Analyzing the Legitimacy of the ETH PoW Fork from a Technical Perspective
0xa3C5
August 15th, 2022

I notice that many frens are often confused about the idea of different “forks” and wondering why ETH PoW is less legitimate. Let me unpack this concept for y'all here.

A blockchain, just as the name implies, consists of many blocks that are cryptographically linked together into a chain. New blocks are linked to the old blocks one by one. If we say that the real world, one second is the smallest unit in time, then  in the blockchain world, a block is the smallest unit in time.

Overall, there are two types of forks:

  • Forks occur when a part of the network becomes desynchronized
  • Forks occur due to protocol version differences

Let’s talk about the first one, the fork caused by the network. This is a very common situation that happens every day. For example, when two miners solve the puzzle almost at the same time and both package a new block, then whose block is eligible to be linked to the old block? Here comes the fork.

Usually miners will consider the first mined block to be the valid one. However, when two miners accomplish the process almost at the same time, with a tiny difference like only a few milliseconds, plus the delay of the broadcast in the network, some will receive block A at first and the others receive block B at first. So shall we regard block A or B as the first and valid block?

Receiving New Blocks at the Same Time
Receiving New Blocks at the Same Time

To solve this problem, the “longest chain principle” was introduced. It stipulates that if the blockchain is forked at a certain height, the blocks on the longest chain will be the valid ones and others will be discarded.

As shown in the figure, starting from the fork height, two miner groups start to compete in hashrate, and all other miners have to choose sides. In general, miners with the stronger hashrate will be faster. But in theory, there is a chance for very fortunate miners with weaker hashrate to overtake the competition.

Hashrate Competition
Hashrate Competition

The winning miners earn all the block rewards + fees, and the losers get nothing to cover the wasted electricity.

Considering this kind of issue, ETH especially designed the uncle blocks. The uncle is the brother of the father, but the uncle has no son. It allows the losing miners to receive some compensation, which can prevent some unnecessary disputes.

If one party is malicious, that’s a 51% attack. If neither party is malicious, it is a normal fork. This is also the reason why crypto exchanges require waiting for several blocks of confirmation when users make deposits.

Then, as shown below, the longest chain is the valid version of the blockchain, and that’s it for the network fork.

The Winner’ Fork Becomes a Part of the Chain
The Winner’ Fork Becomes a Part of the Chain

Now let’s discuss the second type of forks, what’s the fork that occurs due to changes in protocol versions?

In theory, ETH has been forked many, many, many times, mostly because of system version upgrades. For forks caused by version upgrades, the longest chain principle does not apply. It only applies to solving network synchronization problems.

Here we classify forks into two kinds: hard forks and soft forks.

If the hard fork is caused by the upgrade, with the achieved consensus of the whole network, no new fork chain will be created. It is actually a complete replacement of the old chain by the new chain.

A typical example is the “London Upgrade” of Ethereum, the so-called upgrade is actually the London Fork. The London Upgrade brought EIP-1559 (fee burning mechanism), and no new chain was created.

 EIP-1559 is a Classical Hard Fork
EIP-1559 is a Classical Hard Fork

However, if the community does not reach a consensus on the version upgrade, there will be a split. For example, the ETC Fork completely separated ETC and ETH. The latter is the version that has been rolled back due to hackers, and the former is the version that has not. There are some differences between these two versions and both do not recognize each other. This is like the relationship between North Korea and South Korea. They used to be one country, but are now completely two different countries that do not recognize each other.

ETC and ETH Separately Exist Together
ETC and ETH Separately Exist Together

So hard forks are risky, and there is no guarantee that consensus can be reached among the whole network every time. Therefore, Bitcoin prefers soft forks, which is less risky.

What is a soft fork? Although with version differences, it can be upgraded or not upgraded, and everyone will still be on the same network.

As we all know, each state in the United States has its own laws, but they all follow the Constitution of the United States. A soft fork is like a state in the US introducing new regulations, which does not have to be adopted nationwide. And therefore won’t cause any division.

The most famous soft fork of Bitcoin is segwit - Segregated Witness, which can remove part of the “witness” data to mitigate a block size limitation issue, maintaining a 1MB limit and eliminating impacts on security, and could underpin the lighting network.

Soft Fork: Both the Old and New Versions Can Exist Together
Soft Fork: Both the Old and New Versions Can Exist Together

As shown below, if you are a miner, and you agree with the Segregated Witness, then you can upgrade your miner client. But if you do not approve, you can skip it. Although there are differences among various versions, miners can still participate in mining & consensus together.

Soft forks focus a lot on “forward compatibility”, while there’s no way for a hard fork to be "forward compatible”. As you can see, Bitcoin’s reforms “seem” to be small, but Ethereum’s reforms are drastic. This is because it’s a considerable challenge for core devs to conduct a soft fork.

Software engineers may have a deeper understanding of how difficult it is to add new features on the old versions. It’s like dancing among chains, and it’s hard to find the perfect solution. But BTC has made it, which involved a lot of political trade-offs and development wisdom. Each soft fork that occurs on BTC has been exquisitely designed. Very impressive.

This is because BTC is a chain with an absolutely stable “monetary policy”. It‘s persistence on perfection makes you feel secure and safe. In another 10 or even 100 years, BTC will not be much different from today, like real gold. The dollar may disappear in the next 100 years, but not gold.

For ETH, as long as Vitalik is still here with us, everything will be fine. But ETH pursues breakthroughs in the impossible triangle, rather than the ultimate security. They are taking risks to conduct every hard fork upgrade. You know that hard fork can easily lead to a split in the community. ETC in the past, ETH PoW today.

According to previous experience, PoS Merge should be a hard fork similar to the previous one, in which upgrades would be applied to the old version without the introduction of any new chains. Meanwhile, no one has ever asked about the allocation of USDT and USDC before.

A Technical Upgrade
A Technical Upgrade

The ETH PoW team is determined to carry out another hard fork upgrade at the same height as the PoS Merge. This upgrade removes difficulty bombs and so on. Here is a common mistake: the fork of ETH PoW is not on the original PoW chain of ETH. Many frens may mistakenly think as below, one goes straight and the other turns left.

If so, the legitimacy of ETH PoW is unquestionable.

The Misunderstood Situation
The Misunderstood Situation

However, the original PoW chain of ETH should be the one with the difficulty bomb, aka, the PoW chain that “does not upgrade the client”. PoS ETH is a new chain with “upgraded clients”, and ETH PoW is also a new chain that has “upgraded the client”. So the actual situation is as below: one turns left, the other turns right, and no one goes straight.

The Actual Situation
The Actual Situation

ETH PoS and ETH PoW are actually new chains generated by hard forks, which are at the same height. The real ETH PoW, the original chain, will die out because of the difficulty bombs.

So ETH PoW is also weaker in legitimacy. In other words, even if the fork occurs today, it’s perfectly fine. There is no need to choose the same height as the PoS Merge. Then why does it have to fork at the same height as PoS ETH? The answer is obvious.

Tracing block history seems like something that only historians would do, so why bother researching these? If we consider every chain as a country, and if you choose to live in a country, you must understand its legitimacy and institutions. With 0x or .eth on name, as a citizen of the ETH world, we want to get to the bottom of everything.

Consensus determines the success of a chain. Losing legitimacy also means losing a lot of consensus.

Subscribe to Nothing Research
Receive new entries directly to your inbox.
Collectors
View
#1
#2
#3
View collectors
This entry has been permanently stored on-chain and signed by its creator.