Vitalik: Five Regrets About the Initial Design of Ethereum

Original: DL News

Compiled by: Asher, Odaily Planet Daily

Last week at the ETHBerlin event, few would have expected Ethereum co-founder Vitalik Buterin to speak on stage. Even more surprising was that the content of his speech described some regrets about Ethereum's initial design. For many in the audience, his talk not only evoked the glorious days of the Ethereum network's birth in 2014 but also helped outline a roadmap for the next steps of the cryptocurrency, which is currently valued at $448 billion.

Regarding the current state of Ethereum, the U.S. SEC has just partially approved a spot Ethereum exchange-traded fund (ETF) — it has only passed the 19 b-4 and still needs to go through the S-1 filing. The world's largest asset management company, BlackRock, has also launched its own tokenized fund on the Ethereum network. The Ethereum network has spawned a vast ecosystem of developers and financial applications, valued at over $63 billion, and has become synonymous with decentralized finance.

Vitalik: Rethinking the Construction of Ethereum

Vitalik stated that there were a number of things he could have done differently, including the development of the Ethereum Virtual Machine, smart contracts, and the PoS consensus mechanism. He also mentioned that even as Ethereum becomes more mainstream, it is still misunderstood. "The narrative of Bitcoin is simple; it's digital gold. But when it comes to Ethereum, it’s more like 'Wow, what exactly is Ethereum?'"

ETHBerlin organizer Afri Schoeden asked, "Based on everything you know and have learned over the past 10 years, if you could start from scratch today, how would you build Ethereum differently?"

Vitalik discusses the past and present vision of Ethereum at the ETHBerlin conference.

Too Many Bits in the Virtual Machine

The first concern Vitalik mentioned was the Ethereum Virtual Machine, which is key to making the Ethereum network a decentralized large-scale cryptographic computer.

Vitalik stated that the original EVM design chose 256-bit processing instead of 64-bit or 32-bit. The initial design was overly complex for 256 bits, which is very inefficient and can generate a lot of redundant data on the blockchain even when executing simple tasks.

Smart Contracts Still Need Optimization

Regarding smart contracts, Vitalik expressed that early Ethereum developers should have focused on making it easier to write smart contracts with fewer lines of code to increase transparency. With fewer lines of code, people could more easily view and inspect what was happening inside.

Switching to a "Subpar" PoS Sooner

On the transition of Ethereum's consensus mechanism, Vitalik stated that Ethereum should have switched from PoW (Proof of Work) to PoS (Proof of Stake ) much earlier. "When we switched to PoS, we should have been willing to adopt a worse version of PoS sooner, ultimately wasting a lot of cycles trying to make PoS perfect. If there had been a much simpler PoS proof in 2018, it could have saved a lot of trees."

Now, Ethereum is no longer managed by miners but by validators who stake 32 Ethereum (worth about $124,000) to do the same thing — and thus earn rewards. If they misbehave, such as validating fraudulent transactions, they will be penalized. In summary, this transition replaced the original high-energy computing power with economic incentives.

Automatic Logging to EIP

From large token transfers to backdoor traps, users can easily track the flow of funds in cryptocurrencies, thanks in part to automatic logging features.

However, as the industry has evolved, especially with the shift from externally owned accounts like MetaMask to smart wallets like Safe, some important aspects of this logging have been lost. Vitalik stated, "Automatic logging of Ethereum transfers should have existed from the beginning; we could have completed the coding in just 30 minutes, yet it became an EIP (Ethereum Improvement Proposal)."

SHA-2 Replacing Keccak

Vitalik mentioned that if he could choose again, he would use SHA-2 for encryption in Ethereum instead of the currently used Keccak encryption algorithm.

The early Ethereum team implemented a non-standard version of Keccak. Essentially, Ethereum uses an iteration of SHA-3 that predates it, which means Ethereum developers need a custom library — a reusable code collection that doesn't require rewriting from scratch to accommodate SHA-3 and Keccak. However, due to incompatibility with other systems using SHA-3, the Ethereum network had to support both algorithms in the EVM.

Overall, this issue has largely been resolved, and current developments are certainly not affected by it.

Conclusion

In the end, despite a series of small design missteps, Vitalik stated that any project is bound to encounter such situations. He remarked, "I’m really glad; I feel that the core Ethereum developers and their execution capabilities seem to be improving every year. The team now has the ability to effectively and safely correct some of these mistakes."