Why a Former Chinese Ethereum Community Leader Thinks POS Is a Lie and Joins Bitcoin Education Work

Sep 28, 2022

As almost everyone cheered the successful Ethereum merge, a 26,000-word article was widely shared in the Chinese community. As one of the most famous leader of the Chinese Ethereum community, he bid farewell to the Ethereum community and joined the Bitcoin community because of the implementation of EIP-1559 and the abandonment of POW in Ethereum.

This is a key summary of Fork It #23. In this episode, they have invited Ajian, the head of content at BTCStudy, to talk to you about the PoW vs PoS debate, his views and position. （This article does not represent the views of WuBlockchain, but only shows his personal views and transformation process）

Ajian’s Intro: a very hardcore researcher, translator, content contributor in the Chinese ethereum community, Editor-in-chief of EthFans and has been contributing to BTCStudy since the end of 2021. He decided to abandon Ethereum after EIP-1559.

PoW is overall better than PoS

Ajian: First of all, regarding the position, I think PoW is an all-round better thing than PoS without a doubt. It is all-round. There has been a debate about the superiority of PoW vs PoS for a long time. However, in the whole process, the argument of PoS on the superiority of PoS has contained many impurities, and I even think most of them contain lies. A lot of people think that bitcoin came up with PoW before someone came up with PoS. In fact, technically, I think it’s a error. Byzantine fault tolerance algorithm is the predecessor of the so-called PoS algorithm, which appeared much earlier than PoW. However, in the earliest form of Byzantine fault tolerance algorithm, it defaults that the signature weight of all participants is the same, and there will be no case that the voting weight of participants is not the same. The conceptual advance of PoS over the Byzantine fault-tolerant algorithm is that it relaxes this point so that signatures of different participants have different weights. However, we can say that PoS is an older thing that doesn’t actually give us the consensual mechanism that everyone can participate in without permission like Bitcoin. This is a very important feature.

Trilemma

Security

Ajian: In any dimension, there is no argument that PoS is better than PoW. In terms of security alone, PoS mechanism still has Stake Grinding Attacks, Long Range Attack, and 51% Attack Cost. These attacks are just an opinion, because some opinions of some people are pooled. One side says this is an attack, but the other side says no, this is an act of justice. I am imposing sanctions on you for the safety of chain and consensus. There is no way to defend against this attack unless you also have enough agreement to fork another chain. It is because we do not have so much consensus that we have to create a variety of technologies, including blockchain, to restrict the process of participation in expressing opinions or participating in consensus. However, PoS is subject to the will of some individuals, so it cannot play the effect of protecting individuals as we want.

Why do some people believe that PoS is more secure? Because they do not understand this set of theories, a more intuitive feeling is that there are many PoS chains, and the attack you mentioned does not happen. On the contrary, some PoW coins often have 51% attacks, so it is easy for them to conclude that PoS is more secure. When people say PoS is secure or can’t come up with an example of an attack, a big part of the reason is that they haven’t been exposed to it. For example, Stake Grinding Attacks, some PoS blockchains have been killed by Stake Grinding Attacks, like the Balancing Attack I just talked about, which is a new form of Attack that seems to be proving to be worth defending against.

For all blockchains, when a new node joins the network, it needs others to provide it with blockchain data for synchronization. The key is whether you can verify the data given by others. We often say, “Don’t trust, verify.” You have to verify it yourself. Whether it’s Bitcoin or Ethereum, there are seed nodes, these nodes give you the current state of the blockchain, and you might find a node that is closest to you to download the block data.

One of the characteristics of the PoW blockchain is that its blocks are very difficult to produce, so every time you receive a block, to verify its proof of work, as long as the verification process continues and you are connected to enough nodes, your trust in this process can be minimized. In other words, the element of trust has been reduced to a minimum in the way we can think of. So, we can say that PoW itself is trust minimization, but PoS clearly does not conform to this.

Scalability

Ajian: PoS did not improve scalability, the same circumstances, the condition of the hardware node is the same amount of resources, the more you spend resources on validation of consensus, the less resources on processing trade, PoS consensus validation overhead, a lot larger than PoW, because the PoS validation consensus, is one of the so-called status verification, Validation is not the same as PoW. The overhead of PoS on consensus verification increases as the number of verifiers increases. Another aspect that many people don’t notice is that the communication overhead of the verification block also becomes much larger than that of PoW.

Censorship-resistant

Ajian: Your public key can be changed every day in PoW, but you can’t do that in PoS. You can keep switching websites, but your identity is always that identity, so in PoS, you can be continuously tracked based on this identity.

The Merge is taking the big block route

Ajian: I think The Merge is a big block. Large blocks come at a cost, which means higher operating costs for nodes. Ethereum Layer 2 currently has no clear and verifiable scaling effect. In my opinion, all these rollup solutions are only interim solutions, and Ethereum has not really stepped out of a fully foreseeable and bright future of Layer 2 solutions. So it comes down to one word — big blocks. To increase so-called throughput through large blocks is, in my opinion, a very sad thing.

Why leave the Ethereum community?

Ajian: Essentially, it’s a disapproval of PoS and EIP-1559. For EIP-1559, I strongly disagree with the idea that it is only related to miners, because it is not targeted at miners, it is targeted at what conceptually should be called the block producer, that is, the people who are originally eligible to get the part of the fee revenue, except that the consensus mechanism of Ether happens to be the PoW mechanism, so it is the miners who are weakened. If we switch to PoS, it is actually the witness we mentioned earlier, or the participant of PoS consensus, who is weakened.

Everyone is also very excited, thinking that you miners make so much, so after pulling out your earnings, it becomes deflationary, and then everyone is happy. That’s actually a very short-sighted, very extreme, and irresponsible kind of thinking because you’re actually undermining this incentive for everyone to pledge to the system consensus to provide it with security.

But I also have another feeling that people are obviously giving so much to Ethereum, so why are these contributions not taken seriously? It’s not just me, but others as well. Ultimately, those of us who are passionate about joining the historical process are forced to compromise with what is actually clearly a more centralized process, or a more centralized governance process.

I was doing my own research on PoW and PoS, and around 2020 or 2019, after I felt that I had a little bit of an answer, my focus started to shift to research on Ethereum itself, and this research eventually led me to the conclusion that the architecture of Ethereum, whether it is its account model or its on-chain computing paradigm, ultimately it has not thought about how to protect those users, how to accommodate those users who do not recognize a process.

Ultimately, you’ll find that Ethereum is a political consensus without a technical consensus, and it can only be pushed with a hard fork, and that’s something that I ultimately I want to leave.

Why start researching Bitcoin and running BTCStudy？

I liked it a lot after I discovered how protective Bitcoin was for users. Of course, there are original reasons for this liking, such as the fact that Bitcoin proponents actually recognize the true merits of PoW as a consensus mechanism and the fact that it is indispensable. They understand this, and have sorted out as much of the argument as we can imagine in terms of thoughtfulness to the point of pushing the limits of what we can achieve at the moment.

Bitcoin will tell you that these things come at a cost, and if we can, can we choose a smarter approach, or can we choose a more long-termist approach, and this to me is what appeals to me the most about the Bitcoin idea.

We didn’t find any good information in China at that time, so we had to re-read from these English materials one by one and try to make up for our knowledge of Bitcoin, and this was the original intention of running BTCStudy.

BTCStudy’s mission is to be a gathering place for bitcoin thoughts in Chinese.

Learn more: https://talk.nervos.org/t/fork-it-23-pos/6751

Wu Blockchain

Discussion about this post