EthCC #3 | "StarkWare isn't Leaving Ethereum" - Eli Ben-Sasson and Uri Kolodny

🇫🇷 Welcome to the Bankless EthCC 2022 Experience 🇫🇷 With fantastic guests from all corners of the ecosystem, this 8-part series is an exploration of crypto culture and the current state of Ethereum. In episode 3, we cover the recent announcement of the StarkWare token launch, the exploding zkEVM ecosystem, and the future of StarkWare. ------ 📣Rhino. Fi | Massive Mystery Airdrop https://bit.ly/3o9trRE ------ 🚀 SUBSCRIBE TO NEWSLETTER: https://newsletter.banklesshq.com/ 🎙️ SUBSCRIBE TO PODCAST: http://podcast.banklesshq.com/ ------ BANKLESS SPONSOR TOOLS: 🚀ROCKET POOL | ETH STAKING https://bankless.cc/RocketPool ⚖️ARBITRUM | SCALING ETHEREUM https://bankless.cc/Arbitrum ❎ACROSS | BRIDGE TO LAYER 2 https://bankless.cc/Across 🦁BRAVE | THE BROWSER NATIVE WALLET https://bankless.cc/Brave 🌴MAKER DAO | DECENTRALIZED LENDING https://bankless.cc/MakerDAO 🔐LEDGER | SECURE STAKING https://bankless.cc/Ledger ------ Topics Covered: 0:00 Intro 2:30 Eli and Uri 4:52 zkEVM Explosion 10:30 The StarkWare Approach 16:20 StarkNet on Ethereum 19:35 The StarkNet Token 24:30 Token Models 26:53 The Future of StarkWare ------ Resources: Eli on Twitter: https://twitter.com/EliBenSasson Uri on Twitter: https://twitter.com/ukolodny StarkWare: https://twitter.com/StarkWareLtd ----- Not financial or tax advice. This channel is strictly educational and is not investment advice or a solicitation to buy or sell any assets or to make any financial decisions. This video is not tax advice. Talk to your accountant. Do your own research.

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:03David Hoffman?

  Welcome Bankless Nation back to this third episode of the ETH CC experience. In this episode, we're talking with Uri and Ellie of the Starkware team. And this is coming off of their Starknet token announcement, which is of course the subject matter of this particular episode, but not the only subject matter, of course. There are many different things to talk about. I get their perspective on the ZK EVMs that were released, not from the Starkware team, but from three other competitors, all having to get their ZK EVM released in the same week. Scroll, Polygon, and ZK Sync. But since Starkware is peripheral to this, they are a ZK rollup. I get their perspectives on this whole role and where they fit into this conversation. Of course, like I said, ask about the Starkware token. And then just really clear the air as to where will Starkware be in the future?

• 0:48David Hoffman?

  Are they ever going to make their own Layer 1? Are they ever going to use an alternative Layer 1 other than Ethereum? And I'm too good at producing content to tell you that answer here in the intros. I'm going to wait for you to find that one out in the actual show. But first, we're going to speedrun the sponsors all here right now, so we don't have to interrupt the conversation later on. And we're going to start with Rocketpool. Rocketpool is Ethereum's decentralized validator network where you deposit ETH and you get our ETH in return. And you can take that our ETH into DeFi. And if you run a node, you let other people deposit their ETH into your node and you get to charge them 15% for staking rewards as payment for your node operating services. And then you can take that our ETH and use the across bridge to join over $2.3 billion in the Arbitrum ecosystem. Across, of course, is the fastest, cheapest, and most secure cross-chain bridge

• 1:33David Hoffman?

  and one of the safest ways to bridge your hard-earned assets in the world of crypto. And of course, Arbitrum is where you want to be. With over 35,000 contracts and 1 million unique addresses, Arbitrum is leading the way into the age of rollups. You know how I know these numbers? Because I watched Arbitrum's talk at ECC, which you should also go watch, but not before you're done watching this. But when you do watch that video, make sure it's on a privacy-first browser, which is why you should be using Brave. It blocks all the browser ads. It's got a native Web3 wallet and it puts the user first the Web3 way. You know what else puts the user first? Ledger hardware wallets and the Ledger Live ecosystem. I know Ledger is a great product because I have literally nine of them and I have 99% of my crypto assets on my Ledger wallets.

• 2:14Speaker 01

  And the Ledger Live ecosystem is all you really need when it comes to living a bankless life, such as buying with fiat, swapping and staking. But for the rest of you that are sitting on stablecoins because the market is scary right now, make sure you're using a decentralized bankless stablecoin. May I suggest DAI from Maker DAO, the most tried and tested stablecoin in existence. But with Maker, you don't have to hide your stables on the Ethereum Layer 1. Maker is building DAI portals, so you can mint DAI natively across all the Layer 2s, so you can make sure that your assets don't go down another 90%, but you still get to go do all the DeFi things across the L2-verse. And I hope you use all of these sponsors on your quest for going bankless. And now I bring you my in-person conversation with Uri and Eli of the Starkware team. Eli and Uri at Starkware, what's up, guys? How's it going?

• 2:57Speaker 02

  Hi, hi. Hi, pretty good.

• 2:59Speaker 01

  Yeah? Is this your guys'... Not your first ETC, certainly.

• 3:02Speaker 02

  Uh, actually, it is mine, but it's been here a few times.

• 3:05Speaker 01

  Yeah?

• 3:06Speaker 02

  Yeah, at least three.

• 3:07Speaker 01

  Yeah? How do you like the city? How do you like Paris?

• 3:10Speaker 02

  Amazing. What's not to love? I hear they have good food.

• 3:14Speaker 01

  The current state of the crypto world, we're in this like bear market. Whether or not you think Ether at 1,500 or Bitcoin at 22,000, wherever it is, is a bear market is up to be decided. But the current like meta that we've going into is that we need to build our way out of the bear market. We're not going to just like magically come out of it. We have to actually do something to get out of it. And Starkware has certainly been heads down. Guys, of course, started with StarkX, the proprietary app-specific change, we would call them. But we are now going into the more generalizable ZK rollups, which is what we're doing with Starknet. So let's talk about Starknet first, because that's the thing that's getting everyone

• 3:58Speaker 01

  super excited. Where would you say in the grand roadmap of Starknet, where are we and where are we about to be soon?

• 4:06Speaker 02

  We think about this in terms of three phases, overlapping phases. The first phase is the feature completeness or the feature phase. We're pretty far down the road there. People have been able for a good number of months now to build whatever they want on Starknet. The second phase is what we refer to as scalability phase, and there the intent is by the end of the year to reach 10x Ethereum's throughput at 1/100 of the transaction cost. And the third phase, which started a long time ago, the decentralization effort, and this is a gradual phase, and we're making sort of incremental progress there. Day in, day out, this will culminate in the ability of anyone, anywhere, being able to run a prover or a sequencer on StarkNet.

• 4:51Speaker 01

  There's a recent tweet from John Adler, which was pretty funny because Polygon just announced their ZK EVM and then Scroll announced their testnet and then ZK Sync is like, "No, we were first." So there's like a lot of hype around a ZK EVM. I'm wondering, where does StarkNet fit into this whole story?

• 5:15Speaker 03

  We, okay, we're proud of the actual in-production achievements that we made. I actually do not want to go into this fight of who did something first. We have been operating in production for now more than two years using what some people call ZK technology. We prefer to use the validity technology because it actually is not ZK. So, you know, our numbers and performance speak for themselves, so we prefer not to enter this fight, but or pick who is the real first.

• 5:51Speaker 03

  We've been at it for a pretty long time and I think the world, especially developers, knows this.

• 5:56Speaker 02

  So I do want to add that on the ZK EVM front, I think that the excitement is terrific. I think we're all sort of waiting for the real production systems to be out there with proofs turned on, et cetera. We have yet to see that and that's a big achievement we're looking forward to. I would add that in the context of StarkNet, one of the things that a bunch of people have expressed interest in and I think we're likely to see is people deploying an emulation of the EVM as a layer three settling on StarkNet, which would be sort of Asher-esque in nature, right? StarkNet settling over Ethereum with its EVM and then the EVM being simulated as verifiable computation.

• 6:38Speaker 03

  Layer three on to a Starknet. To what extent will that draw the attention and excitement of developers? It's a very interesting, open question in my mind.

• 6:48Speaker 01

  Okay, so is that part where you can... Is that where we get the EVM expressivity built on Starknet? Is like, it would be a layer three, not a layer two?

• 6:59Speaker 03

  Yes.

• 7:00Speaker 01

  Okay. Okay, so the technical details are something that challenges me in my understanding. So Starknet, not a ZK EVM, but you could do a ZK EVM as a layer three on Starknet? Is that correct?

• 7:15Speaker 03

  I want to clarify something that...

• 7:17Speaker 01

  Sure.

• 7:17Speaker 03

  That getting, emulating the Ethereum virtual machine, which is what is often referred to as ZK EVM, is not Starknet's end goal.

• 7:31Speaker 01

  Right.

• 7:31Speaker 03

  Our end goal is to provide the best scaling solution for Ethereum. And we strongly believe, based on our deep understanding of both EVM and how to build Stark systems, that you are far better off in terms of scaling if you use a different virtual machine. And ours is Kyro. So I think what's going to happen is that even though you can emulate EVM inside Kyro, and this will happen, the real scale will be derived from native Starknet Kyro-written smart contracts.

• 8:08Speaker 01

  Okay, so this... I start to understand this a little bit. A lot of developers talk with very frustrated tones about the EVM. It limits them, there's... It doesn't... It's not parallel. I know some of these words. I just generally know that developers are... Feel very limited by the capacity of the EVM. So if I'm tracking correctly... Cairo is something far more optimal and also more expressive and allows something like the EVM, which people are already familiar with, but also much more than the EVM, to also achieve scalability while also having the Ethereum security.

• 8:47Speaker 03

  I actually beg to differ in the following sense. First of all, I think EVM, Ethereum and Solidity are amazing. They are what delivered the gold standard for a general computation blockchain. And Solidity was and the EVM were built under this set of constraints that says we need a Turing complete programming language that will be best fit for blockchains. And it's an amazing achievement of Vitalik and, you know, anyone who worked on it in the early days and advanced it since then, Gavin Wood and others, anyone who was involved in this. This is amazing. Now, for better or worse, when you build scalability through validity rollups and in particular STARKs, you just have a very different set of constraints.

• 9:32Speaker 03

  And you need, for scaling, you need a different virtual machine model. So Cairo and Starknet, they are not better than Ethereum. They are built to address a different set of constraints to allow Ethereum to scale.

• 9:46Speaker 02

  I want to add to that, that the question of, say, assuming the existence of highly performance ZK EVMs, okay, say five years out, the question of whether, at that point in time, if you're a new developer building some massive hit, whether specifying, you know, coding that for the ZK EVM makes sense or not is an interesting question, meaning that code as such, almost by definition, will not run on a Layer 1. It will be too computationally intense for a Layer 1. And so at that point, will that be the optimal tool chain for whatever scaling solution, you know, is developer's first choice? I don't know. That's an open question.

• 10:28Speaker 01

  And certainly Stark X, the first implementation of Starkware, has generated some absolute hits. Immutable has blossomed into an ecosystem and just seems ready to roll into the world of traditional gamers with traditional gaming companies. And so their BD skills are absolutely insane. We had DYDX, one of the biggest volume exchanges in all of crypto that's also decentralized because of Stark X. I'd like to take a peek into the world of BD for Starkware. Like, how do you guys get this done? And why do you guys have so many killer hits?

• 11:05Speaker 02

  Well, we actually don't think of ourselves in those terms. Meaning, we would really like to be the technology developers that we are by nature. And so focus on the research and then the engineering effort and the product effort. But at the end of the day, hopefully leave it to the people building on top of us to blossom and profit and, you know, sort of build all sorts of innovative stuff and bring it out there. And as you mentioned, Immutable is a wonderful example. DYDX is, of course, they've really built a powerhouse with the help of Stark X. And Sareer is a very interesting sort of real-world application

• 11:50Speaker 02

  that many people have told us feels like Web 2. I know of no bigger compliment than someone saying to us that felt like Web 2.

• 11:57Speaker 03

  Certainly. I want to add that even though, I mean, I thank you for the compliment of like doing amazing on BD and having so many successful companies deploying on it. And of course, we're very proud of that. Our feeling is that even just the StarkX systems have so much untapped capacity that we would love, we would love far more, you know, many more companies and entities to deploy on them. And I don't know, if you're out there and you need scale and you like what DYDX or Sareer or Immutable have been doing, then definitely talk to us. There's a lot more room on StarkX.

• 12:37Speaker 02

  I mean, I think StarkX is wonderfully useful for the, you know, the use cases that we've envisioned. You know, for example, minting and trading of NFTs and so on.

• 12:48Speaker 02

  The mind blowing stuff is, once again, we see that day in, day out is the stuff that we never thought of and no one else thought possible that developers are building on StarkNet.

• 13:00Speaker 01

  And certainly StarkNet and that generalizability and composability is where that serendipity happens between DeFi apps, right? We all love the word composability in this industry. But I'm getting a hint that you guys think that the StarkX side of things, the more like app-specific chains that are built for one company each one at one time, also still have a lot of potential left in them that is left. Can you hint, can you talk about that a little bit? What's left on the table?

• 13:28Speaker 02

  Well, I can't, I can't announce, we won't go into the announcement of an announcement of an announcement thing, but there are a whole bunch of things in the pipeline on the StarkX front that's going to come online in the coming period, which is very, very interesting. But once again, limited to the use cases supported by StarkX. Say a bit about L3. So, you know, StarkX is going to move up to Layer 3. So what is Layer 3?

• 13:54Speaker 01

  Ah, right.

• 13:55Speaker 02

  Which is another Starkware innovation. Which is, which, yes, is another Starkware innovation. Recursive Starks are coming to mainnet to support our production systems in a short number of days. But, so what is Layer 3? So instead of having the StarkX systems that today are proven off-chain and then verified by a smart contract on Ethereum, instead of having that verification done on Ethereum, it could be done on the public StarkNet. And by merit of doing it on the public StarkNet, you could sort of conceive of these applications as having moved one layer up. And that's why we refer to this as Layer 3. So I think the first things we're going to see at Layer 3 are the StarkX instances.

• 14:40Speaker 02

  And for those applications, the benefit is trivial. For one, it's just lower gas spending, so better economics for them and their users. I think we're going to see a lot of excitement on Layer 3 with custom StarkNets. And I think that's where a lot of sort of the app-specific action is going to take place. I think it's going to strike a very interesting balance between offering sort of So the best of both worlds, meaning the best of decentralization and censorship resistance and general computation and composability, yet at the same time offering those applications, the certain degrees of control that they deeply care for, which are essentially taken away from them on a public chain. It doesn't matter whether that

• 15:25Speaker 02

  public chain is Ethereum or the public StarkNet.

• 15:27Speaker 01

  Right. There's been growing interest in the Cosmos ecosystem lately.

• 15:31Speaker 02

  For sure.

• 15:32Speaker 01

  And this starts to look a little bit like Cosmos, right?

• 15:35Speaker 02

  For sure, for sure. This was, yes, yes. This is a vision that Zaki described to us three years ago.

• 15:42Speaker 01

  And so the strategy is you have the generalized settlement layer, which most people assume I'm talking about Ethereum, but you guys are actually talking about StarkNet, the generalized, the D K rollup that is the foundation for the layer threes, the more proprietary StarkXs, right?

• 16:01Speaker 02

  Well, it's a two-layer sort of foundation, right? There's Ethereum providing its security and support at the bottom. That's the massive, those are the shoulders of the giants we're standing on. And on top of that, we have the public StarkNet, on top of which the layer threes are going to be instantiated, yeah.

• 16:20Speaker 01

  And there's a growing conversation in the Ethereum community is like, okay, well, all these StarkX or just like general layer threes settling on the StarkNet layer two makes StarkNet its own ecosystem, its own vibrant ecosystem. And also, you guys released the token details, which we'll talk about in a second. But the token had native fees payment in its own token, as opposed to something like Arbitrum or Optimism, which take fees in Ether. And so there's a growing conversation in the Ethereum space about, is the StarkNet layer two interested in actually becoming its own kind of sovereign chain that disconnects

• 17:02Speaker 01

  from the Ethereum layer one? The technical details, a little escape me as to how this would work, but is that a future world that we might see where?

• 17:10Speaker 03

  Absolutely not. We do not foresee this. We're not, we don't have any plans for such a thing. We are relying in so many aspects on Ethereum, and that's our strategic decision from day one and...

• 17:27Speaker 02

  It's a four-year-old strategic decision. Yeah, yeah.

• 17:30Speaker 01

  Okay, so, okay, so there's no plans for the Starknet to become its own Layer 1? What about settling upon a different Layer 1, like in addition to Ethereum?

• 17:38Speaker 02

  We've been approached many times by anyone you can think of. Okay, our answer is simple and straightforward. You know, we're a Layer 2 on Ethereum. That's what we've been building for years now, and we're very proud of that.

• 17:57Speaker 02

  You know, so this is where we are, you know, this is where we're at. This is where we exist.

• 18:02Speaker 01

  And what are the properties of Ethereum that really make that true? Like, what about Ethereum keeps you so aligned with it?

• 18:10Speaker 02

  Well, I, two things. Fundamentally, culturally, okay, there's this developer focus and mindset. And that's a sentiment we share. And that's something that sort of runs in our veins. And we're very happy to see in that regard the fast-growing ecosystem around Starknet, which is really sort of developer-focused and developers-first. That's really top of mind for us. At a more technical, or call it technical economic level, the security offered by Ethereum is unmatched, right? There's just, cannot be compared with anything else other than, I don't know, Bitcoin.

• 18:50Speaker 03

  I want to add to that, that everything about our team and our technology and the way we work is about future-proofing. So the technology is the most future-proof of all the, you know, options of proof systems out there. And the most future-proof Layer 1 we see today is Ethereum. So it's a very obvious choice.

• 19:14Speaker 01

  Yeah, certainly, the idea of assurances is very, very important in crypto. Settlement assurances, but also just assurances that the ecosystem that you build upon sticks around for the long term.

• 19:26Speaker 02

  For sure.

• 19:27Speaker 01

  And so, especially in these very nascent years of crypto, I mean, we're 12 years into it, but, you know, we know that this is a century-long endeavor. Having long-term assurances is super, super important.

• 19:37Speaker 02

  Of course.

• 19:38Speaker 01

  Okay, so let's get into the... Details of the token because the token was recently announced and I have to I have to start with this. Suzy accidentally leaked, well not accidentally, he leaked the details of the token and then the token blogs, which were already, it seemed like they were already ready to go. Were released 24 hours later. Was Suzy anything to do with the timing of when those articles went out?

• 20:00Speaker 02

  So yeah, unfortunately, it had nothing to do with it, meaning that was actually the game plan. And as you indicate from the three-part series, this was a well-thought-out.

• 20:12Speaker 01

  Right.

• 20:14Speaker 02

  Sort of.

• 20:15Speaker 03

  Yeah, maybe no one will believe us, but the target date many weeks before was July 13, Wednesday.

• 20:21Speaker 02

  And we're working towards it. I'd like to hire anyone who can produce that kind of content in 24 hours. That sort of clarity of thought and expression.

• 20:30Speaker 03

  Yeah, and get everyone to sign off on it, you know, inside the team.

• 20:34Speaker 01

  Did you guys feel like rugged by Suzy for?

• 20:37Speaker 02

  No, no, no. I think, you know, I think they're in a situation of extreme distress.

• 20:44Speaker 01

  Sure.

• 20:45Speaker 02

  Right, in finance class, you know, that you're taught about the cost of distress. We're seeing this sort of extreme behavior. Unfortunate.

• 20:56Speaker 01

  There's a part of the Ethereum community is actually very happy about the things that have gone down with Ethereum's capital because they think that the decentralization of SarkNet is super important. And so they're actually kind of happy to see the tokens that Three Hours Capital owns to be distributed amongst almost anyone else other than Three Hours Capital. Is this a sentiment that you guys share or any thoughts on this?

• 21:20Speaker 02

  I gotta say, the schadenfreude is not something that I sort of, I can't sort of.

• 21:28Speaker 01

  Sure.

• 21:28Speaker 03

  But on a more like formal and technical level, the whole, you know, legal process that is unfolding with Three Hours is not something that we are, you know, in control of or aware of all its details.

• 21:47Speaker 02

  Or a party to.

• 21:48Speaker 03

  Yeah, yeah, so like we are.

• 21:52Speaker 01

  So a- about the the actual token launch, so let's run through the next weeks, months of Starknet and the token and the roadmap for that. Uh, can you guys, for people that haven't read the blog post, or perhaps they only consume data through uh video and audio, uh, what what's the roadmap, what's the launch process for this token?

• 22:12Speaker 03

  Okay, so 10 billion tokens have been allocated, minted, and their allocation has commenced. Um, allocation has been, uh, has begun to investors, to core contributors, which include Starkware employees, consultants, and developer partners. Um, in parallel to that, a foundation is being set up that, once it is set up, I mean, it is being allocated slightly over 50% of the tokens.

• 22:49Speaker 03

  And those tokens are gonna be used both for user, um, user-driven, um, rebates and provisions uh, to the, um, to around 18% in total of all of the 10 billion tokens. In addition, there are grants and the foundation is gonna have a strategic reserve and other things that you also see in other, um, Layer 2s and in other, um, chains. Now, this is being done right now, so the foundation is being set up now. Um, by September, uh, the tokens which are now off-chain are gonna be, uh, placed on-chain in their ERC 20 token.

• 23:32Speaker 03

  All of the tokens that thus far have been allocated to investors, core contributors, are gonna have a four-year schedule of release, uh, one-year cliff, and then gradual release beyond that. Um, looking a little bit later on, um, sometime, shortly after they are placed on-chain, they will start, uh, they will be requested for, um, governance and voting on upgrades to the network operating system. Um, and then, next year, at some point, we will transition to allowing fees and then demanding fees be paid only in the Starknet token, and...

• 24:15Speaker 03

  Further down the line, within 2023, these tokens will be used for staking for the purpose of participating in consensus and leader election for operating in a decentralized way the system.

• 24:29Speaker 01

  And so the economic model that I've come to like understand generally about Layer 2 tokens, like optimism, Arbitrum, they don't have their token yet, but I'm assuming it's going to follow this path too, is that you stake the token to validate the chain. Same thing with Ether on the Layer 1 for Ethereum. But the difference between the StarkNet token and the other concepts or strategies of the Layer 2s is that Layer 2s, optimism, Arbitrum, at least the generalized optimistic rollups, pay for the transaction fees in Ether. But the difference here with StarkNet is that you pay for the transaction fees in the StarkNet token. What's the thought or strategy behind this? Why this decision?

• 25:12Speaker 02

  Well, our focus, as I mentioned earlier, our focus is on developers and developers first. And we believe that this mechanism that we've put in place that would allow people to pay via Paymasters to pay ETH for transaction fees, but the native token of this layer will be the StarkNet token. We think that this creates an economic ecosystem that would give developers the environment that they need to build and thrive in. We think that long term, this is the healthier ecosystem that needs to be built.

• 25:48Speaker 03

  I want to just add a few more details. So it's really important for us that developers further down the line experience inclusivity, which means, you know, suppose there's probably someone, she's now in college or, you know, maybe high school, and three years down the line, she's going to write some amazing smart contract that's going to be really important and valuable. And now, how are we going to find a way for her to have a say in the governance, for her to participate in operating the network? Well, the mechanism we want to come up with, and we're very proud of trying to innovate on that as well, is one in which a portion of the fees And these are governance tokens, right? And also used for staking for operating the system,

• 26:32Speaker 03

  are going to flow automatically to smart contracts as a portion of the fees being paid. So now, this imaginary smart contract developer, she will start receiving tokens that allow her to participate in the governance and in operating the system, which can only be done if you have a native token and demand the fees to be paid in it. That's why we're doing it.

• 26:53Speaker 01

  Certainly, certainly. And we're, of course, like I started this interview, in this bear market, which is also a build market. And we need to build our way out of it. And so just lay down the vision for how you hope that StarkNet progresses. Where are we in six months? Where are we in six years?

• 27:09Speaker 02

  Where are we in six months? I think where we were six months ago, where are we in six years? Same answer. Writing code, deploying systems, educating developers, onboarding more people into the blockchain ecosystem, into the broader Ethereum ecosystem. That's where we want to be.

• 27:27Speaker 03

  In six years, maybe also hardware.

• 27:30Speaker 02

  May well be.

• 27:30Speaker 03

  Not just software.

• 27:32Speaker 01

  All right, well, I hope you guys have a fantastic rest of your ECC. And enjoy Play. So thank you for coming on and joining us.

• 27:40Speaker 02

  And don't forget the sunscreen. I think it's appropriate.

• 27:42Speaker 01

  Certainly. Cheers, guys.

• 27:43Speaker 02

  Okay, thank you.

• 27:44Speaker 03

  Cheers.