Bankless:重审以太坊Rollup路线图

Odaily星球日报Опубліковано о 2023-12-24Востаннє оновлено о 2023-12-24

Анотація

以太坊扩展为全球金融层的目标仍然遥不可及,但我们正在逐步实现这一目标。

原文标题:Recapping the Rollups Roadmap

原文作者:Arjun Chand

原文编译:Sharon,BlockBeats

以太坊的扩容一直是加密领域中最受争议的话题之一。在经过多次猜测后,社区确定了以 Rollup 为中心的以太坊发展路线图。但为什么选择了这条道路呢?

随着像 Solana 这样的区块链链在其卷土重来的过程中势头迅猛,现在是重新审视以太坊社区为何选择 Rollup 和模块化,并以此来实现全球规模的良机。

今天的文章,重点介绍了引导以太坊扩容工作朝着以 Rollup 为中心的路线图发展的关键理念和发展情况。接下来一起深入探讨一下。

以太坊的去中心化扩容愿景

以太坊的最终目标是成为全球协调的金融层。为了实现这一目标,它必须支持不同类型的应用程序,并对所有类型的用户都可访问。

在 2020 年,Vitalik 指出了以太坊网络的一个紧迫问题:他在 Augur 上下注时,不得不支付 17.76 美元的 Gas 费。这些高昂的费用表明,以太坊在面对不断增加的需求时难以扩展。结果,它变成了一个「为小众而非为全世界」的平台,也偏离了其最初的目标。

问题是显而易见的,以太坊需要能够以更低的成本处理更多交易。然而,解决方案可能会更为复杂,因为必须考虑和平衡许多因素。

一个主要考虑因素是优化三个属性:去中心化、安全性和可扩展性。这三者共同形成了「可扩展性的三难困境」,也就是任何区块链所面临的「不可能三角」困境。

Bankless:重审以太坊Rollup路线图

可扩展性三难题可以理解为一种平衡行为——在一个属性上的改进,通常意味着在另一个属性上的权衡。

随着时间的推移,许多被贴上「以太坊杀手」标签的区块链,一直在处理三难困境,通常会在追求可扩展性的同时,牺牲去中心化和安全性的方面。

然而,对以太坊社区来说,妥协去中心化从来都不是一个选择。这也是因为它与安全性密切相关,并赋予以太坊网络中立性、抗审查性和无许可性等性质,这些性质与可扩展性三难题中的性质同等重要。

在以太坊需求层次结构中,如果去中心化、安全性和可扩展性形成了「生理」需求,那么中立性、抗审查性和无许可性可以说形成了「安全」需求。这两个层次可以被称为以太坊网络的「基本需求」。

在认识到这些优先事项后,以太坊社区采取了一种长期的开发方法,该方法可能在中短期内牺牲可扩展性三难题的某些方面,但使以太坊能够「壮大到足以帮助全人类」。

这个策略就是模块化区块链论。它涉及在 L2 扩容解决方案,也就是所谓的 Rollup 上执行交易。由于这些交易发生在以太坊之外,它们可以更快速、更便宜。然而,由于它们的数据返回到了 L1,它们便继承了以太坊广泛的安全性。

Bankless:重审以太坊Rollup路线图

为什么选择 Rollup 来扩展以太坊?

到了 2020 年底,以太坊社区基本达成共识,采用 Rollup 作为中短期未来的主要扩展解决方案,形成了以 Rollup 为中心的以太坊发展路线图。使 Rollup 成为首选方案的关键因素有:

1. 与 Eth 2 (PoS)的前向兼容性——ArbitrumOptimism 等 Rollup 背后的公司分别在 2018 年和 2019 年推出。在那个时候,以太坊仍然使用 PoW 共识机制,而 PoS 正处于提案阶段。Rollup 与 PoW 和 PoS 兼容,为以太坊开发者社区提供了在关键网络升级(如从 PoW 切换到 PoS)的时间表方面所需的灵活性。

2. 技术可行性——到了 2020 年,Rollup 已经取得了实质性的发展。正如 Vitalik 在他的论坛帖子中所指出的,OptimisticRollup 团队(如 Optimism 和 Arbitrum)早在 2020 年初就推出了他们的测试网,并发布了在 2021 年主网启动的路线图。

与此同时,像 zkSync 和 Loopring 等 zkRollup 已经投入使用,并在其主网上拥有用户,尽管功能有限。这种进展使以太坊社区对 Rollup 技术的成熟度,及其从 2021 年起有效扩展以太坊的潜力充满信心。相比之下,其他扩展解决方案,如执行分片,仍然需要数年的时间。

对于在前一轮牛市期间以太坊主网的状态,迫切采用 Rollup 的程度进一步增强。Gas 费创下新高,使用户每笔交易花费数十美元。此外,一些应用程序,尤其是那些非金融用途的应用程序,由于高昂的费用而被迫关闭。

这些因素共同影响了以太坊社区,使其坚定地将 Rollup 作为可预见未来的扩展策略,认识到其即时影响和长期潜力。

尽管 Rollup 的直接好处显而易见,但多 Rollup 生态系统也带来了一些挑战。以用户视角为出发点,以太坊社区提出了一些关键问题,以下是对这些问题解决进展的概述。

Bankless:重审以太坊Rollup路线图

以太坊 Rollup 路线图的现状

我们处于一个多 Rollup 生态系统中,许多 Rollup 已经取得了显著的发展。

在这个情况下,以太坊通过 Rollup 进行扩展,充当清算和数据可用性的基础层。所有 Rollup 都继承了以太坊的安全性,并将其用于确认交易和存储数据。

在安全性方面,以太坊无疑是生态系统中最安全的区块链之一。已有超过 3350 万枚 ETH 用于保障网络安全,使得网络在当前 ETH 价格(1 ETH= 2000 美元)下拥有超过 670 亿美元的经济安全性。

然而,随着 L2 中的 TVL 已经超过 160 亿美元,抵押品创下历史新高,社区对 Rollup 排序器的中心化和运营商提取 MEV 表达了一些担忧。尽管有不少 Rollup 吸引了大量用户,但正如 L2 BEAT 对 Rollup 成熟度的分析所示,这些系统中的大多数仍处于早期发展阶段。

Bankless:重审以太坊Rollup路线图

另一个积极发展的关键领域,是增强数据可用性。在这一努力的关键部分是 Danksharding,这是以太坊路线图中「激增」阶段的一个关键功能。Danksharding 旨在显著降低 Rollup 在以太坊上验证交易的成本,从而推动网络的扩展能力。为此,一些创新项目,如 Celestia 和 EigenDA,专注于为 Rollup 提供廉价且去中心化的数据可用性层。

在 Danksharding 完全实施之前,以太坊社区正在进行 proto Danksharding,即 EIP-4844 ,作为一个中间步骤。这一步骤以简化形式引入了 Danksharding 的一些概念,如 blob。预计 proto Danksharding 将很快实施,而完整的 Danksharding 还需数年时间。

结束语

将以太坊扩展为全球金融层的目标仍然遥不可及,但我们正在逐步实现这一目标。最令人振奋的是,社区在路线图和愿景上保持了一致,一些最杰出的思想家正在共同努力将这一愿景变为现实。

展望未来,成千上万个专业化的 Rollup 涌现的潜力令人难以置信,每个都满足独特的用例并推动创新。这种演进将使以太坊能够成为其所渴望的全球金融层,支持各种各样的应用程序和用户。所有这些 Rollup 都将支付费用,使其数据在以太坊上得到结算和安全保障,从而提高以太坊的加密经济安全性。

Пов'язані матеріали

If the AI Bubble Is Already Bursting, Who Will Truly Survive?

If the AI Bubble is Bursting, Who Will Remain? The debate over an AI bubble is intensifying, with figures like Ray Dalio warning of high levels and Jensen Huang seeing immense, early-stage opportunity. Both views hold truth: a speculative bubble in capital markets likely exists, mirroring the dot-com era, but the underlying technological shift is real and transformative. History shows that while bubbles burst—wiping out overvalued companies and speculative capital—they often leave behind critical physical and digital infrastructure. The dot-com bust, for instance, eliminated many firms but left the global fiber optic networks and data centers that enabled the rise of Amazon, Netflix, and cloud computing. Today's massive AI infrastructure investments (projected at trillions by 2030) in data centers, power, cooling, and GPUs may follow a similar path, creating the foundation for future applications. A key divergence from past bubbles is the "Jevons Paradox" effect in AI. As the cost of AI inference has plummeted by over 99.7% since 2023, enterprise spending on AI has skyrocketed. Cheap "tokens" have unlocked vast, previously uneconomical use cases, moving AI from simple chatbots into core business workflows—code generation, legal document review, scientific simulation, and financial analysis. The market is now in a phase of self-correction, weeding out superficial "API-wrapper" startups, but this cleansing process strengthens the ecosystem. The long-term trajectory is clear. The value is gradually shifting from capital expenditure (CapEx) on hardware to operational expenditure (OpEx) on transformative applications. As AI becomes a utility, the winners will be firms that deeply integrate it to solve vertical industry problems in law, healthcare, finance, and manufacturing. The泡沫 will recede, but the foundational shift towards an AI-powered era across all sectors is irreversible. The underlying productive force of AI contains no bubble.

marsbit6 хв тому

If the AI Bubble Is Already Bursting, Who Will Truly Survive?

marsbit6 хв тому

If the AI Bubble Is Already Bursting, Who Will Truly Remain?

**Summary: If the AI Bubble is Bursting, What Will Remain?** The debate around an AI bubble is intensifying, with figures like Ray Dalio warning of high valuations while Jensen Huang sees immense opportunity. This echoes the dot-com bubble, which saw massive wealth destruction but ultimately left behind critical infrastructure like undersea cables and broadband, enabling future giants like Amazon and Netflix. Similarly, today's AI boom involves trillions invested in data centers, power, cooling, and GPUs, while application-layer revenue remains comparatively modest. This investment-disparity signals a bubble. However, the core technological progress is real and accelerating. AI inference costs have plummeted by over 99.7% since 2023, making intelligence increasingly cheap and accessible. This cost collapse is unlocking vast new demand. Instead of reducing spending, enterprises are tripling their AI cloud expenditure. Cheap "tokens" enable AI to move beyond simple chatbots into complex workflows—automating code writing, legal document review, financial analysis, and scientific research. This follows "Jevons's paradox": improved efficiency leads to greater total consumption. The market is now undergoing a necessary purification, weeding out "API-wrapper" startups with no real moat. The deeper evolution involves a shift from capital expenditure (CapEx) on infrastructure to operational expenditure (OpEx) on value-creation in applications. While hardware vendors currently profit most, long-term value will migrate to AI-native firms solving vertical industry problems. Ultimately, a market correction will cleanse speculative excess but will not reverse the AI+ trend. The massive physical and algorithmic infrastructure being built will endure, becoming a cheap, utility-like foundation. Just as the internet became indispensable to all industries post-2000, AI is poised to empower and redefine every sector, moving society irreversibly toward an intelligence-augmented era. The bubble may burst, but the underlying productive momentum is solid.

链捕手13 хв тому

If the AI Bubble Is Already Bursting, Who Will Truly Remain?

链捕手13 хв тому

Microsoft CEO: In the AI Era, How Do You Define a Company's Moat?

Microsoft CEO Satya Nadella argues that in the AI era, a company's true competitive edge, or "moat," is not determined by choosing the single most powerful model, but by its ability to build a continuous "learning loop." This system integrates and evolves by connecting human workflows, domain expertise, organizational judgment, and employee experience. He posits that future companies will accumulate two types of capital: Human Capital (employee knowledge, judgment, creativity) and "Token Capital" (a firm's own built and owned AI capabilities). Importantly, AI amplifies rather than devalues human capital. Human direction is essential to guide progress, as computational power alone is aimless. The core opportunity lies in creating a closed-loop system where human and token capital reinforce each other in a compound, self-improving cycle. A company must be able to preserve its unique institutional knowledge—its "company veteran" expertise—even if it switches underlying general-purpose AI models. This requires private evaluation benchmarks, reinforcement learning environments based on internal data, and queryable knowledge bases. Nadella warns against a future where economic value is concentrated by a few dominant models that commoditize entire industries' knowledge. Instead, the priority should be building a broad "frontier ecosystem" where every company, industry, and nation can own its learning loop. This allows organizations to retain control of their intellectual property, amplify employee capabilities, and ensure the economic value created by AI is captured within their own businesses and communities. True corporate sovereignty in the AI age comes from turning organizational knowledge into a compounding system that creates enduring, defensible value.

marsbit48 хв тому

Microsoft CEO: In the AI Era, How Do You Define a Company's Moat?

marsbit48 хв тому

ETFs Are Just the Ticket: The True Institutionalization of Bitcoin Is Happening Where You Can't See It

Beyond the Bitcoin ETF spotlight, a deeper institutionalization is underway, leveraging Bitcoin as a foundational financial primitive. Institutions are using Bitcoin for purposes long reserved for assets like U.S. Treasuries and gold: as collateral for loans, insurance reserves, and the backbone of rated bonds. Examples include a Barbados-based insurer capitalizing with $40M in Bitcoin reserves and Ledn's $188M securitization of Bitcoin-backed loans, which received the first-ever investment-grade rating (BBB-) from S&P for a digital asset-backed security. This structure was stress-tested during a 27% price drop in early 2026, triggering automatic liquidations that functioned as designed but revealed the systemic risk of synchronized selling across leveraged positions. Infrastructure is evolving to support this, with platforms like Anchorage Digital's Atlas network enabling secure, institutional-grade settlement and collateral management. Strategies like basis trades and corporate treasuries (exemplified by companies like MicroStrategy issuing billions in equity and debt to fund Bitcoin acquisitions) further integrate Bitcoin into financial mechanics. While ETFs solved "how to own" Bitcoin, these developments answer "what to do with it," embedding the asset into the working machinery of finance—as collateral upon which loans, derivatives, and structured products are built. The real, enduring institutional shift is happening in these largely invisible plumbing and financing systems.

marsbit54 хв тому

ETFs Are Just the Ticket: The True Institutionalization of Bitcoin Is Happening Where You Can't See It

marsbit54 хв тому

ZEC Co-Founder Responds to Orchard Vulnerability: No Signs of Theft, Orchard Pool to Be Sealed

ZEC Co-Founder Addresses Orchard Vulnerability: No Signs of Theft, Plans to Sunset Orchard Pool A security vulnerability was recently discovered in Zcash's Orchard shielded pool, raising key concerns. The primary questions are whether the flaw was exploited, if user funds are safe, whether users can verify the total ZEC supply, and if other similar vulnerabilities exist. Analysis suggests the vulnerability was likely not exploited prior to its discovery. It was found proactively by a researcher using specialized tools, not due to an active breach. The development team and mining pools acted quickly to contain the issue. Typical financially-motivated attacks would likely have left visible on-chain evidence, which has not been observed. User funds in Orchard are considered safe and should be recoverable, assuming no prior exploitation. If the flaw was never used, all legitimate funds can be withdrawn. The article outlines risks associated with moving funds to transparent addresses or other pools, but concludes that leaving assets in place is a reasonable option. Currently, users cannot independently verify that the total ZEC supply hasn't been inflated due to this bug. However, the planned Ironwood network upgrade is designed to resolve this. It will permanently close the Orchard pool to new deposits and internal transfers, allowing only withdrawals. This mechanism will cap total withdrawals at the amount of legitimately deposited funds, enabling anyone to cryptographically verify the supply post-upgrade. Multiple teams, including Shielded Labs, have conducted extensive audits focused on counterfeiting vulnerabilities, assisted by advanced AI tools. No additional flaws of this type have been found so far, increasing confidence that no other similar undisclosed vulnerabilities exist. In summary, evidence indicates the Orchard bug was probably not used, user funds are secure, and no other counterfeiting flaws are currently known. The upcoming Ironwood upgrade will restore users' ability to independently verify the total ZEC supply, closing this chapter.

Foresight News59 хв тому

ZEC Co-Founder Responds to Orchard Vulnerability: No Signs of Theft, Orchard Pool to Be Sealed

Foresight News59 хв тому

Торгівля

Спот
Ф'ючерси
活动图片