Vitalik Buterin: The challenge of Ethereum POS vs L2

Achievements and challenges after the Ethereum transformation

Vitalik first reviewed some of the key achievements Ethereum has made since last year's Merge upgrade, and the consensus mechanism shifted from PoW to PoS:

The first merged hard fork succeeded.

More than 25 million ETH are being staked in proof-of-stake.

After the switch and upgrade of the consensus mechanism, Ethereum has been running stably for more than a year without major accidents.

But at the same time, Vitalik also directly threw out the current challenges of the PoS consensus mechanism:

Staking centralization: Since staking has become a necessary element of Ethereum's daily operation, staking service providers can help users to staking, but this inevitably leads to the centralization of staking.

Transaction processing efficiency: For optimal efficiency, the ETH network may want to process no more than approximately 10,000 transaction signatures per time window (Slot).

Note: The time window (slot) should be a predetermined period of time, usually a few seconds, used to propose and validate blocks. Each time slot has a pre-selected validator whose task is to propose a new block within that time slot. To ensure the efficient operation of the network and to prevent potential spam attacks or congestion, there may be an upper limit on the signatures or transactions that can be processed in each time slot.

Account Abstraction and Social Recovery

Vitalik has repeatedly emphasized the importance and necessity of account abstraction. He believes that the promotion of account abstraction is based on two considerations:

Convenience considerations: users can pay gas fees with any ERC20 token, as well as perform multiple operations (signing, authorization, verification, etc.) in one transaction;

Security considerations: Vitalik has also talked with a number of multisig wallet and MPC wallet projects, and believes that the mnemonic-based scheme is still not the most perfect choice, and the loss of mnemonic phrases and private keys will lead to the loss of assets.

At the same time, Vitalik continues to favor social recovery. The main reason is that Vitalik starts from the habits and experiences of ordinary users, who may not understand the private key, or lose the mnemonic phrase.

In this case, a "guardian" of your assets to help you is crucial. Considering that you may forget or lose it, your friend, your other device can become an asset keeper in case of emergency; At the same time, on the basis of trust, institutions that carry out encryption services can also play the role of this guardian.

In addition, account abstraction presents its own challenges. This challenge is not just an isolated technical issue, but involves the entire crypto ecosystem.

For existing wallets, applications and development tools, their choice to use account abstraction to improve user experience or developer experience necessarily involves technical adjustment and adaptation. Account abstraction is a technical concept, but it is bound to encounter problems when it is actually adapted.

And, as more L2s emerge, whether and how different L2s support account abstraction is also a challenge, and users may encounter situations where one L2 supports but another does not.

Privacy infrastructure is one of the newly introduced parts that is closely related to account abstraction. This means that in order to achieve account abstraction, transactions and operations need to be privately ensured, so that different authentication methods and account types can work without exposing user privacy.

While advancing broader account abstraction, it is likely that there will also be consideration of how to deal with MEVs to ensure the fairness, security, and healthy development of the system.

Therefore, Vitalik generally believes that the promotion and popularization of account abstraction requires the joint efforts of characters in the whole ecosystem, and the overall goal is to make the on-chain experience and centralized service experience consistent.

Scaling, L2 & Privacy

With the advent of L2 and the advancement of Ethereum's own technology, the scaling problem has been solved to some extent, but Vitalik believes that there are still the following challenges:

When L2 submits valid proof of transaction to L1 package, how to ensure the security and decentralization of this proof system?

In the technical component of L2, the sequencer for most transactions is centralized and potentially risky

L2 has different technology options and development directions, how to build wallets and addresses across L2 to provide users with a better experience?

How to solve the data availability problem of transaction records that requires space for storage and storage?

Due to time constraints, Vitalik did not give a very detailed answer to the above questions, but more was throwing out all the problems faced by the current expansion.

Interestingly, Vitalik also admitted in his speech that there are too many L2s in the market at present, but it is not necessary to do rollups when it comes to expansion.

He also gave some personal advice:

1. Projects should clarify their needs for security and scalability, what level they are, and then decide whether to go on Rollup.

For example, scenarios such as account key storage and high-value financial assets have extremely high requirements for security, while games or non-financial applications may have higher requirements for scalability.

2. Clarify the difference between Validium and Rollup, and take a seat.

Validium is a Layer 2 scaling solution. It stores data off-chain but verifies on-chain. This means that it can handle a large number of transactions, but may sacrifice some decentralization and security -- especially for scenarios such as gaming;

Unlike Validium, Rollup (specifically zk-Rollup) keeps data on-chain, which can be detrimental in terms of privacy or efficiency. But if you want reliability or more security, you might choose zk-rollup.

Judging from the graph shown by Vitalik live, for many projects, the current choice to use Validium may be a reasonable choice, as it provides a solution to achieve the availability of off-chain data while guaranteeing the validity of transactions.

But with the expansion of the data space of blobs (used to compress large amounts of data into smaller "blobs" to store data more efficiently in Ethereum), projects can turn to Rollup, which provides more security and reliability by keeping data available on-chain.

Finally, Vitalik also mentioned three types of privacy issues facing the current Ethereum ecosystem:

Coin Transfers: Privacy issues in the transfer process of cryptocurrencies. For example, while many blockchain transactions are open and transparent, there are still needs and technologies designed to make these transactions more private to protect users' financial data.

Guardians: Guardians, as mentioned earlier, are authorized by the user to make certain decisions or perform certain actions. The privacy of guardians may have to do with their identity, the assets they control, or the actions they perform.

Dentity/reputation and similar apps: On a blockchain, identity and reputation systems can help verify the authenticity and trustworthiness of a user or entity. However, these systems may also involve privacy issues, as they may need to collect and display sensitive personal information.

Conclusion

At the end of the speech, Vitalik summarized the general idea of the future development of Ethereum:

Maintain robustness and solve carefully: Maintaining the robustness of Ethereum's basic layer is a prerequisite on which we can carefully expand other aspects of capabilities, such as scaling, user experience, and privacy.

Balance between experience and decentralization: Ethereum must become "user-friendly" for the average user, but the benefits of decentralization must not be lost in the process.

Technology and ecology are equally important: the development of technology points out the direction, but how to implement and promote it still requires the joint efforts of the whole ecosystem.