1/30 V God recently updated the ETH roadmap, which contains some [interesting points] and [hidden passwords] for everyone to interpret:
It is divided into six stages (black box). Each stage has several milestones (blue box). The completion of each milestone requires several technologies and proposals (green box).
2/30 Milestone A of the first phase of Merge, "conversion to PoS", was successfully completed, and PoW withdrew from the stage of history. As for the withdrawal function of Staking, judging from the progress bar, it is 80% complete. It is estimated that withdrawals will be opened soon. We look forward to the "Shanghai upgrade".
The B milestone of 3/30 Merge "Single Block Final Confirmation (SSF)" is newly added. What is block finalization? To give an inappropriate example, in the past PoW era, it usually took 12 blocks for exchanges to post accounts, but this was only a loose demand, and there was still a very low probability of forks. After switching to PoS, theoretically 32 blocks will be finally confirmed, and it is absolutely irreversible/forked.
4/30 SSF is often translated as "single slot" final confirmation, but I think it is more smooth to say "single block". As the name suggests, the final confirmation can be completed in 1 block, reducing the confirmation time from 32 blocks (6.4 minutes) to 12 seconds.
Paths toward single-slot finality - HackMD
https://notes.ethereum.org/@vbuterin/single_slot_finality
5/30 This is difficult. You must know that in the traditional BFT consensus, although a single consensus is irreversible every time, it cannot support many nodes. Once there are more nodes, the performance requirements for node equipment will increase exponentially; while the traditional PoW consensus can accommodate many nodes, it will take a long time to achieve 99.99% irreversibility.
6/30 And SSF "want both and want more". It requires a large number of nodes to participate, a low threshold for node participation, and quick confirmation in a single block.
This depends on the advancement of signature technology (such as BLS), so that a bunch of nodes can sign together to save bandwidth, coupled with the committee mechanism, then there is hope for single block confirmation. However, looking at the progress bar, there is still a long way to go.
7/30 The second phase of Surge is very interesting. It adds a very intuitive KPI: TPS must reach 100,000. But I added parentheses a little guiltily and said that rollups are considered haha. Milestone A is to complete the initial expansion of rollup, which is completely dependent on the EIP-4844 proposal.
8/30 EIP-4844, proto-danksharding, was explicitly put into the roadmap. 4844 is a Danksharding front-end solution, which is less difficult. To put it simply, blobs are first connected to the ETH main network, so that all rollup data is stored in blobs and not in the main chain. Similar to logistics warehouses that prefer to be built in Kunshan rather than Shanghai. This allows you to receive more rollups and at the same time reduce the cost of each rollup. https://www.eip4844.com/
9/30 In addition, since 4844 is a simple version, the blob is still managed by PoS nodes. In this way, the originally planned 16MB would be unbearable for the node, so it was changed to 1MB blobs. Of course, 1MB is not bad, and the rollup fee can be reduced a lot. The progress bar is about 60%, so the progress is considered smooth.
10/30 Milestone B of the second phase is the full expansion of rollup. The focus is on the DA layer, which also stores data.
An inappropriate metaphor is that it would be overkill for ETH to store some data in person, so ETH decided to find an outsourcer. So in order to prevent this outsourced data layer (DA) from doing evil, data sampling (DA sampling) is also needed. This requires the introduction of many cryptographic algorithms. The progress is impressive, and it is probably still in the theoretical demonstration stage. Haha.
11/30 In addition, from the official perspective of ETH, the technical maturity of ZK is only 1/3, and the OP system is only 1/2. In other words, one is in kindergarten and the other is still a toddler, far from it. to the ideal point. Therefore, it is understandable that the several ZK series that will be launched soon are basically the product of rushing to launch on time, and they probably made a lot of centralized assumptions and compromises.
12/30 Scourge is newly added in the third stage, which was not included in the previous roadmap! This little zombie avatar represents that this stage is mainly aimed at MEV, and I am very pleased that ETH attaches so much importance to this. MEV is bad. Nansen labeled an address of mine as a heavy on-chain trader and victim of a sandwich attack. My trading heads were squeezed every day, and no user liked MEV.
13/30 However, MEV is a good business for PoS nodes. The argument for low EQ is that PoS nodes connecting to MEV are actually "competing with the public for profit" on the chain. Back to the roadmap, fortunately, ETH still believes that transaction packaging needs to be neutral. The currently selected solution is to use PBS to reduce the impact of MEV.
14/30 PBS means the separation of block producers and orderers. The sorter is only responsible for sorting and does not put it on the chain. The block producer does not care about transactions and directly selects the package prepared by the sorter and puts it on the chain. In this way, the bribery chain will become longer, which will be somewhat better. But it cannot cure the root cause, it can only "mitigate" it. After all, PBS only weakens the power of nodes and makes quotations public.
15/30 In addition, there are a few small questions left in the roadmap, which may be the next research direction: For example, can the App directly help you with pre-confirmations? For example, Uni helps you make an internal reservation and the transaction is successful; or can the App publicly ban sandwiches to protect users (frontrunning-protection)? For example, Uni banned sandwich bots.
16/30 Btw, PBS’s work is basically only 20-40% carried out. Only one has been completed, and that is the external MEV market. This probably refers to flashbot.
17/30 The fourth phase of Verge (the third phase of the old roadmap) has not changed much. The main milestone is building the Verkel tree. If this thing is developed well, the verification block will be very fast and smooth. There is some progress in this work, even faster than the progress of PBS in the previous stage.
18/30 However, there is a change. Before, ETH only considered fully ZKization (fully SNARKEed ETH) to be an interesting topic. Now, L1 ZKization has been clearly added to the Verge roadmap. Note, it is not L2, it is L1 and it needs to be ZKized. Of course, work related to it has not yet begun.
19/30 In addition, the first four stages also slightly involve quantum safe algorithms, but quantum cracking... is still in the very early laboratory stage, and is an important but not urgent matter.
20/30 Purge, the fifth stage, means that the Emoji is a broom. The most important thing at this stage is to complete some "technical debt" owed in the early years. This term is really down to earth. In the early years, many ETH structures were not thought through clearly, such as the preservation of historical records & status records. Now we have decided to improve them all in one go.
21/30 The change compared to the past is that EIP-4444 is explicitly written into the Roadmap. The core essence of EIP-4444 requires the client to prune some meaningless historical data, such as data from one year ago. Note that this does not mean directly cutting off the blockchain ledger, but some bloated historical data. The blockchain ledger is the foundation and can never be touched.
22/30 Here’s a fun thing: EIP-4444 officially requires that its pronunciation must be “four fours”.
EIP-4444: Bound Historical Data in Execution Clients
https://eips.ethereum.org/EIPS/eip-4444
Same goes for 23/30 expired status. In fact, in theory, as long as there is a ledger, all transactions in each account can be reversed. But in practice, it would be too slow to push back every time. Therefore, Ethereum also maintains a status tree. For example, at a certain moment, how much money is in account A and how much money is in account B. This will make checking much faster.
24/30 But the price of "fast" is that ETH has too much status data. Several TB of status data is too much for many nodes. Allowing more home-level nodes to access is always the biggest pursuit of ETH (and BTC). Therefore, in order to achieve this goal, the status must be cut. State expiry is to do this, but I understand that the current plan... is actually not very clear yet, and it has not been started yet.
25/30 In addition, in the fifth stage, some EVM optimization work needs to be done, such as disabling self-destruction, which is a mechanism to return GAS, and optimizing the gas model and so on.
26/30 The final sixth stage, Splurge, has a special goal: fix everything else. For example, EVM requires many things to be optimized; VDF, for example, allows true random numbers to be generated on the chain. There is actually a previous roadmap for VDF, but it seems that implementing VDF requires hardware cooperation (emphasis added).
27/30 In addition, EIP-4337 is also a new proposal in this route. Account abstraction has been introduced by many people. Let’s put it simply: in the past, smart contracts could only be used for on-chain transactions, but now we also hope that smart contracts can be used directly in your wallet.
For example:
-Restore wallet directly via social media
-The project party will help you advance the gas fee
-USDT as gas etc.
28/30 The path to implementing it requires a new special transaction mempool. Those advance transactions are all posted here and are advanced by others. Did you find out? This does not require modifying ETH. There were actually many alternatives to the 4337 plan back then, but the key to ultimately defeating other proposals was that it did not require modification of the consensus layer. Everyone likes to apply patches, and hard forks are too harmful to the blockchain.
ERC-4337: Account Abstraction Using Alt Mempool
https://eips.ethereum.org/EIPS/eip-4337
29/30 Account abstraction is actually a good thing, although the name is too abstract and difficult for many people to understand. However, this is the key to mass adoption. Without this thing, ETH would never be able to break through the bottleneck of tens of millions of users. However, it should actually be placed at a earlier stage. Although the overall development is parallel, the priorities are obviously different.
30/30 Endgame! What is the Endgame of Ethereum like? $ETH will be a very high-performance, secure, numerous-node, and censorship-resistant public chain. At the same time, it also has a very easy-to-use front-end and back-end, which may lead us away from this trust-based chain. Black box world. Looking forward to that day.
