Ethereum's Layer-2-Centric Approach Offers A Unique Benefit: Unrestricted Creativity
The growth of Web3 over the last decade has been nothing short of awe-inspiring, especially looking at the last 2-3 years. Use cases never considered are now mature, value-added parts of the Web3 ecosystem. The growth is expanding across chains, with nearly all areas of blockchain technology growing.
However, this process has very much been by design, and there is a critical element that has allowed this much growth. In May 2024, Vitalik Buterin spotlighted this element in Ethereum’s design, and discussed what a difference it has made. In the post, he discusses the role of security vs. scalability in the L2 implementation of Ethereum, and while much of the focus has been on the EVM, the flexibility of other L2 developments has been quite fortuitous, as this free market of innovation has allowed different platforms to build their L2’s with considerable freedom and independence, able to optimize what they need depending on critical aspects like security, time windows per transaction, gas fees, and more.
What is especially exciting about this turn of events is the breadth of innovation that has been created out of necessity for change, but also in a place that has been free to experiment and create. For Web3, this is absolutely necessary as it has not yet reached mass adoption. For it to reach the majority of users, it needs to innovate even further, tackling issues such as security, ease of use, and finding more ways to create value that its users need. Stagnant innovation is unacceptable in a rapidly growing field, and that stagnation would have likely happened on Ethereum without the L2 model allowing platforms to define their own parameters for success, while still being able to rely on the L1 security and stability.
Exploration vs. Exploitation
There is a concept in optimization problems (like an AI algorithm trying to find the best path from one point to another) called the Multi-Armed Bandit. Picture a slot machine, but instead of one arm to pull, there are ten different arms. If your goal is to maximize your payout, how do you accomplish that not knowing which arm is the luckiest? This is where Exploration vs. Exploitation comes in, and it involves weighing how many times you are going to play in total, then finding the right point where you stop exploring (playing different arms to test which might be lucky), and where you start exploiting (focusing on the one or two arms with the best payouts). It’s a tough problem, but it lays out a problem we face in many different areas. For Web3, we are living through this problem right now. However, given how many new and useful innovations we are developing daily, it’s clear that we are still in the “explore” phase of building fast, cheap, secure scalability on-chain.
Vitalik says it best when discussing how the L2-centric model is best for independent exploration:
“The fact that each L2 is isolated also means that deploying new approaches is permissionless: there's no need to convince all the core devs that your new approach is "safe" for the rest of the chain. If your L2 fails, that's on you. Anyone can work on totally weird ideas (eg. Intmax's approach to Plasma), and even if they get completely ignored by the Ethereum core devs, they can keep building and eventually deploy.”
If we were still held to an L1-centric model, the risk appetite for exploring new things would be a lot more conservative. In an L1 model, innovations with unknown risks can negatively affect the entire chain, and any risk to the chain has a high cost. The “politics” of the chain can come into play as well, as many different parties have a stake in what other platforms are doing on the chain. However, the L2-centric model solves this by creating a safety barrier that only intersects during rollup.
Designing for Safe Exploration
A key example of this is the project Vitalik mentioned: Intmax's approach to Plasma. This project, called Plasma Next, actually takes an idea that Vitalik had years ago, and uses modern innovations to revisit the feasibility of the plasma concept. In this particular case, Plasma Next is designed specifically to create massive scaling capabilities. It is built to take advantage of the high security of Ethereum’s L1, while creating a safe place for processing complex transactions quickly (up to 80k tps), with constant near zero gas fees. It also offers open source, customizable development so it can be used for countless applications. With the transactions occurring on L2 but rolled up to L1, it has the added benefit of full security while keeping the transparency of individual transactions hidden on-chain. This creates even more privacy-centric use cases.
Plasma Next is a strong case for an L2-centric system, but is just one example of the exploration the industry continues to pursue. With more and more exchanges, P2E copycats, and other clone-like platforms, we need to ensure that the freedom to take risks and innovate is both present and encouraged. We aren’t ready for mass adoption because the masses have let us know (if anything, through their lack of mass adoption) that we have to make Web3 easier to use, cheaper, and there need to be more use cases that we can’t live without (but can’t get anywhere else).
If we continue exploring the limits of Web3 at this pace, the next few years will yield massive innovation and improvements across the industry. Thanks to Ethereum’s L2-centric model, we can make that happen.