Panel: Reasons for the Smooth and Successful POS Upgrade of Ethereum

Foreword: The most important upgrade in the history of Ethereum, from POW to POS, was simply completed on September 15th. Many concerns from the outside world (such as vulnerabilities, miner attacks, node synchronization issues, etc.) have not materialized. Ethereum developers have previously stated that what they hope to achieve is to make the transition to POS a smooth, nonsensical transition. Looking back, as an economy with hundreds of billions of dollars, it is worthy of applause to be able to complete the POS upgrade so smoothly. What is the reason for the success is also worthy of discussion and analysis.

Gary Ma:

At 14:43 on September 15, 2022, Ethereum completed the merger of the mainnet and the beacon chain, successfully transitioning to PoS. Maybe many community users are excited but puzzled: the merger transition has been successful, and the big move that has been brewing for so many years has been completed so smoothly, and the market price is calm. Even now, it feels a little like Sell The News.

At the beginning, we all took the transition metaphor of this consensus mechanism as a moving train to replace the engine, which made us consciously consider PoS transition is beset with difficulties. In the end, faced with such a smooth and even user-insensitive upgrade transition, users are inevitably a little surprised. And all of this actually comes from the efforts and contributions of the Ethereum community.

You must know that before the mainnet merge, the developer community has prepared several testnet upgrade previews (Kiln, Ropsten, Sepolia, Goerli) and more than ten mainnet shadow fork tests to fix potential bugs.

At the same time, the merged upgrade also achieves the most comfortable user experience. For ordinary users, there is no need for any operation, and they will hardly feel the trouble caused by the upgrade (except for the unilateral temporary prohibition of deposit and withdrawal by the exchange, or brief suspension of use of some agreements . Even for developers, minimal changes have been basically achieved, such as subtle changes in block structure, block time, opcodes, etc., which basically do not affect the main function of the DApp, and upgrade modificationcan of related contracts can be left after the successful upgrade.

In addition, the important factors that make this transition so smooth are also inseparable from the long-term stable operation of the beacon chain and the cooperation of node operators. You must know that the beacon chain was launched on December 1, 2020, and has been running stably for nearly two years. On the day of the mainnet merger, more than 88% of the nodes’ execution clients were ready, which was one of the important conditions for a smooth transition.

Finally, at the block height of 15537394, Ethereum officially entered the PoS era, and the network indicator data was very healthy. The participation rate of validators was as high as 99%. The first epoch also achieved finality after 12 minutes as scheduled, and there was no abnormal block loss. According to the developer terence.eth, the combined block loss rate is about 1.73%, which is within the expected range.

ChainFeeds founder Zhixiong Pan:

I think there are three factors: the test network has been tested many times, the upgrade content is user-insensitive, and the problem may not be exposed in a short time. In fact, this upgrade has already been a little later than the expected launch time, so the Ethereum Foundation has gone through a lot of testnet verification, and some problems have been encountered in several tests, which can be corrected as soon as possible. It is worth noting that within the day before the Merge, the execution client Geth was still releasing a version update to fix some minor problems while many problems can still be found on the testnet. In addition, this upgrade is inherently insensitive to users, just to update the consensus while the execution layer has not changed, so there will be no perceptible difference in user experience. Finally, this upgrade hasn’t been around for a long time, so the problem may not have been discovered or exposed. Although a lot of discussions have been brewing in the community about PoS being easily censored, more impacts may take time to accumulate, not instantaneous, and require long-term observation.

Solv Protocol founder Yan Men:

The first is the smoothness of technological change, which has been widely analyzed. The Ethereum core development team has established an engineering system that is comparable to classic open source projects such as Linux and Python, and has formed a mature innovation, development, testing and deployment process. The Merge proves that the Ethereum engineering system can support major technological changes. In the future, we don’t have to worry about the risks in this regard.

The second is the smoothness of institutional change. The shift from POW to POS is actually a change in the entire Ethereum token incentive system, taking away the cheese of a large number of vested interests. From a token economics perspective, the motivation for doing this is not difficult to understand. But it can be implemented relatively smoothly. On the one hand, it shows that Ethereum has always been an innovator-led ecology, and the governance mechanism is strong. On the other hand, it also shows that the POW miner groups havery made a lot of money in the past seven years, but in terms of governance rights, or political rights, there is no weight, there is no organized or decent resistance, and even the volume of complaints is not loud. This shows that the previous participation of POW miners in Ethereum was purely driven by economic interests, and the attention and contribution to ecological innovation was little, and there was no close relationship with Ethereum developers and innovation communities, which in turn proved that the proposition ofPOS upgrade leader : Incentivizing others is more reasonable than incentivizing POW miners.

ColinWu:

Previous research by Amber Group pointed out that miners may become more hostile, severely degrading the user experience of Ethereum before the Merge. For example, miners can steal MEV. Theoretically, miners can also start running clients, reorganizing the Ethereum blockchain to retroactively capture MEV. The reason why this did not happen is that, firstly, the Ethereum community has a high degree of trust in the developer community and Vitalik, and the transition to POS has also been paved for a very long time while the miners and mining pools have not expressed any intention to attack or be unfriendly; more importantly, in the outbreak of DeFi and NFT in the past two years, Ethereum miners have obtained extremely high profits with cheaper machines. A friend of mine from outside the circle once made a 100% profit in 3 months. Therefore, except for those who have been “fudged” into the market recently (developers have warned countless times not to buy mining machines and will need POS, this kind of continuous reminder is also very worthy of praise), most of them have earned enough profits. As Yuchi said in the farewell letter to ETHPOW: We acknowledge that the era of ETH PoW is over and will become history. Let us cherish and bid farewell to the glorious years and move towards a new era of PoS.

Joey Wu:

In order to complete a major upgrade like the Merge, sufficient consensus among stakeholders is needed. The main stakeholders are nodes, applications and miners. Since the network needs to be supported by enough nodes to function, the primary client must complete the upgrade. In early September, the Ethereum development team launched the last major upgrade before the Merge: Bellatrix. According to the ethereum node website ethernodes.org, about three-quarters of the nodes had already upgraded their software at the time (more than two-thirds should theoretically be enough). Since then, although there has been a small setback, that is, 9% of the blocks have missed node verification, the remaining nodes have also kept up with the progress after the disclosure of the matter by the core developer of Ethereum, Marius Van Der Wijden. Until the eve of the Merge, almost all clients had completed the upgrade. It can be said that the Bellatrix upgrade is similar to Powell’s anticipatory management, exposing unstable factors in advance.

The application aspect is the most indisputable, with almost all major applications supporting the Ethereum merger. In particular, Tether, the issuer of USDT, and Circle, the issuer of USDC (stable coins are the blood of DeFi), clearly expressed their support for the new consensus mechanism, rather than any fork. The only people who might be constrained are the miners. However, miners are an extremely loose interest group. When the Ethereum development team announced the Merge time this summer, Ethereum’s entire network hashrate experienced a sharp drop in mid-to-late June. You can’t completely blame the bear market for this phenomenon, because Bitcoin’s hashrate didn’t drop that much during the same period, while ETC’s hashrate was rising sharply, so this just proved that miners had capitulated. This does not mean to belittle the miners, but just proves the greatness of PoW — it can use the selfishness of the miners to decentralize the power, make them fight against each other, and finally achieve decentralization. This is indeed the case. When miners claim to unite to defeat ETH PoS, what they do first is not “resistance”, but “annex”. So we have seen various ETH PoWs, and their respective leaders such as Justin Sun or Chandler Guo have no strong appeal, but even have a bad reputation in the community.