是的，是的，你聽到我對嚴格的全球共識以及它如何成為區塊鏈的唯一獨特屬性的咆哮和狂熱。但我並沒有過多談論替代方案，因為我有意識地避免談論基礎設施。基礎設施痴迷並消耗了加密空間，以至於其他一切都被忽略了。然而，我突然想到，這只是區塊鏈基礎設施，沒有人真正談論無狀態基礎設施——因為沒有無狀態基礎設施包可以涵蓋。因此，我對談論無狀態基礎設施並不感到內疚。
什麼是無狀態基礎設施？
首先，什麼是「無狀態基礎設施」。有人會說 UTXO 或 DAG 或其他鍊是無狀態的。其他人會說具有無狀態客戶端的鍊是無狀態的。即使是一些我稱之為無狀態的解決方案也具有狀態——儘管其形式與區塊鏈不同。所以也許我們需要一個更好的術語來描述這個基礎設施，歡迎提出建議，@polynya on Farcaster。
但目前，在本文中，我們將堅持區塊鏈=任何實現即時嚴格全球共識的鏈（包括rollup）；無狀態基礎設施 = 不是一條鏈，沒有共識（或鬆散共識），但與上面相同 對區塊鏈交互的去中心化補充。
當我們進入一個大膽的混合應用新世界時，區塊鏈僅用於嚴格的全球共識- 主要用於金融和身份- 我們將看到幾乎所有計算、數據等都發生在非區塊鏈基礎設施上。
但首先，我們在討論無狀態基礎設施時需要改變我們的觀點。在區塊鏈世界中，我們已經適應了誠實多數的假設，因此運行盡可能多的節點非常重要。然而，在區塊鏈之外，大多數事情都是基於誠實的少數人的假設——只要有誠實的一方，一切都好。
我想我們可以大致分為三類：
1) 伺服器－這是一個顯而易見且眾所周知的數量。
2) 冗餘伺服器 - 添加冗餘以及任何人都可以啟動伺服器的能力，您可以獲得一定程度的去中心化，同時也保留了傳統伺服器的最大效率。
3) Stateless Infrastructure - This is a broad category in itself. These are the most decentralized structures, running peer-to-peer but without consensus (or loose consensus). So you run an honest minority assumption that even a network with 10-100 nodes is fully decentralized. But there may also be certain types of stateless infrastructure where there is only one node but they don't need more nodes due to validity proofs. (These are called ZK coprocessors.) Stateless infrastructure enables many of the features erroneously attributed to blockchains—decentralization, privacy, auditability, durability—but without the burden of consensus. By calling it "stateless", we can feel that they are related to the blockchain, but they have no state [PS: They can have state, but not a real-time strict consensus state, and are not related to what we associate with the blockchain Strict state transition], so no consensus will be reached. Again, this is a very broad category and I'm sure we'll see many different solutions. IPFS, BitTorrent, Farcaster hub, ZK coprocessor - all of these can be considered forms of stateless infrastructure.
Why do we need stateless infrastructure?
Today, blockchain is very inefficient, requiring tens of thousands of times more overhead on the server for the same calculation. With new technologies like validity proofs and data availability sampling, we can be thousands of times more efficient, but even with endgame ZK rollup settlement using DAS, the overall overhead can be 100x.
Today, "web2" runs on more than 100 million servers. How optimistic we are about decentralized computing depends on who you ask. Single-chain maxis is the most bearish, believing that one chain can handle most of it. The more optimistic people will tell you that 1% of computing will be moved to decentralized computing, and the most optimistic people will say 10%.
But let's take 0.01% as an example - not as pessimistic as a single maxis, but still quite pessimistic. This would require the equivalent of 10,000 servers. Add in the inefficiencies created by all the chains and you could be looking at a million chains.
Using ZK technology, it is possible to scale to literally a million chains with universal synchronized composability and shared security – which is literally a million-fold improvement over monolithic chains.
But this is not enough. Let's take a multiplayer game like Palworld as an example. According to the developers, their monthly server costs are $500,000. To have it all run on endgame ZK Rollup - which is the most efficient frontier of blockchain - would cost millions of dollars, which is not economically feasible. (As mentioned above, such a use case is completely impossible without using ZK technology)
Instead, we must take a hybrid approach while retaining decentralization. Multiplayer games can be run largely peer-to-peer, with game clients running on everyone's computer or console, effectively running a giant decentralized network. Some things need to be coordinated on the server - these things can be moved to a peer-to-peer stateless infrastructure. Things that need to be verifiable but not consensus can run on the ZK coprocessor. Finally, things that require strict global consensus can run on ZK rollup/validum.
Finally, it's important to note that in a stateless infrastructure, parallel processing across multiple cores and multiple machines is much easier, which can lead to exponentially higher scale.
It will still cost well over $500,000, but by using the most efficient solution for each task while retaining decentralization, we can enable new use cases that simply aren't feasible today using just blockchain, even is the most advanced form of ZK and will not be viable in the future.
It's all about enabling new applications
Today we are in a very limited paradigm. It doesn't matter whether it's L1 or L2 - the hardware being limited to one chain - one server, actually - is very limiting. We have seen that L1 after L1 and L2 after L2 reach their limits and become crowded very quickly, with even the fastest chains only able to accommodate a few thousand transactions per second at most.
Using new technologies like proof of validity, we can start using multiple chains while maintaining security and composability, which is already a huge, infinite improvement.
Using stateless infrastructure, we can go one step further.
By combining all of these, we can start to implement applications that wouldn't be possible without them.
But what about decentralization?
Stateless infrastructure can actually be more decentralized than blockchain. Reaching strict global consensus is an extremely difficult process, requiring an expensive sybil resistance mechanism and the assumption of an honest majority. The block production mechanism is a plutocracy (proof of stake) at best and a corporatocracy (proof of work) at worst. These blockchains mitigate this problem by running on nodes, but they still run on the assumption of an honest majority and you need thousands of nodes to achieve any degree of resiliency in the consensus formation process.
While stateless infrastructure will come in a variety of different forms, most of them can satisfy an honest minority of assumptions - so as long as you have a node, everything is fine. Therefore, they can be more decentralized with fewer nodes and have fewer points of failure and centralization.
In short, peer-to-peer stateless infrastructure is more efficient and decentralized than blockchain if strict global consensus is not required.
in conclusion
It’s time to stop fighting for blockchain infrastructure and start building applications that enable new applications through stateless infrastructure.
Great to discuss more on Farcaster @polynya. By the way, Farcaster is a good example of something that can only use stateless infrastructure and why it succeeded while its blockchain counterparts like Steem or DeSo failed. More Farcasters, less L1 and L2, please.
