Recap of PoW vs PoS discussion on Fork It Podcast — Part 1

Crypto currency mixer Tornado Cash has been sanctioned by the U.S. Department of the Treasury’s Office of Foreign Assets Control (OFAC). Currently it looks like over 66% of the beacon chain validators will adhere to OFAC regulations.

The Merge is on the horizon and Ethereum is about to move its consensus mechanism to Proof of Stake (PoS). Will Ethereum face protocol level censorship after the Merge? If so, what will this mean? Would it be different if its consensus mechanism was PoW?

In this episode of Fork It, a Chinese podcast program for blockchain technology, our host Terry (co-founder of Nervos Network) invited Cipher, founder of Nervina Labs, to talk about PoW and PoS in depth.

Here is the full recap:

1. Introduction

Terry: Good day to everybody. In this episode, we’ll discuss what you might consider a cliché, PoW vs. PoS. It has become a hot topic due to recent events like the sanction against Tornado Cash and the upcoming merge of Ethereum. We have our old friend Cipher here to discuss this topic this time. Cipher, please say hi to everyone.

Cipher: Hello, everyone. I’m Cipher.

Terry: Although Cipher is an old friend, there could be some new listeners; therefore, I’d prefer if you started by sharing your blockchain-related experiences.

Cipher: I am a veteran, although a pretty inexperienced one. I learned about Bitcoin in 2013, but I was not a believer then. I regarded bitcoin as a practical T+0 investment at the time because I was working on other endeavors full-time. I created a trading bot at the time and earned about ¥300,000 from only a few thousand.

Terry: A quick question. Were you superior, or were the trading competitors too weak then?

Cipher: At the time, there were few trading bots. Hence there weren’t many rivals in the market. My approach was fairly simple: purchase after two consecutive upswings and sell after two straight downturns. I earned money that way. There was a window of time, perhaps two or three months, after which this tactic ceased to be effective. Then, I stopped for several reasons, including regulatory concerns.

I didn’t enter the blockchain industry until 2016 when I started to work on research and products of a consortium blockchain project. After that, I joined the public blockchain project until last year; then I began working on an independent NFT project. I started a new project this year and have moved to Singapore to do something big.

Terry: Cool! I will first provide some background information to set the stage for our discussion on PoW vs. PoS today. The recent sanction against Tornado Cash protocol by the Office of Foreign Assets Control (OFAC) of the US Treasury Department and the arrest of one of its core developers in the Netherlands caused a stir. It’s a big deal, and some centralized organizations, including exchanges, wallets, and even the front ends of some blockchain applications, are active in banning addresses that connect with Tornado Cash.

Another background is the upcoming merge of Ethereum, where its consensus mechanism will be upgraded to PoS.

The third background is that it currently looks like over 66% of the beacon chain validators will adhere to OFAC regulations. A poll was launched on Twitter on the possibility of these validators censoring Ethereum at the protocol level in the future. Since there is a significant difference between the two in terms of how they handle protocol-level censorship, the debate between PoW and PoS has become more intense.

Therefore, there is a necessity to produce an episode to talk about PoW and PoS. This is also an opportunity to dispel some common misunderstandings regarding PoW.

2. The Relationship Between PoW and PoS

Terry: First, let’s limit the scope to layer 1 because there may be different choices beyond this scope. Cipher, should our future global money be built on PoW or PoS? What is the relationship between PoW and PoS? Is PoS a progressive, a regressive, or are they not progressive or regressive, but rather like left and right?

Cipher: I think PoS and PoW are two choices of the path. There is no progress or regression.

I want to clarify that PoS and PoW are two methods of selecting the node to produce blocks. They are not the entire consensus mechanism. The consensus mechanism should include at least other equally important things, such as which chain is the right one when there are forks, block rewards, punishment, etc.

From the perspective of selecting the node to produce blocks, PoW is connected to off-chain variables, such as hash rate, energy consumption, chips, electricity prices, local regulations, etc. In contrast, PoS is more connected to on-chain factors, such as token distribution, staking percentage, etc.

So it is easy to understand that PoW and PoS have nothing to do with who replaces whom or who is more advanced. They choose two separate paths, one relying on off-chain variables and the other depending on on-chain factors to determine which node will produce the next block. They are designed to fit different use cases.

I wrote a blog post titled Mining to the Left, Staking to the Right a few years ago, and I used left and right to describe the relationship between PoW and PoS. My opinion is unchanged.

2. Two Types of Public Blockchains

Terry: I vaguely remember you mentioning two types of public blockchains in that blog post. Can you explain a bit more?

Cipher: There are two typical application scenarios for blockchains. One is anti-censorship, similar to bitcoin fundamentalism, which entails seizing control of your property rights. The ability to build dApps or DeFi applications on the blockchain and improve TPS is less important than the core task — anti-censorship. This is one direction for public blockchains.

Another direction is designed for open finance, which is not necessarily decentralized finance. Compared with the current financial system, open finance is already a very revolutionary thing. We can build incredible financial applications on it, and third parties can be added to the monetary system. The so-called DeFi can significantly improve productivity and economic efficiency. Users may not be so concerned about whether it is resistant to censorship and may even embrace the regulation.

In short, the first type of public blockchain mainly solves the problem of how to resist censorship. In contrast, the other type mainly solves the problem of global collaboration via open finance. Can these two types of public blockchains be merged? It is possible, but they must be in different layers. If you want to merge them all in the same layer, such as in layer 1, I don’t think you will succeed as there is a technical limit. There must be a trade-off.

Terry: When you wrote that blog post, you must have classified Ethereum as the first type of blockchain, right?

Cipher: Yes, at least the notion that Ethereum Foundation and Vitalik were promoting to the public tells us that Ethereum belongs to the first type. However, it turns out to be more like the second type of blockchain. The result speaks louder.

Terry: So, after the merge to PoS, Ethereum is more likely to be the second type of public blockchain, right?

Cipher: Yes. The reason why there are now so many arguments about Ethereum is due to the path it chooses. It is not a technical problem but a choice problem from different approaches. For example, there’s also a lot of controversy on Ethereum EIP-1559. You can find out that EIP-1559 does not help to improve TPS, nor does it reduce the gas fee. It only weakens the rights of miners. This is the choice Ethereum made, and it has nothing to do with technology but political things.

4. PoW or PoS, which is more suitable for layer 1 blockchain?

Terry: We narrowed down the topic to layer 1; we want to serve as global money and be the first type of public blockchain you mentioned above. I believe you also favor PoW over PoS; am I correct? What are your thoughts about that?

Cipher: Yes. I need audiences to think about one question: What is blockchain’s core value? Or, what kind of problems does blockchain solve?

This question was discussed frequently before the massive emergence of DeFi applications. At that time, the blockchain only had a payment function, which was largely weakened due to various reasons. For example, many users regarded bitcoin as a SoV (store of value) and were less willing to pay for things in bitcoin. There were few applications, and users repeatedly asked “why should I use Bitcoin?.”

The application layer has been the center of attention for the previous two years. As NFT and DeFi applications grow in scope, users appear to be less concerned with blockchain’s core value. That is a meta question and one of the most important ones in the blockchain world. Many later things are genuinely in the air if that question is not fully discussed.

In the past years, there were many labels for the blockchain, such as a non-reversible public database, a global computer, etc. All these labels are right, but one definite key word is missing: no permission is required. Blockchain is a license-less, tamper-evident global public database, a license-less global computer. All of these labels are accurate but missing one important word — permissionless. Blockchain is a permissionless non-reversible public database and a permissionless global computer.

If you remove the word “permissionless,” you’ll find that you don’t need a blockchain to achieve it. A non-reversible public database can be created via hashed digital signatures, and a global computer can be achieved with the help of Amazon cloud service or some open APIs. Permissionless can only be achieved through a blockchain.

I believe that the blockchain system is the first and only open system created by humans that is permissionless. All other systems require permissions; to use them, you must, for instance, use your ID card to create an account or have someone else make an account on your behalf. Therefore, things like performance improvements, privacy protection, cross-chain bridges, sharding, NFTs, and so on are the tall buildings sitting on top of the foundation of blockchain technology, which must first be established well.

PoS outperforms PoW in several aspects, including performance, energy consumption, etc.

However, nearly all PoS proponents, even the most aggressive defenders, have to admit that PoW works better in terms of permissionless. Anyone can send a transaction in PoW since there is a considerably lower entrance barrier, while PoS weakens the permissionless for various reasons. So, in my opinion, PoW maintains the ideological stance of permissionless, and it has to make sacrifices in other aspects.

PoS has gone farther and farther down the road of practicality. Delegate service was unavailable in the early PoS blockchain projects, but it was eventually added. The later PoS blockchain projects, like Solana, are becoming increasingly centralized. You will find out that they have started to operate more and more like the traditional financial systems.

I think PoS blockchain projects are more like the next generation of central banks or Wall Street, with greater openness. Data is more accessible as anybody can access and read it on-chain, but there is no assurance that you can write data on-chain or won’t be censored when you write the data.

The PoW mechanism has always represented a rebellious force against financial tyranny. You can go to Wikipedia and search for “Cantillon Effect,” an effect related to the money supply, for a better understanding of economic tyranny.

Therefore, I believe that the PoW mechanism will probably never become mainstream because of its limited performance, even though it is incredibly hardcore. However, I would find it quite dull if PoS were the only consensus mechanism surviving in a decentralized world. That’s why I am firmly in favor of PoW.