L2s have gotten hyperfocused — that’s bad news for Web3

In the Netherlands, we have a saying: “Meten is weten,” which, literally translated, means “measuring is knowing.” We prefer to look at situations directly, because only then do we know what we’re up against — and what needs to be changed.

And it does not take much measuring of Crypto Twitter to know that we are deep in the L2 hype cycle. The sense is that L2 blockchains are more user friendly, safer and better for scalability, and we (as a collective) should focus on their development.

Forgive the directness — but that’s a mistake if we want Web3 adoption to increase.

Here’s what we know keeps users from Web3: security risks and poor performance. Now, L2s on their own do not bring these problems, as technology is neutral. It’s our own hyperfocus on L2s that is the issue. This hyperfocus has created an environment where security risks and poor performance go from minor speed bumps to major adoption obstacles.

It is counterintuitive that a high amount of focus on a technology would lead to security risks, but it’s true. 

It’s called inattentional blindness — when we fail to see a “stimulus” (like an exploitable smart contract or overcentralization) even when it is right in front of us. A famous experiment was done to prove this effect, where participants were asked to perform a visual task. Afterwards, they were asked if they noticed the person in the gorilla costume. Half of them did not. 

The same effect is happening with L2s. We’re paying so much attention to their potential that an unacceptable level of centralization (the main in the gorilla suit) has gone right past us: centralized sequencers.

Sequencers are responsible for batching transactions and submitting them to the chain. As a single point of failure, they should evolve from their current design to something more robust. However, what is even more concerning is that they are often controlled by the L2 team and, therefore, can be stopped at will. Just this week, Ethereum L2 Linea decided to “pause its sequencer” to halt an exploit on its chain. 

Situations like this draw the attention of regulators, and for good reason. It is their responsibility to prosecute when their citizens suffer theft (the exploit) and to create and enforce laws for consumer protection. If Visa or Mastercard stopped running transactions for a day because they wanted to prevent being hacked, they would almost certainly be fined to compensate for the missed bill payments and late fees experienced by their users. 

And if we want to hit billions of users, we need both an answer to this centralization problem and to be ready for the increased oversight when (not if) someone decides to stop the chain.

The other area of “hyperfocus” that has led to blind spots for Web3 adoption is our overestimation of L2 transaction capacity. 

We know that L1s have difficulty processing many transactions per second on the scale of millions, much less billions, of people. This makes Web3 truly unscalable — and L2s were a welcome addition to the ecosystem to help solve the issue. But now that we are very much in the superchain future, there’s still insufficient capacity (even when the sequencer is on). 

Read more from our opinion section: Don’t fossilize Bitcoin

In the desire to reach the number of transactions per second of Mastercard or Visa, the focus was on developing more L2s: In reality, we should have also thought about developing more L1s. The immediate effect of doing so would mean more base chains on which to build layers and rollups, which is good for transaction capacity and performance. 

However, the longer term and more important effect of focusing more on L1st is innovation. Just look at Solana’s outsize effect on our ecosystem and user growth. The process of innovation involves changing what seems unchangeable, and in doing so, creating new technological foundations, different value capture mechanisms and better ways of operating. This diversity is good for transaction capacity, but it also hedges us against the possibility of a monolithic technological future.

And it is a monolithic future we’re at risk of creating. The more we emphasize layer 2s, the more we actually centralize ourselves around one chain. Besides the security risk and poor performance this will inevitably create, there’s just no fair and equitable future where one chain is the fundamental basis of the entire digital economy. 

The good news is that if we are serious about Web3 adoption, we don’t have to kill off the L2s. We don’t have to unroll the superchain. We don’t have to focus only on the person in a gorilla suit. All we need to do is dedicate just a part of that hyperfocus to L1 development. 

Start your day with top crypto insights from David Canellis and Katherine Ross. Subscribe to the Empire newsletter .

Explore the growing intersection between crypto, macroeconomics, policy and finance with Ben Strack, Casey Wagner and Felix Jauvin. Subscribe to the On the Margin newsletter .

The Lightspeed newsletter is all things Solana, in your inbox, every day. Subscribe to daily Solana news from Jack Kubinec and Jeff Albus.