The Delphi Podcast Host and GP of Delphi Ventures Tom Shaughnessy sits down with Barney Mannerings, co-founder of Vega, a protocol for creating and trading derivatives on a fully decentralized network.
The Delphi Podcast Host and GP of Delphi Ventures Tom Shaughnessy sits down with Barney Mannerings, co-founder of Vega, a protocol for creating and trading derivatives on a fully decentralized network. The two discuss the protocol’s ability to allow for any derivative contract to be built, dynamic incentives, and much more.
We would like to thank Cosmos for making this podcast possible through their sponsorship! Cosmos is on a mission, link one million blockchains. Name brand projects like Terra, Band, Kava, and Secret use Cosmos and the Cosmos Hub to connect to every other chain in our network. The Cosmos Hub is the port city that delivers Inter-Blockchain Communication today. Learn more at cosmos.network.
Every Delphi Podcast is dropped first as an audio interview for Delphi Digital Subscribers. Our members also have access to full interview transcripts. Join today to gain exclusive early access to every new interview before it’s published.
(1:37) – (First Question) – Barney’s Background and what brought him to crypto.
(8:33) – Vega’s Elevator Pitch.
(10:08) – Problems in the centralized financial system that Vega is attempting to solve.
(12:46) – Where is Vega in its development timeline.
(14:51) – How Vega works and use cases.
(22:43) – Modularity of Vega / how easy it is to implement a derivatives contract on Vega.
(27:01) – Vega’s Competition.
(30:19) – Why Vega chose to build its own chain.
(39:29) – Incentivizing liquidity providers.
(43:23) – Risks and initiatives.
(46:44) – Vega’s fee structure.
(49:56) – What is “Wendy” / how it tackles the MEV problem.
(55:35) – Barney’s thoughts on inter-blockchain communication (IBC).
(57:15) – Dealing with lag between moving collateral between chains or cascading liquidations.
(59:18) – What’s keeping Barney up at night.
- Our Video interviews Can Be Viewed Here: https://www.youtube.com/channel/UC9Yy99ZlQIX9-PdG_xHj43Q
- Access Delphi’s Research Here: https://www.delphidigital.io/
Disclosures: This podcast is strictly informational and educational and is not investment advice or a solicitation to buy or sell any tokens or securities or to make any financial decisions. Do not trade or invest in any project, tokens, or securities based upon this podcast episode. The host may personally own tokens that are mentioned on the podcast. Lets Talk Bitcoin is a distribution partner for the Chain Reaction Podcast, and our current show features paid sponsorships which may be featured at the start, middle, and/or the end of the episode. These sponsorships are for informational purposes only and are not a solicitation to use any product, service or token. Delphi’s transparency page can be viewed here.
- Cosmos by From The Dust | https://soundcloud.com/ftdmusic
- Music promoted by https://www.free-stock-music.com
- Creative Commons Attribution 3.0 Unported License
Hey, everyone. Welcome back to the podcast. I’m your host, Tom Shaughnessy. I help run Delphi Ventures and host the podcast. Today, I’m thrilled to have on portfolio company CEO, Barney of Vega. Barney, how’s it going?
Great, thanks. Great to be here.
It’s great to have you on, Barney. So always start off with a bit about the founder. How did you get into crypto?
Yeah, so I was a computer scientist and ended up working in sort of traditional finance, ended up working in the city, in London. I worked for Accenture, and worked for them and some other consultancies, for a load of top-tier banks, for the London Stock Exchange and others, designing sort of trading and risk systems and implementing them and project managing, and all of that. So I was sort of super into finance, and I got pretty deep on matching engines and order books and market microstructure as well, and pretty into that stuff.
But equally, as a computer scientist, I’d always been kind of into the cypherpunk stuff on the side, very into cryptography and encryption. And when Bitcoin came along, that was interesting to me. Eventually by 2013, I’d got interested enough to stop doing on my laptop and do a little bit of mining. Ended up investing in the Ethereum presale, and kind of a few others, very early token projects, sort of using counterparty on Bitcoin even before Ethereum was a thing. So really kind of dipped my toe in then and started thinking about how to marry these things together. I think I’d always sort of wanted to do something in this space, wanted to do something more start-up and entrepreneurial, and started thinking about trading and getting that onto the blockchain.
Although I was excited when Ethereum came along, when I actually tried to think about what I could do in terms of the stuff I knew, the order books, the matching engines, the derivatives, the complex risk, it became apparent that a lot of that stuff wasn’t going to happen on chain, on Ethereum with kind of the performance and other properties of that chain. So sort of left it behind a little bit, and then a couple of years back after I’d actually left that world to start a different company, I met my co-founder Ramsey, and he introduced me to the researchers at UCL who ended up founding Chainspace, which sort of exited to Facebook a little bit later. And they basically talked to me about proof-of-stake and some of the research they were doing, and convinced me that the things I’d been thinking about doing in Ethereum were actually possible using a sort of proof-of-stake network, and the design that became Vega.
So that was really the genesis of it in 2017, and then Ramsey and I started Vega at the beginning of 2018 and sort of fleshed out the research and ended up, the rest of the story sort of came from there.
That’s incredible. Any wild war stories from back in the day? I mean, crypto was a much, much different world back then, especially when you did the ETH resale.
Yeah, it’s interesting because it was so different. I mean, I think a lot less people were really in it for money. It was kind of, I was in the ETH presale for amounts of money that were pretty inconsequential, really. But I was in it so that I could develop on the beta and play with it and use it. A lot of people back then were sort of thinking that way, and no one really knew how far it was going to take off. People sort of had these wild speculations, but nothing had come true yet, I guess. So we’re all kind of in it for the ideology and the technology and the fun. It was really interesting, I sort of forgot about the … Well, I didn’t forget about the ETH presale, but I forgot that it had value in money, until I sort of looked at it about two years later and was like, “Oh, wow. This is money worth paying attention to now.” And it sort of all came from there.
So it was a pretty interesting time. It was sort of very technologically driven. People were very skeptical of anything other than Bitcoin as well. There was a lot of Bitcoin maximalism. It was kind of the default position I think back then. It was interesting watching all the sort of, the different innovations come up and how things happened.
Did you have any idea back then, or I guess, what was your take on what would actually be built on ETH? Did anybody really have a good concept back when you did the presale or two years later? Because I mean, three, four years ago, DeFi was just [inaudible 00:04:07]. The only thing they really did was [crosstalk 00:04:09]
Yeah, I don’t think anyone really … It’s interesting, because actually some of the first stuff in the Ethereum whitepaper was like, “Yeah, we’re going to build derivatives, and we’re going to build these applications.” It was basically someone had just taken a list of the most interesting technology sort of server or internet-type projects. I said, “Yeah, we can do all this on Ethereum.” And it was very sort of wide-eyed, optimistic, but actually, not really grounded in the reality of the performance of what’s going to happen.
So I would say, people had a lot of expectation. They were kind of building this internet computer, if you like. But I don’t think anyone really knew what was going to take off. It’s interesting that the thing that clicked for me was when the ERC-20 standard got created, and you had this standardized way to launch tokens. And the thing about launching tokens is it’s actually kind of very, very similar to sort of loans or equity issuance. And you can’t necessarily create more Bitcoin from nowhere or more ETH. But you can create a token, and if create 100 tokens and sell them for 100 ETH, you’ve now got 200 ETH worth of stuff, if the tokens keep their value. So kind of sort of mirrors the way that loans create money supply in the traditional economy.
And once that became standardized and easy to do and got picked up by sort of business people and entrepreneurs, if you like, it really sort of started the engine going, and then obviously, you had the sort of 2017 stuff, which went up and then back down, and started to introduce some realities to the situation. But I don’t think anyone really knew what was going to go on and how it was going to work. I think there still are people who think that anything like loans and under-collateralized stuff or margin trading on Ethereum or Bitcoin is bad and shouldn’t happen, or even people who think it can’t happen because of the tech, and that’s just not really true.
So I don’t know that anyone really predicted it, and even when we started Vega in 2018, DeFi wasn’t a term yet. We were sort of talking about the parallel financial system, as we would call it. And we were trying to convince people that it wasn’t good enough just to have Bitcoin or Ethereum, because when you look at finance and the jobs that finance does as a tool, finance is not primarily about having money and moving money around. It’s primarily about all the products and tools that you can build and do with that, and the price discovery and the economics and the resource allocation that those tools bring. So we sort of had a bit of a time, I guess, initially convincing people that finance and decentralized finance needed to exist in order for Bitcoin and Ethereum and these sort of cryptocurrencies to actually be successful.
That’s awesome. I’m kind of getting the feeling from your statement … It’s a great way to put, that finances isn’t just about moving the money around. It’s not just about storing, let’s say, Bitcoin. It is all the apps built on top, which is kind of DeFi. Is that kind of a good, quick way to think about your answer there?
Yeah, absolutely. Having a base of that monetary value and Bitcoin being, perhaps even Bitcoin changing its narrative from internet money to the sort of digital gold, the store value on which everything else can be built is interesting. But yeah, I think like you say, building those apps and use cases around it so that you actually can replace finance is probably the key.
I love that. So Barney, let’s dive into Vega. I’d love to get the elevator pitch from you, and then we can kind of dig into all the different specifics.
Yeah, so Vega, it’s a protocol and a blockchain that’s designed for trading. Specifically, it’s designed for trading derivatives and sort financial products that are margined or have a lifetime, because they’re the more complex and difficult things to do in a decentralized way. The goal of Vega is to make decentralized versions of these products be as good or better than the centralized, traditional finance version. So it’s to make it, the idea that the people that I used to deal with back in the city, who would be trade and who would know customers who need these products would be able to say with a straight face, “Yes, we could use this,” when they look at these decentralized alternatives we’re offering.
So that was really the way we think about Vega. It’s not so much about servicing directly the crypto sort of industry directly. It’s about actually building realistic real world finance. That comes with solving a whole load of challenges that Ethereum and other general purpose chains present. It’s our general opinion that by being sort of more specific and by tackling the problems head one, we can design something that really does that, and that’s really what Vega’s designed to do.
It’s a protocol also that works sort of halfway between a layer one and a layer two. It’s sort of an app-specific sidechain that optimizes performance for derivatives and other financial products, but it’s also designed to work with bridges to initially Ethereum, and then other chains as well, so that people aren’t needing to issue new assets just to use Vega and actually can work with the ecosystem that exists.
You mentioned Vega’s attempting to solve a lot of problems. I guess, what are the problems that you see in the centralized financial system that Vega’s attempting to solve?
Yeah, so there are a few major problems in the centralized financial system. And one of the, I guess, nice things, if you like, about the problems in the centralized financial system is there’s a really high amount of value attached to them, which means that in some cases where you sort of look at the use case of blockchain, that was an expensive database, actually even though blockchains are more expensive than a server, the amount of waste and inefficiency in the financial system is so high that the blockchain solutions are likely to be much cheaper and better overall.
The first of those problems is probably access. So there’s a hugely uneven amount of access to markets. And in fact, that also comes into a fragmentation problem. So even if everyone in two countries can access the same market, the likelihood is they may be accessing a different market that’s selling the same thing, or via different sort of entities or exchanges. Some people simply have no access at all. Either their governments don’t allow it, or no one’s offering the product where they are.
And then the people who do have access, they’re in a very tiered system where it’s a bit like going to a sort of nightclub in Vegas, where there’s all these different sort of velvet ropes. And you have to be rich enough or be doing enough business with them to get to the next layer, and to get the next level of cheapness or access. That effectively means that individuals and small business particularly massively, massively overpay sort of 25 to 50 times more than the large companies accessing these markets, even though there’s no technical reason they should need to do so. So that’s a big problem.
Then, the other one is the speed of innovation. Effectively, and we’ve seen this with the internet and sort of information type services and systems. When you have a small number of people controlling something, then what tends to happen is they’re gatekeepers, they choose what happens. But also, the new ideas and the innovations come slowly, because they’re all sort bottlenecked through there. And so the creation of new markets only happens when the demand is high enough and everyone is sort of crowding outside Goldman Sachs or whatever, shouting and demanding this market. Whereas with the internet and sort of .com businesses and all of that, it’s been possible for people to literally sit in their bedroom and say, “Hey, I can see if this is necessary,” and on a shoestring budget, create something really new and exciting.
I think that’s a huge problem that we solve, is not only do we give everyone the equal access and reduce the overall cost on average for users, we also solve this problem, where now, anyone on a reasonable low budget will have the ability to design products and create markets and take those things to the world, which is something completely new. We’re very excited to see actually what people do with this.
No, same here. I guess you said that you guys founded Vega in 2018, so this clearly isn’t something new. Where are you guys in the timeline? And apologies for playing the dumb, devil’s advocate asking questions. I know we’re investors, but would love to go through it step by step.
No, very good. No problem with that at all. We are at probably the most exciting part of the timeline, because it’s the part where we start to see other people learn and use the thing that we’ve created and see how excited they are by it. So this is, as you’ll probably have noticed with other layer ones as well, when you’re building a blockchain, you’ve got to think about security and sort of economic mechanisms, and the design of the chain and the network and all of these different things that come together and then implement those. It’s a lot trickier than building a protocol in a smart contract on a chain that’s already built. There’s a lot to think of.
So we had to do that, and we also had to solve a few sort of research problems around … fairness was one, so we’ve solved the MEV problem, which means that Vega people can’t effectively pay for ordering and pay for front running. We’ve also solved some problems around measuring and funding and recognizing and rewarding liquidity, which is really important to market growth. So we’ve gone through that phase of prototyping, solving problem, identifying problems, building, testing. We’ve been in testnet since the beginning of last year, sort of late Q1, early Q2 2020, our private testnet. We opened it to the public in Q4. We’re now in an incentivized testnet, so we’re starting to pay people to test Vega and to act kind of in a more realistic way and get us even closer.
And then in the next few months, we’ll be launching the first alpha main net. So super exciting time, because even with the public testnet and with the recent CoinList sale bringing a massive number more people into the community, we’re starting to see people get Vega and understand what it can do, and get excited by it. And that makes us very excited, but it also starts to validate the decisions we’ve made, and show us where we’ve maybe not got something quite right and we need to go back and tweak some of the design as well.
That’s awesome. You guys are clearly on a journey to learning more and you’re implementing those changes, so it’s awesome. So I guess, let’s dive into it a bit. I’d love to kind of unpack how Vega works. What can different users do with Vega? I’m not sure if it might make sense to kind of walk through a user journey. But whichever way you’d like to take, I’d love to kind of go through that workflow.
Yeah, so I think the best way to do this is probably to do two things. The first point is maybe to give a little bit of a high level conceptual overview, because Vega’s quite different from a normal chain. And then maybe I can walk through what that would mean if you’re a user.
So on the conceptual side, the way that Vega is designed is, it has an understanding built in of some of the primitives, if you like, of trading. And the huge advantage of that is that you can optimize the design of those, but it’s also, that everyone uses the same primitives, so they’re very well-tested. So if you think about some of the hacks and stuff that you see on Ethereum-based protocols, often they’re in fairly standardized bits of functionality that are going on, that are just been re-implemented for that protocol, and not been quite right. Whereas Vega has an implementation of an order book. It’s highly optimized, but it’s also going to be tested by every market. So there’s a bunch of advantages there.
The way that Vega’s designed is very similar actually to how you would design a trading platform running in a bank, but you’ve decentralized and automated the whole thing. What that means I you have the trade determinations, so where you actually enter order into a book, and it can be in an auction, in can be in a continuous trading. In future, there’ll be other modes like request for quote, modes for more bespoke markets as well. When the trade is determined, you kind of have the trade lifecycle management, which manages the lifecycle of the product, it manages margin calculations, it manages risk. And then you have the settlement engine as well, which manages the settlement of the product. For instance, when you get the oracle data and its expiry for a [inaudible 00:15:16] future. Those things all kind of work together in a workflow.
And the other advantage of this for market creators in that instead of having to code all of this stuff from scratch and write hundreds of lines of sort of boilerplate code to implement the management of a lifecycle, or taking in an oracle value, or changing the trading state. Those things are all kind of provided. So what you focus on as a market and product designer in Vega is actually just defining how to calculate the cash flows. So In Vega, you simply tell Vega via your product definition how to value the product, given a certain price, and also how to calculate the cash flows when an event occurs. And those events can be anything. They can be the ending of a race in The Grand National, they can be the settlement data coming from oracle. They can be multiple stream of settlement data in the [inaudible 00:16:00] they can come perpetually, can come from oracles, they can come from other markets. You have all the power of smart contracts to do that kind of stuff.
But then when you want to sort of code out the product, rather than coding out the mechanics of how the system works, you simply calculate the cash flows that will be transferred from buyers and sellers, and the system is able to do the rest with that and is able to run the risk models and calculate the margins and everything. So effectively, it gives you a much higher level set of constructs to build products with. We typically look at the number of lines of code to implement. Something like a perpetual is in the tens of lines, if that. It may even be singe digit lines of code to implement perpetuals, as an example.
Option’s the same, it will be a few tens of lines rather than hundreds of lines. And that just gives you a huge advantage as a product creator, both from a speed and rapid sort of development point of view, but also from a risk point of view. If you’re writing 10 lines of code, it’s a lot easier to audit them and know they’re right than if you’re implementing thousands of lines code. So that’s kind of how it works technically, and that sort of modularity of those built in things that allow the flexibility, but also the sameness in the structure and rigidity that gives it a very robust thing will be very valuable in future.
So then, let’s maybe talk about an example from a user point of view. And the nice thing from a user point of view is that this works quite a lot like centralized finance, in that the system understands things like order. And so where when you’re a user and you’re developing with Ethereum, you’re kind of interpreting things on the Ethereum chain, sending data in and out, connecting to the Ethereum sort of event system, and then turning that into your protocol, and you’re having to sort of interpret the chain in that way. The transactions on the Vega chain represent things like orders and trades, so it’s actually quite easy to program for. We’ve got sort of really [inaudible 00:17:50] APIs.
As a user, you would deposit money onto the bridge, so we’ve got an Ethereum bridge. You effectively send some tokens that say you’re going to send some [di 00:18:00] to the Ethereum bridge. You can do this through our console app, and the UX is basically the same as Uniswap. The console app is going to be connected to your Vega wallet, and also have MetaMask. Sign in with your Vega wallet, then you basically just click a deposit button, MetaMask pops up. You say, “Yeah,” deposit 100 di. That waits a few blocks for the 100 di deposit to sort of be confirmed and secured in the bridge, which is the link between the two chains. Then you can see that in your wallet, and the network’s recognized it.
So then, you’re effectively free to place orders on any of the markets, and effectively, you can go browse markets, look at the IDs, use the APIs, place an order on a market. A certain amount of margin is reserved when you place an order, in case that order executes, unless you’re already got kind of a net order on the other side or something, we can net these things out a little bit. That margin’s reserved. So if the order trades, so either it trades on entry, or someone comes and hits your passive order later, that all happens on chain. And then a position is generated, you have a position on the Vega network. And effectively what happens is that position is marked to market constantly. So every time the market prices changes, your P&L for that position is reevaluated. That’s adjusted from your margin account effectively, and then your margin is maintained.
So if your margin goes up, so if your position is in a large profit, then there’s what we call a release level. And at a certain point, some of that margin’s released back to your sort of main wallet, which allows you to use that margin elsewhere. And then in the other way, if the position goes against you and you get sort of below the search level, or you’re close to your maintenance margin, then money will be swept from your wallet into the margin account. That’s all done automatically on chain.
Because of that sort of sweeping back and forth of gains and losses, if you kind of see you’re long Ethereum and you’re short Bitcoin, some sort of pairs trade like that, if one of them’s gaining and one of them is losing, then actually, the gains from the gaining position will be swept back into your account and become available to the losing position. So you get this kind of ability to net margin across positions, which makes for good capital efficiency on the system. And then whenever you close your position or if you hold to maturity, then effectively, the position’s marked to the oracle price at maturity, or to the market price when you close it. That’s settled using the system’s settlement engine, and the funds are swept back into your wallet or you to do as you please with.
And then ob, when you’re done on Vega and you want to release those things from the bridge, you effectively ask the Vega network which controls the bridge, you ask the validator on the Vega network to sign a message for you saying, “Release this collateral.” And as long as you’ve got the funds available, they mark the funds as withdrawn and sign that message which you submit back onto the Ethereum chain. The Ethereum smart contract maintains a list of the validators, and therefore, when you submit that, the Ethereum contract lets you pay out the money back to your Ethereum wallet.
Barney, wow. That’s incredible view on user journey. I’m so glad we’ve transcripts so I can go back and read it. One of the first parts you mentioned on your first part of your answer was just the modularity of Vega, and I think that’s really important because I guess a lot of people in this space view drag and drop features for developers as super useful, because it lowers the barriers of entry to experiment, to release new products, etcetera. There’s a lot of examples, like Polkadot, Substrate, stuff like that.
I think it’s kind of lost on people though. When you say something like Bitcoin is a stored value, and then you see something like ETH, you can build the world on it. I think people don’t really realize that until it’s live, until they see it. Do you think people understand the modularity of Vega and what you can actually do with it? And I guess, the other question is how easy is it to implement a derivatives contract on Vega if I want to? I know the lines of code is much lower, but what’s the experience level that in eed to do this?
Okay, so there’s a couple of things in there to talk about, I guess. First is sort of how do people understand it, and what does it mean? I think one of the nice things about Vega is because we have some very specific use cases, we’re not presenting people with kind of an empty text editor saying, “Write any code you want, build anything you want,” which is maybe something that Polkadot or others have to deal with. We’re kind of saying, “Hey, you’re designing financial products. Here are some of the ones that exist in the world, and here are some examples of how to do that.”
So we have an advantage in that respect, I think in that it’s easier to kind of say, “This is what’s going on.” Well, the other reason this is very important is that people who are going to be good at designing financial products might also not be people who are particularly good at writing code. And we’ve probably seen that to an extent in DeFi so far, where a number of the first attempts, particularly around the sort of EtherDelta time and Bancor and others, that first very early attempts … A number of those things were not necessarily particularly useful for anyone who really had a need for financial products, because it was sort of built by technologists trying to see how far they could push things, but maybe without the financial background. So I think it’s really important to make it accessible build things on Vega, to people who understand the needs of the users of this stuff and the finance.
In terms of how easy it actually is, the answer is sort of twofold. Because in the one hand, it’s super easy. I can define some cash flows really, really easily and define a new product quite easily. Just to be clear actually, the first version of Vega is going to have built in products, and you would write them in Go and submit them to the Vega repo as a sort of new open source product components into the system. That’s not the long-term way that this will happen. That’s kind of the first MVP. The real sort of long-term by sometime early next year, hopefully, is tat you will write code in something like WebAssembly and compile it actually submit it as a transaction on the network, so you won’t need to do anything on the open source project. We’re just kind of launching the MVP first without that, because it just simplifies the launch and de-risks it.
But it’s actually very easy to do that. The simple version is to calculate the cash flows for a simple futures contract, where literally, you’re basically saying the value of the product is the size times the price. So if I have bought 100, and the price is 500, then I have 50k value. And that’s literally the single formula, the single line that goes in the valuation function. And then there’s like three lines that go in that say, when you receive data from the settlement oracle, settle at the price for that data, which is one line, and then change the status to sort of settled, expired, close the market … And that’s basically it. It’s really is that simple. Other products are a little bit more complicated, but still, that very high level of simplicity certainly many of the products that people sort of trade today in the centralized markets will not be much more difficult.
The thing that makes it difficult to create markets is that to create markets, you need to find good oracles, you need to understand risk models, you need to be able to calibrate the risk models. So the coding bit’s kind of easy, and then what you’ve got to go and do is get the data to launch a well-functioning market, which means understanding what volatility is and things like that and how to calculate it, and how to analyze it. It also means knowing what the risk model for a particular product and particular market are. Those are things that require a bit of skill in terms of knowledge worker type skill, but they don’t necessarily require huge amount of effort in terms of implementation.
And so actually, a big part of what Vega as a project team have to do and we as a community have to do, is to educate people how to do that well, how to actually calibrate the parameters for a market and launch a market that’s going to work and is going to be well-specified. Because if it’s not, it’s either going to be constantly kind of entering a liquidity or a price auction, or it’s going to constantly hit the insurance pool, and people are getting closed out all over the place or something like that. Os that’s actually the community governance thing we have to get right is how to manage those parameters well, and how to give people tools to do that.
That’s an awesome answer, Barney. This might be somewhat of a basic question. Vega offers so much customization, modularity to create markets, to create different types of derivative contracts. Is there an issue though, or is there competition between that and existing derivatives exchanges? Let’s say a very popular market is on DyDx or on NCDEX, or something like that. Does Vega in essence compete with them, or plug into them? I guess, what’s the dynamic there?
I think in those cases, it probably competes. You look at what they’re doing, and they’re effectively doing a subset of what Vega does. They’re offering some of the functionality Vega has, and different trade-offs in each case. Some of the cases you have, like PERP Protocol I think has an AMM-based perps with 10x leverage. And DyDx, I can’t remember the leverage they offer, but they have a sort of, I think, and off-chain order book and on-chain settlement.
A lot of these protocols … one of the weirdest things I always find very confusing is a lot of DeFi sort of exchanges of protocols act like centralized exchanges when it comes to market creation. They kind of decide what to list and what people can trade. That seems weird to me, because actually, the part of being a centralized exchange is easy to scale as running a matching engine on a server. You can run thousands of markets on one server. The London Stock Exchange, I think when I was working with them, had six servers, and they ran 30,000 markets on them. So really, that bit was pretty scalable and not a problem. But the business work of finding liquidity and launching markets to specifying them was hugely time-consuming, expensive and risky. And so they had a sort of relatively slow cadence of launching new markets.
To me, that’s what the decentralized blockchain world is good at, is good at saying, “Here’s the incentive to launch a market. HEre’s how you get rewarded if you provide liquidity. Go and launch many markets.” In a way, I think that’s what people like Uniswap have done very well with, giving that freedom to people. So it always sort of surprises me when you look at DyDx or whatever, and they kind of slowly list new products rather than giving that to the community.
But the other thing here, when it comes to that competition that Vega’s doing with those product is, those products are sort of in a way stuck on Ethereum, and you have relatively low leverage compared to other options, so relatively low capital efficiency. One of the things I think’s incredibly risky and people don’t really have a good handle necessarily yet on the risk levels of things like liquidation auctions. So on a lot of Ethereum-based protocols, when someone needs to be liquidated, someone else needs to notice and put a transaction in. And presumably, someone can block that by paying higher gas for their transactions.
Whereas on something like Vega, the liquidations are atomic, which means as soon as the market price is one that makes your mark to market require a liquidation, you’re going to be liquidated by the protocol, and there’s nothing you can do about it, and there’s no need for anyone else to step in with a transaction. That’s simply because Vega has the sort of computational capacity because it designed in a way that optimizes for these things, and it compiled in code risk models and all of that. Because we can do that atomically, it means that we don’t have the risk that your position moves further and further against you, and you start to owe everyone in the network money, effectively. So that’s one of the things that allows much better leverage and much better capital efficiency on Vega, just the fact that you can have a bit more guarantee that you can close someone out.
Makes sense. That’s interesting to hear the differences. And I mean, a couple of the perp plays I mentioned … we’re investors [inaudible 00:29:20] disclosure there as well as always. But the next question for you, I guess, on this is, Vega has taken an approach of building its own custom chain, whereas I guess a lot of the perp plays we’ve mentioned and that are quite popular, are basically competing to implement different L2s, whether it be StarkWare, Arbitrum or others. Can you kind of walk me through why you guys chose to build your own app chain?
Yeah. It was very a pragmatic choice rather than an ideological one. It was simply looking at what was going on and saying … As I mentioned to start with, we’ve always been looking at how do people use these products in the real world, and how do we build a version of that that when some of the connectivity and on-ramps are available, will actually be usable for that? Think of an example, like an importer and exporter of goods who wants to hedge their currency risk. If they’re sort of having a collateralized position or only getting 10x leverage when other currencies are at 100x. That’s a real difference. It’s probably a commercial decision that they won’t use the decentralized platform if they need 10 times as much collateral.
We really always looked at it in that way. And we looked at how do we best optimize for, and best do this? And the answer was to build something that works for that. There’s probably two sort of anecdotes I can give that demonstrate this probably being the right choice in my view. The first one is to look at trading and finance in the real world. There are many, many online applications, electronic server-based applications. And for sort of a lot of generic use cases, people will use WordPress or Ruby on Rails or some framework-based system to build their software. When it comes to trading, they hire unbelievable expensive developers to absolutely optimize the hell out of the trading system in C++. Why? Because every bit of optimization makes things cheaper and faster and better, and increases the profit and efficiency of the market.
And I can’t possibly see a way, that if crypto and DeFi’s successful, the amount of money in it will mean it will be worth doing that. And so if you have an app-specific chain, you can optimize for the use case. If you’re sharing your chain with other applications, you won’t get to, and you’ll eventually want to build something app specific. On the blockchain side, that’s kind of how that sort of mirrors the way that finance developed for me. And then the other way I think about it is looking at something like IPFS. You can imagine the guy who looked at the first blockchain and said, “We want to store data in a decentralized way. What do we design?” And I’m sure they very quickly realized that it was too expensive to do a Bitcoin-like thing, even building your own chain, and actually designed a network that was optimized for that use case.
We think about Vega in that way. We don’t worry about not having native assets on Vega and only using assets via bridges, because we think of Vega as this kind of finance or derivatives layer for the decentralized web. So we think it the same way that IPFS gives you this storage capability, so you no longer have to host Uniswap or whatever, or Vega’s console app on a centralized server, you can host it on IPFS. We’re thinking the same way. When you want derivatives and when you want trading primitives, what you want to do is going and find a network that optimizes for that and gives you a great experience that could be plugged in.
So we really see Vega as sort of sitting as that layer, rather than maybe if you’re Polkadot, your goal is to usurp Ethereum and take over. We’re sort of saying, “Actually, Ethereum’s good. You can issue assets on Ethereum, that’s great. It works really well. ERC-20s are good. There’s lots of other protocols on Ethereum which we think are good.” Bitcoin’s great to store value. You’d be able to plug that into Vega eventually, but if you want to trade derivatives and if you want to do margin trading and you want a highly capital efficient financial products and trading, go to the system and the networks that’s optimized and designed for that.
Makes a lot of sense. I guess two maybe lower level concerns, they might not actually be concerns, but we’d love to get your take on it. I guess the perp plays, deriv plays building on Ethereum, and I totally understand that Vega is way more than just one product, but just an easy comparison for this episode. I guess one thing they benefit from is the Ethereum community and having a specific one product. Obviously, if you guys nail it, you can offer way more products. I guess, do you think of it as a concern though on messaging or the narrative? How do you overcome the we can build everything and we are our own chain? Do you think that it’s a messaging issue or do you think that people will get that? I don’t know the best question to ask. I’m trying to formulate that a little bit, but I think you know where I’m going with it.
Yeah. I think there’s a few different ways to look at this, and the first one is to say … This is something that Ethereum and Polkadot and others have had to do. They’ve kind of got to say to people why use our chain. And in fact, it’s interesting, when watching Algorand developers, as Algorand kind of got proposed and everyone got excited. The sort of later, it was the DeFi chain. Well, that was the thing they were sort of trying to do. And it was really interesting just watching how that emerged as a thing, and I wonder if it’s sort of a similar question of, “Well, you can do everything, so what are you actually for?”
It’s definitely very important that we find that. Obviously, we have slightly less of a problem than a completely general purpose chain. And I think there’s two ways. The first way is to find some markets that people just love trading. So that someone launches those markets, and then everyone starts trading them and everyone gets it. And once you see people making money, proving liquidity, and launching markets on Vega, everyone else will want to make money doing that themselves. In the one hand, if you find one or two things that work, and it could be as simple as someone launches a US dollar Bitcoin with 50 to 50x odd leverage, it could be that just launching that will be enough in a decentralized way. And as soon as someone does that, it gets picked up, or it could be something more nuanced or different.
So that’s one sort of side of that question. And the other one is that actually, the Ethereum community is both a blessing and a curse. And what I mean by that is, if you look at the way that people talk about DeFi and yield and AMMs and LPs and the products and Ethereum, there’s tons of risk, there’s all kinds of weird words, and there’s unicorns and there’s rainbows. And this stuff is great. If you’re into crypto, it’s really fun. It’s a real cool community. It’s a real, great sense of something exciting happening. But it also means that you go and talk to someone who just wants to access a financial product just to hedge a risk in their business. And you show them this stuff, they don’t know where to start. And so it’s not very accessible for the next level.
And I think actually in some of the guys like Compound, for instance, I’ve seen them sort of say things like, “We’re actually aiming to be very boring and to move slowly, and to just be this sort of approachable money market rate, approachable lending protocol, because we want to take over real, traditional finance, and we want to be this kind of slow and stead thing.” I think there’s a lot of value in something like that, because I do worry that some of the things going on in DeFi will be very difference to translate to the real world use case.
So I think some we have to do is convince people, we don’t need to necessarily sit there and try and become the rainbows and unicorns of Ethereum. We need to convince people of our value and set out store well, and say, “Actually, this is what we’re for. This is what Vega can do.” And in fact, I’ve spoken to people at even some sort of the major Ethereum-based protocols, have said, “Yeah, once your two-way bridge is in place and you can take out Vega positions from an Ethereum smart contract, which you’ll be able to do,” they could see using Vega to hedge things in their Ethereum smart contracts to make Ethereum DeFi protocols more efficient. So Vega has a place even in there. But I think it’s a case of us building a real good narrative for what Vega’s for and what it can do, rather than trying sort of shoehorning it into the DeFi narrative that currently exists, that maybe doesn’t quite fit Vega.
That’s a really strong answer. I loved your take on Ethereum being a blessing and a curse, and it is pretty interesting to think about being able to build anything and then dominating a niche. But you’re basically able to dominate a niche and offer anything. So that is really [crosstalk 00:37:13]
Yeah. And look, I mean, I think something that I would love to see is I think there’s the opportunity to take some of the things like Vega that exist, and to build products that people can use in the real world now that may be tactical solutions from a full decentralization point of view.
And what I mean is, let’s say you wanted to offer hedging products to people. You could build a front end, you could integrate with credit card providers so that people could actually pay for it from their bank account or with their credit card. And you could build a business on the back of something like Vega or offering them a better price or more access to these products. But you could just build it for the US market, you could be regulated in the US. You could use Vega products on the back end, but you could just present it to people and give them things they know and love, and ways to integrate with the current financial system they have. So you can kind of bridge those worlds.
And something I’m really keen on in the Vega ecosystem is finding people who want to build those kind of thing, and really helping them sort of think about customers and their needs in the real world, and think about how to use Vega for that, and really empowering them to do that. Because I think if we could get even just one or two services like that to take off in one or two places, and actually get real world use, I think it would be really transformational and really good for DeFi to start being held to the standards of that kind of user, as well as our sort of internal crypto and DeFi narratives.
Awesome. No, that makes a lot of sense. So I have some more specific parameter type questions. So just diving in. So anyone can launch a market on Vega, and the markets, for there to be created new liquidity, which requires incentives to kind of attract those liquidity providers to the new markets. How do you plan to reward those liquidity providers and track them in the protocol, and I guess, to retain them too?
Yeah, that’s a really great question. It’s one of the most important things, actually. One of the things that I realized really quickly when sort of trying to pull together the high level design for Vega was that although I wanted to remove these middlemen that were the exchanges, and they do do some things which no one else was doing. And one of those things was to create enough liquidity to launch something and to ensure there was kind of liquidity and to match those needs.
A lot of the particularly early … if you ever remember EtherDelta and the sort of 3,000 markets, none of them with anything on the order book, type of experience. A lot of people overlooked that initially, when they built decentralized exchanges. So something we thought about from day one is, how do you make participants in the network take on the role of the exchange? And roughly speaking, that role is build a market, attract liquidity before there’s proven demand, and then profit it there is a lot of users.
So the way that we think about this is that liquidity providers in Vega commit liquidity to the order book. In future, they will also be an AMM style thing, to allow what you might call passive liquidity commitments, so people can kind of just dump their collateral in for yield. But liquidity providers commit to a market, and when they commit to a market, the size of their commitment effectively buys some number of LP shares, in terms of a share of the revenue that comes from the trading fees on that market.
But the thing that’s interesting about the way Vega does this, is it treats the market as a bit like investing in a business. If you invest in a business as a seed investor, you’re going to get a much larger share of the business than if you then come in with the same size investment in a series A, assuming the business is going up in value. And so, we wanted to capture the same thing in a market, to say if you take more risk, if you create a market or you come in early, you get more of it for the same size of commitment than someone who just sees a giant liquid market and comes in.
So what we effectively did was we looked at what’s a good proxy for value of a market, and we decided that trading volume is the value of a market. Because why is the market there? Well, it’s there to create the opportunity to trade and so the more opportunities people have to trade, the more trading that happens, the more valuable the market is. So we effectively measure the kind of average trading volume per unit time at a given point in time, and that’s the sort of proxy for the value of the market. So when you commit to a market, you commit and you get your LP shares based on the amount of trading going on.
So if you create a market of come in early, you put something in, you might get 20% of the LP shares. Even though in six months’ time, the market might have 10xed in size, and you might be only providing 5% of the liquidity. But because of your sort of lower valuation when you came in, if you like, you may still own 15% of the LP shares. What that effectively means is that as well as just market making itself and liquidity being profitable as an activity, picking good markets early on, or creating new markets that then do well is profitable in the same way that being a VC is profitable.
And what we expect to happen is that when people realize this and they realize that … they watch a market go from 10k daily notional volume to 10 or 100 million, and they watch that person who provided a relatively small amount of liquidity is suddenly receiving multiples of it, everyone is going to want to create markets, because they’re going to realize the size of the opportunity for creating them. And so it’s that kind of capitalist, business-like way of rewarding people for creating markets and providing liquidity that I think is one of the most exciting parts of the Vega design.
Yeah, not only incentives, it’s so important, so I love that you guys are extremely thoughtful about it, and you’re incentivizing people at the right stage and in the right amounts. I guess one of my other critical type questions, just to run through them, is how do you assure that traders, that their money is safe and that the system’s a fair one? There’s a lot of risks here, like counterparty risk, the efficiency of price discovery. I guess there’s a lot of things to think through, but I guess, how do you think about those?
Yeah, and the first thing to say is you can’t just reassure people, I don’t think. You have to prove it, and sometimes, proving it’s going to take time. So the first thing we’re doing, we want to engage the community a lot on this. We recently completed a raise, and we’re sort of in an expansionary mode as we get close to the main net launch. And one of the sort of capabilities we’re building is an internal research capability, that rather than being focused on research of the development of the protocol, is research in how the things we’ve built performs, and what it’s like, and where the risks are.
So when you look at the great work that people like Gauntlet are doing, and we hope to work with those guys too at some point, and get that sort of community input, and you guys as well at Delphi. But when you look at all that great work, we think that it’s not just enough, that the community and Excel people are doing that. We also think that it’s sort of a responsibility of the Vega team to be doing that research and putting data out there. So that’s the first thing. Obviously, we’ll publish everything we do around testing. We’ll be open sourcing everything around the protocol at the right time. We’ll be getting audits done on a lot of this stuff. So all of that stuff as well.
Then there’s just the amount of time that passes. Sometimes, you just need to see that the protocol has worked flawlessly for one year, two years, five years. And I think the more money that’s been in the protocol and gone through it, the sort of more secure you feel. The larger the use cases become, like if the whole protocol is doing a million dollars of trading a day, you probably as an individual or as an entity who’s trading don’t want to do 500k, because you’re like half of the value of the protocol, so it feels risky. Whereas if you’re only 1% of the value of a fraction of that, then you’re taking less risk compared to what’s going on, so it’s a little bit safer. So that’s another way to think about it.
And then the other thing we’re doing is we’re sort of allowing the community to have a bit of control over this. So we’re building some limits into the system, which is effectively, when we launch Vega … and we got the idea from … actually, it was the Lightning Network, I think. When they first launched Lightning, I think it was about $300 worth of Bitcoin, was the most you could move around in one transaction. It’s just designed to stop people doing crazy, big things on something a bit untested.
And we’re sort of doing the same thing. We’re going to have a few limits around the deposit amount for any given key. And the goal of that is to basically say, “Well, look. You can Sybil attack. You can just make 1000 keys and do it if you want. We’re not … It’s decentralized, we can’t stop you.” But what we do want to do is say, “Well, as a community, we think that right now, taking 1000 bucks worth of risk on Vega is probably okay to do, but don’t take more than that, because it’s new.”
And then the community through the governance process will be able to vote to raise those limits, and eventually remove them. And so we want to put some of this stuff in the hands of the community to say, “Actually, we don’t want to tell you you should trust Vega. We want you to tell us that you do trust Vega.” And we want to build that kind of relationship with the community where we’re constantly doing what we can to demonstrate it, and the community’s kind of coming back to us and saying, “Okay, we’re ready to take this a step further.”
That’s awesome. Again, extremely thoughtful. And I guess fees play into this a lot. You touched on them a bit, but I guess, how do you implement fees, and how is the fee structure better for users on the Vega protocol itself? Sorry to jump all around, but your answers incredibly thoughtful, so I want to make sure we get to all of them.
No problem. Fees are really, really important. One of the sort of the problems with the generic blockchain is, got all these different competing use cases, and so you end up doing things like measuring how much computational power they use. And then sort of charging them the gas and all of the problems we see with high fees and MEV and everything on Ethereum. And so we totally rethought that when we were thinking about Vega.
Because firstly, if you’re running a market, you actually don’t want to disincentivize placing an order. Because a limit order provides information to the whole market. If I place a limit order saying I’m willing to buy for 100, that tells everyone in the market something. It allows the market to be more efficient at price determination. If you create a system where it costs 30 bucks to put a limit order in, you dis incentivize providing that information to the market, which is not generally what we consider a good thing, as people with a sort of trading pedigree. And so the way that we’ve designed Vega is that actually, you don’t pay to place an order onto the blockchain. You pay when your order trades. There’s a fee for trading, like on a centralized chain.
Now, there’s a risk here, which is that if placing an order’s free, I can just start spamming the network with hundreds of orders and fill up the blocks. So we have a neat solution to that, which is basically proof of work. But it’s not proof of work on the nodes, it’s proof of work by the client. So if you place one order, one transaction on the block, it’s basically no effort. If you start putting more transactions into a block, it becomes increasingly more effort in the proof of work you have to complete for your transaction to be accepted into the mempool. So effectively, the nodes will start turning your transaction down unless they’re accompanied by a proof of work. And if you try and fill two thirds of the block, the proof of work is going to be incredibly expensive.
So effectively, we have this sort of system where we can limit how much you can do and how much spam, and put a cost on that, without actually charging you to place an order. And that I think is a really neat solution, and it also allows us to make order placement free. So fees are charged when an order trades, and in fact, the fees are charged to the aggressive order. So if you’re providing liquidity, if your order was on the order book, you don’t actually pay a fee at all. In fact, you get a small rebate. So you actually get paid for your order trading. But if you are taking liquidity off the book, then you pay for the liquidity.
So it’s pretty simple, maker taker, but it’s kind of very pure form of that because there’s no exchange that wants a profit from everyone. The maker actually receives money and the taker pays. So the taker’s buying the liquidity from the maker. And then the fees are effectively then split between the maker taker relationship, like I described there. So the taker provides, pays the fee. A small portion of it goes to the price maker, another portion of it goes to the validators in the network, and then the remainder goes to the liquidity providers. And the amount that goes to each liquidity provider is, like I mentioned before with that kind of sort of equity-like share, the LP share that you have in the market, that determines how much of that fee you get as a liquidity provider. So really, the whole thing is based on that sort of fee.
It’s incredible to think through. Yeah. No, I love that. And just to jump around one last time, because I know we’re running low on time, but I want to make sure we get to it. Can you kind of dive into Wendy real quick, and why it’s important for Vega, and I guess, just DeFi broadly?
Absolutely. This is something actually I’m really, really happy we managed to do when we first started, when I was talking to [George 00:49:29], and [Dave 00:49:30], and the guys about sort of the proof-of-stake, and thinking about how to do Vega. One of the things I said was I’ve been in the Ethereum ecosystem long enough to know that … Ethereum lets you pay for front running. You can literally just say, “Hey, I’d like to be ahead of this order because I know what this person’s trying to do,” and you can pay for that. Now we have MEV and Flashbots and all of that stuff that are trying to research and deal with this. We sort of posed the question, how can we fix this problem?
The good answer was that because we control the consensus layer and we control the blockchain, and it’s optimized for this use case, we can make other trade-offs to let it happen. The obvious one was effectively where you do everything over two blocks. So that you put your order in, in an encrypted way, and then in the next block, you provide the decryption key so that effectively, there’s a commit and reveal scheme, so that no one can see what the other orders are at the point where they have to place their orders. That’s a horrible, horrible thing from a latency point of view, and there are some of other sort of issues with it. It’s not a very elegant solution, but it was the starting point.
But we got Klaus on board onto the team, and he’s a blockchain researcher. He’s been designing kind of distributed systems and consensus since, in fact, before Satoshi’s Bitcoin whitepapers, so he’s a really sort of serious academic in this and a researcher. And we basically posed this question for him, and what he did was he designed a protocol, which is what we’ve called Wendy, which runs at the consensus layer, and it means that … If you can make the same trust assumptions about the validators that you have to make for the proof-of-stake anyway, i.e. two thirds plus one honest validators, then by running this protocol alongside the consensus … it’s an ordering protocol. If those validators are honest about when they see messages in relation to each other, then we can kind of guarantee that if the majority of honest validators saw your message before mine, that the block will include your message ahead of mine.
That’s really important, because it now means that effectively, you don’t have this way to come in, look at the state of the mempool in the block, and just say, “Hey, I want to get in front of that and take that value of the protocol.” So it’s designed to completely combat the MEV problem.
Can you go a little more into it, Barney? How significant is the MEV problem today? I mean, we all see the success of Flashbots on Ethereum, but would Wendy totally solve that? And I’m not as up to speed on, I guess, the different concepts of fairness here.
Yeah, I believe it would totally solve it. There are some bits where my knowledge gets to the end of its knowledge in terms of the computer science, of whether there are situations in which someone could sort of force Wendy into a state where it’d be an undecidable problem and where a random choice would be taken that could give some advantage. But in general, it certainly solves the vast, vast majority of it. When you sort of say how significant a problem is, I think it’s a very significant problem for two reasons.
Firstly, as more and more value moves on chain, a small percentage of that value leaking out to validators or whatever by MEV becomes a bigger and bigger issue, because it’s not only an issue in terms of the perceived fairness, but it’s also an issue in terms of the efficiency. And blockchain’s already on the back foot compared to centralized servers, because of the processing needed to run a decentralized system. You don’t want to be tipping yourself towards the point where your cost to trade is higher and higher because of MEV.
But the other problem is that it pushes up gas fees. So if I can extract $100 from your transaction by going ahead of it, then I’m willing to pay 99.9 dollars in fees to get ahead of you, and so I’m going to be bidding a higher fee, and so the fee needed to get into a block is just going to go up and up the more DeFi gets used. And we’re already in a situation where fees aren’t great, so I think it’s a big problem from both of those perspectives.
But more to the point, when we get to this kind of crossing, the chasm thing that Vega’s always been angled for, I can’t persuade a trader or a commercial user of derivatives in the City of London or in New York or wherever, to come and start using protocols where people are paying to front run them and undercut them. It’s just not going to fly, either regulators will be very unhappy about it, and would not let many types of institution even potentially trade on those kind of markets.
It’s going to be a nightmare. And so I really think that it has to get solved. And I think the Ethereum guys and community know that as well. Flashbots and others there, I think they’re looking for solutions. We just have this advantage that we control the whole blockchain of Vega, and we have the ability to implement something that much more quickly, because we’re not worrying about legacy users or other applications. So we just have this ability to maybe execute a bit faster and move quickly on this.
That’s awesome, yeah. No, MEV’s definitely top of mind. So after this podcast, I’m definitely re-read that Wendy paper, and probably have to re-read it again and shoot you a couple questions.
Absolutely, please. Please send them through.
It’s really interesting to think through. Yeah. No, I’ll tag them in the podcast show notes for those interested too, and maybe we’ll do an AMA with your team to kind of think through that a bit more too, and just [crosstalk 00:54:36]
Absolutely. And the other thing that we’d be very happy to do is, Klaus, I’m sure will be happy to come on and talk in-depth about Wendy and MEV, if that’s of interest to your listeners.
Yeah. No, it totally would be. I’m glad we’re planning new ones while we’re still on it. And I guess, just a couple closing questions for you, Barney. Wilson from Messari asked your take on IBC. I guess, how bullish you are on it, and if it’ll be inclined to connect solely with the Cosmos Hub, or just on top of an SDK? I guess, what’s your take on that whole dynamic?
Yeah, I think so for me, IBC is a bit wait and see. Maybe if it had existed when we’d started Vega, we would’ve built some things using it. But there’s a lot of, say, sort of make certain requirements of a network that are not necessarily super aligned with the stuff that we need. It may add a bunch of overhead. It’d be interesting to see if people start using it, because if it really becomes a kind of … If protocols get built on top of IBC that let us integrate with many other chains quickly, or let us do some really cool stuff, then I think we would probably pretty quickly move to support it. But at the moment, it’s an interesting solution, bit wait and see.
And as for the other chains, we’re actually super interested. We’re working reasonably closely with the team at Terra to think about ways to integrate the Terra chain, it’s something we’d like to do as one of the earlier integrations with Vega. We’re certainly looking at other chains like Polkadot as well. So we’re certainly not limiting ourselves to Cosmos or IBC, but if IBC helps solve a problem and gives us access to … Ultimately, the way I think if Vega is, as a derivatives layer, it needs collateral to be traded. And so wherever the collateral is in the blockchain ecosystem, wherever the funds that people want to trade with are, whether it’s on the Bitcoin blockchain, Ethereum, Terra, Cosmos Hub, wherever that collateral is that people actually really want to trade and wants to access derivatives with, that’s where we need to support. And if IBC makes it easier to do that, then we’ll certainly support it.
Awesome, Barney. And one last question from him on this topic, or actually bit of a different topic we kind of already covered in part. But I guess, just managing the crosschain collateral, I know you touched on it earlier, but what’s the plan if there’s a lag between moving collateral between chains or cascading liquidations? How do you think through that?
Absolutely. So I guess Vega’s designed at the moment that it is independent of other chains when it comes to liquidations. If there’s liquidation on Vega, that happens on Vega, and the amount of funds on the Vega chain are enough to deal with that, in that it’s all calibrated that way. So what could happen is you have a position that you want to keep, and you need to transfer money, and then if there’s a lag, then your position could get liquidated before your funds are transferred in because you’ve got that sort of lag on the bridge. But what’s not going to happen is that that’s going to … the product’s sort of interconnected in that way across the chain, so it’s not quite the same as two smart contracts that need to talk to each other in that way.
Lag is obviously non-ideal, and I think there’s everything from some of optimistic roll ups, where you have a sort of seven day period of withdraw, which we don’t have, it’s kind of more like a number of confirmations, like you might have on a centralized exchange. But one of the things that we’re looking to research in general is increasing the ability to interconnect between chains to build things in that work on Vega, from Ethereum smart contracts to speed up to interactions between chains. We may be able to do certain things much more optimistically. We may be able to wait for fewer confirmations if some other parameters are true. There’s a bunch of bits of sort of research we can do, and are doing around latency and lag and interactivity between chains.
But in general, on the Vega side, it shouldn’t cause a lot of risk. It might mean that if something on Ethereum is dependent on a Vega position, then that lag is an input into sort of how it manages or thinks about the risk in that Vega position though.
Makes a lot of sense, Barney. And just to close out, what’s something with Vega that’s currently not solved that you’re attempting to solve, or maybe a risk that you want to tackle this year? What’s keeping you up at night?
What’s keeping me up at night? That’s a good one. Honestly, I think the most terrifying thing in the world is putting something live that you don’t control. Or in fact, we won’t even be putting it live. We’ll be putting the software out. Other people will be running it. It will be live, and then other people will be putting their money in it. The thing that you always think of is stuff like the DAO’s hack early on in the sort of history of Ethereum, and how giant that was, and how that set that team and a bunch of things back.
And so I think the big thing that keeps me up at night is how on earth do you test and ensure quality, and how do you do that? And one of the reasons we’re sort of doing this sort of cautious launch, we’re going to have these limits is because we know we can’t guarantee it. So everything we can do, whether it’s incentivizing testnet, whether it’s bug bounties, whether it’s more QA and audits, whether it’s more analysis, more integrations. Anything we can do to de-risk that and to try make sure that the launch is safe for people, and that there’s trust in the team and the project and the network is probably the thing that we’re sort of constantly thinking about right now.
That’s incredible, Barney. It’s been incredible having you on. You’re super well-spoken. You understand your project. It’s great to kind of hear the story on how expansive Vega is, and why you chose to build your own chain, and the things that people can build on it easily, and kind of how you handle incentives. It’s really kind of cool to think through all that, and it’s obviously super well thought out. So love having you on, definitely want to do another one sometime soon.
Yeah, thanks, Tom. It’s been great to be here.