|
As the merge gets closer, more news will start to emerge, and many ill-informed crypto journalists will publish more articles. This signals to more people that the merge is indeed close, which will lead to more questions and discussions about PoS, while also repeating the same topics and misunderstandings. I have already seen this to a certain extent when the successful Kiln testnet merge was announced last week, and I expect to see more of the same in the future. So here are some of the common points I make, just like I did in my previous article (which was more about the overall situation of Ethereum), I will boil it down to a few simple points. I plan to list these points as I see them raised, and I encourage you to do the same, and please correct me if I'm wrong or suggest any additions. What is a merger? More information can be found on the website ethmerge.com, so I won’t spend too much time on this part. After the merger, Ethereum will be secured by PoS consensus instead of PoW. After the merger, it will not be "Eth 2.0". There is no such thing as ETH 2.0. It is an outdated term. If you hold ETH now, you don't need to do anything. You will still hold the same amount of ETH after the merger, there will be no such thing as "ETH2 tokens", you don't need to migrate anything, etc. Everything remains the same, only the underlying consensus mechanism has changed. It is called a “merge” because the upgrade will merge the beacon chain (consensus layer) with the current chain (execution layer) and discard the PoW part of the execution layer. “Consensus” is just a fancy word for ensuring that transactions are ordered and that the ordering is economically guaranteed not to change. PoW and PoS achieve consensus through different means: PoW: “The cost of disrupting the block order is too high, and it is more profitable to participate according to the rules.” PoS: “It’s too expensive to disrupt the block ordering because if I do that I lose all my staked assets.” PoS itself will not significantly reduce gas fees because only the consensus mechanism changes.
Why merge? Maintaining security will be less expensive because less energy will be consumed to achieve consensus. PoW: requires miners to at least cover the cost of all hardware and energy usage, otherwise no one will mine. This requires a large amount of issuance and quick sale in exchange for fiat currency to pay the bills. PoS: Just give stakers some yield to make people willing to deposit assets instead of investing directly elsewhere. There are no expensive bills to pay other than an ordinary computer and a stable network connection, so the yield only needs to reflect the opportunity cost and risk involved.
More sustainable:
The security of a blockchain is basically proportional to its coin value. This is true for both PoW (more valuable token rewards = more reason to follow the rules = more miners = harder to disrupt consensus) and PoS (more valuable staked tokens = more reason to follow the rules to avoid losing staked tokens).
Newly issued tokens essentially take value away from all holders and redistribute it to certain people. All else being equal, converting tokens to fiat currency extracts value from the network.
This will pave the way for many scaling solutions, including data sharding, statelessness, light clients, and more.
By separating responsibilities between the execution layer and the consensus layer, it can help reduce some of the complexity of code updates.
Supporting environmental protection and appeasing gamers is certainly a nice side effect, but it is not the main reason for switching to PoS. This is mainly affected by external factors, and Ethereum as a protocol does not have much control (such as the source of energy production, GPU supply chain, etc.).
When will the merger take place? The official date has not yet been announced. However, developers and the community have expressed cautious optimism that the merger will take place in June. For reasons, please refer to the article "Five Signs That the Ethereum Merge Will Happen in June 2022". Testing is still ongoing, and nothing will be merged until the developers are completely confident that nothing will go wrong. I personally wouldn't count on a merge in June, but I'm sure it will happen by summer unless something goes terribly wrong during testing (e.g. a critical bug that takes weeks to fix, or a hole in the spec that takes months to fix and reimplement). The difficulty bomb is set to explode in June, so a hard fork upgrade will occur at that time, whether merged or not. I recommend that you bookmark wenmerge.com to quickly check the latest status of each testnet merge to estimate the mainnet merge status.
Don't be silly, they will delay it like they always have. The merge was promised years ago, but it's not here yet. First, some useless nonsense: the official merger date has not been announced, and it has never been announced. The deadline has never been set, so how can it be postponed?
The quote “Ethereum will switch to PoS in 2018” comes from an extremely optimistic attitude, which underestimates not only the complexity of PoS security design, but also the complexity of the transition from PoW to PoS security. The work done at that time was equivalent to partially completing the specification of Casper FFG, a hybrid PoW-PoS mechanism, which was eventually abolished. But now there are many different designs. These new designs were not applicable at the time, but they are applicable now.
After years of research, analysis of possible attack vectors, and continuous improvement, a complete protocol specification was written. The client implementation is all done, now all that's left is testing Everyone is working on the merge, and there is basically no other work going on. The necessary preparation steps for the merge have been completed. It's not that "they've finished complex work like EIP-1559, so now they can focus more on the merge", but that "all attention will be focused on the merge". It can't be that the merge "gets postponed again" and developers start working on something else. There is nothing else to do until the merge is completed. PoS has actually been running in the form of a beacon chain since December 2020. This means that Ethereum's PoS has been tested and run in the form of a product for more than a year, and there are currently more than 10 million ETH staked on the beacon chain. It's just that the current PoS has not yet produced blocks for the execution layer.
Once the staked ETH assets can be unlocked, the price will collapse Of course, many stakers will ultimately want to profit, especially those who locked up their stake when 32 ETH was worth only $10,000. But there are many other aspects to consider: The merger will not unlock any ETH. Withdrawals will be made in the first hard fork after the merger, which may be 6-8 months later. In these months, there will be no additional PoW tokens (about 13,000 ETH/day) being sold, and there will be no additional PoS tokens circulating in the market for the time being.
Just like staking ETH requires queuing, withdrawals also require queuing. Assuming a massive sell-off, everyone is in the queue, and the exit rate is limited to 1,125 validators per day. So there is no "open the floodgates" moment. It will take more than a year for all stakers to unlock. Over the course of a year, about 38,000 ETH will flow back into the market every day (or... about 1% of the average daily trading volume).
After the merger, validators will also start receiving fee rewards, and it is estimated that the yield will double. (Estimated ETH issuance and staking yield after the merger). There are thousands of people queuing up to join the staking. Since they accept an annualized yield of 5%, I don’t think they will give up staking when the annualized yield becomes 10%.
By far the biggest risk involved with staking is a merger. Something catastrophic could happen. Yet despite these risks, and despite the uncertainty of the date to unlock ETH, people have staked and locked up their ETH for over a year. How many people/institutions are sitting on the sidelines, waiting for this risk to fade before joining the staking fray?
And don’t forget, stakers exiting means fewer validators, which means that validators that haven’t exited will have higher returns. Likewise, for those who haven’t participated in staking before, there is more incentive to join staking.
But, of course, this is crypto, and it's a crypto that's not going to change. There are some exciting and destabilizing possibilities with the merger, or there could be a drop when all the good news comes out, who knows. I won't pretend I can predict the future, but it seems to me that there's probably going to be more ETH joining the staking than exiting.
If PoS is so good, why didn’t Ethereum use it from the beginning? PoW is easy to conceptualize and implement, PoS is more difficult, especially in 2014 when it was mostly a theoretical concept still being researched, with some blockchains implementing some version of it.
Before considering implementing PoS, there are several fundamental issues that need to be overcome from a research perspective.
There is no universal PoS solution. Each PoS blockchain has its own PoS specification, and each has its own advantages and disadvantages. So many people say "other blockchains have done it, why can't Ethereum do the same thing", it's not that simple.
One benefit of starting with a PoW chain is that it allows anyone to mine without permission. This allows Ethereum to have a much better token distribution than chains that are PoS from the beginning (they still have to decide how to distribute the initial tokens, and there is really no way to do it permissionlessly).
Related to the above: yes, there was a premine/presale for Ethereum, but after years of mining and multiple bull-bear cycles where ETH was constantly changing hands, the premine tokens are now diluted to half of the total supply, and the top 10,000 ETH holders hold nearly as much as the top 10,000 BTC holders. So in 2022, when Ethereum is an extremely liquid and easy-to-buy asset, this isn’t a big deal.
This is really a trick that ruined the miners' years of hard work. PoS has been the ultimate goal of Ethereum since its launch, and every miner knows it will end one day. There is no scam or unfairness here.
Economic factors trump any form of miner loyalty to the blockchain. You can think of the blockchain as a company and the miners as employees.
Miners/employees are already compensated for their services (i.e. providing secure consensus) in the form of block rewards. Salaries are paid by the employer (the blockchain) by diluting the value of the tokens currently held by holders (see "Why merge?" above).
Miners generally look for the chain that offers the highest reward, and most of them would immediately abandon Ethereum if another coin that can be mined by GPUs offered more rewards.
Likewise, if stakers can complete their services at a lower price, Ethereum will also pay less in fees.
This is not entirely exclusive. Miners can also be holders of tokens blocks and users of the blockchain. There is nothing stopping them from holding the rewards they receive and participating in staking.
If you don’t expend real-world energy to mine, the token loses its inherent value I don't quite agree with this. There's nothing magical about computing hashes over and over until you find one that meets arbitrary requirements. I mean, imagine there was a PoW blockchain where work was done by solving Sudoku puzzles; that works exactly the same way: in an NP-complete problem, it's hard to compute the value of one side, but easy to verify once you find the value of a side. This doesn't mean that solving Sudoku itself adds value to the world. Making it harder to mine a certain coin doesn't magically make everyone richer, it just makes mining less profitable - unless, of course, demand also goes up, but so far that hasn't been much of a problem in the cryptocurrency world. I think the value of a token ultimately comes from supply and demand, and demand comes from the value of block space. People need ETH to buy block space, whether the ETH is generated by miners or stakers. Of course, the more miners there are, the higher the security/decentralization, further increasing the value proposition of block space in a positive feedback loop. But this feedback loop also exists in PoS Ethereum, and they are super cool too! PoS is a recipe for complete centralization It's basically the same as PoW, but slightly different. "Better" or "worse" really depends on your perspective. In my opinion, PoW is really just an additional step to implement PoS.
Ethereum as a community places a high value on decentralization, and for any potential centralization vector, the research team will find ways to mitigate it, even if it means sacrificing other important things like scalability (for example, limiting the gas limit to a relatively low level so that more nodes can participate in the decentralized network, even if this leads to network congestion and high transaction fees).
There are some shortcomings, decentralization is a spectrum and a process, and we are not there yet. Right now we still rely on a lot of centralized things, and in the long run these will need to be gradually replaced. That said, none of these centralized obstacles represent a risk to the network, and in fact, for the statement that "this network is centralized because of X", there will be a corresponding solution in Vitalik's latest roadmap. I personally think it's more exciting to come up with a bunch of solutions to solve problem X than to just give up on decentralization and say "we can't achieve decentralization because of problem X".
Ethereum Protocol Development Roadmap There is something interesting about Ethereum’s PoS design that is often overlooked: quadratic penalties. A single validator node that goes down, has issues, or directly attacks the network is not penalized very severely. If a thousand validators do this at the same time, they will be penalized much more severely.
➤ That is, if you are a large organization operating thousands of validator nodes, it is better for your own benefit to spread them out, avoid cloud hosting, use different clients, etc. Sure, the assets are still centralized, but at least the points of failure are dispersed, which is good for the overall health of the network. ➤ Some large mining entities rely on a central location to amortize costs, and authorities can monitor specific locations based on energy usage and shut them down. It is difficult to move mining equipment around the world, but staking relies only on private/public keys and consumer-grade computers. PoS actually makes the rich richer Yes. Unfortunately, we live in a world with extreme wealth inequality. Blockchain is not going to solve that.
This is also true for PoW. Whoever has the money can buy more mining machines and make more money. In addition to mining, the return on investment is also affected by economies of scale: centralized mining has a lot of money to buy hardware at a certain discount rate and move to places where electricity is cheap. Individual miners cannot compete with it in reality. In PoS, whether their stake amount is $10 or $10 million, everyone can get the same rate of return in proportion.
"These large mining operations may be centralized, but they have no reason to attack the network because they have invested millions of dollars in infrastructure...". So what you are saying is that you are fine with large centralized operators as long as they have a certain size share of the network?
Is this passive interest on your deposit? Printing money out of thin air? Isn't this just like the central bank issuing more fiat currency? You have to stretch the argument to make that point, but I’ve seen people do it (laughs). These arguments usually start with “PoS is nothing new.”
Validators are still doing some "work": creating blocks and validating other blocks. It's just that this work consists entirely of actual useful work (the blockchain needs to reach consensus), rather than calculating hash values over and over again until one of them meets arbitrary requirements.
This isn’t really “free money printed out of thin air”; there are still costs involved in staking assets, they’re just more abstract and less intuitive than an energy bill.
➤ Opportunity cost – If another investment can give you a higher rate of return, why would you pledge? ➤ Poor liquidity - From the moment you deposit, your funds are locked up, waiting in line for your validator to activate, and then when you withdraw, you have to wait in line again. ➤ Inherent Risks — Staking is still fairly new and things can go wrong. A critical bug could occur, the network could be attacked, your staking hardware could break, etc. ➤ Volatility — It is still a volatile asset after all, and if you are the type of investor who invests in your national fiat currency, it is not very attractive when an asset can drop 30% overnight and the yield may only be 5% (although, once the asset doubles, a 5% yield is very good, turning a 100% gain into 110%). ➤ Maintenance costs — You still need to maintain and secure your validator nodes, ensure 100% uptime, update software, etc.
Here’s an interesting point: the more stakers there are, the lower the staking rewards an individual gets. This basically means that all the costs listed above will be priced by the market itself. The reason is simple: if the staking yield is too low and the rewards are not enough to cover the maintenance costs, then people will quit staking and invest elsewhere. With fewer people staking, the yield will rise again. Similarly, if the yield is too high, it will attract more capital to join, and the yield will fall again.
As for inflation: Assuming that the market as a whole has an ideal rate of return of 5%, 3% of which comes from token issuance. This means that about 30 million ETH are staked and 900,000 ETH are issued each year. With a total supply of 120 million ETH, the inflation rate is 0.75%. As long as the gas fee is at least 23 gwei, the inflation rate is lower than the ETH destruction rate brought by EIP159. (I cannot emphasize this enough: Ethereum will soon become a profitable deflationary asset)
“The math is good, but there is no supply cap and they are always changing monetary policy” ➤ For many years, the goal has been to "achieve the minimum viable issuance while ensuring network security", with Ethereum prioritizing the security of the network over setting an arbitrary supply cap. ➤ As for monetary policy changes, none of the updates increased inflation. Low inflation (especially deflation) has been the community’s goal since day one. ➤ Once EIP-1559’s burn rate matches the issuance rate, an equilibrium point will emerge that serves as an effective supply cap — again determined by market forces valuing Ethereum blockspace.
So, there is no such thing as an “Ethereum Central Bank” that arbitrarily adjusts inflation/deflation rates and prints money to its cronies. The market itself determines inflation/deflation, and no single entity can control it the way a central bank controls the inflation rate of a fiat currency.
Whales have enough funds to control and change the rules and confiscate honest validators There is no such risk, and Ethereum does not have any form of on-chain governance for this reason. Protocol updates are the result of community efforts (Layer 0), and you do not need to stake any assets to report some bad attention and participate in the process.
This is exactly the same as PoW: even if you have 99% of the hashrate, you can’t make invalid transactions, steal other people’s assets, or change the protocol rules without your private key. There’s really nothing you can do except reorganize the blocks. 1% of honest nodes will reject any blocks that don’t follow the rules, so the bad guys will be mining on an invalid/useless chain. The same is true under PoS consensus, now just replace “hashrate/mining” with “staking weight/staking” (the difference is that the bad guys who reorganize the blocks will be fined and all staked assets will be confiscated if they are found, but the blockchain cannot completely destroy the mining equipment).
Simply put, there is a lot of ETH involved on-chain. There is over 10 million ETH currently, and that’s before the merge. At current prices, that’s about $30 billion. Both “amount of ETH staked” and “value of ETH” are expected to rise, so attacks become increasingly unlikely because the economic cost of launching one is too high. And if the attacker is from outside Ethereum, it’s ridiculous to get that much ETH in the first place (where do you buy 10 million ETH to reach 51% stake? Or 20 million?)
32 ETH is too much, most people don’t have that much money I agree that this is a big problem. There are some proposals to lower the barrier to entry for staking (better signature aggregation or setting a cap on active validators and rotation mechanism), but they don’t seem to be a high priority at the moment, and it’s more important to ensure that the base layer can be secure.
The reason why such a high ETH stake amount is needed is because this value needs to just meet a technical point. In short, it needs to be low enough for everyone to participate and have enough validators to ensure the security of the blockchain; but it needs to be high enough to avoid too many validators and make the blockchain too expensive. And the stake amount of each validator node is the same, so that each validator has exactly the same weight when deciding who produces blocks in the distributed random process, reducing a lot of complexity.
From a technical perspective, a lot of math was involved to arrive at the 32 ETH threshold, which was worth about $7,000 at the time. Early math in 2017 even suggested a minimum of more than 1,000 ETH.
Thankfully, just like mining pools in PoW, PoS also has staking pools that allow for staking of small amounts of ETH. This doesn’t necessarily go against the “not your keys, not your coins” mantra, thanks to schemes like RocketPool and Secret Shared Validators (not yet launched) that leverage smart contracts to be permissionless, decentralized, and non-custodial. And because of the quadratic penalties mentioned above, I believe decentralized staking will outperform centralized staking schemes in the long run. I recommend reading superphiz’s staking guide for more information. Staking through an exchange is a terrible idea if you value decentralization.
Related to the above, it’s best to think of solutions like Rocket Pool as a higher-level abstraction of basic staking, rather than just a “staking pool.” I wrote more details about this here for those interested.
PoS has not yet been proven to work, but we know that PoW works This argument is actually completely reasonable and obviously we can't really refute it. Only time will tell. I just think it's irrelevant in the context of Ethereum switching to PoS and having decided to switch all along. If you don't believe in it, don't participate/invest in it. I personally believe in a long-term sustainable PoS Ethereum. These are all part of the great cryptocurrency experiment of our lifetime. Either it will be a flash in the pan and fail so badly that no one will ever know about it, which will be a shame, or we will succeed in creating a strong network that will outlive human civilization. To achieve this, prioritizing decentralization is key. I see this decentralization primarily in Bitcoin and Ethereum, even though their philosophies are very different. That’s why, long term, I’m excited to see how both end up playing out. Source | www.reddit.com/r/ethereum |
