RSK, federated sidechains and Powpeg - NADO 20

Aaron van Wirdum Live from Utrecht, this is the Van Wirdum Sjorsnado.

Sjors Provoost Hello.

Aaron van Wirdum Sjors, I have another announcement to make.

Sjors Provoost Exciting, tell me.

Aaron van Wirdum Did you know you can find the Van Wirdum Sjorsnado on its own RSS feed?

Sjors Provoost I did, yes. And this is actually a new recording of that announcement. This is not spliced in by the editor.

Aaron van Wirdum The thing is, the existing Bitcoin Magazine RSS feed also has the Van Wirdum Sjorsnado on it. But that feed is going to be broken up. And all podcasts are going to have their own feed, and then there's going to be an other aggregated feed. But the point is, our listeners that are listening to the Bitcoin Magazine main feed, and they want to keep listening to us, now they really need to switch to the Van Wirdum Sjorsnado RSS feed, or the Van Wirdum Sjorsnado podcast thingy in their app. Or maybe they can switch to the new aggregated feed when it's there, but it's going to take a month.

Sjors Provoost No, you should switch immediately.

Aaron van Wirdum Yes, switch now. Switch now.

Sjors Provoost Cool. I always listen to dedicated podcast feeds, but it's just my own listening habit.

Aaron van Wirdum Especially [unclear].

Sjors Provoost Yes.

Aaron van Wirdum Sjors, what do you think of Ethereum?

Sjors Provoost It exists.

Aaron van Wirdum Yes. What do you think of DeFi?

Sjors Provoost I've heard of it.

Aaron van Wirdum Wouldn't you love it if you could use DeFi on Bitcoin?

Sjors Provoost No.

Aaron van Wirdum No. Good. That's a great start of this podcast, because we're we're gonna explain how to do that.

Sjors Provoost Exactly.

Aaron van Wirdum Here's the thing, oh wait, have you ever been to the…

Sjors Provoost I'm always in favor of explaining how one might do something even if it's not something I want to do. For example, I can… we can explain how to burn coins you know, yeah, chain, we've done that with Ruben too.

Aaron van Wirdum You don't have to do that. You don't have to… have to actually burn coins. Well the other thing I want to ask, have you ever been to the… laBITconf, I think that's what it's called the… the Latin American one?

Sjors Provoost No, I heard it's great fun but unfortunately not. Yeah, [unclear] in Latin America period.

Aaron van Wirdum So I've been in Suriname for a while. Unless you count Mexico, I've been there as well. Anyway this is getting off the rails. Okay so here's the point. I'll get to the point, dear listeners, I'm gonna get to the point right now.

Sjors Provoost This is actually take four.

Aaron van Wirdum RSK, you know about RSK?

Sjors Provoost Yeah.

Aaron van Wirdum Okay, so RSK is a Ethereum-like sidechain for Bitcoin. And on laBITconf, they announced a new pegging system. Do you know what pegging means?

Sjors Provoost Well, yeah, it means trying to keep the exchange rate the same, right? Central banks are very good at that.

Aaron van Wirdum Apparently, it's slang for fornicating. But that's not what we're talking about.

Sjors Provoost Exactly.

Aaron van Wirdum We're talking about a different kind of pegging where you… like, Sjors said, you peg coins to other coins. Okay, so RSK does this. We're gonna explain how the new pegging… peg in and peg out mechanism works. Okay, first of all let's start at the start RSK is a sidechain.

Sjors Provoost What are sidechains?

Aaron van Wirdum Great question. So a sidechain is a blockchain, it's a separate blockchain from the Bitcoin blockchain but in some way or another you can essentially move bitcoins to the sidechain and then move them back to Bitcoin. Slightly more technical there are different tokens on the sidechain so there is a different blockchain with different tokens, but somehow you can always exchange one of the tokens on the other blockchain for a Bitcoin. Therefore, they should be worth the same because if you get one of these other tokens, you can always just get a Bitcoin. So, you know, they should be valued the same. And as long as you have a system where you can do that both ways, so you have a Bitcoin and you can exchange it for a token or you have a token you can exchange for a Bitcoin, then that's what we would call a sidechain, I think. There's some discussion about the definition. That's my definition of a sidechain. I call that a sidechain if it does that.

Sjors Provoost I think that's fine. I mean, others might have all sorts of arguments why maybe the thing on the side is not a chain, or maybe… the thing on this… or maybe… you know it… because it involves more trust on the other side, it's not like the original Satoshi's vision of a sidechain, even though I don't think Satoshi was the one envisioning the sidechain.

Aaron van Wirdum No, that was like the original Blockstream vision, I guess.

Sjors Provoost Right. I mean, there was an original vision where the sidechain would almost be an equal partner. And, there are variations on it where the sidechain is less and less that. I mean, PayPal could be a sidechain, too, if they wanted to, right?

Aaron van Wirdum Yeah, well, according to my definition, they would definitely need some sort of token that's pegged with Bitcoin in one way or another. And I think there needs to be something. There needs to be a blockchain as well.

Sjors Provoost So what I would imagine is PayPal could say, well, you send us some Bitcoin to this address, and now it's on the PayPal sidechain, and you can send it between PayPal users. And then if you want to redeem it, we'll send it back to you, and we'll do that according to our database.

Aaron van Wirdum Yeah, I still think there needs to be something of a blockchain, and I guess we get into the definition of a blockchain.

Sjors Provoost I agree. I tend to agree, because otherwise, why not just call it the custodial wallet?

Aaron van Wirdum Yeah.

Sjors Provoost So then, you know, are you building… I… think some of the solutions to this problem are… what's it called? It's called a Rube Goldberg machine.

Aaron van Wirdum Right.

Sjors Provoost So it's one of those super complicated machine where you can move all sorts of objects around and they fall and then something rolls and a rabbit comes out of the hole and then all of a sudden, look, it's a sidechain. So, you know, I'm not saying that RSK is that, but I am saying that's why you can have very long discussions about whether something is a true sidechain or not.

Aaron van Wirdum Yeah, I think it was Jorge Stolfi or however I announced that, who argued that his couch was a sidechain.

Sjors Provoost Like I said, you can probably argue anything is anything.

Aaron van Wirdum Okay, let's get to RSK.

Sjors Provoost The principle here, yeah, is you put something into RSK, you park some Bitcoin, you're going to explain it.

Aaron van Wirdum I'm going to try to explain how RSK works, roughly. Okay, first of all, RSK is a merged mined sidechain. So in RSK's case, that means that, well, I guess, well, there are different ways of doing merge mining, but in RSK's case, I didn't know this I thought it was kind of interesting it's that it works sort of similar as mining or I should say pools mining where all miners in the pool they are sort of trying to find the next block and now every block they try is going to be a valid block but there's sort of a subset of blocks that are kind of almost valid and then they're still sort of valid within the pool to determine who gets the rewards and RSK is sort of a similar thing where not all blocks that are valid for Bitcoin are valid for RSK but there are like a bunch of sort of blocks that are almost valid for Bitcoin that are still valid for RSK so there's more…

Sjors Provoost Based on proof of work?

Aaron van Wirdum Right, it's based on… it's just Bitcoin miners who are mining this.

Sjors Provoost Yeah, because with Bitcoin indeed if you're a pool you produce a proof of work and in order to get accepted by Bitcoin nodes the proof of work has to be for example 100. But miners can say, well, as long as you produce 10, then we know that you're actually mining and not cheating. And so we're going to allocate you 10% of whatever the next block is.

Aaron van Wirdum Yeah. So RSK is sort of a similar thing where they have more blocks than Bitcoin and they sort of use the same trick. Okay.

Sjors Provoost But they're actually mining blocks in between then as well.

Aaron van Wirdum Yes.

Sjors Provoost Okay, so that's different. Yeah, there are some proposals for Bitcoin too to do that where you would have intermediate blocks.

Aaron van Wirdum Yeah, so RSK has that. So they have a block like Ethereum every 15 seconds or something like that. And this is how they do it. So these miners are just Bitcoin miners and they are producing valid RSK blocks. And these include transactions and they're doing whatever miners are doing. And they're doing whatever miners are doing on Ethereum, apparently. But they're producing blocks anyways.

Sjors Provoost And they're also putting these blocks, they're referencing these blocks in the Bitcoin blockchain, right?

Aaron van Wirdum Yes. Every time a Bitcoin block is mined, it's also an RSK block. So yes.

Sjors Provoost But not every Bitcoin block.

Aaron van Wirdum No, every Bitcoin block.

Sjors Provoost I mean, not every RSK block is referenced in a Bitcoin block.

Aaron van Wirdum Correct.

Sjors Provoost Because there would be gaps.

Aaron van Wirdum Yes.

Sjors Provoost But that's fine. Yeah.

Aaron van Wirdum Okay, is that part sort of clear?

Sjors Provoost Yeah, it is. And if people want to learn more about merge mining in general, I think they should also listen to our episode with Ruben, where we kind of talk about another way to do this merge mining that does not need the help of Bitcoin miners.

Aaron van Wirdum Exactly. Okay, so this is part one. Okay, then we get to how do you actually get Bitcoin onto RSK? And then after that, how do you get RSK coin back to Bitcoin? By the way, RSK coin is called Bitcoin on RSK. Or RBTC, RBTC, is not… not a subreddit.

Sjors Provoost So much censorship.

Aaron van Wirdum Yes, they pick the same name for their coin RBTC, but that in this in this context of today's podcast that means this token okay now we want to send Bitcoin to RSK so okay so this is what we're doing right now and then we'll get to the new change right now you send a Bitcoin to a multisig address and this multisig is controlled by a group of Bitcoin companies or cryptocurrency companies. Once the coins are sent to the multisig, these companies control it by majority. They control the coins. There's a smart contract on the RSK chain that keeps an eye on this multisig. And once coins are sent to the multisig, the smart contract on the RSK chain knows, okay, time to issue that amount of new coins. And they are issued to the person that sent the money, the Bitcoin, to the… multisig. And then they have RSK coins from that point on.

Sjors Provoost Yeah.

Aaron van Wirdum That part is pretty clear, right?

Sjors Provoost Yeah. And I think we also talked about that with Ruben, right? You can do that. And on the Bitcoin side, it just looks like money going into a multisig. Nobody cares. That's fine.

Aaron van Wirdum Yeah.

Sjors Provoost The question is how to get it out.

Aaron van Wirdum Yeah. Okay. So how to get it out: right now, the way to get it out is you send… while these coins, while the RBTC are on RSK, the coins in the multisig cannot move… or shouldn't move. You know, the companies that are controlling this multisig should not move the coins until the RBTC are sent back to the smart contracts. And at this point, these companies know, okay, now these coins are locked in the smart contracts. Now we're going to release them from the multisig. So then they're sent back to whoever wants the coins on Bitcoin's main chain.

Sjors Provoost Right. So it's kind of similar to like they would get burned and created out of thin air. Just a slightly different…

Aaron van Wirdum Yeah, a slightly different way of doing it. Slightly different terms. But it's sort of the same.

Sjors Provoost So the key is that it has to stay in balance. Because you do not want a situation where, you know, less RSK, where there is more RSK… than… The amount of Bitcoin in a multisig contract because then you get a game of musical chairs. So how do we prevent or how do they prevent the game of musical chairs? Because the Bitcoin side doesn't care, right? When the multisig address is empty, Bitcoin just keeps going. But some people may be very upset.

Aaron van Wirdum So right now there's no real way of preventing that. It's trusting that these companies aren't going to cheat on the RSK users. That's how it currently works. You just got to trust that they're not going to collude against you, essentially.

Sjors Provoost And I think it's also that there has to be a threshold, right? Because it's a multisig that has a certain threshold.

Aaron van Wirdum Yeah, it's a majority.

Sjors Provoost Right, so you want to make sure the majority doesn't collude against you.

Aaron van Wirdum Yes, exactly, right.

Sjors Provoost Or is forced to, or is hacked.

Aaron van Wirdum Yeah, also possible. But, you know, and the theory is, like, these are sort of well-known companies, so you could probably trust that they're not all going to collude against you, hopefully.

Sjors Provoost Yes, but you understand why I might make an analogy with a custodial wallet?

Aaron van Wirdum For sure, yes. I get why it's not ideal.

Sjors Provoost Not a legal analogy, but a physical analogy.

Aaron van Wirdum Yeah. Okay, but RSK is upgrading. So now they're going for a new solution. The new solution is this. The pegging in part is still the same. You still send coins to a multisig address and the smart contract issued… now you're sending back coins to the smart contract… well, the pegging in part is not exactly the same either. But here's the change. All of these companies that are controlling the multisig, from now on, they are using special hardware modules that have private keys embedded in them. And these companies shouldn't even themselves be able to extract these private keys. Okay, these modules, they are programmed in a certain way, namely, they are programmed to only ever release coins from the multisig, still by majority, but only do it if the coins are sent to that smart contract. That's the only way they could possibly send the coins from the multisig, if they are in fact sent to the smart contract. And they have 4,000 RSK block confirmations, which is equivalent to, I think, 200 Bitcoin blocks.

Sjors Provoost Okay.

Aaron van Wirdum So now you're not just trusting these companies not to collude against you. You are… Trusting the hardware modules I would argue and then also these miners if you are trusting the modules then there are actually blocks being mined that are actually confirming that the coins have been locked back into the smart contract which requires energy and you know hash power so it's not something they can do for free there needs to be actual energy expenses plus if someone would try to cheat on you somehow it would be obvious because everyone can see what's happening on the blockchain. Does that make sense?

Sjors Provoost Yes, I mean you know hardware modules are pretty cool, but they are not infallible. So I very briefly read on their own blog post, they were talking about the problems with the Intel SGX system which was a sort of a secure hardware module on site inside the Intel chips. And there have been lots of scandals. But in 2015, Intel was saying, oh, my god, we have this amazing secure enclave thing where, you know, you can put trusted code and the whole world can trust it. And there were proposals to put Bitcoin private keys inside of those hardware modules. And basically, you wouldn't need lightning anymore. And it was just, you know, it was going to be amazing.

Aaron van Wirdum That's Teechan. You're referencing Teechan.

Sjors Provoost Yes, I'm referencing Teechan. And it was already at that point clear, there were books written about that Intel module. People spent, I don't know, some PhD spent lots of time reverse engineering how the Intel stuff works in the inside because it wasn't very well documented. And basically the introduction said something along the lines of like no self-respecting security conscious engineer should ever use this. And then a couple of years later, there are all sorts of massive vulnerabilities with it. The risk with these systems in general, I think, is that they can have serious bugs in them or backdoors. Now, in this case, the RSK team, we're not using that particular system. I don't know what they are using. I mean, there are devices out there, right? Like hardware wallets, the Ledger wallet that has one of those modules in it, the COLDCARD too. It's a very cool concept. I mean, in principle, the chips are designed in a way that… you need very sophisticated equipment to get into them. And as long as you really can't get into them, they can hold on to a secret for you. That's why they're great for hardware wallets, because they can hold on to a secret. But they can do more. They can be given instructions. You can give them a piece of code and say, the only way you're going to sign this thing is if these conditions are met. And then you can never change the conditions again, in theory, unless you have… well, here's where the gotchas come in, right? Unless you have an upgrade mechanism where some arbitrary person can sign a software upgrade. Oh, but the firmware upgrade. Okay, who can sign the firmware upgrade? Okay, now that's the guy the government goes after. So there are the gotchas. There could be a backdoor in it. There was, I think, the crypto company in Switzerland that was backdoored by the NSA, and they were happily eavesdropping on their own allies.

Aaron van Wirdum Right.

Sjors Provoost For a very, very long time because there were just a couple of engineers that they were corrupt enough or patriotic enough, I don't know, depends on your perspective, to cooperate with that. So there's that risk.

Aaron van Wirdum So you're making it sound like a… bad idea. Is that your intention or how do you…

Sjors Provoost No, no, I mean, you could… you could argue that it's a strict improvement over just trusting the companies right you could say well you're already trusting the companies now at least the companies have to put in significantly more effort to run off with the coins so I'd say that's good as long as you're honest about how you present it the other problem would be a denial of service okay so if I am a government and I don't like this project I walk into these companies, I tell them to turn off the machines or I shoot the machines physically. Now the private keys are physically gone. Does that mean the coins are really gone? Or is there some backup system that now becomes alive? I think there was something like that with Liquid.

Aaron van Wirdum Yeah, Liquid has some sort of backup. I didn't read that for RSK.

Sjors Provoost Yeah, but then you can choose to make that backup happen and then go after the backup. And of course, RSK itself. As far as I know, I could be wrong here, relies on node software that honors certain private keys and certain behavior. So you could issue a hard fork. But then, of course, it depends on how many people run their own node and whether they'd go along with that. Similar discussion is with Bitcoin. I don't know how the incentives are in that system, so I don't want to comment on that. So there are all sorts of risks, but I think it's reasonable to argue that it's better to have the modules. Than not to have them. Maybe. Maybe not, because that backdoor could actually be worse than not having the modules. And just using the old laptop as someone likes to say.

Aaron van Wirdum But, um, yeah, well, in either case, I mean, RSK is a work in progress they would still like it to become a drivechain.

Sjors Provoost Yeah, so that's an interesting topic that we should dedicate a podcast to.

Aaron van Wirdum Yeah, we'll discuss drivechain in an episode one time but… but the… essence the essence in one sentence is that in that case it would well, can I summarize it in one sentence? It would just be hash power that controls the pegs, pegging in and pegging out.

Sjors Provoost The drivechain is not the end-all, be-all either. There are still sort of caveats. I think we should discuss that in another podcast.

Aaron van Wirdum That's for another episode, yes.

Sjors Provoost But it could be arguably maybe a nicer solution than hardware modules because you're really moving away from these companies completely. You're saying we're going to just integrate this with the Bitcoin protocol in general. Now we have all these Bitcoin miners that you don't have to trust. That's the idea, but that might not be possible either, but we'll find out.

Aaron van Wirdum Yeah. Then again, I mean, at this point, you kind of… well, is that right? but how are they… there's sort of… there's sort of different levels of trust, like you're sort of trusting the hardware, but then… like the miners can't cheat on their own and the hardware… well, the hardware [unclear] cheat on its own I guess.

Sjors Provoost Well I guess, if they want to move to drivechain they'd move away from the hardware.

Aaron van Wirdum Sure, sure, I know, I'm saying… right now…

Sjors Provoost [unclear] different security trade-offs that you can make I think the most important thing is to be honest about them. Yeah but what I was asking is how are they going to make money because it sounds like if you peg a currency you can't pump it so is there a pre-mine? Or are they like honest people?

Aaron van Wirdum The company? Yeah, IOV Labs, I don't know.

Sjors Provoost Okay. Because one of the things, of course, that is nice about currencies that are pegged to Bitcoin is you can say, well, there's no speculative value in the exchange rate, so you can focus a little bit more on the merits of the project itself. That was one of the ideals behind a sidechain. And the other might be to sort of try and piggyback on the proof-of-work security of Bitcoin somehow. Well, that turns out to be complicated.

Aaron van Wirdum Yeah. Well, one of the nice things about a sidechain like this, arguably, is that it allows Bitcoin miners to earn more fees, which improves Bitcoin security. Even for those of us that aren't using the sidechain. I guess the counter argument would be that running a mining pool or being a miner… Probably would come with a little bit more [unclear] because you could be outcompeted.

Sjors Provoost Yeah, so there could be a centralization problem there. If the miners have to run all these sidechains in order to still have a good margin, now one of the sidechains turns out to be Ethereum 2.0 and you have to process terabytes of data and you have to do it in a big data centrum. You can't do it hidden away behind a Tor connection. That could actually be bad. Right? Another problem. I forgot what the other problem is.

Aaron van Wirdum In general?

Sjors Provoost Yeah, yeah. Another problem is if there is more value in one of these sidechains or smart contracts or whatever it is, if, say, the entire, let's say the gold bugs say, hey, you know what? We kind of like this blockchain technology. We're going to put the entire ownership record of gold on the planet on the blockchain and protect it with Bitcoin's proof of work. Well, now you have a tiny thing, Bitcoin, protecting a very large thing. Because probably most of the… I don't know, but my guess is there's more custodial gold out there than Bitcoin.

Aaron van Wirdum Is that bad?

Sjors Provoost If all the custodial gold in the world is backed by a sidechain on Bitcoin, then creating a reorg on the physical gold chain could be worth a very, very expensive 51% attack on the Bitcoin chain. You don't care about the Bitcoin chain. You care about the gold sidechain. So yeah, that would be very bad. That would be like being the minority hash power. Maybe not that bad because, I'm a guess is… if you do this custodial gold sidechain, there's going to be some boss because you're still protecting a physical asset. So you could just ignore shenanigans on the chain. And so the attack might not be worth it because of something I would like to call proof of prison.

Aaron van Wirdum Do you think something like RSK could obliterate Ethereum? That wasn't the word I was looking for. Replace Ethereum?

Sjors Provoost Does it pump?

Aaron van Wirdum It doesn't pump.

Sjors Provoost And how is it going to replace Ethereum?

Aaron van Wirdum I agree. That's probably the main use case for Ethereum and it doesn't have that.

Sjors Provoost I don't want to say that.

Aaron van Wirdum [unclear]…

Sjors Provoost If it doesn't pump, it's not going to replace a thing that at least has pumping as a feature. Ethereum may have other features and maybe it can compete on that one.

Aaron van Wirdum Yes.

Sjors Provoost But so do spreadsheets.

Aaron van Wirdum Yes.

Sjors Provoost I don't know.

Aaron van Wirdum I don't know either. I don't know. I think if something like Ethereum needs to exist, I'd prefer to see it as a Bitcoin sidechain.

Sjors Provoost That's not how the world works. The world doesn't say something needs to exist and therefore we get to decide what it is.

Aaron van Wirdum The world doesn't form around me?

Sjors Provoost No.

Aaron van Wirdum Well, damn.

Sjors Provoost All right. Anything else?

Aaron van Wirdum No, I think that was the podcast.

Sjors Provoost Okay. Thank you for listening to the Van Wirdum Sjorsnado.

Aaron van Wirdum There you go.