OCLUB AMA RECAP: Orbiter X zkSync (16/02/2023)
Continuing with OCLUB AMA episode 2, we have hosted zkSync to toss around their scalability, exhilarating updates awaiting to be announced and their roadmap ahead. We were joined by zkSync’s head of BD, Marco.
Full Recording
Click to play our AMA recording on Spotify.
AMA TLDR
- Exhilarating updates on exciting milestone — Fair Onboarding Alpha.
Marco: Today (16th Feb) is the day that we open to project so it’s what we call fair onboarding. And the idea is that zkSync has been investing since February last year openly and then in main net since October last year. Finally now open the door to projects so any project, big or small can register for deploying, so we are very excited to see what happens now. We have hundreds of registrations, and we are just waiting for the flood of questions and requests and support. But you know, we love the idea of now finally being able to really engage with the community on on their own projects. The second thing I can announce is that we have decided to make the entire code open source. This is a big deal. So at the moment, most of our code has been developed in a closed fashion. And the idea there is that we’re developing an open source is time consuming, there’s a lot of effort. But it’s also usually important to give everybody comforted that what’s happening on chain is verifiable, and also to see what happens. So once the community starts to contribute to our code, and I’m sure we will grow much faster than what you’re growing now. And nobody has done anything like this, most of our competitors have announced that they tended to open source, but they have not done it yet. Some competitors made something open source, which is still unfinished, to maybe more testnet version of the infrastructure. We see ourselves as first and foremost a security company. So the most important thing we are trying to do, which is in line with the essence of zkrollup, zkrollup running the same exact security of Ethereum, so we want to make sure that zkSync is as safe as possible. We spent the last few months testing the system, we are now moving into the second phase of the opening, which we call Fair Onboarding. The idea is that we can allow now projects to deploy on chain, but only projects, not users. Why we are doing it, I think there’s a couple of reasons. One is we want when users come they have a full ecosystem fully formed. So some of the other rollup, when they open, they only had one application. And then after a few months, they have two applications. And this doesn’t lead to a nice network effect because users can only do one or two things that doesn’t lead to a proper ecosystem experience. What they would like is that when users arrive, there are hundreds of application that are ready. So it’s a fully formatted ecosystem. The second is we want to make sure that project they have the ability to deploy in a secure way. So we want to give them time to deploy, make all the tests without having to rush to deploy because there’s other maybe competitors that are deploying faster. Obviously, when we go into the next phase, which is that is make the system permissionless, anybody will be able to do everything, anything they want. But at least in this initial phase, again, security in mind, we are trying to make sure that everything is as safe and practical as possible.
2. Thoughts on how zkSync advances Ethereum scalability and how to guarantee the security.
Marco: So first of all, maybe I get to start with a problem. So the problem that we have in minority blockchain which includes also Ethereum, is that they try to do too many things at the same time. Put it in a better way, there is a trilemma called blockchain trilemma. And the idea is that blockchain has a lot of properties, but three key ones, one is decentralisation, the other is security and the third is scalability. And in general, you can get one of these properties very easily. You can get two with a little bit of effort, but it’s impossible to get all three. So if you want something that is safe and decentralised, you have to sacrifice scalability. That is because especially in order to have decentralised, we need to have small hardware that is able to run the system, which is the case of Ethereum nodes. In which small hardware where they simply cannot process enough transaction at the same time. And that’s the problem that we have at the moment right now. How do we solve it with zkSync in general, rollups solve it by moving computation of chain. So we take one of the key things that the blockchain is doing. The general purpose of blockchain like Ethereum is like computers. The distributed computer, which has been computing is doing storage is doing everything in a safe and decentralised way. So what we do is we take computing and we move it off chain, and you can do it in two ways. One is the optimistic approach where you do it through fraud proof. The idea there, it’s quite nice actually, and is that you can take a validator, is to verify transaction off chain would then send them back on shade with a confirmation that they’re valid, and then sometime is allowed to pass for anybody permissionlessly to provide fraud proof. So the idea is that the validator because he or she knows that anybody can prove that they have done something wrong. And because anybody can do this proof, then result someone will do it all the time. So there’s almost no point in being malicious. There’s no way that you can be malicious because if you create an invalid block, you will be punished and you will not really achieve anything with that. So now this is all kind of nice. It leads to a dilemma where everybody reacts well, without pretty much anybody have to do anything for for people to behave well. If we move to zkrollup, what we do is we take the transaction of chain like optimistic rollup, but then we calculate the proof of it. So we don’t only verify them, and we allow people to say no, you’re wrong. We calculate that in irrefutable mathematical proof that all the state changes are byte and we send that this proof together with the final stage back on chain. So this is a much better system. Basically, the only problem is that the math is much more difficult. And so it took a little bit longer to get there. But pretty much everything else is better, we are faster. So once the proof touches the chain, and the system is immediately Finally we don't have to wait a week. So it’s faster for message passing is faster for NFTs. And it’s faster for also fungible token is safer, because you don’t need to trust any human behaviour. It’s simply impossible for people to create invalid blocks. And so that’s strictly better. You just have to trust the math behind the system. And then more importantly, zkSync starts to scale, like a computer system.
3. Deep dive and envisions on zkEVM: technologies, status quo, what are ahead etc.
Marco: zk-EVM is a general broad word that actually we came up with a long time ago but then it has been changed recently to zk-EVM equivalents. So in general, there are maybe three broad flavor of zk infrastructure as I was saying before. One is the starknet flavor where you ignore EVM and you just create your own language. The other is like the scroll approach where you try to be totally EVM equivalent. If you create your own dedicated language, you get out of performance, but very small compatibility to know compatibility. If you go for transpiling, essentially the solidity code into something that can be used with zk proof, then you get this huge penalty in terms of performance, but you have a very, very high degree of compatibility. We think our approach is in between. So we think both of these characteristics are necessary, not sufficient. So they are not just nice to have, they’re both necessary. And so you need to have both at the same time. And our approach, it gets to a very high degree of performance, again, we will need to see live, we think we can be more performant than the starkware implementation. And we will always be more performant than equivalenct zk EVM just because we directly combined, so it will always be better in terms of performance. While at the same time where we are extremely high in the compatibility space, we also have a couple of ideas on how we can implement the full bytecode compatibility. So equivalence, stay tuned on this will probably be released something in the next one or two quarters. So this is a little bit the development space that we have at the moment. And we think that our approach is the more pragmatic one is the one that will lead to the highest degree of adoption, because you get something that at the same time is very performant and is very compatible.
4. How to attract more developers to deploy in zkSync in the future?
Marco: The first thing we did is really this EVM forms and also we are live today open for any project to deploy. At the same time you have an infrastructure that is compatible, so usually you don’t have to change anything to deploy in zkSync or you may have to make very small changes that are very easy and very simple to do. So that’s one one first way to attract developers. And also, most of the other roll ups, especially the optimistic rollup is being considered by some project as a stepping stone. I know this is not the future state. But in the meantime, I’m deploying on an optimistic rollup because it gets me some scale while I wait for zkrollup. So we basically saying you don’t have to wait anymore, you can just deploy and this will be potentially the place where you can stay forever. And then we were thinking about how to support our ecosystem with grant programs. We will direct the support to the developers, not just in terms of financing, but also in terms of advisory on how to set up the company behind the project, in advisory on how to use certain tools to improve the quality of the code or the infrastructure. We are partnering up with a lot of our infrastructure providers to create the teachings so that our partners can go into an AMA with our infrastructure service provider, and learn how to use them better and grow and become more efficient and faster. And slowly, we will also start to pre negotiated deals with some of these infrastructure providers so that our partners have a simpler way of integrating technology like privacy, like digital identity etc. So this will be all the program that will be put in place, at the moment, we are having a huge response from the ecosystem. So we are of course, providing all the technical support to the partner. And we are starting to do this the second thing that I mentioned, but we don’t have a grant program. And despite this, there is a huge amount of projects that have deployed in mainnet like NFTs, games, identity privacy. So we are very excited that the ecosystem will be quite nice and interesting in terms of network effects for anybody to join.
5. Elaborations on the reference for Layer 3 as ‘HyperChains’, and views on will it be an ecosystem of customisable and trustlessly linked blockchains.
Marco: One thing that zero knowledge proof enabled through zero knowledge proof is the ability to launch a roll up on top of itself. Essentially, zkSync is a Layer 1 smart contract that is a server of chain that verified the transaction. Once again, I think you can copy the server and we could be giving this to Orbiter, for example, or to a large corporation. And now you will have your own layer that is totally sovereign. You can change the consensus mechanism. You can use your token you can do a lot of different things with these layers, totally independent in terms of sovereignty. You can choose how to run it completely. So what you can do is you can have a very cheap, and very fast and the compatibility across different layers. So imagine a layer that is done for a large payment company. And then a layer that is done for a larger E commerce company, imagine a layer for MasterCard and layer for Amazon. And then what you can do is you can buy things, verify them all the mastercard here, send this confirmation to the Amazon layer who will give you your assets. We should use it will confirm that you have made a payment and then we will recall that you have to receive some assets later right in the immutable data base. So that’s pretty cool and it’s something that can only be done with this web of layers on top of zkrollup. So that’s a little bit what we are working on at the moment.
Marco: Also, chain narrative is something happened with dydx, a very successful project. The problem that they have with this is that at least 21 is that they need to create a bridge between that chain and Ethereum. And that bridge is not native I mean Orbiter is the first one to integrate. And it’s a non easy problem to deal with, so they will have this into to work hard. And the second is that dydx needs to bootstrap the security of the chain completely, this means that the security is the token value, the tokens by these days worth the hundreds of millions of dollars, which of course is not secure because the hedge funds, regulators, they could have enough money to just be covered the consensus and do whatever they want in the blockchain. So with hyper chains, dydx will have a native bridge to Ehereum. So they don’t need to care about this problem while retaining all the good stuff that they wanted when they created their own layer, they are sovereign, they can choose the latency. The second thing, what I mean that we like hyper chain is that they are the answer to the older private chain stories. So if you’re an enterprise today, you are scared to go direct to a chain. What you do is to typically work with IBM or with the three consortium, which will give you some private block chain infrastructure. One of the biggest problems we spoke with many of them is that these are private block chain, don’t scale, and we think it’s fundamental. They don’t scale because they’re private. So if I am a bank, I will not to want to use my competitors bank blockchain because they can mess up with the state. They can do whatever they want basically in a private blockchain. So it’s really not useful to have the best private blockchain. We have a hyper chain instead is that this will get two things. One, again, they gained the native bridge to Ethereum which they may not want to use, often they want to any private permission infrastructure, at least at the beginning. But it’s easy to take a system that is permissionless and make it permission and then later removed the permission. And its impossible to be the opposite which is to start the permission you cannot make it permissionless later. So very exciting to see what happens when bigger enterprises start to shift to some of their business structure to blockchain. We think it’s absolutely essential that they will do something like an hyper chain.
About zkSync
zkSync is a ZK rollup, a trustless protocol that uses cryptographic validity proofs to provide scalable and low-cost transactions on Ethereum.
About Orbiter Finance
Orbiter Finance is a decentralised cross-rollup bridge for transferring the Ethereum native assets, which is the infrastructure of Layer 2, we offer low cost and almost instant transfers.
