Ethereum Casper 101 以太坊 Capsper 初步 (上)

本文翻译自：https://medium.com/@jonchoi/ethereum-casper-101-7a851a4f1eb0

译者：区块链中文字幕组 oscnet

翻译时间：2017-10-25

Ethereum Casper 101 以太坊 Capsper 初步

Casper will implement proof of stake in Ethereum. We begin with a review on why proof of stake matters and continue with its strengths & weaknesses. This post aims to provide a broad overview of Casper and clarify some of the confusion with respect to the two protocol design efforts related to Casper. The two proposed implementations share the same core design principle: applying cryptoeconomic mechanism design to secure the network while managing challenges regarding liveness, safety and synchrony assumptions. This post is also an overview of the progress so far and the challenges that lie ahead. Most importantly for fellow newcomers, the post identifies & defines key concepts and ties together various helpful resources under one context. The overarching intention is to make Casper and proof of stake more approachable to everyone in the community.

Casper 将在以太上坊上实现 POS (权益证明) 协议。这篇文章首先回顾了 POS 的重要性及其存在的优缺点，
并且对 Casper 进行了广泛的概述，澄清了与 Casper 相关的两个协议设计上的一些混乱。这两种提议的实现使用了相同的核心设计原则:应用加密经济机制设计来保护网络，同时管理有关活动，安全和同步假设的挑战。这篇文章还概述了迄今取得的进展和未来的挑战。最重要的是，对于初学者，本文将各种有用的资源结合在一起，对关键概念进行了说明和定义。

Enjoy and please don’t hesitate to reach out with questions, corrections or feedback

文中如有不当之处，请随时向我联系提出问题、更正或反馈。
(email, twitter).

Outline 大纲

Introduction 介绍
Proof of Stake 股权证明
A Tale of Two Caspers 两个 Caspers 的故事
Why Casper Matters 为什么 Casper 重要
Design Principles 设计原则
Challenges 挑战
Future Work 未来的工作
Glossary 词汇表
Conclusion 结论

Introduction 介绍

Casper is hardly a new project as it dates back to 2014, so fortunately it has a lot of related information online. However, despite the wealth of information out there, there’s no easy jumping-off point for a beginner to ramp up and start engaging with the project. That’s why I’m writing this post.

While other posts, videos and papers focus on the specification, implementation and verification of Casper. This post focuses on the various guiding design principles of Casper, how it differs from competing approaches, why you should care about it, and how you can contribute to this project.

Casper 最早从2014年出现，并不能算是一个新的项目，所以幸运的在网上能找到很多有关于它的文章。虽然有关的信息非常多，初学者却没有一个足够简明的学习起点，来参与到这个项目。这就是我写这篇文章的原因。

不同于其它的文章、视频和论文，主要集中讨论有关 Casper 的规范、实施和验证。这篇文章着重介绍 Casper 的各种指导性设计原则；它与其它竞争替代方案的差异；以及为什么我们要关注它；如何为这个项目做出贡献等内容。

In addition, Casper has maintained an exceptionally open and collaborative culture among Ethereum researchers, developers and community members. I hope this post can continue that tradition by introducing you to Casper and convincing you why it’s important to Ethereum as well as the larger public blockchain ecosystem.

此外，Casper 的维护保持了非常开放和相互协作的文化，它由以太坊研究人员，开发人员和社区成员共同维护。我希望这篇文章可以继续保持这个传统，向你介绍 Casper，并试图阐述 Casper 对于以太坊以及更大的公链生态系统的重要性。

To summarize, this post is:

A quick and broad introduction to PoS and Casper.
Discussion on why Casper matters, its design principles and its challenges.
A list of key resources and terms to get you started with PoS and Casper.

总而言之，这篇文章是：

对 PoS 和 Casper 的快速和广泛的介绍。
讨论为什么 Casper 重要，其设计原则及其面对的挑战。
能让你开始学习使用 PoS 和 Casper 的一系列关键资源和团队的列表。

This is not:

A full history of Casper.
Specification of Casper.
Implementation details of Casper.
Formal verification of Casper.

而不是：

Casper 的完整历史。
Casper 规范。
Casper 的实现细节。
Casper 正式验证。

If this is review for you, a list of resources are linked throughout the post.

With that, I hope you enjoy this beginner-friendly introduction to Casper: Ethereum’s proof of stake research and implementation.

注意：文章中随处可见相关资源列表的链接。

通过这个，我希望你喜欢这个适合初学者的对 Casper 的介绍：以太坊 POS （权益证明）的研究和实现。

Proof of Stake 权益证明

(If you’re already familiar, please feel free to skip ahead to the next section: “What is Casper?”)

（如果您已经对 PoS 了如指掌，请阅读下一节：“Casper 是什么？”）

Proof of Stake (PoS) is a category of consensus algorithms for public blockchains that depend on a validator’s economic stake in the network.

In proof of work (PoW) based public blockchains (e.g. Bitcoin and the current implementation of Ethereum), the algorithm rewards participants who solve cryptographic puzzles in order to validate transactions and create new blocks (i.e. mining). In PoS-based public blockchains (e.g. Ethereum’s upcoming Casper implementation), a set of validators take turns proposing and voting on the next block, and the weight of each validator’s vote depends on the size of its deposit (i.e. stake). Validators are rewarded for their service to the network, but the stake also acts as an economic disincentive for bad actors.

Significant advantages of PoS include security, reduced risk of centralization, and energy efficiency.

股权证明（PoS）是公链的一种共识算法，其实现基于验证者在网络中的经济利益。

在基于工作量证明（PoW）的公链（如比特币和当前实施的以太坊）中，通过对解密码拼图的参与者进行奖励，来验证交易并创建新的块（即采矿）。而在基于 PoS 共识的公共块（如以太坊即将推出的 Casper ）中，验证者轮流被提出并对下一个块投票，每个验证者的投票权重取决于其存款的大小（即股权）。验证者对于区块链网络的服务是有奖励的，而且这种奖励也实现了对攻击者的的经济制约。

PoS 的显着优点包括安全性、降低集权风险和提高能源效率。

Explicit Economic Security 显性经济安全

In PoW, your downside is capped at how much energy cost and hardware depreciation you incur and therefore has an implicit cost that adjusts dynamically (via 10 min block time target). While PoS has its own challenges (discussed later in the post), one of its major strengths is the flexibility to explicitly design the penalties of Byzantine behavior (i.e. not following the protocol). This gives the protocol designer increased control over dialing in the “shape” of the asymmetric risk & reward profile of various actions within the network. One proxy for security is the cost of doing damage to the utility and correctness of the network, and therefore the ability to have explicit penalties (perhaps at levels that are more draconian than PoW) can increase the security of the network (i.e. economic security).

PoW 工作量证明受到你能承受多少电力成本和硬件折旧的限制，所以具有（通过10分钟出块的目标）来动态调整的隐性成本。虽然 PoS 有自己的挑战（在后面有讨论），其主要优点之一是可以灵活地明确地设计对拜占庭行为（即不遵循协议）进行的惩罚。
这使得协议设计者能够对网络中各种行为的不对称风险和收益回报情况进行更多的控制。
安全性的另一个方面是对软件和网络精密性进行破坏的成本，因此具有明确惩罚（可能在比PoW 更严重的级别上）的能力可以增加网络的安全性（意即经济安全）。

On a related note, Vitalik further argues that PoS has better recovery properties. In PoW, there’s an issue of “spawn camp” attacks that can render a blockchain unusable. In PoS, the network can delete the attacker’s stake and prevent repeated attacks. The economic analysis further clarifies this concept. The marginal cost of repeated attack is the same as the first round in PoS. Whereas in PoW, the main marginal cost for another round during a 51% attack is the power cost (the incremental hardware depreciation & physical space cost is minimal for repeated attacks). To drive the point home, let’s paraphrase Vlad Zamfir, the cost profile of a repeated 51% attack in PoS is as if “your ASIC farm burned down” with each additional round.

相关说明，Vitalik 进一步认为，PoS 具有更好的恢复性能。在 PoW 中，存在一个可以使块链不可用的“51%算力攻击问题。在 PoS 中，网络可以删除攻击者的攻击，防止重复攻击。经济分析进一步澄清了这一概念。反复攻击的边际成本与 PoS 的第一轮相同。而在 PoW 中，51％攻击期间另一轮的主要边际成本是电力成本（重复攻击引起的的硬件折旧和物理空间成本增加很小）。简明如 Vlad Zamfir所说，在 PoS 中重复的 51％的攻击成本就好像每一轮都要“你的 ASIC 农场烧毁”一样。

Mitigation of Centralization 减少集中化

Proof of Stake mitigates the impact of economies of scale in making consensus. In proof of work, we can already see that the network trusts a relatively concentrated set of mining pools (e.g. AntPool) to secure the network. Larger proof of work mining pools can lower the unit cost of their technical and operational infrastructure (datacenter costs, power costs, personnel) by both (1) amortizing a fixed cost over a larger operation and (2) having bargaining power by operating as a larger entity.

为了达成共识，PoS 股权证明减轻了规模经济的影响。在 PoW 工作量证明中，我们已经看到，为保护网络安全，相对集中的矿池（例如 AntPool）比规模小的更受到信任。大矿池可以通过（1）在较大的运营中摊销固定成本，（2）通过运营较大的实体来提高议价能力，来降低其技术和运营基础设施（如数据中心成本，电力成本，人员成本）的单位成本。

This means that two sets of mining pools with equal economic cost, one may be able to achieve a higher hash rate and have more influence in the network, dollar for dollar. For example, 10,000 miners that each spend $1/min ($87.6M/yr) may have less hashing power than one mining pool that spends $10,000/min (despite also spending $87.6M/yr). (Further work: quantifying the benefits to centralization in PoW mining would be fascinating. i.e. is it 1bps, 1% or multiples in hashing power per dollar invested?)

这意味着两个成本相同的矿池，取得的算力会不一样，一个能实现更高的算力，并在网络上有更大的影响，即钱能生钱，例如，10,000名矿工单独挖矿，每个矿工花费1美元/分钟（8,860万美元/年）的成本，比一个同样花费10，000美元/分钟(尽管也花费了8760万美元/年）的矿池的算力要低得多。(进一步：量化 PoW 挖矿中心化的好处将是令人着迷的，~~即它是1bps ，1％还是算力以投资每美元比率的倍数？~~）

However, in proof of stake, a dollar is a dollar. The benefit here is that you can’t pool together to make a dollar worth more. Nor can you develop or buy application-specific integrated circuits (ASICs) to have an advantage technologically. So, PoS intends to mitigate the regressive distribution of PoW mining rewards and move directionally towards proportional distribution. (Going beyond proportional to progressive distribution will require mature decentralized reputation/identity management services).

然而，在 POS 股权证明的情况下，一美元就是一美元。这样的好处是，你不能通过汇集在一起，使得一美元值得更多。您也不能开发或购买专用集成电路（ASIC），从而在技术上占有优势。所以，PoS 不同于 PoW 挖矿收入的累计分配方式，采用了比例分配。（成熟的去中心化的声誉/身份管理服务为按比例分配收益成为可能）。

Energy Efficiency 能源效率

Proof of work relies on wasting resources to secure the network. Bitcoin currently uses over 20 TWh per year of power, which is close to the power consumption by the entire country of Ecuador. In order for Bitcoin to have broader adoption and operate at Visa’s scale, it would have to waste as much power as a much larger country. Extrapolating it’s current pace, one can see why proof of work may not be a sustainable path forward.

PoS 工作量证明机制通过浪费资源来保护网络。比特币目前每年使用超过20 TWh的电力，这已经接近厄瓜多尔整个国家的电力消耗。为了使比特币在 Visa 的规模上得到更广泛的使用，它必须浪费象更大的国家一样多的电力。根据这样的发展，我们就可以明白，工作量证明可能不是合适的发展方向。

While Bitcoin may serve an important social function that may indeed exceed its financial costs and environmental externalities (i.e. Nick Szabo’s social scalability argument), proponents of PoS believe it is possible to replicate the incentive mechanisms of a PoW blockchain without wasting as much energy (by orders of magnitude). Alternatively, some may argue there exists a price of negative externalities where even the benefits of social scalability might be outweighed by the environmental costs.

虽然比特币可能具有超过其财务成本和外部环境（即Nick Szabo的社会可扩展性论证）成本的重要社会功能，但PoS的支持者认为借鉴 PoW 块链的激励机制，而不用浪费太多的能量是可能的。~~或者，有些人可能认为存在负面外部性的代价，即社会可扩展性的好处甚至可能超过环境成本。~~

While the exact answer is hard to ascertain a priori, I believe the crypto ecosystem as a whole has the responsibility to explore all promising consensus mechanisms to weigh their pros/cons and feasibility. (For example, it’s great that other projects are testing the benefits and feasibility of various forms of proof of storage, among others.)

虽然确切的答案很难得到确定，但我认为整个密码生态系统有责任探索所有有希望的共识机制，衡量其利弊和可行性。（例如，其他项目正在测试各种形式的存储证明的好处和可行性等等）。

Taking a step back, it’s also worth noting the two kinds of costs to PoW issuance. There are internalized costs, which are paid by miners and passed on to currency holders. Then there are externalized costs, such as the environmental costs and subsidies from governments (most likely in the form of cheaper electricity). In PoS, there are low consensus costs (no power & hardware costs), which allows for low issuance. As the network matures, it may even allow for negative issuance (net of burned tx fees as well as any penalties and slashing), which can have a price-stabilizing effect.

退后一步，值得注意的是 PoW 存在两种成本。内部成本，由矿工支付，并转交给货币持有人。外部成本，比如政府的环境成本和补贴（很可能以更便宜的电力）。在 PoS 中，共识成本较低（无电力和硬件成本），从而允许低发行量。随着网络的成熟，甚至可能会出现负发行（扣除销毁的交晚费用及罚款），并形成稳定的价格。

Therefore, not only is lower energy consumption better for the environment but it also enables simpler mechanism design. That is because lower energy consumption allows for substituting realized costs (power and depreciation costs are nonreversible) with potential economic value-at-loss (i.e. believable risk of unrealized costs) in order to secure the network. This is a key underlying hypothesis of PoS: both realized costs and prospect of losing money can incentivize participants to secure the network. Therefore–while difficult–it is possible (and thus preferable) to secure public blockchains via loss aversion, which can decrease the public costs and deadweight losses in a system.

因此，不仅能降低环境的能耗，而且还能实现更简单的结构设计。这是因为较低的能源消耗允许将潜在的经济价值损失（即可信赖的未实现成本的风险）代替实现的成本（电力和折旧成本是不可逆的），以确保网络的安全。这是 PoS 的一个关键的基本假设：实现成本和亏损的前景可以激励参与者保护网络。因此，虽然困难 - 但是有可能（因此更好）通过避免损失来确保公链安全，这可以减少系统中的公共成本和无谓损失。

Proof of Stake: Summary PoS 总结

That sums up the main benefits of PoS. While Casper provides specific benefits to Ethereum (discussed below), a significant portion of its importance are the general benefits of PoS: explicit economic security, mitigation of centralization and energy efficiency.

So, now that we’ve discussed PoS, let’s dive into Casper.

这里总结了 PoS 的主要优点。虽然 Casper 为以太坊（以下讨论）提供了具体的好处，但其重要性的很大一部分是 PoS 带来的好处：明确的经济安全，减轻集权和能源效率。

所以，现在通过 PoS，让我们来看看 Casper。

A Tale of Two Caspers 两个Caspers的故事

In simple terms, Casper is Ethereum’s Proof of Stake work stream.

简单来说，Casper 是Ethereum 的 PoS 股权证明工作流。

Casper is not a specific implementation but a family of two main projects under active research by the Ethereum team. Informally, there’s “Vitalik’s Casper” aka **Casper FFG ** and “Vlad’s Casper” aka Casper CBC (explained below). This subtlety isn’t clear until you start diving deep into Casper materials online, which can be very confusing to the “uninitiated.” (In fact, that is a main motivator for this post). While independent implementations, they have the same goal in mind: moving Ethereum over to proof of stake.

Casper 并不是一个具体的实现，它是实际上由以太坊团队正在积极研究的两个主要项目组成。非正式地， “Vitalik's Casper” 又称 Casper FFG，还有 “Vlad's Casper”，又名Casper CBC（见下文说明）。细微之处一直到你开始深入 Casper 的在线资料才能明白，这可能对“末启动”非常不解（事实上，这就是写这篇文章的主要动力）。虽然他们是独立的实现，但他们有着一样的目标：将以太坊的工作量证明转到 PoS 股权证明。

(Despite a surprisingly common impression that Ethereum is already implementing PoS, it is still a PoW chain (using ethash). While being memory hard and more ASIC-resistant than Bitcoin, Ethereum is a PoW chain nonetheless and has the same drawbacks with respect to energy efficiency.

So with that, let’s briefly discuss the two Caspers.

（尽管有一个令人惊讶的共同印象是，以太坊已经在实施PoS，但它仍然是一个PoW链（使用ethash挖矿算法），虽然比比特币更内存相关（挖矿的效率基本与CPU无关，却和内存大小和内存带宽正相关）），更耐 ASIC(即使用专门优化的芯片产生的挖矿优势尽可能的小)，但是它们是 PoW 链无疑，并且在能源效率上具有相同的缺点。

所以我们来简要的讨论这两个 Caspers。

FFG vs CBC

Caveat: both of these projects will present more detailed papers and proof of concepts the weeks following Devcon3, but here’s a quick preview of their approaches.

注意事项：这两个项目将在Devcon3之后的几周内提供更详细的论文和概念证明，这里对其实现方法进行快速预览。

Casper the Friendly Finality Gadget (“FFG”) — aka “Vitalik’s Casper” — is a hybrid PoW/PoS consensus mechanism, which is the immediate candidate for Ethereum’s first bridge to proof of stake. More specifically, FFG implements a proof of stake mechanism as an overlay on top of a proof of work chain (such as Ethereum’s ethash PoW chain). Simply, the blockchain would grow every block with the familiar ethash PoW algorithm, but every 50 blocks is a PoS “checkpoint” where finality is assessed via a network of validators.

实用的终结工具 Casper（“FFG”） - 又名“Vitalik's Casper” - 是一种混合 PoW / PoS 的共识机制，它是以太坊首个通向 PoS 股权证明的候选方法。更具体地说，FFG 在工作量证明（如以太的 ethash PoW 链）的基础上，实施了股权证明。简单地说，块链将用熟悉的 ethash PoW 算法增加区块，但是每50个块有一个 PoS “检查点”，通过网络验证者来评估区块的最终有效性。

Casper the Friendly Ghost: Construction by Correction (“CBC”) — aka “Vlad’s Casper”—differs in approach from traditional protocol design: (1) the protocol is partially specified in the beginning and (2) and the rest of the protocol is derived in way that is proven to satisfy the desired/requisite properties (typically protocol is fully defined then tested to satisfy the said properties). In this case, one way to derive the full protocol is to implement an estimate safety oracle (“an ideal adversary”), which either raises exceptions of a fault of a justified estimate or enumerates the potential future faulty estimates. More specifically, Vlad’s work focuses on designing protocols where one can extend local views of a node’s estimate of safety to achieve consensus safety.

鬼马小精灵 Casper（其名字源于上世纪90年代的一部电影《鬼马小精灵》） ：使用正确的建设（“CBC”） - 又称“Vlad's Casper” - 采用传统协议设计的方式：（1）协议在开始时部分指定，（2），其余部分协议以证明能够满足所需/必需属性的方式得到（通常协议被完全定义，然后被测试以满足所述属性）。在这种情况下，得到完整协议的一种方法是实施估计安全性（“ 一个理想的对手 ”），或者提出合理估计的错误的例外，或列举潜在的未来错误估计。更具体地说，Vlad 的工作侧重于设计协议，扩展节点对安全性估计的本地视图，以实现一致的安全性。

Taking a step back, FFG focuses more on a multi-step transition to introducing PoS for the Ethereum network. This prepares for an iterative implementation that increases the role of PoS in the network over time. (PoS will claim a smaller portion of the rewards to start). In contrast, CBC focuses on formal methods that derive safety proofs “by construction” from first principles. Albeit confusing, the different approaches to solving this problem created two distinct work streams that complement each other well. The final form of Casper will likely draw from learnings from both FFG and CBC.

退后一步，FFG更侧重于通过多步骤过渡为以太网络引入 PoS。通过准备的迭代实现，增加 PoS 在网络中的作用。（PoS 将从较小部分的奖励开始）。相比之下，CBC 着重于通过第一个原则“通过建设”得出安全证明的正式方法。尽管令人困惑，解决这个问题的不同方法创造了两个不同的工程。Casper 的最终形式可能来自对 FFG 和 CBC 的互相学习。

Next steps 下一步

While substantial progress was achieved, many of the details — both at the higher mechanism design level and at the lower implementation level — remain to be finalized. This is openly acknowledged by both Vitalik and Vlad, and this can be an invitation for more of the community to engage to move the conversation forward.

In sum, both research projects are very active, and more updates should be available at Devcon3 in November. The role of this overview does not include going into more detail, but please feel free to dive into more implementation and design details in the links below (This doc will likely be updated or followed up once FFG and CBC papers are released).

虽然取得了实质性进展，但是在高层的机制设计层面和底层编程实现层面的许多细节仍有待确定。维塔利克和弗拉德都公开承认了这一点，并且邀请更多的社区参与推动和讨论。
总而言之，这两个研究项目都非常活跃，11月份 Devcon3 将有更多的更新。此概述的作用不包括更详细的内容，但请随时在以下链接中深入了解更多实施和设计细节（ FFG 和 CBC 论文发布后，本文档可能会更新或后续）。

Links for FFG FFG的相关链接

Casper FFG Basics Paper Casper FFG基础教材 by Vitalik Buterin and Virgil Griffith (Draft)
Casper Github Repo (FFG)
Karl Floersch’s FFG Implementation on pyethereum

Links for CBC CBC 的相关链接

Will be updated after Devcon3 将在Devcon3之后更新

Why Casper Matters 为什么Casper重要

Now that we have decoded what this mysterious Casper project is. Let’s synthesize what we learned about PoS and Casper to understand why this matters.

现在我们解码了这个神秘的 Casper 项目。让我们综合一下我们对 PoS 和 Casper，来理解 Casper 的重要性。

In simple terms:

Decentralization (covered in PoS)
Energy efficiency (covered in PoS)
Explicit Economic Security (covered in PoS)
Scaling Ethereum
Gentle Transition from PoW

简单来说：

去中心化（PoS）
能效（PoS）
显性经济安全（PoS）
以太坊的扩展性
从 PoW 温柔过渡

Because of Proof of Stake 因为权益证明

The first three points are covered in the Proof of Stake section. However, it is worth mentioning that at ~$28B, Ethereum is the second largest cryptocurrency and represents ~18% of the total market cap of the space. So any incremental decentralization and gains in energy efficiency can have a non-trivial impact today and a very significant impact in the future.

前三点在“权益证明”部分中介绍过。不过值得一提的是，以太价格在280美元上下时，它是第二大的数字货币，约占总市值的18％。因此，任何增量的去中心化和能源效益的提高都可能产生不平凡的影响，并对未来产生非常重要的影响。

As a review, (1) PoS has less available economies of scale because — “dollar for dollar” — a miner/validator can’t achieve outsized influence on the network. In PoW, a large mining pool might get more hashing power per dollar than an individual miner, whereas in PoS, a dollar is a dollar, which will likely mitigate the centralizing forces in mining.

作为回顾，（1）PoS 具有较小的可用经济规模，因为 - “美元兑美元” - 一个矿工/验证者不能对网络产生巨大的影响。在 PoW 中，一个大型矿池可能比单个矿工获得每美元更多的哈希算力，而在 PoS 中，一美元是一美元，这可能会减轻挖矿的中心化。

And (2) whereas PoW relies on energy wastage for network security, PoS relies on the potential of losing a deposit for network security. The challenge then becomes (3) how we mimic (and enhance) the positive features of PoW and mitigate the weaknesses of PoS with economic mechanism design.

（2）PoW 的网络安全依赖于能源的浪费，PoS 则依赖于保证金的损失。那么挑战就是（3）我们如何模仿（和增强）PoW 工作量证明机制）的优点，并通过经济机制设计来减少 PoS 的缺点。

Because of Scaling 扩展性

Next, let’s talk about something new: (4) scaling. The key to understanding this is two-fold: (a) Casper is about establishing explicit finality (as opposed to probabilistic finality) and (b) explicit finality enables maintaining network security while scaling via sharding.

接下来，让我们来谈谈一些新的东西：（4）扩展性。理解这一点的关键是双重的：（a）Casper关于建立明确的终结性（而不是概率终结性），（b）明确的终极性可以通过分片进行扩展来维护网络安全。

In PoW chains, finality is implicit (just as “skin in the game” is implicit via power wastage). The implicit nature of finality in PoW chains is apparent when you examine how transactions are finalized in real use cases. Depending on the magnitude and importance of the payment, you wait for additional number of block confirmations (number of block times since a transaction appears in the longest chain). For example, for a coffee, you might be okay with a few confirmations, but for buying a car, you may need more than the average number of
confirmations.

在 PoW 链中，终极性是隐含的（如“游戏中的皮肤”特效是通过花费电力进行渲染）。当您检查交易在真实用例中的最终确定时，PoW 链中的最终性的隐含性是显而易见的。根据付款的幅度和重要性，您可以等待额外的块确认（自从最长链中出现交易以来的块次数）。例如，对于买咖啡，您可以使用较少的确认，但是为了购买汽车，您可能使用比平均确认数量更多的块次数来确认交易。

In contrast, Casper provides a concept of explicit finality. For example, Casper FFG begins to overlay finality on top of the PoW chain. So the underlying chain continues to have an implicit way of determining how final a transaction is. However, Casper FFG provides explicit finality after about 2.5 “epoch times.” (Each epoch time being a sequence of 50 PoW blocks. A checkpoint is the last block in an epoch. It is first “justified” then “finalized” by a validator set. More details available in the paper linked above and potentially a future post.)

相反，Casper提供了一个明确的终极概念。例如，Casper FFG 开始将重点放在 PoW 链上。因此，基础链继续有一种隐含的方式来确定交易的最终结果。然而，Casper FFG 在大约2.5 epoch 时间窗口之后提供了明确的终局性（每个 epoch 是一个50个PoW 区块，一个检查点是一个 epoch 的最后一个块，它是第一个”对齐“区块然后被验证人确定，在上面链接的文章中或以后的文章中可以看到更多的细节。）

At that point, with certain Byzantine Fault Tolerance assumptions, we can be sure that either our assumptions have been violated or that the checkpoint is final. Since we are also aware of the validator set a priori (which can also be dynamic), bad actors are penalized through fault attribution.

在这一点上，使用某些拜占庭容错假设，我们可以确定我们的假设是否被违反，或是检查点是最终的。既然我们也意识到验证者设定了先验（也可以是动态的），则不良行为者将通过分析故障来源而受到惩罚。

So how does this relate to sharding & scalability? Having this explicit finality allows more flexibility in how much (more accurately, how little) each node in the network has to do. Having more regular explicit finality allows further exploration of questions such as: what if not every node had to hold all of the state or all of the transactions? what if not every node had to validate every transaction? These questions of having heterogenous “responsibilities” of nodes in public blockchains are tackled by a workstream around blockchain sharding.

那么这与分片和可扩展性有什么关系呢？具有这种明确的终结性提供了
每个网络节点必须做的（更准确地说是更少的）工作的灵活性，有更多的定期明确的终局允许进一步探索问题，如：如何处理如果不是每个节点都保存所有的状态或所有的交易？如果不是每个节点都必须验证每个事务怎么办？公链上有关区块的这些问题都在区块分片工作中解决了。

So to go back to the point, if we are to explore each node in the network “doing less” or “knowing less,” it is hugely beneficial to consider only the past few epochs of finality rather than the entire probabilistic chain since the genesis block. Therefore, at this epoch time interval, finality doesn’t actually help with confirming a simple transaction, since transactions clear in number of confirmations fewer than the epoch time. Instead, finality will enable public blockchains to scale beyond the current ~10 transactions per second to larger orders of magnitude.

所以要回到这一点，如果我们要网络中的每个节点“做得更少”或者“知道得更少”，
那么只考虑过去几个 epoch 时间内的确定性比考虑创世块以来的整个区块链的确定性有巨大的好处。因此，在这个 epoch 时间间隔内，确定一个简单的交易实际上并不能提高最终性，因为确认次数比 epoch 时间少。相反，最终性将使公链扩展到比现在每秒至10个事务更大的数量级。

Because Ethereum is worth ~$28B 因为以太的价值

The final point is (5) a gentle transition from PoW. Here’s the context for newcomers such as myself. Ethereum’s explicit goals to move to PoS predates the significant increase in the value of ether this year. The plan is to start with a hybrid PoS overlay on top of the ethash PoW chain and to ramp up gradually towards “pure” PoS. Given the significant increase in ETH network value, this gradual transition to PoS is a prudent strategy to prevent potential value destruction while transitioning a significant piece of the underlying Ethereum infrastructure.

最后一点是（5）从 PoW 平缓过渡。象我这样的新手需要了解的以下原委：以太的明确目标是在今年大幅度提高以太网的价值。该计划是在 ethash PoW 链的基础上混合 PoS，然后逐渐转向纯粹的 PoS 实现。鉴于 ETH 网络价值的大幅增长，逐渐过渡到 PoS 是一种谨慎的策略，用于防止潜在的价值破坏，同时转移重要的以太坊基础设施。

以太坊 Capsper 初步 (上)