Popular Science | What is EIP-1559 arguing about?

Since the previous expansion series was a bit difficult to deliver (I felt that I had delved too much into technical details), let’s digress and return to the style of the earliest “blockchain romance” to talk about the now very popular EIP-1559. To be more precise, let’s eat some big melons about the quarrel between the two factions of Ethereum.

In fact, regarding this topic, I have already written about the Hong Kong consensus, the Segregated Witness upgrade, UASF, and the final BCH split in my Zhihu column "Blockchain Romance". This time, we will talk about almost the same story, but it is a story that happened on Ethereum.

What is EIP-1559?

EIP refers to proposals from the Ethereum community to make various improvements to the Ethereum algorithm. Some of these proposals have been adopted into Ethereum upgrades, some were not adopted for various reasons and were submerged in the history of Ethereum, and some did not even cause any waves in the community.

But EIP-1559 is clearly not one of those that remain unknown. Regardless of whether it is ultimately accepted or not, it has still caused a stir in the community.

So, what is this EIP-1559 that has caused such a stir?

Since EIP-1559 has now become a political issue, almost all introductory articles about EIP-1559 are biased because the explanations are not clear enough - those who support EIP-1559 are reluctant to talk about the hidden motives of EIP-1559; and those who oppose EIP-1559 are reluctant to talk about the original purpose of the EIP-1559 design.

So, in my opinion, there is not a single article that clearly explains the mechanism and design reasons of EIP-1559.

So, here, from a neutral perspective, I will explain what the core essence of EIP-1559 is, how it came about, and what it wants to do.

Lower transaction fees

In terms of purpose, it is a proposal that attempts to reduce transaction fees.

If you ask what the biggest problem with Ethereum is right now, nine out of ten Ethereum users will say it is congestion, and the remaining one will say that the confirmation time is too slow and the transaction fees are too high - in other words, it's still congestion.

Ethereum’s congestion is already a commonplace, and the most fundamental solution to congestion is capacity expansion.

The permanent solution to capacity expansion is called sharding, but this permanent solution seems to remain in the "next phase plan" forever; therefore, the community, unable to wait, came up with the rollup solution. Although sharding cannot bring unlimited output improvement, whether it is dozens or hundreds of times improvement (depending on the rollup project), it is a timely help for the current congested Ethereum. Of course, the above two solutions are not the focus of this article.

EIP-1559 has nothing to do with capacity expansion. It does not solve the capacity problem, but solves the congestion problem from another angle.

For example, the traffic congestion in Beijing is caused by too many cars on narrow roads. The fundamental solution is to build roads or to limit the number of cars. But from another perspective, although too many cars on narrow roads are the main cause of traffic jams, are they the only cause? No, uncivilized driving behaviors such as frequent lane changes, cutting in, and driving at low speeds in the fast lane are also causes of traffic jams. Such behaviors may make roads that are not congested become more congested, or make roads that are already congested even more congested.

In fact, if we regard all cars queuing up in an orderly manner during a traffic jam as an efficient use of traffic resources, then frequent lane changes and squeezing in, which are "taken for one's own benefit in the competition with other vehicles without information exchange," are inefficient uses of traffic resources. Therefore, in the case of congestion, that is, when traffic resources are very tight, using certain methods to urge everyone to use limited traffic resources more efficiently is actually a way to alleviate congestion.

And EIP-1559 is such a solution.

In Ethereum, if we consider “transaction/fee”, that is, “on-chain computing resources available per dollar” as the efficiency metric, then Ethereum’s efficiency is indeed low during times of congestion. The reason is Ethereum’s “first price auction”.

First-price auction

The first-price auction is actually the most general auction model we understand.

When transactions are congested, Ethereum uses the first-price auction principle to select transactions. In other words, "the highest bidder wins". This "win" refers to the opportunity to be packaged and put on the chain by miners - for example, if there are 5 opportunities to be packaged and put on the chain, if 10 people bid 1 to 10 respectively, then the miners will naturally choose to put the five transactions with bids of 6 to 10 on the chain.

So what about the remaining 5 transactions? They will be left in the transaction pool. If there are 5 more positions in the next block, and the bids of other transactions are less than 1, then these 5 transactions will be on the chain.

The above situation rarely occurs in reality - when there is no congestion, everyone will only offer a commonly agreed basic price, such as 1. At this time, ordinary people have no reason to offer a price higher than 1.

But when congestion occurs, for example, as mentioned earlier, there are 10 transactions waiting to be included in the chain but only 5 can be packaged. If everyone bids the same, then the choice is on the miners' side - there must be 5 transactions that will have to wait until the second round to be packaged.

At this time, if 10 traders all want their transactions to be included in the chain, they have to raise their transaction prices. But the question is how much should they raise? Because Ethereum does not have a real-time public bulletin board to tell you how much others have quoted, so everyone will think, "If I raise it to 2, what if others think so, then I can't guarantee this round of chain?"

It's like when there's a traffic jam, if everyone wants to be faster than others, they will try to change lanes, squeeze in...which makes the jam even worse.

But there is no upper limit to the price increase. If everyone thinks that others will increase it to 2, then if you want to put it on the chain immediately, you have to increase it to 3, and so on... In addition, due to the difference between Ethereum and centralized platforms, others cannot see your price increase, and naturally cannot accurately judge other people's current bids. Therefore, everyone can only bid by guessing other people's bids.

Have you ever had the experience of taking Didi during rush hour? It shows that you can get a car faster by increasing the price, but after you increase the price by 5 yuan, you wait for 10 minutes and the car still doesn’t come. At this time, you don’t know whether it is because there are no cars or because you didn’t increase the price enough. Canceling at this time means that you have waited in vain, so the remaining options are to wait or continue to increase the price.

The end result is that you may get a ride, but the fare you pay may be more than the price you need to pay - for example, maybe you can get a ride without adding any fare, or maybe you have to add 20 yuan to a ride that you could have gotten by adding 10 yuan.

It is similar in Ethereum, and you cannot lower the transaction fee in Ethereum, because even if you lower the transaction fee, miners can still publish transactions with high transaction fees. Another consequence of this setting is that if you are worried about misjudging the transaction fee and are unwilling to increase the transaction fee, it may cause the transaction to be unable to be put on the chain for a long time.

Therefore, users who are eager to put transactions on the chain may pay higher transaction fees than necessary in this competition with incomplete information. In other words, in the Ethereum network, which is already congested and inefficient, the cost of transactions has become higher due to the premium of competition.

Flat Price Auction

The starting point of EIP-1559 is to change the above situation through the form of first-price auction - this method is a unified price auction.

Let’s continue to consider the previous situation. 10 people bid 1-10 respectively, and the miner chooses 5 to package.

In a unified price auction, each person who successfully puts a transaction on the chain does not need to pay his own bid, but only needs to pay the bid of the person with the lowest bid among all the packaged transactions, and then the miner gets the money as a reward. In this way, people who bid 6-10 still get packaged and put on the chain, but everyone only needs to pay 6.

This proposal is good for users. For them, transaction fees are reduced, and it is very psychologically friendly. If you think about it, if it is the former auction method, the person who bids 10 will definitely regret it after seeing the final result - he should have bid 6 instead of 10, and paid 4 more in vain. But the uniform price auction will not have this problem. Even if you bid high, you only need to pay the bid of the person with the lowest bid in the end.

For miners, this auction method is definitely not good, because obviously their income has decreased. But in fact, in the long run, this method is not so bad. First, when there is congestion, miners' income will still increase, but they will not earn as much; second, in fact, in the above scenario, the extra money earned by the four people who bid 7, 8, 9, and 10 is actually earned by taking advantage of excessive competition caused by information asymmetry: if the bids are transparent and can be changed in real time, after seeing each other's prices, the five people only need to change the price to 6 to successfully upload to the chain.

Therefore, if this situation continues, users may eventually create a platform for them to exchange bid information with each other, or switch to a chain that allows them to exchange information transparently, so miners still cannot make money from the information asymmetry. Instead of doing this, it is better to simply give up the money in advance, which is a user-friendly consideration.

But there is a problem with the uniform price auction. Let's consider the following situation:

After packaging the previous block, there are still transactions with bids 1-5 left in the transaction pool.

At this time, if the miner packages these 5 transactions, the transaction fee he can receive is 5, because the minimum bid is 1.

But at this time the miner came up with an idea - if I generate a transaction with a bid of 2 and pack it into the block, then my income will become 8, because the minimum bid is 2.

This is the problem caused by the unified price auction - miners may generate transactions to raise the price. In the case of network congestion, these transactions generated by miners occupy the already limited resources. For example, the transaction with a bid of 1 that was originally able to be put on the chain now has to wait for one more round.

Basic fee + tip

But in fact, after using the unified price auction, miners still benefit from the users' bidding - because in fact, even a bid of 6 seems too high in hindsight. If ten people can sit down and discuss for a while and make a "simulated bid", then those who bid 6-10 will say to those who bid 1-5: "We are willing to bid higher, so your bid can only be put on the chain in the second round. Why don't we just not bid such a high price for the sake of competition - we bid 2, you bid 1. If we use this kind of negotiated bidding method every time in the future, miners will not be able to keep making so much money from us!"

Most people agreed, and they formed a price alliance:

If congestion occurs in the future, the transaction fee will increase by a fixed value; but if there is no congestion, the transaction fee will drop back to normal levels.

But some people expressed dissatisfaction:

“I just have money, and I want to spend money to get my transactions on the blockchain sooner, why can’t I do that?”

Others said of course it’s okay, but the result of you spending money is that everyone will get caught up in internal competition, and in the end even those who don’t want to spend money will have to pay extra fees.

Therefore, a basic fee + tip plan came into being:

Everyone privately agrees on a price based on the congestion situation, and everyone abides by this agreement to form a price alliance so that in a congested seller's market, miners cannot make money from everyone raising prices in competition.

Then, if someone is really anxious to get on the chain, everyone will tacitly agree that you can do it by giving tips privately, and on the surface everyone will still maintain a price alliance.

Therefore, in order to prevent the above situation from really happening, where users flip the table and say "Fuck your auction, I'm only paying this much, take it or leave it", Ethereum chooses to compromise: because although Ethereum is large, there are also many competitors eyeing it. Now it seems that Ethereum's resources are a seller's market and supply exceeds demand, but if high transaction fees make Ethereum an "uneconomical choice", then it is very likely that Ethereum will become the party that begs traders to send transactions on the chain in an instant.

Therefore, before this happens, the way to prepare for a rainy day is to come up with this plan before the traders do.

"You guys don't have to form a price alliance anymore. We won't be auctioning anymore. The price will be fixed according to your plan. It's called the basic fee."

"And you don't have to tip secretly, just give it directly. We have stipulated that tips will not affect the basic fee, and we will not raise the price just because the tips are high."

Destruction basic fee

However, the above mechanism still has some problems, because the aforementioned miners can still send transactions to increase the basic fee. When the network is not congested or the congestion is not serious, miners can always generate some transactions to create the illusion of network congestion, thereby increasing the basic fee. In other words, miners should be able to design an algorithm and then calculate when and how many transactions to generate to make themselves the most profitable based on the number of transactions in the network.

So what to do?

There may not be many ways to do this in reality, but in blockchain, there is indeed a way:

Miners do not receive the basic fee, but instead destroy it. If the basic fee needs to be destroyed, then it is not feasible for miners to issue transactions themselves - in the past, it was just a matter of transferring money from one hand to the other, sending some transactions and then collecting the transaction fees after the transactions were uploaded to the chain; now issuing transactions has become a real waste of money, and the increased basic fee will not fall into their hands.

The word "destruction" may seem a bit strange, but it is not new in the blockchain.

What everyone is more familiar with is Bitcoin's block reward, which is also called coin minting. The essence of coin minting is that all current coin holders dilute the price of their own coins to attract new users and miners to join, thereby increasing the value of the entire blockchain.

The essence of burning coins is actually to use your own coins to distribute dividends to all coin holders. Because burning coins will cause deflation, and deflation can increase the price of coins, and the increase in coin prices means popularity. No one will think this is a bad thing for users, miners, developers, or the entire industry. All problems such as insufficient liquidity and decreased transaction volume are not worth mentioning compared to this benefit. Therefore, many blockchain teams have long regarded the regular destruction of some of their own coins as a way to increase the price of coins.

This time, in Ethereum, a system that was originally inflationary, this proposal was a complete shot in the arm for the entire community, and almost no one could resist the charm of the word "deflation".

But the only exception is miners - because what is destroyed is the transaction fees originally given to miners, which is equivalent to using the miners' money to distribute dividends to coin holders...

Of course, from another perspective, this cannot be completely said to be the miners’ money, because no one says that Bitcoin’s transaction fee model is the only truth. This is just another transaction fee scheme, that is - part of the transaction fee is used to distribute dividends to all coin holders, and miners only take the tip part.

In a way, it's like you're a waiter at a very busy restaurant: you get a base salary whether you're busy or not. Then, because of a big event, the restaurant's business has recently increased and it's full every day, but the increased transaction volume has nothing to do with you, but it increases the restaurant owners' dividends. However, it's not all bad for you, because you can still earn more tips from more transactions.

This... sounds reasonable - if we position miners as waiters in a restaurant.

What do the miners think about this?

The Situation of Ethereum Miners

Before talking about the opinions of miners, we need to explain what miners do and the status of miners in Ethereum.

Algorithmic consensus and social consensus

Theoretically, for all blockchains that use PoW, miners are the core of the blockchain because miners actually maintain the ledger generated according to the original algorithm, which is the consensus. Let's call it "algorithm consensus".

The fundamental reason why we use blockchain is that we hope to use this "algorithmic consensus" to replace "social consensus", because "social consensus" is the consensus of people, and human consensus needs to be based on trust, and the cost of trust between people is usually very high. In comparison, once the "algorithmic consensus" is determined, it will be automatically executed, saving the cost of supervision and conflict in "social consensus".

But from another perspective, we live in society after all, so "social consensus" exists objectively. The so-called "social consensus" is the view of most people on a certain matter.

So, we are faced with a somewhat philosophical question:

What is Ethereum?

In other words, for all blockchains, we are always faced with a soul-searching question: "If this chain is now forked, and one side conforms to the 'algorithmic consensus' and the other conforms to the 'social consensus', then which chain is this chain?"

In fact, this is a very realistic question, because every time we have a hard fork, we are faced with this choice. And the conclusion is that in most cases people will choose social consensus.

On the one hand, at least for now, "algorithmic consensus" is far less popular than "social consensus" - although "code is law" has been discussed for so long, it may still be only some geeks who truly believe in it. On one side is the correct thing given by the algorithm, and on the other side is the thing "officially certified" by the founder, core team and investors. Most normal people will choose the latter.

On the other hand, the "algorithm consensus" of blockchain is actually a kind of "social consensus", because whether it is Ethereum or most of the public chains that appear later, the consensus reached by people is not based on a certain algorithm or a certain code, but the recognition of the white paper, the founding team and the future prospects, so it is also a "social consensus". Using a certain algorithm to reach a consensus at a certain stage, and in the future, when the founding team updates the algorithm, the new algorithm will be adopted, which is actually part of this "social consensus".

Therefore, in Ethereum, although miners in principle reach an "algorithmic consensus" for all users every day - they sit together every day to vote on newly released transactions, then stamp and publish them, as if the correctness and authenticity of the entire ledger is the result of their consensus. In fact, they are just part of a larger "social consensus" - the founding team and developers set a rule, a room, and let the miners sit together for meetings and agree that the results of their votes and stamps are correct and true. Whether it is the community, developers, users, investors or exchanges, the consensus they actually believe in is "the founding team will definitely select the right people and use the right methods to reach an algorithmic consensus", rather than the "algorithmic consensus" itself.

Moreover, in fact, the status of the two is a relationship of one rising while the other falling:

If the founding team is not active, does not frequently modify the "algorithm consensus" or does not emphasize its own "social consensus", then over time, the "algorithm consensus" will become more deeply rooted in people's hearts, such as Bitcoin.

On the contrary, if the founding team is very active, always emphasizes "social consensus" and incorporates the modification of "algorithm consensus" as part of "social consensus", then naturally the "algorithm consensus" will become very weak.

Ethereum Miner Status

Therefore, because Ethereum has a very active and energetic founder, Vitalik Buterin, and a very outstanding team of researchers and developers, the miners who reach the "algorithmic consensus" in Ethereum are themselves in a subordinate and rather marginal position.

In a sense, miners are not the main body of consensus. They are just people who are temporarily selected to run Ethereum in a larger consensus on Ethereum reached by the community, developers and Vitalik Buterin.

If you think about it this way, then there is nothing wrong with thinking that miners are equivalent to waiters who earn a basic salary and tips.

However, in addition to the low status of miners due to the strong "social consensus", the status of miners in Ethereum is even lower than that of other PoW chains. There are two reasons:

First, because Ethereum's PoW algorithm is Ethash, which is a memory-dependent hash algorithm, and it is difficult to design ASIC mining machines for memory-dependent hash algorithms, GPU mining machines are still the mainstream of mining to this day. In other words, the threshold for ordinary users to enter mining is still lower than that of ASIC mining machines. Although it is said that the culprit of PoW centralization is mining pools rather than mining machines, compared with miners using ASIC mining machines, GPU mining miners are obviously less organized and centralized than Bitcoin miners. Therefore, although the main body of Ethereum miners is definitely large mining pools, the participants of large mining pools are retail investors, and compared with ASIC mining machines, the interests of retail investors are not as strong and consistent as those of ASIC miners, especially GPU miners actually have other alternative currencies to mine. Therefore, the constraints of mining pools on retail investors are not that strong, resulting in mining pools having less confidence in front of developers.

Secondly, Ethereum has had a "difficulty bomb" since its inception. In fact, you don't need to worry about what this "bomb" is. In fact, Ethereum's initial "social consensus" and "algorithm consensus" both stated that "Ethereum will eventually change from PoW to PoS." Later, although the "difficulty bomb" was repeatedly postponed, that is, the "algorithm consensus" changed again and again, the fact that PoW will eventually be replaced by PoS has not changed. In other words, from the very beginning, Ethereum miners have agreed to a consensus that "I am only temporarily managing Ethereum and will give way to others in the future."

So, the miner is not only a waiter, but also a temporary worker.

The essence of EIP-1559

Putting all of the above together, let’s imagine Ethereum as a centralized Internet company, and then restore the logic proposed by EIP-1559 from the boss’s perspective.

1. Well, users complain that the transaction fees are too high, but there’s nothing we can do about it. That’s how blockchain platforms work, you can use it if you want or not.

2. What? A competitor has already appeared, claiming high output without congestion? That doesn’t matter, we will also publish a roadmap - sharding arrangement.

3. Competitors have already started to grab the market and many users have already wanted to switch platforms due to congestion? It’s too late to shard, so let’s arrange the rollup first. Let everyone know that the rollup is coming soon, with a 100-fold expansion.

4. But I still feel uneasy seeing the loss of users and other projects fanning the flames of poaching. So let me see what else I can do?

5. Let me see the transaction fees everyone paid... Hmm? How did the miners make so much money? It's clear that you can get on the chain by paying 6 yuan, but some people paid 10 yuan in transaction fees. This user experience must be poor! No, this is obviously caused by the miners hiding the quotation information from the public. Miners are just a group of workers. Using information asymmetry to encourage users to drive up prices is not a loss of public benefit for private benefit, how dare you? ! The miners should spit out this money to improve the user experience - change the first price auction to a unified price auction.

6. Wait, if we change to a unified price auction, miners can still earn money from users bidding against each other. When there is congestion, it used to cost 2 yuan to be put on the chain, but now it has been raised to 6 yuan due to competition. This user experience is still not good, and miners should also spit out this part of the money - we don’t do auctions anymore, users are the gods of Ethereum, why compete with each other? Change to charging a fixed basic fee based on the congestion situation, and then for users with special needs, they can give tips to increase the speed of on-chain.

7. But in this case, miners seem to be able to do something behind my back, and deliberately send transactions to cause congestion to increase the basic fee. This seems a bit difficult to solve... That's right! We can just let the basic fee burn, which solves the previous problem, and will also cause deflation and increase the price of the currency, which is perfect!

8. No! It’s even better than perfect - this solution will be a perfect match for PoS in the future. Now, it actually burns miners’ money to subsidize coin holders, and it can also encourage everyone to participate in PoS staking, so that everyone can evolve to PoS in a natural way. Everyone is a beneficiary! As for miners, what are miners? Do miners dare to rebel?

From the boss's perspective, this logic is impeccable.

In other words, from the perspective of management consulting, if my boss came to me after point 4 and asked me what to do if the transaction fee is too high and the user experience is poor, and if I regard miners as employees of Ethereum, then I would also come up with three suggestions: 5, 6, and 7 - because it is obvious that as an employee of the company, you don’t do your job and take your salary and tips, but always want to make money from users by letting them compete with each other based on information asymmetry. What’s wrong with that? In the past, the boss tolerated it because it didn’t affect the user experience, but now it affects the user experience and even causes the company to lose to others. Is there any law that doesn’t punish you?

But all of the above logic is based on the premise that “miners” are employees of Ethereum…

But is this really the case?

——————————————————————————

Of course not. As mentioned before, miners are actually the real masters of Ethereum in the "blockchain logic".

From the perspective of miners, they are not someone’s subordinates or hired by someone - they are one of Ethereum’s most important partners, because they provide the most important computing power in the PoW blockchain:

“Although we don’t usually talk, look at Bitcoin. Isn’t computing power king in PoW blockchain?”

It’s like in the minds of miners, they have always been the legal wife and the good helper of Ethereum. In the past, they always said, “Well, my husband treats me very well, just a little bit chauvinistic, but that’s fine, we don’t need to show up in public, anyway, husband and wife are one.”

So, it’s okay to be excluded from almost all decision-making.

The husband is obsessed with praising other women every day, publishing articles that PoS is better than PoW, and then pushing for the conversion to PoS in a high-profile manner, which is equivalent to finding a mistress and preparing to make her his wife. The miner feels bad, but considering that he does have a bad background and there is an agreement on the "computing power bomb" after all, he just endures it.

Even relatives and friends in the community often proposed proposals that infringed on the interests of miners, such as ProgPoW, but the husband did not say anything that time, and after a long time the miners just pretended that nothing had happened.

But this time, EIP-1559 suddenly came out, which not only deducted the monthly wages of miners, but also gave the deducted monthly wages to the mistress. Even the husband publicly supported it, saying that "this change will bring huge benefits to the family."

The most important thing is that from the beginning to the end, no one asked for the miners’ opinions - all outsiders and relatives expressed their support and directly regarded the matter as a fait accompli. They even gave advice to the miners while looking forward to the bright future of Ethereum.

Then the miners suddenly realized:

"I thought I was the legal wife, but it turns out you've always treated me like a maid..."

The (Potential) Miners’ Fightback

At this point in the melodramatic family ethics drama, the ordinary miners should have woken up and started to take revenge.

But what can miners do?

Delete the database and run away, change the number of your own accounts, or just be passive and lazy?

These can indeed be done, but these are all life-or-death threats, equivalent to taking a pair of scissors to fight with Vitalik Buterin and the core development team who represent the official Ethereum. If miners want to do it, they have to be ruthless and do it together, and once they do it, it will be a lose-lose outcome. It is precisely because of this that the authorities are not worried about the threat from miners. First, although doing this is a lose-lose situation, as mentioned before, Ethereum's "social consensus" is actually stronger than its "algorithm consensus". Although miners can seriously hit the "social consensus" in this way, I believe that Ethereum supporters will soon be able to form a new "social consensus", such as building a PoS chain and rolling back the state, but the miners' "algorithm consensus" will be completely bankrupt, and miners cannot afford such consequences. Second, miners are distributed and have different minds. In addition, miners are actually represented by mining pools, and they cannot completely control their miners. In addition to the first point, even if some mining pools raise their arms and say "Are kings, princes, generals, and ministers of different races?", as long as there are miners under their command who disagree, the final result is likely to cause a short-term fork, and this result is like stabbing the other party's hand with a pair of scissors, and then being caught and beaten to death. This is obviously not what the big mining pools want.

So, is there any way for miners to not harm their own interests but cause damage to Ethereum's interests, and then use this handle to threaten Ethereum officials?

Actually, there are.

Fork

Forking is of course the most natural choice for miners - this is how the blockchain world works, if you don’t agree to the route fork, just do it.

There are actually quite a few articles in the community saying things like “miners don’t dare to fork” and “no one will use it even if it’s forked”. In my opinion, these are just self-encouragement.

Miners can fork, of course. Although the result of the fork is unpredictable, it is actually an official problem rather than a miner problem. Whether it is a lesson learned or a precedent, the BCH example is there. If miners feel that their interests have been infringed, then they know that their final option is to fork. Supporters of EIP-1559 can say, "Look, the big Bitcoin mining pools tried to compete with Bitcoin in a big way, but failed. Look at the current market value of BCH, which is only 10% of Bitcoin." But the same thing from another perspective becomes: "At worst, we can be the leader of a chain with a market value of 10%, which is better than being treated as a maid in Ethereum."

It is true that if miners fork Ethereum, it will just switch to PoS, but that being said, if miners really fork today, are Ethereum developers and the community really ready to seamlessly switch to PoS? After all, the current Casper algorithm cannot be used as a pure PoS algorithm on the blockchain.

Moreover, it is purely self-deception to say that miners dare not fork. When Ethereum switches to PoS, will you let so many Ethereum mining machines starve without forking? Or which other coin can make their investment back? By then, there will be a 10000% chance that a forked coin of Ethereum will appear, but the size is unknown. So if miners want to fork because of EIP-1559, it is just doing something they have to do in advance.

Therefore, for miners, even if the fork does not happen now, it does not affect them from now on. They can prepare for the forked coin - attracting ecological projects, improving algorithms, promoting and waiting for opportunities... These are not inconsistent with the temporary adoption of EIP-1559. But at the same time, has Ethereum made these preparations for the miner fork? Let the temporary worker manage the warehouse, and then tell him that the contract will not be renewed and the salary will be reduced. Are you really not afraid that he will start a fire?

Form a price alliance

As mentioned earlier, EIP-1559 is essentially a price alliance initiated by Ethereum officials on behalf of users to take into account user experience. The intention is to encourage everyone to pay the transaction fees they can accept according to the reference price when there is congestion, and not to raise prices to give miners more opportunities to make more money.

But the other side of this problem is that miners can also form a price alliance during congestion - for them, the base fee will be burned, so for them, they can ask the trader "Unless you give us enough tips, even if you give the base fee, I will not accept your transaction."

In other words, it is a boycott of EIP-1559 - I don't care about the base fee, I only know that what I can get is the tip, so I will still collect the tip according to the previous standard. If your transaction tip does not meet the previous standard, then I will not accept it.

Obviously, the two sides could not agree on the price at this time - the transaction fee that the user was willing to pay was originally 5 yuan, but now it has become a basic fee of 1 + a tip of 4, but the miner said I want to charge 5 yuan, and now it is only 4 yuan, I won’t do it.

The final result is obvious. It depends on which side compromises first, or whether both sides take a step back.

In the end, it is more likely that miners will take a step back. After all, maintaining this price alliance means losses for both parties, but the loss for users is that their transactions cannot be put on the chain temporarily, and many users do not really care about this, or would rather put the chain on later than pay an extra 25% transaction fee. But for miners, maintaining this price alliance means losing a lot of money from transactions that do not pay enough money in tips, and miners are not a monolithic group. If you don’t make this money, others will.

But in fact, miners may not gain nothing - because there are many types of traders, and there will definitely be users who are sensitive to the immediacy of transactions and will increase their tips to ensure that their transactions will be packaged no matter which miner receives them.

So, what happens next depends on how the two sides play. But in fact, it doesn’t matter what the final result is. What’s important is that if the miner firmly acts as “we don’t want our income to decline” from the beginning, it’s hard to say whether the miner’s income will drop in the end, but the user’s transaction fees actually increase instead of falling, because it means paying more basic fees.

This is obviously an act of killing one thousand enemies and losing 800 themselves - after all, miners will definitely be criticized for monopoly and centralization, and the increase in transaction fees will definitely cause Ethereum to lose users, which will ultimately affect the price of Ethereum.

But from another perspective, if miners were excluded from the future of Ethereum and even the current Ethereum ecosystem, then why should they care about the long-term interests of Ethereum? From your own perspective, regardless of whether you used EIP-1559, whether your original intention of EIP-1559 is to reduce transaction fees, I just care about not reducing my income. Isn’t this natural?

Miner's attack

Someone has actually said the above two points, but this is a problem that both sides ignore.

Here, I am not referring to the risk of 51% attacks, because as I said before, 51% attacks are the choice to kill the ball.

I'm talking about the risk of other attacks that miners can do.

Someone here may ask:

"Wait? Is there any other attack risk for PoW besides 51% attacks?"

Of course, there are more attacks except for 51% attacks, and many of them are actually something that has happened or has happened in reality.

First, it is active attack, that is, it does not include normal transactions for a period of time.

This kind of attack occurred in FOMO3D. At that time, in order to take away the rewards from the FOMO3D bonus pool, the initiator wrote a contract with high gas. Then, when others wanted to buy, they used high transaction fees to induce miners to package these high gas transactions to fill the entire block, making it impossible for others to buy FOMO3D.

Last year's March 12, MakerDAO's liquidation incident was a non-human active failure. Due to the sharp drop in Ethereum, it triggered a large number of liquidated assets, and there was a large number of Ethereum transaction demand. This was not taken into account in the MakerDAO asset auction mechanism, so no one bid within the limited time, resulting in a large number of assets being sold to some default asset custodians at extremely low prices. Although this is not an active attack, it is obvious that asset custodians can actually use this method to conduct active attacks to obtain profits.

Second, it is censorship attacks, that is, attacks that do not package certain specific transactions, which is equivalent to active attacks on specific transactions.

This attack has not yet caused very serious consequences, but it can be fatal after the deployment of the optical rollup, because the security of the optical rollup is based on "if someone publishes a false status, someone must provide 'false proof' on the main chain for a period of time and take away the counterfeiter's deposit." If the message providing the false proof is censored, the security of the optical rollup will be gone because someone can commit fraud on the side chain without punishment.

How to conduct a censorship attack? In fact, some people have proposed a practical method - the counterfeiter deploys a smart contract to bribe miners on the chain, and then publishes a reward transaction that can take away part of his deposit (less than the deposit lost by exposing the fraud) as a transaction fee. It stipulates that if no one challenges during the challenge period, the miners can take the reward when publishing the transaction in the future. Then, under the role of this reward, if miners are all profit-seeking, their most rational choice is to review any transaction that provides "fake proof", because then they have the opportunity to get the reward after the challenge period is over.

The third is not an attack, but an arbitrage method, that is, front-running.

In the current DeFi ecosystem, many decentralized exchanges adopt the AMM model, that is, the pricing of currency transactions is determined by the supply and demand relationship, that is, they will always fluctuate by the influence of new transactions. Therefore, for example, if I want to buy currency B with currency A, although it is now 1 for 100, it is very likely that this ratio will not be the case when trading, so I will set a range, for example, as long as the transaction price is between 1:90 and 1:110, I will accept it. Of course, the smaller the range, the better, but if it is set too small, for some coins with large currency price fluctuations or when the currency price fluctuates violently, it may not be possible to complete transactions.

Since the issuance of transactions itself is open, front-running is given room for front-running: if a large customer suddenly wants to buy a bunch of B coins, then the price of B coins will definitely surge after this transaction. Then the person who learned about this will make a fortune by buying B coins in advance and selling them after this transaction. Of course, the amount of B coins cannot be too much, otherwise the price of B coins will exceed the range of the large transaction.

Since the order of transactions and transactions on the chain are decided by miners, so if you want to achieve front-running, you need to bribe miners and add their transactions before that large transaction.

Of course, buying miners is not insured - the more insured way is to get the miners themselves. In fact, miners can go front-running as long as they receive large transactions, which is how to maximize their own profits.

——————————————————————————

The above three attacks can be done by miners, and in fact, if we assume that miners are just seeking profits, then many rational miners will inevitably do it.

Therefore, some people are already doing it.

So, what prevents miners from doing these things?

This is because miners are not just selfish, but also have "altruistic" behavior. In addition, in addition to their interests on the chain, miners also have interests outside the chain, such as reputation and long-term interests.

Or to put it simply, it is because miners think that their obligations for Ethereum and their longing for the future, and their simple moral concept of "not doing evil".

Therefore, everyone believes that on the one hand, they suppress the profits of miners in Ethereum, deny their obligations to Ethereum, completely exclude them from the future of Ethereum, and say "You are just temporary workers." On the other hand, they stigmatize miners and say they are hindering the development of Ethereum. What will this behavior lead to?

Obviously, the result of this behavior, that is, the possible consequence of EIP-1559, is that miners are no longer bound by this "altruism" and simple moral concepts, and act on the standard of simple profit-seeking. The first problem is front-running - "Since you don't give us transaction fees, why don't we grab it based on our ability? You are unkind and I'm unrighteous!" The next question is the censorship attack - can rollup be deployed with confidence when miners may not cooperate?

Moreover, the most important point is that miners can do the above three methods at the same time:

On the one hand, fork, or prepare for fork.

On the one hand, this adversarial attitude is adopted on Ethereum today:

"You forced me to lower the transaction fee? Sorry, I can't lower it, we have to support our family. What is the basic fee? We only accept tips."

"You want to deploy a rollup? Sorry, I don't understand. I really don't know what censorship attacks or something. I haven't heard of it. The problem must be not the reason for my censorship."

"Why are front-running more common? I don't know, the order of transactions is completely random. Maybe the transactions in the mines I dug happen to have one transaction ahead of large transactions."

Then, when it really forked, the miners can say:

"Welcome to the real Ethereum! Low transaction fees, rollup, no front-running!"

This is a key hidden danger behind EIP-1559 being pushed now-

In the eyes of supporters, especially developers, unless the miners can become tough and form an alliance to attack 51%, we will upgrade EIP-1559, and they will eventually be able to accept their fate for their own interests.

But in fact, if EIP-1559 is deployed, then miners can actually do more than just 51% attacks. Moreover, many things they can do will not deviate from their own interests at all, and can even increase their short-term benefits...

As for the long term... you have all excluded PoW miners from the future of Ethereum. So what if I care about PoS in the future?

Summarize

Seeing this, my tendency is already very obvious - like the previous article about Bitcoin core developers and miners, I am still on the side of miners. The reason is very simple, because as a theoretical researcher, it is difficult for you not to stand on the side of miners - because in theory, miners are the owners of blockchains. Although in fact, "algorithm consensus" is only part of a larger "social consensus", but if there is no "algorithm consensus", what is everyone's "social consensus"? Code is law is rarely taken seriously now, and it is not realistic, but if we do not respect the rules stipulated by code and algorithms, what other blockchains do we need? It would be better to use a centralized system.

: : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : :

Therefore, either, they are so arrogant that they have begun to regard themselves as the only truth; or, this is just a political struggle, and the high-sounding words they say are nothing more than using their own expert identity and status to guide public opinion and thus use any means to crack down on dissidents.