Foresight News: Recently, Movement Labs announced a new round of funding based on the Move language, which garnered significant attention. Similarly, you’ve chosen the Move language, but opted for L1 infrastructure, while Aptos made a similar choice. From your current standpoint, how do you view the different development models of L1 and L2?
Adeniyi Abiodun: Fundamentally, I believe the more developers using the Move language, the better it is for the ecosystem. Solidity, being a very difficult language to master, has hindered ecosystem development. Even talented developers face challenges and vulnerabilities when using Solidity. We’ve seen even the best developers in the EVM ecosystem struggle, with continuous hacking incidents. Move eliminates many of these security issues inherent in smart contracts. It’s a safer programming approach overall, and we’re pleased to see Movement Labs promoting Move, and Aptos supporting Move’s adoption. Over time, ecosystems like Avalanche, Solana, and others will adopt Move, expanding the pie for everyone involved. This means a larger share of the pie for all.
In Web3, there are only 20,000 developers, while JavaScript developers number in the millions. If we can get millions of JavaScript developers to use Move for coding, Move could become the JavaScript of Web3. With an easy-to-learn and error-resistant language, we can create better applications, attract more users, and build a more secure overall ecosystem. This translates into more developers entering Web3, reducing bugs and hacking incidents. Ultimately, we believe superior architecture and infrastructure will make our Move ecosystem stand out.
Foresight News: But which path do you think is better, L1 or L2?
Adeniyi Abiodun: I don’t think L2 is viable. I firmly believe that—it’s not about building a fast chain, but rather about constructing something complex on top of another chain that’s not scalable. Because each layer fails to scale, we have L2 now, and we’ll have L3 in the future. The fundamental issue is that each layer fails to scale, and adding another layer won’t solve this flaw.
Solana doesn’t believe in L2 because they believe the solution lies in building a very fast L1. We agree with this approach—just build the fastest L1 possible for scalability.
Sui happens to be the only chain that allows infinite scalability. If you need higher performance, just invest in more hardware—double the hardware, double the TPS—without increasing latency, and with horizontal scaling capability like the internet. There’s no need to build additional complex layers; we see that as regressive.
I fundamentally don’t believe L2 solves the problem. In fact, they create more problems than they solve. Because Ethereum will never scale, you’re stuck with Ethereum’s L2. But Sui doesn’t need L2 at all because Sui itself is scalable. So, you don’t need to add another layer for higher performance because the base layer can already achieve it. L2 will always have different roles in the Ethereum ecosystem. L2 is valuable in low-bandwidth, small-scale chains. But for high-bandwidth, large-scale chains, it’s not an issue—L2 has no value.
But Move is open source, and the more builders, the better.
Foresight News: Could you elaborate on how Sui achieves its self-expansion?
Adeniyi Abiodun: In Sui—actually, in most programming languages—you have a contract where bytes in the contract indicate ownership. However, due to data structure limitations, you can’t know if your transactions relate to each other. This forces us to queue everything sequentially, bottlenecking transactions. Because Sui is an object-oriented system—where your phone in your hand is a different transaction than his pen in his hand—we can handle them separately. So, the more CPUs we have, the more tasks we can handle in parallel. Sui allows us to understand if one transaction relates to another. Knowing this beforehand, we can process all transactions in parallel, meaning there’s no throughput limit.
Other blockchains have fundamental issues with transaction collision, like Solana, Aptos, and EVM. But they also can’t scale beyond a single machine. Because we have an object model, we not only differentiate between two objects but can also place an object into another machine and process it faster. This is a new scalability model. In fact, this model is used daily by companies expanding their business. With a background in building Facebook and Google infrastructure, our team understands how to scale infrastructure. So, this approach is unique to Sui. Solana’s issue is a CPU limit. You’ll eventually exhaust CPU capacity. For Sui, this isn’t a problem. We just add more hardware to increase CPU. Doubling hardware doubles TPS. There’s no limit.
In essence, we’ve solved scalability issues. Hardware issues, like faster signature verification, are irrelevant to us. Other chains might need it, but we don’t foresee bottlenecks.
Foresight News: Mentioning hardware issues, the development of Generative AI has also posed significant new demands on hardware. How do you foresee this potentially impacting Sui?
Adeniyi Abiodun: AI and blockchain will definitely interact. But I think it’ll be agent-to-agent communication. My agent needing to negotiate with your agent would incur costs.
That’s where AI’s importance lies. In fact, today we announced a partnership with Atoma Network, which builds disruptive technology networks directly on Sui. So, we believe AI and blockchain will integrate well.
I think much of the current hype around AI and blockchain is speculative. People buy tokens, but it won’t succeed. What interests us are partners solving fundamental problems for consumers and businesses.
So, we’re optimistic about the future.
Foresight News: Recently, within the Ethereum ecosystem, researchers like Justin Drake have sparked controversy for accepting significant EigenLayer tokens. Some view this as “corruption” potentially influencing Ethereum’s development direction. As Sui is also an L1, maintaining fairness and health of the ecosystem is crucial. What are your thoughts on this?
Adeniyi Abiodun: Sui is a Delegated Proof-of-Stake (DPoS) network, serving anyone with sufficient stake. In fact, even rules on the Sui blockchain can be decided through community voting. Since mainnet launch, we’ve received several community proposals, which have been approved and improved the protocol. So, Sui is fully decentralized.
With time and more developers joining, Sui will become increasingly decentralized. This means we can promote change, and the community can promote change. This is important to us. We encourage a secure ecosystem where people have the freedom to build.
This isn’t just talk; it’s proven with submitted proposals. Proposals like launching DeepBook were externally driven. Changes to validator stake requirements were community-driven proposals.
The community will gradually propose more changes to steer the basic direction of the Sui protocol. Isn’t that the essence of Delegated Proof-of-Stake? However, the challenge with DPoS is ensuring everyone is interested in voting. That’s always been difficult. We haven’t yet found the right answers to all these issues. Eventually, delegation is the solution. It’s reasonable to delegate my stake to someone to vote on my behalf. We’d love for every user worldwide to vote, but not everyone understands the technical nuances.
Tags:
AI, Aptos, Move, Mysten Labs, Solana, Sui, WEB3, Ethereum, public chain