Original Article:
Author: 0xTodd, Founding Partner of Nothing Research
I have seen many of my friends often confuse various types of "forks" and wonder why ETH PoW is an unorthodox fork from a technical perspective. Today I would like to explain a little bit about the concept of blockchain.
Blockchain, as the name suggests, is a chain consisting of many blocks being strung together. New blocks are connected to old blocks endlessly. Say the real world calculates time in seconds, the blockchain world calculates time in the smallest unit of blocks.
There are two types of forks in the world:
1. Fork due to network desynchronisation.
2. Fork due to version differences.
Let's talk about the first fork — the fork caused by network problems. This is a very common phenomenon that happens every day in the world of blockchain. For example, when two miners solve a math problem almost at the same time and mine a new block, which miner is eligible to receive the said block? Fork then occurs naturally.
It is generally agreed by the miners that the first block is the valid block. However, what if two miners discover the valid block somewhat at the same time (perhaps a difference of a few milliseconds) in addition to the delay of the broadcast across the network? It will then split into two groups naturally. Some miners will receive block A while some will receive block B instead.
To solve this problem, the Longest Chain Rule applies. It stipulates that whenever the blockchain forks at a certain block height, the longer and more meaningful blockchain is to remain, and the shorter one is to be discarded.
As shown in the figure below, starting from the new block, the two miner groups start to compete in computing power, and all other miners have to choose sides. It’s all about hash power, hence the group with the stronger computing power usually calculates faster and wins. But in theory, there are also chances the weaker party may overturn the result with a great amount of luck.
Winners take all, both the block rewards and transaction fees! Losers go home empty-handed, not to mention the wasted electricity.
Of course, ETH specially designed Uncle Blocks to solve this matter. Uncle is the brother of the father, but he has no son. It means losers are able to at least get a little compensation, which dissuades them to pursue further.
If one of the parties is malicious, it is a 51% attack. If neither party is malicious, it is simply a normal fork. This is also the reason why crypto exchanges have to wait for several block confirmations before crediting your deposits and top-ups.
As seen in the figure, the longest blockchain becomes the main blockchain, and the fork issue is resolved.
Ok, it’s time to talk about the second fork — a permanent divergence from the previous version of the blockchain.
In theory, ETH has gone through numerous forks, most of which were due to version upgrades. For forks caused by version upgrades, the Longest Chain Rule is not applicable as it can only apply to solving network synchronisation problems.
From here, the fork is either becomes a hard fork or a soft fork.
For the hard fork that happens due to upgrades, no new alternate blockchain will be created if the whole network agrees. It is a complete replacement of the existing chain. One of the notable cases is the Ethereum’s London Upgrade. “Upgrade” is such a euphemistic term here, in fact, it should be called the London Hard Fork. Anyway, the London Upgrade introduced EIP-1559 (fee burning) and no new chain was created. This is like the Qing Dynasty inheriting the territory of the Ming Dynasty, and subsequently, the Republic of China inheriting the territory of the Qing Dynasty.
However, if the community does not reach a consensus on the version upgrade, it will then cause a split. Say the ETC fork, for example, the original chain is completely separated into ETC and ETH blockchains. The former blockchain is the version that has not been rolled back due to the DAO Hack, while the latter is the rolled back version. These two versions have differences and do not recognise each other. This is similar to the situation between North Korea and South Korea. They originated from one country, but are now completely two countries that do not recognise each other.
Therefore, hard forks are risky, as there is no guarantee that the entire network will come to an agreement every time. Thus, the world of bitcoin prefers soft forks in which the risk will be much smaller.
What is a soft fork? It implements changes to the current version, upgraded or not, everyone is still in the same network. Just like how everyone knows that each state in the U.S. has its own law, but all of them follow the Constitution of the United States of America. When one of the states is introducing a new regulation, other states do not have to comply, hence there will be no division at all.
The most notable soft fork in the history of Bitcoin is SegWit. It removes part of the witness information to save block capacity, without exceeding the 1MB limit and also without compromising security. As shown in the figure below, if you agree with SegWit, you can go ahead and upgrade, otherwise you can don’t upgrade. Although Bitcoin clients would then have differences, miners can still mine together in the same chain.
Soft forks pay great attention to keeping backward compatibility, while hard forks have no way to be backward compatible. As you can see, Bitcoin's upgrade seems to create minimal impact, while Ethereum’s changes are always drastic. Soft fork updates can be a test of the wisdom of core developers.
Those who have experience in software development would know how difficult it is to add new features while ensuring backward compatibility. It's like dancing with shackles on, extremely difficult to design for the best of both worlds. Against all odds, bitcoin succeeded, with a lot of political trade-offs and wisdom in development, the design of each soft fork is exquisite and admirable.
One of the reasons is that Bitcoin is a public blockchain with an absolutely stable “monetary policy”. This gives users a sense of reassurance that Bitcoin is in pursuit of perfection. In another 10 or even 100 years, Bitcoin probably will not be much different from today, just like gold. The U.S. dollar may disappear in the next 100 years, but gold will not.
On the other hand, Ethereum does not pursue ultimate security but to break through the impossible triangle instead. Therefore, Ethereum takes huge risks every time when making hard fork upgrades. As you know, a hard fork can always lead to a split in the community. ETC of yesteryear, ETH PoW of today.
Learning from previous experience, Ethereum’s PoS Merge should have been a hard fork similar to previous upgrades, thus no new chain would be created, and the original version will be inherited. At the same time, no one has ever asked where USDT and USDC belong to.
Yet the ETH PoW team is determined to carry out their own hard fork upgrade at the same block height as PoS Merge. This upgrade will remove the difficulty bomb and more. Here comes a common mistake from most people – thinking that the ETH PoW should look like the figure as shown below, one goes straight, and one turns left. In fact, the ETH PoW’s hard fork is actually not created out of the original Ethereum PoW blockchain. If so, the orthodoxy of ETH PoW would be unquestionable.
The original Ethereum PoW blockchain should be the chain with the difficulty bomb, also the chain that its client has not upgraded. The new PoS Merge blockchain will then be the upgraded client, as well as the ETH PoW blockchain. Therefore, the actual hark fork should be looking like this — one turns left, the other turns right, and no one goes straight.
Therefore, the actual situation is: both ETH PoS Merge and ETH PoW are actually new blockchains created as a result of hard forks, which just happened to be “forked” at the same block height. The current Ethereum PoW blockchain that is running as of now, will eventually halt due to the difficulty bomb.
In this case, the orthodoxy of ETH PoW is rather weak. The truth is, ETH PoW’s hard fork can happen anytime, it's perfectly fine not to choose the same block height as PoS Merge. So, why does the ETH PoW team have to choose the same block height as PoS Merge? Well, their intentions are no open secret.
Some people may ask, Todd, what on earth are you doing to clarify these things? I believe that every blockchain acts as its own country. If you are living in that country, you must understand its legitimacy and institutions. Since my wallet address starts with 0x, as a citizen of Ethereum, it is natural to get to the bottom of it.
Consensus determines the ceiling of a blockchain. Losing orthodoxy will end up losing consensus as well.
Translated by: [Coinlive] Nell