Exclusive Interview with Mysten Labs Co-founder: L2 Has No Value

Interviewee: Adeniyi Abiodun, Co-founder and CPO of Mysten Labs

Interview and writing: Wendy, Foresight News

"L2 has no value" ------ While acknowledging the significance of Movement Labs in promoting the adoption of the Move language, Adeniyi Abiodun, co-founder and Chief Product Officer (CPO) of Mysten Labs, was outspoken when discussing the choice between L1 and L2 paths. He told Foresight News: "I don't think L2 solves the problem at all. In fact, they create more problems than they solve."

The choice between L1 and L2 has been a long-standing topic in the public blockchain space. As one of the representative projects of high-performance public chains, Sui, developed by Mysten Labs, chose to build an L1 using the Move language, while the relatively younger Movement Labs opted for a completely different path by building an L2 using the Move language on Ethereum.

This means that the Move language is gaining broader attention, which is beneficial for the overall ecosystem development. Meanwhile, the overall development of high-performance public chains still faces challenges. In this regard, Sui needs to enhance the popularity of Move as a foundational language and also requires a surge of high-quality projects.

In the competitive and incentivized high-performance public chain space, attracting quality developers and projects is no easy task. Recently, Adeniyi Abiodun accepted an exclusive interview with Foresight News regarding Sui's development over the past year and the competition in the high-performance field.

In the past year, you have held many hackathons globally. What achievements have you made? In which regions is Sui's adoption rate the highest?

Adeniyi Abiodun: We have a high adoption rate in Asia, which is our primary market, with some of the largest projects and developers in Asia using Sui. We also have projects that were previously built on Arbitrum and Solana, but due to many things not being feasible, they transitioned to Sui. Additionally, some people want to develop across multiple channels, while others prefer to focus on a single ecosystem. Notably, our ecosystem has a very low developer churn rate, especially among those who have been building on Sui from the beginning and are now very successful. This is a positive sign; they recognized Sui from the start and are now reaping the rewards, such as with protocols like NAVI, Aftermath, and Scallop.

Now, we also welcome newcomers like Bluefin and Sui lend, who have transitioned from the previous Arbitrum and Solana ecosystems to Sui.

I believe Sui is very successful in Asia, but now North America and Europe are also starting to catch up, which adds more diversity. However, our focus remains on Asia, continuing to increase our influence in that region. Before the mainnet launch, we held hackathons in various places like Vietnam, South Korea, and Japan, and participated in Taipei Blockchain Week. We have done a lot in Taiwan, so our user base there has also seen significant growth. More importantly, we have already seen the success of builders there. If we enter a market, even if there are ten thousand developers, if they are all working on obscure applications, it’s meaningless. What we want to see is developers achieving results, which is why developers are saying, "Look, Bluefin can succeed on Sui, Suilend can succeed on Sui, I want to see if Sui is a protocol worth considering." Interestingly, many EVM projects view Sui as the next chain to build on, which is a strong testament to our success.

In the public chain space, many projects offer substantial bounties and rewards for developers. How do you attract more developers into your ecosystem?

Adeniyi Abiodun: First, we must lower the learning curve for developers. We believe Sui is currently the easiest blockchain to build on. In fact, most developers can fully understand and start building or migrating (to Sui) within four days.

Second, we focus on the quality of projects rather than the quantity of builders. Many on-chain projects are just copied and pasted from another chain without real innovation. Sui provides many primitives that cannot be built elsewhere, allowing builders to fully unleash their creativity. For example, many projects are now building decentralized limit order books directly on Sui, rather than centralized mechanisms. Others are building new bounty systems directly on-chain, which was previously impossible. Some are even building random number generators on-chain, which used to require services from Oracles, but now you don’t need that. In fact, on Sui, we support any cryptographic primitives.

Now, you can log into Sui using your Google account or Facebook account, and Sui can help you build things that were previously unachievable.

Because of this, Sui is the fastest-growing non-EVM compatible chain. We have many builders, all of whom are very high-quality builders, who are able to capture a large market share from other chains. In some Sui protocols, our transaction volume and TVL exceed those of entire other chains.

This is thanks to the success of the builders. So, we haven’t spent much effort convincing them to join. Our hackathons have already attracted over 2000 developers. In the coming months, we will announce the winners of the global hackathon. We expect around 200 builders and over 200 projects to win awards. This shows that we can attract high-quality builders.

We have never tried to buy developers or make them sign exclusivity agreements. That doesn’t work. Ultimately, developers have the freedom to choose where they want to build, and we agree with that. In fact, some competitors have made developers sign exclusivity agreements, promising not to build on other chains, specifically stating not to build on Sui. We have never done such a thing. In fact, after the exclusivity agreements expire, most of these builders have come to Sui.

Therefore, creating a compelling platform that helps builders truly achieve their business goals is the way to win over developers. We have done this, not through shady incentives or means. In the long run, the latter simply doesn’t work.

Within the Sui ecosystem, which areas of projects do you focus on more?

Adeniyi Abiodun: Gaming, finance, and commerce are the three main categories we focus on in Sui. Gaming will be an opportunity that naturally brings millions or even billions of users into Web3.

Before the official launch of Sui, we had already signed contracts with 50 gaming partners, and now we have over 70 gaming partners. These partners include billion-dollar South Korean gaming company NCSoft, game studios with decades of development experience, and startup game studios. We believe gaming is a great way to attract users into Web3. In fact, we believe DeFi will further develop alongside gaming growth, as millions of gamers can freely exchange between native game tokens and stable assets.

This is our goal. If you purchase an NFT or buy assets in a game, you will be using some form of decentralized finance. Therefore, merely targeting the existing Web3 user market is not enough; we must expand that market pie further. In fact, we are already starting to see signs of this. For example, one wallet in our ecosystem, WaveWallet, has attracted over a million users into Web3, while the usage of DeFi has significantly increased, along with transaction volume, and users may not even realize they are using DeFi. Because we focus on bringing a larger market into Web3, we are confident that Sui will have more daily active users than any other chain.

In addition to gaming, finance is also a growing field. Gaming needs finance to facilitate all transactions and exchanges.

Commerce refers to the ability to check inflow and outflow channels on-chain, whether using your existing assets as sales terminals or exchanging points, rewards, memberships, and fan engagement, which were previously impossible. Companies developing on Sui are working to achieve all of this. A large number of gaming companies will bring millions of daily active users to Sui. We also have a plethora of DeFi protocols and financial applications under construction that will serve this user base.

Many companies are also building fan interaction tools on Sui. For example, we are collaborating with the Red Bull Racing Team, which is one of the top Formula 1 teams in the world today. We are building a lot of tools around fan engagement for that brand. We are also working with other sports brands like the NBA. We see this as an opportunity for businesses to interact with their fan base in a more engaging way on Sui compared to existing methods.

So, we will announce a series of significant partnerships with Korean gaming companies at Token 2049. There will also be some exciting news about SuiPlay0 x1, a gaming console we announced a few weeks ago at Basecamp.

The SuiPlay0 x1 gaming console allows both Web2 and Web3 players to enjoy gaming. You can play PC games at high resolution and high frame rates on a portable device, earning rewards in both Web2 and Web3 games. We believe this will revolutionize gaming.

It is not a phone, but a low-barrier gaming console that everyone can buy. We will also announce more updates regarding stablecoins. Overall, we are very excited about the progress we have made in the ecosystem so far.

I just announced the number of developers in the ecosystem, and as mentioned, over 2000 developers have signed up for our hackathon. We believe that a continuous stream of developers in the Sui ecosystem will launch new, exciting, and innovative things. This makes me very optimistic.

From day one, Sui has been listed on every major exchange, including Binance, OKEx, Kucoin, and any exchange you can think of. So, from the very beginning, Sui has received the most exchange support in history.

Previously, Movement Labs, based on the Move language, announced a new round of financing that gained considerable attention. You also chose the Move language but opted to build an L1. Aptos made a similar choice. From today’s perspective, how do you view the different development models of choosing L1 and L2?

Adeniyi Abiodun: Fundamentally, I believe that the more developers using the Move language, the better it is for the ecosystem.

Solidity, as a very difficult language to master, actually hinders ecosystem development. Even as an excellent developer, (when using Solidity) you will still encounter some troubles and pitfalls. We even see that the best developers in EVM are not adapting, and there are continuous hacking incidents. Move eliminates many of the common issues encountered in smart contract security. Comparatively, it is a safer programming approach, and we are very pleased to see Movement Labs promoting the Move language, and Aptos is also helping to promote Move. After some time, Avalanche, Solana, and other ecosystems will adopt Move, enlarging the pie. This way, everyone gets a bigger slice.

There are only 20,000 developers in Web3, while the number of JavaScript developers is nine million. If we can get nine million JavaScript developers to use Move to write code, then Move will become the JavaScript of Web3. If they have an easy-to-master and error-free language, we will have better applications, better users, and an overall safer ecosystem. This means more developers can enter Web3, and the number of bugs and hacking incidents will decrease as a result. Ultimately, we believe that better architecture and infrastructure will allow our (Move) ecosystem to prevail.

I believe we will enter a world of coexistence of multiple chains, but I believe we will be the best. Otherwise, I wouldn’t be here. Because we strongly believe there should be an object-oriented language suitable for most developers. It has been designed as a whole by leveraging the advantages of the object model from the ground up. This means we have unparalleled scalability. It also means that Move on Sui will be more scalable than Move on any other chain.

In summary, Sui is more scalable, safer, and faster, enabling you to build fundamental components that were previously impossible. The creator of the Move language is Sam Blackshear, who is a top engineer at Facebook and our Chief Technology Officer. So, in terms of Move innovation, we will always be ahead of anyone in the ecosystem.

But which path is better, L1 or L2?

Adeniyi Abiodun: I believe L2 doesn’t work. I am convinced of this ------ that building another complex thing on top of a chain instead of building a fast chain is not feasible. Because it cannot scale, that’s why we have L2, and now there will be L3. The fundamental reason is that each layer cannot scale, and they try to compensate for this flaw by building another layer.

Solana does not believe in L2 because they think the solution is to build a very fast L1. We agree with this statement, which is to simply build the fastest L1 possible to achieve scalability.

Sui is precisely the only chain that allows you to scale infinitely. If you need higher performance, you just need to invest more hardware ------ double the hardware, and TPS doubles; triple the hardware, and TPS triples ------ without increasing latency, and with horizontal scalability like the internet. There is no need to build additional complex layers; we believe that is a step backward.

I don’t think L2 can solve the problem at all. In fact, they create more problems than they solve. Since Ethereum will never scale, you cannot escape Ethereum’s L2. But Sui does not need L2 at all because Sui itself is scalable. So, you don’t need to build another layer to achieve higher performance because the base layer can already deliver that. I believe L2 will always have different types in the Ethereum ecosystem. L2 will have its value in the Ethereum ecosystem and low-bandwidth, low-scale chains. But for high-bandwidth, high-scale chains, this is not an issue; L2 has no value.

But Move is open-source, and the more builders, the better.

Can you elaborate on how Sui achieves its self-expansion?

Adeniyi Abiodun: In Sui ------ in fact, in most programming languages, you have a contract, and the bytes in the contract represent who owns what. However, due to the data structure, you cannot know whether your transactions and each other have any relationship. This leads to the necessity of lining everything up, executing one transaction at a time, which becomes a bottleneck. Because you cannot know whether your transaction will trigger other transactions, we can only handle one thing at a time.

And because Sui is an object-oriented system ------ the transaction you have with your phone is different from the transaction he has with his pen ------ I can handle them separately. Therefore, the more CPUs we have, the more tasks we can process in parallel. Sui allows us to know whether one transaction is related to another. Knowing this in advance allows us to process all transactions in parallel, which means we have no throughput limitations.

Other blockchains have the fundamental problem of not being able to determine whether transactions collide, such as Solana, Aptos, and EVM. However, they also cannot scale beyond a single machine. Because we have an object model, we can not only know that two objects are different but also place that object on another machine and process it faster. So, this is a new type of scalability model. In fact, this model is used daily by companies that scale businesses. Our team has a technical background in building infrastructure for Facebook and Google, understanding how to scale infrastructure. So, this approach is unique to Sui. The problem with Solana is that the CPU has a limit. Then you must acquire increasingly larger CPUs. One day you will run out of CPU processing power. For Sui, this is not an issue. We just need to invest more hardware to increase the CPU. Increasing the number of hardware increases the number of TPS. If you double the number of hardware, the number of TPS will also double. There is no upper limit.

In other words, we have already solved the scalability problem. There are also hardware issues, such as how to verify signatures faster.

But signature verification is not our bottleneck. So, this doesn’t help us. Other chains may need it. But in the end, they may still encounter some other bottlenecks.

Speaking of hardware issues, the recent development of generative AI has also posed a lot of new demands on hardware. How do you view the potential impact this may have on Sui?

Adeniyi Abiodun: AI and blockchain will definitely interact. But I think this will be communication between agents. If my agent needs to negotiate with your agent, it will need to pay a fee.

That’s the importance of AI. In fact, today we announced a partnership with Atoma Network, which is building disruption technology directly on Sui. So, we believe AI and blockchain will have a good combination.

I think a lot of the current content about AI and blockchain is just hype. People are buying tokens, but it won’t succeed. What truly interests us is collaborating with partners who solve fundamental problems for consumers and businesses.

So, we are very optimistic about the future.

Recently, some researchers in the Ethereum ecosystem ------ such as Justin Drake ------ have faced controversy for accepting valuable tokens from EigenLayer. Some view this as "corruption" because it may influence the direction of Ethereum's development. Sui, as an L1, also needs to maintain the fairness and health of its ecosystem. What are your thoughts on this?

Adeniyi Abiodun: Sui is a Delegated Proof of Stake (DPoS) network that serves anyone with sufficient stake. In fact, even the rules on the Sui blockchain can be decided through community voting. Since the launch of the autonomous network, we have received several community requests for changes, which have been approved by the community and improved the protocol. So, Sui is completely decentralized.

Over time, as more developers join, Sui will become increasingly decentralized. This way, not only do we have the ability to promote change, but the community also has the ability to promote change. This is very important to us. We encourage a safe ecosystem where people have the freedom to build.

This is not just talk; the submitted proposals confirm this. The proposal to launch DeepBook was completed externally. The proposal to change the designated staking amount for validators was also completed by the community.

The community will gradually propose more changes that alter the fundamental direction of the Sui protocol. That’s the meaning of being a Delegated Proof of Stake, isn’t it? Moreover, Sui has only been established for a year, and in the future, more governance mechanisms will enter the ecosystem, allowing everyone to participate.

But the problem with Delegated Proof of Stake is that you must ensure everyone is interested in participating in voting. This has always been a challenge. We have not yet found the right answers to all these issues. In the future, delegation will ultimately still be a way to solve problems. It is reasonable for me to delegate my stake to someone and have them vote on my behalf. We certainly hope that every user in the world can vote, but not everyone understands the nuances of the technology.