Layer 2s: Scaling the Business

Transcript

of course [Music] hello everyone thank you very much for coming I just got to Paris about an hour ago so really excited to see that the weather's a little better than last year and that everybody's so excited for another HTC I'm AJ Warner Chief strategy officer at off chain Labs we only have about 15 minutes and I love to keep a lot of time for question and answers so I'm probably going to move through this pretty quickly because I'd love to talk to you folks a lot more there's a lot of components about scaling businesses with layer 2 roll-up technology the business on top of it I'll tell you the things I'm not going to talk about today a lot of my

colleagues are talking about them on other panels and you know they've been you know widely discussed I'm going to not talk about Mev around sequencer revenues as well as shared sequencing mechanisms those are just my colleagues have been more focused on the technical components have been talking about that the areas I'm going to talk about is sort of building out what it means to have role of ecosystems how they can steal how they can grow and specifically my experiences working with arbitrim in particular um one caveat I want to flag before obviously you know in March we had a transition where off chain Labs was the you know the the stewards of the arbitration protocol and it's been transitioned to control by the arbor stream Dao so I work for off-chain labs which is a service provider so some of that context is is important as to

the forward-looking components of of everything that we're doing very briefly before we hop into the core part what is a roll-up what is a scaling ethereum the key part is is you're trying to separate execution and um on data availability so when you you do a transaction arbitrim it sits on top of ethereum and all transactions are batched together and posted to ethereum this allows the transaction fees to come down somewhere between 25 and 100x depending on what you're doing obviously you have a proving mechanism that remains supported in ethereum so you can have the full experience without compromising on the security levels that ethereum gains once obviously the the Roll Up is fully built out

the differences between a roll-up and a side chain really come down to forking components which is data availability an arbitrary one which is a roller protocol that lives on ethereum L1 L2 bridging when you have a bridge it doesn't a bridge by a multi-sig control over the bridge it's a self-enforcing bridge which you enforce from ethereum itself we utilize fraud proofs for security mechanisms and the honesty assumptions are you only need one honest validator which is a huge distinction from a sidechain protocol which typically has a two-third honesty majority assumption and that's one of the key drivers once right now arbitrim1 the Dow has a white list on on who can be validators but assuming that comes off soon as long as you are running a validator itself you have to trust no one other than yourself to submit a challenge to the ethereum

protocol so arbitrim1 these are the current state of the network today is about six billion dollars of value bridged from ethereum to arbitrim there's a lot of different Roll-Ups that are keep popping up into the ecosystem arbitrim1 currently has over 60 percent of the market share according to L2 beat thousands of applications and almost 10 million different addresses using the network so when you think about going back you know I've been working on this for about two and a half years the flywheel mechanisms that we always believed in was you need to have from an onboarding experience you know traditional layer one applications or other ecosystem applications moving over working with infrastructure and

ultimately the core vision is to have major applications come out of the network itself so you can see in terms of what we call the Blue Chips that everybody knows from the you know crypto ecosystems uniswap Sushi open cave Etc that joined the network together with a lot of infrastructure chain link and fewer Alchemy ether scan really everything that you need in order to have a traditional evm experience but what I'm most proud about is the success of the Native arbitrim networks applications GMX treasure Tower leaders in their verticals and perpetuals and you know gaming ecosystems combined and it's really amazing to see how using layer 2 technology additional computation and throughput has allowed a lot of different verticals particular derivatives in D5 to take off on

arbitrim if you look here just one more chart specifically about liquidity the arbitrary one liquidity it's actually this is about a month old so it's a little bit off arbitrum one has about 1.15 million eth which is almost two and a half billion dollars at today's price uh bridged in and this is a comparison with all other major networks that are focusing on scaling the evm combined and we are very proud of the adoption that arbitrim has seen to date this is an interesting chart to show sort of how we've focused on different applications that we've targeted and I want to I put the charts side by side intentionally because as you can see on the chart on the left if you look at the app the amount of transactions um over

the last let's say three to four months ethereum and Arboretum are actually quite similar um but if you look to the right and you look at the daily gas that's used arbitrum is significantly higher the reason why ethereum's line is flat is because basically it is always a congestion right you know the the fee mechanism the fee mechanism kicks in but arbitrum actually uses despite having the same number of transactions it uses much more computationally intensive transactions and I think that's one of the key things that people don't focus on when they think about the power of Roll-Ups and scaling Technologies is we're not only having more throughput living on ethereum the types of transactions that are taking a lot of gas can actually really only exist on layer 2 Technologies a lot of that is accounted for in derivatives

because they take a lot of gas you know a typical GMX transaction would probably cost 100 to 150 times more if it was done on ethereum mainnet as you can see in terms of volume there's a huge growth in Layer Two technology volume over ethereum this is from uniswap in terms of the amount of volume that layer two they're taking in the Dex markets uh this is an example so you know when we when we started working on the arbitrary Network we were really focused on scaling the web 3 native environment and ecosystem but as the technology has continued to mature we've worked with other institutions for example the Norwegian government is currently testing two products at Arboretum test Nets they're building something called

Brock which is focused on um registration of private companies to make it much more accessible to get this data and are also testing how do they Bridge their cbdc token into a public network from a private Network they're currently working on for deployment purposes um so while arbitrim1 is really great and has been focused on building out the arbitrum ecosystem mainly around defy there are certain things in certain use cases that are still too expensive to post all of their call data to ethereum so that's what arbitrim Nova which is an NHS technology is intended to solve this is really a technology for very high throughput dapps with a consumer Focus or you know gaming basically cheap transactions being a necessity in order for the use case to survive

how do we do this so the reason why Nova Works differently than ethereum than Arboretum one is it doesn't post all the call data to ethereum it posts the call data to a data availability committee and an arbitrary Nova these are the members of the data availability committee and the principles are you need two members of data availability committees to remain honest and anyone has the ability to force the protocol into a roll-up mode so how does it work when you submit a transaction on on arbitrib Nova the transaction goes to these members of the committee they submit certificates to ethereum that they've had the transactions and they're posting the certificates there and they may they ensure and are required to keep that data available again any single member of this committee any valuator

can force the protocol into roll-up mode if the committee starts to act maliciously but the introduction of a data availability committee brings the cost down by another factor of 10 so a typical transaction Arbor treminova is subpenny and this has been really important for gaming and for social apps so I'll give an example that Beacon they've done a lot of work on arbitrary and they've really been one of the examples of the growing gaming ecosystem that lives on arbitrum today they they sold out millions of dollars worth of their initial nfts and in the alpha stage and the game is continuing to to grow we have a number of upcoming titles some very famous ones pirate nation which actually recently just launched forgotten ruins is launching the Goblin

Town z-verse ztx is actually a Korean social uh metaverse that has launched uh that announced last week that they're launching an arbitrine which we're extremely excited about a lot of this is taking advantage of the arbitrary Nova infrastructure stack because the costs are about 10 times cheaper than arbitrum one and probably between 100 and 150 times cheaper than than ethereum mainnet itself knowledge that Reddit chose to use for the scaling their Community points they've started with two subreddits that are currently on chain one for the cryptocurrency subreddit and the fortnite subreddits and then when they initially um introduced this they had 250 000 wallets that they submitted tokens to and it cost about 1600

transactions to fund all 250 000 wallets so really a sixth of a penny per wallet we've spoken a lot about scaling the arbitrary networks and scaling the developer experience right now I want to focus on how we on the on the user experience I want to focus on how we're thinking about scaling at off-chain Labs the developer experience as well and this is the introduction of stylus with stylus we are going to add an additional virtual machine into the arbitrim tech stack which will allow you to program in Rust C and C plus and have those transactions live side by side with the evm so you if you want to continue coding the evm solidity you don't see any changes to your networks but you can interact and having pre-compiles that are written in other languages with significant cost savings as well we

think this is really important for onboarding a lot more devs because right now you have a TR you have to trade off between using an ecosystem like Solana which is only rust based or an ecosystem like arbitrim or ethereum which is very focused on the evm this will allow you to have both synchronously and composibly one other area of scale that we're starting to see is application chains in the RF chain labs in the Arboretum ecosystem we have something called arbitrum orbit which is a set of layer 3 technologies that live on top of arbitrim1 and arbitrim Nova so why would you want your own app chain one of the advantages is you can have more flexibility so for example if you want to introduce stylus stylus on the public change is going to have to get approval from the arbitrim Dao if you want to build your own train take

advantage of arbitrum stylus you want to sort of do whatever you want to that code base you have the ability to do that you also have increased gas reliability even though arbitrim scales ethereum's throughput by about 7x on each chain you might not want to have unpredictability of public networks and congestion so you can have much more control about whether that's the case account abstraction the arbitrum Dow actually just recently approved adding account of traction into the arbitrum sequencer but things like that are possible much more flexible to do on your own chain as well as customizable protocol logic you can do whatever you want to change governance settlement execution introduction of alternative data availability committees all of these things you can do on top of the arbitrary protocol

two others decentralization options you can choose to use arbitrine roll-up Arbitrage of any trust you can do what you want to introduce either right from a regulatory and compliance perspective and it's super easy to spin up and um your own we actually have a one-click deployment tool obviously it doesn't cover the devops side you have to make sure that that's an ongoing component but um we can make it super easy to spin one up security you can use any of these chains right out of the box and if you want to have permissioned access we're seeing this in fintech a lot where they need to have certain requirements for rules and regulations for who can interact with the specific chain you can introduce all of those things into the orbit stack there are some frictions though and I don't want to pretend these don't exist

one is cross-train friction there's a lot less liquidity obviously on your app chain than there would be on the public networks working with liquidity providers to figure out how to bridge that Gap is going to be critical as well as infrastructure complexity the arbitrum foundation the arbitrum ecosystem has block explorers oracles indexers everything that you really need in the EDM experience to really survive and thrive you're going to have to make sure that all of those are available there's a bunch of roll up as a service providers that are doing a really good job with this but I do want to highlight these are some of the complexities of running your own application chain that we're seeing today amongst developers that are looking at it well thank you very much for you know for coming and listening to me and I'd love to if we

have time answer any questions yeah can we have one yeah okay this one will do okay um any questions yep all right hi um just a quick question I was wondering what was the potential future of arbitrum uh one with abichon Nova at the end of the day are you expecting I'm become one to disappear uh about everybody to move to Amazon Nova what

are the use cases that and if it's not the case what are the use cases that will be addressed by ibtrom1 versus uh Nova that's a great question so as blockchains enter into a world of much more modularity one of the key focuses that we believe is that you should make sure that the technology stack that you choose matches the problems that you're trying to solve there are real trade-offs between arbitrary one and arbitrary Nova the introduction of the data availability committee you know well minimized is still an additional security assumption compared to using ethereum for data availability what we've been recommending at off-chain Labs is teams that are very focused on defy the arbitrary one network is still cheap enough for your transactions to be economically viable for your users to be economically

viable for example you know if you're looking at doing a DEX trade transaction fee is one component slippage is another component Etc like seven cents a transaction I think is the typical transaction cost right now on arbitrary one usually is okay considering the size of the trades obviously we want to always continue to bring that down Stylus is going to play a key focus on that eip4844 which I didn't really cover here is also going to be significantly decreasing the cost of using arbitrum one but you know obviously this is up to the Dow again an example where arbitrim you know Dow really has full control I don't anticipate them ever getting rid of the roll-up mechanism for arbitrim1 because that's one of the core security properties in the value propositions especially amongst default users

hi um the title of your talk was about sort of like Partnerships and scaling the business what would you say are you know either like the timeline or the key Partnerships or differentiators you know however you want to answer that made you know has made thus far and will make arbitrim more successful than say like a random other uh LTL yeah so I think the key and I don't I can't speak to the Future because you know crypto's crazy I'll talk to the past so I think there were a couple of things when we launched Arbor trim that were unique um one was the so first of all the evm experience making it extremely easy to onboarding at the time now it seems like a given at the time was really the only roll-up

that had that um so that's you know very looking to the past I think focusing on liquidity especially for D5 verticals and protocols is going to be critical because it creates a massive flywheel effect right you know we see a lot of protocols launching on arbitrim as a function of the liquidity that currently exists there so our team currently is always looking at different ways to ensure that you know the interoperability of liquidity the seamless of onboarding liquidity is critical being there I guess the another thing that I would say that sort of we took a different approach than a lot of our peers and competitors is I would say we probably were over indexed on being very web 3 and crypto native um when you look at 2021 2022 there was a lot of focus on sort of you know

growing into sort of traditional fintech traditional businesses loyalty programs we thought it was very early I would say we're probably looking a lot more now preparing for 2024 2025 to see a lot more adoption there but I think understanding who the core audience the core user base still is and designing a product that speaks to those user pieces are key and that will continue to evolve over time I don't know that fully answers your question but I think that's probably where some of the driving distinguishing factors in terms of how we launched the network was probably you know insatiable focus on attracting liquidity sustainably and sticky and making sure that the builders that we were targeting were very committed to sort of the vision that we

had um because a lot of people come and go hey um actually had somebody reach out earlier today from one of the protocols that was up on your slide and they have a proposal up and they were actually asking if people knew the delegates um so I was just wondering if you had any insights into like what the communication channels are there between yeah so um delegates the voting for The

our token yeah definitely so first of all all the delegates are public and the amount of tokens that they have um are all public I think tally.xyz is like the governance interface of the foundation has established um I don't work at the foundation but reaching out to the foundation I think it's probably I think info arbitrim.foundation they can put you in touch with delegates and they're hap they already are setting up um hubs basically for delegates and you know Dow vote proposals uh to be able to interact but it's much more on the foundation side and sort of the stuff that I focus on which is behind you thank you thank you for the presentation

really interesting and congratulations for a great job done so far thank you going back to the business growth side and and business modeling strategies um how how you guys see potential collaborations with application specifically Road Labs protocols right um it's clear that those roll ups also will increase massively the computation space and black space in on top of a base layer chains so do you have any clear I'll say business model or partnership model to offer those those protocols yeah that's a great that's a great question um says a lot to unpack there the the business model for the Dow Pro the

arbitrum Dow separately which is sort of the orbit model is we can put that aside for a second but the The Core Business there is you know if they live as layers threes you know arbitrary one will act as sort of a central Hub like a Grand Central Station type type model with all these interoperabilities between different protocols that live on top of it um in terms of our business model so at off-chain Labs we're working with certain teams that are looking to build their own application chains and Roll-Ups we announced actually that we're helping the sci Foundation which is a gaming metaverse that's building a number of games and we're going to work for them as a service provider to help them with their rollup as well there's also a bunch of roll up as a service providers

conduit Caldera that are working on alt layer as well that are working on spinning up support for these sort of chains and networks I think the business models across all of these different things are are still quite interesting to be completely on honest I think that the conversations around like everybody needing their own Roll-Ups or application specific chains has probably gotten ahead of where we are from a demand perspective you know arbitrary one I think is typically at like 25 percent of capacity Nova's at five percent of capacity I think a lot of our peers um other than maybe the polygon POS side chain really have a lot of capacity I think what's happening now is you know it's the new narrative it's very exciting people are getting into it but my personal opinion is that a lot of you

know when I focus on all the core infrastructure I think focusing on developer applications building a user base that is very loyal passionate to the product is probably the core Focus that most of these you know application focused teams should be doing and then as their businesses scale they should be thinking about you know how do we launch Roll-Ups how do we work together who you know who could we work with and you know off-chain Labs is here and the Arboretum Foundation is obviously quite a nice framework for for how to build on top of the Dow and work with service providers there I hope that answers the question any other question so I wanted to go back to uh liquidity

you know obviously there's like GMX like I don't know is that like a is there a particular thing that you think was like cemented you guys um and I'm just curious like the story there when were you like okay amazing like we got it we got liquidity this is good I don't think that ever happens first of all I could you know as many people know in crypto there's there's no winners and losers there's only temporary winners and loses right everybody's it's a hyper efficient market where liquidity moves quickly um I think that there's a couple of reasons why Arbor trim as a protocol has continued to hold its market share despite uh competitive headwinds uh one is technical maturity you know if you

look at any of the technical Frameworks you know on l2b there are other you know sort of bipartisan or non-partisan websites you know the arbitrum tech stack is I think by far the most mature of any of the Roll-Ups live today you know the the Dow has full control of of all contract governance there's two points I would say of centralization still one is the whitelist the ballader set um they can burn that if they wanted to and open permission so they're waiting for a technical upgrade around a delayed attack mechanism the second is uh the existence of a multi-sig which is you know a collection of public community members again something that the Dow can control or burn and is up for re-election at all times there's a lot of technical maturity I think for significant liquidity providers you know the average person doesn't know about

this but I think people who are you know bringing a lot of liquidity they do care a lot of the applications I think on arbitrim are very focused and committed to the network and they have sub communities right like a lot of what we're building is sub communities that bring liquidity and I think these things have Network effects and flywheels I think the way we approached working with Builders and developers has probably what started that I don't think it's cemented in any way to use your words but I think it's the theme and the culture of the chain is very developer focused which you know I think attracts a lot of sub communities to make it their primary home but also all the on-ramps right that's really important

we spent a lot of time probably not some much now because it's a lot of that infrastructure is there but if you go back to the work that we did at off chain labs in 2021 was you know binance coinbase all the major centralized exchanges you can go directly into the arbitrary Network you don't have to touch ethereum so that user on onboarding experience is better the Fiat on Rams Etc the third party Bridges so I think it's you know a lot of different things went into it but um I think that you know to think that it's cemented is is a Fool's Aaron but um you know I think I think the maturity of the protocol probably really helped a lot any other questions no I have one so um one of the centralized Vector points

today's yeah the sequencer so I think arbitrum has uh voiced a desire to decentralize that over time so there are a lot of different approaches to do that for example shared sequencers L1 based sequencing white list yeah some sequencers so do you already have like a road map of like what the approach will be and those kind of things yeah it's a good question so one thing I'll flag and plug is uh Harry our co-founder I put out a thread Harry colander put a thread on Twitter actually yesterday about sequencing um I think some of the centralization components are overstated what I mean by that is the arbitrum sequencer really can't do anything other than if it

wanted to is extract Mev or for a certain time-bounded period the layer transaction it doesn't have the ability to set the price of a transaction to get it in it doesn't have the ability to delay it for longer than like I think 20 minutes on a specific transaction if you wanted to do it for longer would basically have to Halt the entire chain and all this in the arbitrum instance for example is all controlled by the doubt so if the Dow ever or the security Council ever wanted to vote to remove the arbitrary sequencer they can do it immediately put in a new sequencer you know Etc that wouldn't be malicious and wouldn't be censoring so there are certain limitations and I think the core focus and reason for decentralization are actually twofold one is redundancies to make sure you know if there's a devops failure

that the chain doesn't go down and the second is you know ensuring that you don't have to trust anyone they're not extracting Mev but that's actually sort of the reality of the of the network where we've been focusing now a lot is and a lot of my colleagues are actually talking about it at the conference where we've been focusing is on decentralizing sequencer by using a committee of sequencers that you don't have to submit certificates kind of like Nova in some ways where they they submit what their order of transactions is and then the consensus is posted and if the sequencer from the committee that posts to the consensus deviates from the consensus of the committee they get slashed that's sort of how we're thinking about it now you know there's other you know trade-offs that people are focusing on

around auction mechanisms we think that auction mechanisms are not ideal and I'm not the most qualified on the team to talk about this to be completely candid but the issues auction mechanisms as you're essentially incentivizing people to maximize Mev so you should really be thinking about like Mev minimalization strategies in conjunction with that and we've been doing a lot of research one of my colleagues cocky put out a paper called time boost together with Ed Felton on our team as well focusing basically on how do you have you know Mev that cannot see into a private mempool but has the ability to submit transactions within let's say a 500 millisecond time frame of an advantage that they submit you know a tip essentially so we've been doing a lot of

research around that but um much more focused on decentralizing and distributing the sequencer than than I would say on you know the Mev components and on shared sequencing we're looking at it there's a lot of trust assumptions that are introduced when you have shared sequencing so that's a much earlier stages of research for how it's not so complicated from what I understand doing it it's complicated doing it without sort of sacrificing a lot of the trust assumptions that you know a roll-up as opposed to institutionalize interesting thank you and ask questions or no okay then thank you for the talking thank you everyone for coming [Applause]

[Music]