How do researchers in the Ethereum community view MEV?

EthSign founder Yan Yan, HashKey researcher Cao a new, Ethernet Square researcher Yao Xiang, DODO CEO, respectively, to share views on the radar bear MEV from their point of view.

Original title: “2021 Ethereum Community Meetup-MEV Panel”

On October 26th, ECN brought tog ETH er’s community partners (ETHPlanet/DAOSquare/Original Language Lane/Hiblock/Rebase/ BTC U) to hold the Ethereum Community Meetup in Shanghai. The atmosphere was active, with dry goods full, and various topics. Discussions were held, including EIP-1559, Ethereum 2.0, MEV, DAO, blockchain security, etc. ECN hereby organizes these contents into articles to share with you, hoping to enhance the community’s understanding and thinking about Ethereum.

The following is the second Panel of this event. We invited Pan Zhixiong, Yan Xin, Cao Yixin, Yao Xiang, and Leida Xiong to discuss the topic of Ethereum ‘s MEV (Miner Extractable Value). The following is the transcribed text of MEV Panel:

How do researchers in the Ethereum community view MEV?

Pan Zhixiong: Today we want to talk about a topic that is more difficult, or less valued-MEV. Today, we invited dedicated researchers who are working on the project to share with us some recent developments in MEV, some sharing, and some views from their perspectives. Let’s first invite the guests to introduce themselves.

Yan Xin: Hello everyone, I am Yan Xin, the founder of EthSign. I came to this Panel because, when I made investments, MEV saw a lot, and when we started EthSign, we also considered how to make hidden users better use Ethereum. So, I am very happy to be here.

Cao Yixin: Hello everyone, my name is Cao Yixin. I usually do industry research, and I collaborate with Teacher Yao on some research projects in the “Language Lane”, as well as MEV-related ones. I am very happy to participate in this discussion.

Yao Xiang: Hello everyone, this is Yao Xiang. We are also mainly doing research to study the public issues of some industries. It has nothing to do with investment, so it is more unique. Then, MEV is also a direction that I pay more attention to, but my personal research is relatively scattered, so I will do some exploration in this area with Yixin.

Leidaxiong: Hello everyone, I am Leidaxiong, the CEO of DoDo.

Pan Zhixiong: Thank you for sharing. Let me start by introducing some basic MEV companies and let everyone have some concepts. Many friends may know what it is, but it should be a relatively new term for many friends, right? I noticed that MEV is actually in early 2020. A very famous concept at the time was called Dark Forest, and there were a series of articles. It is mainly researched by Flashbots and their team. They discovered that the sequence on the chain can be used to attack, whether you go to the market in this sequence, or based on the sandwich attack (the sandwich attack we now know). We can see certain transactions first in the memory pool, and we control the size of this gas according to the order of this transaction. We can construct some other attack methods in front of it, or construct some attack methods after it. So this article in 2020 made everyone know about MEV. But in fact, I, let’s look at this history and see that in 2019, the Flashbots team has actually written some related research articles, but there is no final practice yet. But in 2020 DeFi Summer, everyone can feel how important the sequence of DeFi visits by the ether is.

DeFi 交易的顺序是有非常大的商机,所以说不同的矿工或者是科学家,都可以利用这些顺序来从中获取非常大的价值。那到 2021 年,Flashboys 这个团队就改名为 Flashbots。然后又同时推出了 MEV Geth 客户端,通过他们的一些技术的方式和一些设计,让整个以太坊交易排序变成了更市场化排序的规则。所以 2020 年越来越多的矿池加入到 MEV Geth 的使用,它们的使用量也越来越多了。那这个 MEV 的情况相对而言会好一些,比之前的会好很多。不会有很多矿工或科学家都来竞价之类的。然后,最有意思的是发生了一件 SushiSwap 类似的事情。有一个项目叫 Eden,他们很多业务都使用了 Flashbots 开源的一些东西,然后他们比 Flashbots 更早的去发行了 token。因为市场这个时候并没有相关的东西,导致 Eden 得到了非常大的市场关注和非常好的收获。所以,这个有点像 Sushi 之前对 UniSwap 做的事情一样。所以这个也是 2021 年在 MEV 事件里面非常重要的一件事情。在上周 Eden 决定要更新他们的代币经济学。Flashbots 团队对他们也是有各种各样的吐槽,因为他们觉得还没有想清楚,不应该这么早去做代币经济。这些就是目前我能看到的一些进展。但是呢,其他嘉宾可以来分享一下,从这个大家关注到的进展。以及我觉得闫欣可以聊聊对 Eden 的看法,我刚刚只是讲了一些简单的基础。然后,其他两位研究员可以讲他们的角度。然后 DoDo 可以从应用角度来出发,怎么看待 MEV 之类的,谢谢。

Yan Xin: Ok, thank you. Let me first talk about Eden and related to Eden. I have a few thoughts. First of all, we can kindly observe that in the Western powers, they are often politically correct and politically incorrect, and the two factions fight each other. Playing extremely fiercely, for example, Multicoin and Paradigm. But in fact, I personally support Eden very much, because in fact, the best part of token is to get the community to participate, and to be able to profit in the process of participation, and become part of your project, right? For example, if I stake 100 Eden, then I can use this Network, right? In fact, for users, Flashbots does not have a very good innerface. I am a retail investor, I want to use Flashbots, and then you directly give me the link to Github, I don’t understand anything at all. But Eden is actually solving this problem, and then Eden also really let, although there are not many users, maybe one or two thousand, but it does give these two thousand people more choices.

In addition, I think one of the best things about him is that he is practical in this way. I think the main reason is that he captures more computing power. It is that he may now claim that he should account for more than 50%. Um. After he has these computing powers, one thing he does is he divides the space in a block into several layers, which is like, for example, we take the first-class, first-class, and second-class seats of the high-speed rail. . They are like this too, anyway, one layer is Flashbots, one layer is Eden’s token holder, and then one layer is something. I think this thing is very advanced, and I also think it is very future-oriented. Because our method in the past was to let everyone bid in the wild. He was like a dark forest without any organization. However, I personally think that this kind of public resources should have some top-level design. Especially, for example, when there are more and more rollups in the future, some cross-chain bridges will become more and more. The bidding ability of this kind of transaction is not very strong, but it is also very important, because it is closely related to many things in the entire community. For example, if you have a problem with your data, it is very important that if you have to force him to be certain, if he cannot afford the price, then the security of the entire chain cannot keep up, which is a big problem. So I think in the future, someone will definitely propose a similar way, such as dividing into several warehouses, and then each warehouse has its own different functions. I just share so much.

Cao Yixin: Then I might look at the topic of MEV from a more macro perspective. I will think about why MEV is getting more and more attention after DeFi has become popular. My thinking is that DeFi itself has completed or solved a problem that is to use smart contracts to solve the trustlessness of transaction matching in the transaction process. If you take a step back, Bitcoin uses a blockchain economy to solve this stage of trustlessness in the transaction link. In this case, the matching and execution of the entire transaction does not need to rely on a new party to complete. But there is one of the earliest steps in the entire transaction-the discovery of trading opportunities. It turns out that in traditional financial activities, the discovery of trading opportunities may take a long time, requiring a lot of information sources and more specialized operations. Then now, the blockchain itself has a mempool, and all transaction information is presented there openly and transparently. This has both a good side and a bad side: the good side is that everyone can discover the trading opportunities in it, including good arbitrage and liquidation opportunities, so that the financial system can quickly run in with the information in the middle. Symmetry; in addition, the negative side is definitely that there will be a lot of it. Use this information to do some free-riding or offensive behaviors. Then, I think that in a broad sense, MEV refers to any value that can be extracted without permission from the information presented by the blockchain.

The definition I mentioned will be broader than the definition proposed by @phildaian, but I am willing to expand this value. Because this value has always been there, and it comes from the information of those things. But its circulation will change depending on the mechanism. If an arbitrage robot trades successfully on the chain, then this value is the source of income from an arbitrage; but if the arbitrage is taken away by a runaway robot, then it becomes a source of profit after free-riding . Its nature will change, but its essence will always exist on the chain. So what I have been paying attention to is those solutions. How does he guide the distribution of these values?

In my opinion, the current solution may be divided into three levels. Most of them are at the P2P protocol layer, such as those customized nodes made by Flashbots. The Eden just mentioned is actually similar. It divides the block space into different levels, allowing all robots or participants who want to get more cost-effective to compete for this block space through auctions. This is a way of benefit distribution. Another solution is application level. For example, those privacy DEXs, including Flashbots, have launched a privacy DEX called mistX, which is also one kind. There are also some application layers that actually use traditional methods to solve the high-frequency quantization that occurs in the middle of the runaway event. A commonly used method is called batchoption (batch auction), a high-frequency batch auction method, such as a solution such as CowSwap. But what I want to see is some solutions at the protocol level. Through some cryptography or privacy technology, the transaction information inside can at least allow the owner who discovers the value of this information to retain one of his own ownership. This is a solution that I want to see, but it may still need a lot of progress in this area. I will share these for now.

The root cause of MEV may be transparency. If it is transparent, once someone pays labor, searchers discover this value. For example, we discovered that the digital world is a new continent with many gold and silver mines. His wisdom or some other means found the location of a gold mine, and he ran over immediately. This information is made public, and everyone else can use it faster, or the person who has rent-seeking ability himself can grab this value in advance. It is hoped that this value can be better preserved through some privacy or software combined with cryptographic technology.

Yao Xiang: Both of them said very well. Yixin said it very comprehensively. It looks like every chapter of the paper, right? My first concern was actually about last year, and then at the beginning of the year, Tina also asked me whether to study this together. I said that I am not good at learning art, so I will help you spread the results. But I just follow Yixin’s words, it is a core issue, what do we think is his essence? There is a saying: Power leads to corruption, and absolute power leads to absolute corruption. So the right to sort, I’m not saying that it leads to corruption, because it’s in this consensus. I didn’t restrict you. We can do this. That means that people who can sort can do anything. matter. Then your power relationship is not equal, right? Then I can only hide this thing and let you not know what I am doing, right? Those routes, for example, a service that no longer exists, is called Tai Chi Network. Has anyone used Tai Chi at the scene? If you use Uniswap , you can connect to the Tai Chi node when you suspect that it might be frontrun. Because this no longer exists a mining pool promise, he will not force you, only when he produces a block, your transaction will go in, he will not broadcast to the entire network, but what is your tradeoff? That is, you only enjoy part of the computing power of the entire network, so your average inblock time will be longer. It is not a protocol level, it is still a moral level, that is, if I am frontrun, it is only me, and I promise you to provide such a service. This is actually a resistance to programs including Flashbots and Eden. This is much like a struggle of an idealist. Then can we do more extreme? Just like I wear this T-shirt today, I also chose it specially. What I wrote above is Being simulated with zero knowledge proof. Only cryptography can give us this weapon, right? If you don’t know what I am doing, you have nothing to do with me, right. Where I am going, you don’t know what I am going to do. If you want to know what I am doing and you have this right, then you can definitely block me in front. Morality is useless, and there is no point in talking about morality here.

So, there is one more thing that I think of, which is more fun, as I read a paper recently. They analyzed the order of each block of Bitcoin. You will find that if you use the generalized MEV definition, its ratio is very amazing. What do you think Bitcoin has in order? No, it’s not like that. Its proportion is amazing. I don’t remember the specific data, but it should be greater than 11%. It must be so. Why are there so many transactions in Bitcoin blocks that are not sorted by transaction fees? They discussed this issue a lot in that paper, but I haven’t read it all yet, so what do I want to say? There are many things that you don’t know. We are here to say that there may be someone laughing behind a computer in the distance. He said, you are really too ignorant to discuss things, you don’t even know that we are doing this. So I think, in the end, cryptography may give us weapons, and cryptography is a tool that we finally try to solve the problem fundamentally. thanks.

Radixiong: As the project party, we should be talking about some applications, but in fact, our thinking about MEV originated from some very metaphysical definitions. The first question raised is whether MEV is a bug or a feature. Vitalik will certainly not admit that this is a bug, but I think he is a bug. As for its origin, it should be said that trusted computing is a valuable resource, and if the miner masters it, he will collect taxes from it. This kind of tax can be collected directly or indirectly. For example, we have heard the story of high-frequency trading. I rent a place next to the server, put my machine, and I will hurry up. The money was actually handed over to the real estate agent next to the server. The real estate agent may also be the brother-in-law of the person who opened the exchange, so he taxed the tax through some indirect methods. Then we look back and wonder if he should accept it? In fact, I think he should not accept it in the consensus of everyone. We are layering, the application layer is the application layer, and the base layer is the base layer. You should not charge for this. It’s like saying that Google DNS resolution service providers can serve your website incorrectly, in order to invite you and ask you to give them money. So is it appropriate for him to do this? I feel that this is inappropriate for common sense, and it is not just. I always think that MEV is not so justice money. We have already paid you gas. As your daily operating expenses, you should not use your power to extort an extra sum. It’s equivalent, it’s kind of like squeezing the drum to absorb water, and I won’t stop until you squeeze until you have no profit. I think this is unjust.

Then speaking of it, most people actually have no room for resistance to this kind of injustice. But there are some Layer 2, which has room for improvement. For example, the sequencer on Arbitrum is the sequencer. Not only Arbitrum, but all rollup schemes, the sequencer now seems to be a standard configuration. Because once the sequencer appears, the MEV problem disappears and no longer exists. The subtext is that, as a provider of Layer2, I took the initiative to give up the extraction of MEV value. Just let the profit be given to everyone, and you don’t have to fight for it, and everyone should not be involuntarily involved. It’s a bit like saying that in the past, the teacher in charge said that whoever gave me more red envelopes, I would sit in front of the child; now the teacher in charge said that I will not accept gifts, and all of your children are treated equally. Of course I think it is a way to avoid involution, and then it is good for all parents and children. So, if we can’t resist, for example, my business is on Ethereum, then we have to resist MEV. In fact, there are two ways to fight MEV: one is that you run faster than others, and the other is that you let others know what you are doing. Run faster than others. This is a lot of work that Flashbots are doing. Then, make it invisible to others, in fact, you only need to go through the private transactions of the mining pool. The two are actually used in different scenarios. For people who really want to arbitrage, ranking first is the most important thing, so I need to pay a lot of money to use this solution. For ordinary traders, he does not have such a strong demand for arbitrage. He just hopes that he will not be attacked, so he should choose the second one and hide himself. So for DoDo, we provide traders with the second type, that is, you are not public in the mempool of the mining pool, nothing more. This can strike a balance between user needs and costs. That’s it.

Yan Xin: Oh, let me talk about it. In fact, I don’t quite agree with the idea that MEV is unjust. Because for example, you just mentioned a problem, you said that the user pays the mining pool, so he should pack your transaction normally, and the mining pool did not say not to pack it. In other words, people who want to use cryptography should use cryptography, just as if you want to use Tornado, you should use Tornado. But we still have to give a chance not to go. This is one of my thoughts, including something like Sequencer. How do you guarantee that Sequencer is very fair? This may require more cryptography, but this one has to be explained by the hourglass brother.

Yao Xiang: I think that, in fact, some of the issues discussed in the end are metaphysical. Metaphysically, we can easily draw conclusions. So my general method is to observe. How do you say it? Why the mining pool must accept MEV? This is very simple, because if you don’t accept MEV, you will have no clients. Because all the mining pools that support MEV will give more income to miners. Then this thing will happen, then what should I do? Right? In such a market, you can only make this choice, or you just say I won’t do it, right? is it possible? Impossible, right. At the same time, does the Ethereum protocol stipulate that this cannot be done? There is no such rule. I think this is the result of evolution at least to this day. In the current state, I think to a certain extent, it is a relatively stable and a balanced state. Rather, we want to use a better system to reduce this. We intuitively feel that we should protect those who make transactions. I also thought of one thing, that is, during the Chinese New Year this year, we had a conversation on clubhouse in February. At that time, Shaoping was the head of Sparkpool. Then there was one, which should be a trader, and he came to ask: “I think your mining pools are very unjust, change my order.” Then Shaoping said, then you go to rob others, do you think you are righteous? He thinks he is very righteous. Therefore, our views on this issue are indeed difficult to reach a consensus. So I think we are facing the status quo.

What can cryptography do? In fact, there are some solutions now. For example, we use this approach similar to a commitment review, right? I will commit to the transaction deadline first, and then open the next block. Of course, you sound like poor financial performance and sacrifice a lot of things, and if I don’t open it, then you have to add a very complicated mechanism. What if you don’t open it, right? Then we will say that we can use VDF, right? What does VDF mean? That’s, I’m sending a secret. This secret won’t be figured out anyway, so you don’t know what I’m going to do when you pack it, maybe ten minutes later, you know what I’m going to do. This sounds like this, too many things to give up, right? The compromise is particularly big. So in the end, if we want to have a more ideal state. I don’t have a solid support at the level of sorting, it’s my intuition. You are on Layer 1, and the less things you make, the better. Layer 1 is simple enough, and you move your business to Layer 2. It is whether you have various ways to solve the problem of disagreeing with the requirements between applications. Because the big problem now is that different applications on Layer 1 have different technical requirements, but everyone will be affected by the same thing. For example, NFT will cause gas war, and you guys make it so expensive for everyone. So I think that Layer1 can be concise enough, these things are put on Layer2.

Second, from my perspective, my feeling is whether the technology of zero-knowledge rectification can provide a solvable solution in the limited future, at least the future we can see. So how do you do it in the end? No one knows what the chain is going to do, he only needs to know what the final result will be. Generally speaking, I think it is to protect everyone, and then let the system develop in a relatively safe state. Then including my recent influence on MEV, I don’t care much about how you solve this solution. Because I think there are too many such paths, it is difficult for you to say which is good and which is bad? But one thing may be more certain, that is, MEV has a major impact on the underlying security. It has caused your previous security assumptions to be completely changed. You used to think that miners only have this income, but now they have new income. If the miner has new income, he may attack the network for this new income and increase his main motivation. This is a lot of money, and I found that I can get it. Why not do this? So, this is some of my thinking, and it is not a very mature idea. In the end, I may follow a new one, and we will throw out a research report. I hope this is a target, let everyone give suggestions, thank you.

Radar Bear: Then I still think it’s unjust, it’s still a very simple example. The teacher in charge received the red envelope and put the children in front. How do you say it? I feel that this is a kind of power rent-seeking right now, and of course we cannot stop power rent-seeking. However, I think the direction of our efforts should be to get rid of the MEV completely; not to seek, so now I have to give gifts to get my child registered. Then we have a market auction, and whoever gives a lot of gifts, I will let whose children go to school. I think this is wrong, you should find a way to put an end to this phenomenon. I prefer the Sequencer program. Of course, this may not be easy to implement on Layer 1, but this is not our research area after all, I just hold a point of view.

Pan Zhixiong: Actually, I want to talk about the second one, some of the latest developments in MEV and possible solutions, but Mr. Yao has already talked about some of them just now. So let’s talk about Eth2.0 related issues. Would you like to ask what do you think there may be a connection between MEV and Eth2, and what impact it may have? Will it increase? Or weaken the influence of MEV?

Cao Yixin: Actually, I just finished listening to the last report about the process of eth1->eth2 Merge. Actually, I still have a question that has not been answered. There are Validator and execution layer in 2.0. I don’t know whether these two kinds of nodes are running in separate roles, or whether they are running on the same role side and the same node. (A guest will answer during the period) That is to say, the validator is generally responsible for collecting transactions, then performing verification, and then packaging. Oh oh, then I get it. Then I think the difference here may lie in a change between PoS and PoW. With PoW, if you want to become a node, it has a very high hardware or technical threshold. That is, I want to enter the role of such a packager, and I have to invest a lot of capital first. Moreover, this capital is a form of hardware, and its exit may involve an exit risk. If I become a PoS, then I only need to have a lot of ETH, and I am a capitalist, so there may be more potential opportunities for collusion. Will everyone make a small sum of money locally, and then if the price drops, I short the futures on the off-exchange market, can I avoid the risk of my own loss? This is a question of my own random thinking. Regarding MEV, I think after PoS, its volume will be larger. Because the scaling problem may be temporarily resolved at that time, various forms of commercial activities, including DeFi, NFT, and other non-financial games, if more and more of these types of transactions are processed on the above, the valuable information in the middle is definitely More and more. Then this MEV is immeasurable. I look forward to seeing some progress in future development.

Yao Xiang: If a friend participated in our event last year, we were actually discussing the switch of this consensus at that time. After the last event, Merge’s work began to accelerate wildly. My feeling is that after the switch, the security assumptions of the entire network have undergone major changes, so that I have no way to analyze what will happen to it. But a few things are certain, that is, under the consensus of Ethereum PoS, you know when and who produced the block. This is a more dangerous thing. What do you mean, if you mine in PoW, you don’t know who the next block will be, that is, you cannot collude with one thing in advance; but under PoS, this happened. I know who the next person is, and I also know who the next person is. In a limited time, I know who should be the next one. Although you cannot violate the consensus, you can create more MEV opportunities, and even a few people before and after you can collude. I think this matter has become complicated enough that there is no way to discuss it. In the end it may be a market behavior. According to Radixian, this is an unjust thing, right. But for them, it doesn’t matter. I just need to strike a good balance between the negative impact on the Internet and my own interests. For ordinary people, I don’t know what happened. Even this behavior becomes hidden. Let me give you an example. Under PoW, if you want to do MEV, you will complete it in your own block, and you will not bet on whether you are right next. But in the PoS network, it’s different. The next one may be me. You don’t even know that it’s my two staking addresses. These two blocks are mine. You can’t see that this is an MEV, but you don’t even know that it’s my two staking addresses. Get up, it is. Of course this is my opinion, thank you.

Pan Zhixiong: Thank you, our MEV topic today ends here, thank you a few guests.

Posted by:CoinYuppie,Reprinted with attribution to:https://coinyuppie.com/how-do-researchers-in-the-ethereum-community-view-mev/
Coinyuppie is an open information publishing platform, all information provided is not related to the views and positions of coinyuppie, and does not constitute any investment and financial advice. Users are expected to carefully screen and prevent risks.

Leave a Reply