Martin Köppelmann, co-founder of Ethereum infrastructure Gnosis, used the topic "Limitations of L2" at the ETHDenver conference to explain from multiple perspectives that L2 cannot actually solve the expansion problem of Ethereum. In addition, it also proposed the concept of "Ethereum Universe" with reference to the design of Cosmos IBC, using ZK cross-chain bridge technology to create a more decentralized and efficient Ethereum.

L2 is practical, but has its limitations and trade-offs

Martin said at the beginning of his speech that due to the lack of scalability of Ethereum, the current mainstream solution relies on roll up technology, which will create L2's current flourishing ecosystem.

However, Martin believes that with new cross-chain technologies that reduce the risk of cross-chain (such as ZK cross-chain bridge), there are new solutions to the problems L2 is trying to solve. This speech will focus on the limitations faced by L2 and why more L1 block space is necessary to make the blockchain more secure and decentralized.

The initial concept of L2: a temporary trading space

Martin then mentioned the origin of the development of L2. Initially, L2 was used to execute batch transactions deposited from L1 and synchronize the results back to L1 for settlement. This makes L2 like a transition space and does not have too many transactions in it. Stay, and the whole process is also the origin of the word Roll up.

However, this transaction processing model is only suitable for some applications where the state will not expand, such as contract exchanges, because no past transaction history is required, only transaction results.

What is the current transaction capability of Ethereum?

Although L2 does increase the transaction limit, Martin said that the scalability that L2 can bring to Ethereum is still limited by L1. When Ethereum can only execute about 1.25 million transactions per day, L2 will eventually reach the scalability limit.

So with the introduction of L2, what is the current level of Ethereum’s transaction processing capabilities? Martin gave the following two examples:

1.ENS: Assuming that 10% of the world's population (about 800 million people) registers Ethereum domain names together, it will take about two years to digest these transactions.

2. Global stock market: If Ethereum is used as the settlement layer of the global stock market, the 45,000 publicly listed stocks around the world can only conduct less than 30 transactions per day.

L2 Transaction Cost Issue

In addition to scalability limitations, L2 transaction costs are also an issue that needs to be addressed. Martin said that L2’s gas fees are still proportional to L1’s fees, which is unacceptable in many use cases when fees soar above $1.

Although gas charges will be reduced by more than 90% after the implementation of EIP-4844, demand will also increase at the same time, and it will still be unavailable for use cases requiring sub-1 cent charges.

In addition, if the cost of leaving L2 is higher than the user's assets, the user's assets will be stuck in L2 and cannot be withdrawn. Even if everyone can afford to leave L2, L2’s bandwidth limitations may cause users to withdraw all their assets at once.

The inherent limitations of roll up technology

Even if transaction costs and scalability issues are eliminated, L2's roll up technology cannot be used on all applications.

Martin cited CirclesUBI and POAP as examples. The state of these applications is too large, and it is not feasible to transfer them out of L1 because the data cannot be compressed and settled on L1.

What if you never leave L2?

After discussing the current problems faced by L2, Martin asked: "What if we never leave L2?"

He raised two questions about this idea:

1. The current L2 block sequencer (Sequencer) is too centralized, and transactions may be excluded, transaction fees may be increased, or some protocols may be treated differently.

Martin cited the L2 Base launched by Coinbase as an example. Because Coinbase controls Base's centralized sequencer, they can guarantee that their transactions will receive priority if they want to.

2. Centralized sequencers may face censorship or require their users to pass KYC, otherwise they cannot conduct transactions.

In addition to the above two issues, Martin also said that if you just want to issue native assets on the chain, is it really so important whether it is issued on L2?

Although native L2 assets do not require cross-chain, they solve the problem of cross-chain risk. However, since the security of L2 comes from the verifiers of L1, L2 cannot function properly without the security of L1.

In addition, since Ethereum itself is still undergoing continuous updates and evolution, the L2 mechanism also needs to keep pace with the times.

Martin takes the on-chain voting protocol Snapshot as an example. Its trustless voting mechanism is performed on L2 and the results are synchronized back to L1. The overall process Snapshot uses Merkle Proof. However, according to Ethereum’s roadmap, it plans to switch the Ethereum client from Merkle Trees to Verkle Trees, which will make Snapshot unavailable at that time.

In this regard, Martin said that L2 must have a corresponding upgrade mechanism, but this will violate its goal of being trustless.

Since L2 has so many limitations, what are the alternatives?

Martin proposed an idea to run an execution layer and consensus layer network that is exactly the same as the Ethereum architecture, and use ZK cross-chain bridge technology to connect the two chains. The example is called "Gnosis". Gnosis will be as consistent as possible with Ethereum and will also be compatible with future EIPs, making running Gnosis just like Ethereum.

On Gnosis, Ethereum's light nodes will be run, and EVM will be used to verify whether the consensus and blocks are signed by the majority of validators to ensure that they are in a consistent state with Ethereum.

Gnosis's ZK cross-chain bridge concept is similar to Cosmos's IBC (inter-chain communication) model, which enables Ethereum and other beacon chain-based EVMs to connect to each other without trust, ultimately forming a "multi-chain" Ethereumverse.

This article, the co-founder of Gnosis talks about "L2 expansion and technical limitations" and launches the Gnosis chain to create the "Ethereum Universe" first appeared in Chain News ABMedia.