction where people say, “We’re going to implement this as a layer two instead of a layer one.” The very fact that it’s now implemented as a layer two solution and is available as a 100% fee pool, as DC Spark announced today, doesn’t that have the same philosophical implication as implementing it as a certificate? I think people are not understanding that contingent staking does not replace normal staking, and you as a delegator still have the ability to delegate to a stake pool that doesn’t use contingent staking.
Another thing is that, in reading through the cons, they heavily imply that contingent staking equals regulation. That has nothing to do with it; it’s a subset of this capability. All we’re saying with contingent staking is that when you send a delegation transaction, instead of it automatically being accepted, it has to also be signed by the stake pool operator. I didn’t say you have to include KYC with it, didn’t say you have to sign a contract with it, didn’t say anything. It just says that the stake pool operator has to sign it, so that construction pattern has nothing to do natively with KYC.
Now, why you’d want that pattern is that you would ideally be able to put stuff on top of it, like a DID, a contract, metadata, etc. For example, let’s say I have a pool that supports artists, and everybody who stakes to me will regularly receive NFTs from those artists. When you delegate to me, I’d like you to include in the payload an address for receiving those NFTs. There’s no KYC; they’re just putting in an added condition, and they just need you to include a payload for an address. They don’t know who the user is; that would be covered under contingent staking.
Now, the proposal from DC Spark today is a little disingenuous because it’s saying, “Oh, okay, we’ve solved the problem as a layer two. We don’t need to do anything; just set a 100% fee pool and then use a smart contract to redeem your rewards, but you need a verified NFT to do that.” Here’s the problem: there’s a world of difference between having a gate condition to do business with somebody and doing business with somebody and then retroactively trying to sort out the common value.
Just think it through. Would you have somebody come to your home and do a bunch of work, like redo all your electricity or plumbing, and then later say, “Okay, now let’s figure out how much to pay you and sign an agreement for all this work that you’ve already done”? That’s a recipe for disaster because what if you have a commercial disagreement about it? The plumber or electrician has one opinion, and you have another. So, we have things like estimates, scopes of work, and contracts as a condition to begin the work.
The point of contingent staking is that it gives you that business pattern where, prior to doing the work together, you establish a commercial relationship and the terms and conditions of that relationship. If you implement a layer two with what they’re proposing, the work is done, the money accumulates, and then to withdraw the money, the person who gets to decide that is the person who maintains the smart contract, which would be the stake pool operator in this case.
In other words, you go from a completely trustless, permissionless system to a system that is trusted and semi-custodial. Somebody can prevent you from redeeming money that’s been produced in a work unit because they never signed your NFT. The work has already been done; they took your resource, your stake rights, and used it to make money, and they’ve locked it in the contract. You see how that’s very different?
Another argument I see a lot is that this somehow changes the values of the system. Given that we can already do what DC Spark is arguing today, and both of these coexist and blocks can be made under that pattern, how does this materially change the values of the system? It doesn’t. If you philosophically feel that the pool should not know who you are or have any upfront gate conditions for doing business with them, wouldn’t you have the freedom to delegate to a normal stake pool?
Another argument I see on the list is that if you have this capability, then governments will use that ability to force them to require KYC. I think this is another misunderstanding about how governments work. The U.S. government doesn’t go and read the Bitcoin protocol, the Ethereum protocol, or the Cardano protocol and say, “This is what you can and can’t do with it.” They don’t say, “We should change our laws and policies accordingly.” No, what a regulator does is decide what they want and then say who’s in compliance and who’s not. They wouldn’t care at all if contingent staking existed or didn’t exist. All a regulator would do is say, “I now want to see this; you must comply.” If the protocol can’t do that, it’s as good as Gary Gensler saying, “Come and register if you’re a security.”
In practice, if there’s no way to register a layer one, it’s not possible. But notice how he still says, “Come in and register.” The fact that the protocol could have some capability to support a KYC scenario in no way would influence a government decision one way or the other. It’s quite the opposite; governments are just going to do what they do and demand that protocols accommodate them. If they don’t accommodate them, they’ll go after the entire protocol, removing all liquidity and getting rid of people.
In practice, what’s going to happen is you’re going to see a movement towards the East and jurisdictions like Abu Dhabi and Dubai that are embracing cryptocurrencies. A lot of stake pool operators are going to offshore and run pools in their own way, just like we saw in many other industries when the U.S. has become very hostile. The point of these types of capabilities is that they give a lot of new business models, like enhanced yields or more complex business relationships or smart contracts, while also giving people onshore more abilities to comply if they need to.
It’s important to understand that nuance. I saw a tweet today saying it’s Chekhov’s gun, and the mere existence of this is so offensive that it will somehow compel the government to act in a certain way. I see a lot in the commentary that talks about how this is a public good, like net neutrality. This is a binary question: push or no push. Contingent staking is turning block production from a public good to a private enterprise. Block production is not a public good; it’s not. There’s no government here.
I don’t understand how you can say that block production is a public good. It is ultimately an act of self-capitalization, an act of self-interest. The whole reason we provide an incentive is to create a marketplace, and by having a marketplace, it makes people chase the money. If it’s a public good, then you don’t necessarily have market mechanics, and nothing in the stake pool model requires you as an individual to stake your ADA. You can create a private pool with nothing to do with a stake pool operator and run it in a peer proof-of-stake style model.
The existence of a stake pool operator adds a marketplace and a competitive element to maintain something that we all benefit from, like a private electric company, a private water company, or private sanitation. I’ll agree that there is public utility and an emergent public good that comes from the maintenance of the chain, but by embracing a free market to do this service, you’re creating competition. The competition actually gets everybody better quality, prices, and reliability.
I’m not really understanding how increasing the level of nuances of the operators—especially given that the operators are going to be maintaining more than just Cardano—will help. The operator is going to be maintaining all of the side chains in the Cardano ecosystem. The absence of contingent staking means that the people doing that will have to follow the exact same staking model for every single side chain.
Let’s say that Maersk and a few of these other giant global logistics companies say, “We’re done using Fabric.” You know, IBM and Maersk worked together to create this big blockchain system for their shipping. It’s a $40 billion-plus company that moves most of the stuff around the world. Whenever you see those boats coming to port, they have the big connexes on the back. Let’s say they want to move their supply chain to Cardano and run it on Cardano, but they need to have more understanding of the people who are running the consensus side of it, the stake pool operators, and how these delegations work.
We have tons of international regulations we have to comply with, so those particular stake pools involved in the Maersk side chain are under higher scrutiny than people who maintain Cardano. Maybe they have a KYC requirement or some sanctions requirements. Okay, great. What you’re basically saying is, “I don’t want that use case at all; it cannot exist.” So when you’re voting on these things, when SIP 1694 comes out, keep this in the back of your mind. Your philosophical stance, which is not even clear to me, comes at the expense of all that Fortune 500 business, all the government business, and politics.
We want to run a political system; we want to do campaign donations, all these types of things. All those things also clip off all these business models where the stake pool operator would like to know a little bit more about the people delegating to it because maybe it’s paying those people something. It wants to have bespoke information channels with those people, not necessarily identifying them. There’s a large universe of discourse that has nothing to do with KYC.
I’d encourage you to think very carefully about this topic. There’s this board now where you can put lots of interesting stuff. Hopefully, this can be feedstock into a debate, and I really want to understand more about these things. You see disingenuous comments like, “Here are the facts: the road to hell is paved with good intentions.” Simply placing a vehicle that will allow KYC on level one will guarantee abuse by future government. The argument here is slippery slope. Almost all these arguments I keep seeing, and again, I keep asking a very simple question: step by step, tell me how that will happen.
You can’t just say a platitude that’s emotionally charged and dramatic. I see it again and again. We have to take a step back in a fact-based way and explain how people will use 1694 to hold a vote to remove normal staking capabilities. Is it, “Well, we’re going to blacklist all those pools for normal staking capabilities”? Okay, well, exactly how? You have to actually explain that to me, and no one has articulated a step-by-step argument that makes any sense at all.
They’re saying all these other business activities don’t matter because philosophy. By the way, stake pool operators are already able to do this because of DC Spark, and they’re actually able to do it in a way that harms people a little bit by putting total control in the hands of the stake pool operator for whether you get paid or not. There’s no contractual upfront relationship, so the stake pool operator has a problem where they’re introducing high liability.
That’s something to really think about. You can’t argue this way. You can’t say, “It just doesn’t morally feel right.” If you’re in charge of something, you have a job of looking at the facts and circumstances, looking at what we want to accomplish. The facts are that people are using Cardano in this way, not hypothetically, but today. The facts are that they have options with smart contracts, though suboptimal, to already do this.
If your point is that we should not compromise the philosophical purity of block production, we’ve already lost that argument because it’s happening today. These capabilities exist today. The question is, should we formalize them in a way to reduce trust and improve efficiency and reset the relationship so that they’re upfront instead of retroactive? That’s what we’re really asking here.
Third, if you’re going to make a slippery slope argument, you have an obligation to show step by step why that slippery slope is going to happen. The whole point of 1694 is to put resilient transnational checks and balances into Cardano so that China, Russia, the United States, the European Union, Brazil, South Africa, Australia, and other nations don’t have the ability to unilaterally step up and say, “You must do this.” If it’s going to get implemented, you have to vote for it.
At the end of the day, if a stake pool operator is sitting in Dubai, Namibia, or Indonesia, and people are using normal staking to delegate to that operator, how is the United States going to show up and extradite somebody they don’t even know in a jurisdiction where they don’t have extradition treaties? In practice, you’re basically saying that the KYC pools will censor the other pools, but then you’re assuming they’ll maintain all their stake. You’re assuming that people will continue delegating to them, but you personally make that decision of who you delegate to and where that stake goes.
Let’s be clear about that: you’re making that decision, not the stake pool operator. They don’t control the majority of the resources of the system; they’re going to move it somewhere else. For your slippery slope to occur, our entire governance system will be co-opted by the U.S. government, in which case we have a philosophical question of whether we have the right governance system.
Is 1694 actually proper? Are we all mindlessly going to go through KYC and only delegate to contingent pools? Will all the non-contingent pools just disappear? Will the contingent pools have some sort of censorship mechanism for the non-contingent pools with input endorsers? We move even further away from that being a possibility with the design of the system.
That’s what you’re saying when you say it’s a slippery slope: that the existence of this capability is so dangerous that we should ignore what people are already doing and all that economic activity and all the side chain potentials, especially for consortium chains. We should ignore the entire security tokens market, which by 2030 could be a trillion dollars a year, and we should also just allow people to do this with smart contracts, even though it’s not as safe and has some business reasons for why people would not want to do that.
This is what I’m talking about when you talk about a philosophical, fact-based, and reality-based debate where you can look at your values and the facts. This is just a small thing; it’s just multi-sig. I’m excited about the debate, and I’m happy that we’re at this moment. I think it’s a very teachable moment. This is showing people how hard governance is, and I’m very glad that SIP 1694 workshop is happening soon and that we, as an ecosystem, are moving along.
I’m excited that there’s going to be a broad group of D-reps to work with and talk to. These debates are going to get exciting because the D-reps will discuss this stuff on your behalf as you delegate to them. It’s going to be fun and a little challenging, and I’m glad this debate is happening right now. It’s very timely, and it’s good to get it out there.
Just like we need to learn how to be developers and write Plutus smart contracts, we need to learn how to be stake pool operators. People really have to learn how to live in an environment of argumentation and make tough decisions. This is governance. Effectively, when you think about SIP 1694 and Cardano, on-chain governance is so hard. This is why Ethereum has passed on it, and Bitcoin has passed on it. What they do is ossify. In practice, nobody can make a controversial decision, so they don’t.
The problem is that every single thing in business, politics, and life that requires you to get ahead and compete requires some form of controversy. There are no exceptions. This is not the only trade-off. For example, right now we’re looking at how we would introduce some cryptographic primitives into Cardano that would make roll-ups and recursive SNARKs very easy to implement.
There’s a huge technological and scientific debate, but once that’s settled, it still has to get into Cardano, and there are trade-offs there. All kinds of decisions have to be made around that. That’s just one example. Input endorsers are enormously complex and fundamentally change the incentive model of Cardano. The benefits are that if you pull it off, you get infinite throughput; it scales with your network growth, so you’re always keeping up with the demands of the system.
You get massive TPS, which is what everybody wants. But we think we’ve come up with a way of doing it that actually decentralizes instead of centralizes. However, it will change the staking model a bit. Is that a good idea or a bad idea? D-reps, when they come in, need to be paid. Where does that come from? My view is that some should come from the treasury. Right now, taking 20% of every block and putting it into
1:07:35
26:05
26:19