What is an EIP (Ethereum Improvement Proposal)?

Here's the beautiful irony of Ethereum: it's this massively valuable, globally distributed computer worth hundreds of billions of dollars, yet anyone with a GitHub account can propose changes to it. That's essentially what an EIP is - a standardized document where developers, researchers, and community members propose improvements to Ethereum. Think of EIPs as the democratic legislation system for a digital nation, except instead of politicians, you've got cryptographers, developers, and occasionally some very passionate community members debating the future of money and computation.

The EIP process is how Ethereum has evolved from a simple smart contract platform in 2015 to the sophisticated ecosystem we see today in 2025. Every major feature you've heard about - from the token standards that made DeFi possible to the merge that took Ethereum proof-of-stake - started as someone sitting down and writing an EIP. It's decentralized governance in action, and unlike many blockchain governance systems that promise decentralization but deliver oligarchy, the EIP process actually works.

Origins and Structure

Ethereum's founders, having observed Bitcoin's sometimes contentious upgrade process, wanted a more structured approach to protocol evolution. The first EIP, appropriately numbered EIP-1, was authored in October 2015, just months after Ethereum's launch. It outlined the EIP process itself - a meta-proposal about proposals.

An EIP isn't just a random wish list. It follows a specific structure: a preamble containing metadata, an abstract explaining what the proposal does, a motivation section explaining why Ethereum needs this change, a specification providing precise implementation details, a rationale explaining design decisions, and crucially, a backwards compatibility section. Breaking changes require extraordinary justification and careful migration plans.

As of 2025, the EIP repository contains over 7,000 proposals in various states. Some have fundamentally changed Ethereum's architecture. Others proposed ideas that seemed brilliant but were eventually abandoned.

Types of EIPs

Core EIPs change Ethereum's core protocol - adding new opcodes to the EVM, changing the block gas limit, or modifying consensus rules. These require consensus from client developers and network-wide hard forks to implement.

EIP-1559, implemented in August 2021, is the poster child for transformative Core EIPs. It fundamentally restructured Ethereum's fee market, introducing a base fee that burns with each transaction. As of January 2025, over 4.2 million ETH (worth approximately $8.4 billion at $2,000 per ETH) has been burned through EIP-1559, making ETH potentially deflationary.

ERCs (Ethereum Request for Comments) define application-level standards for how smart contracts should behave. These don't change Ethereum's protocol but create conventions making different applications interoperable.

ERC-20, proposed in November 2015, might be the most economically significant technical standard ever created. It defined a simple interface for fungible tokens. Before ERC-20, every token implementation was different, making integration nightmare fuel. After ERC-20, anyone could create a token that automatically worked with every wallet, exchange, and DeFi protocol. As of January 2025, there are over 500,000 ERC-20 tokens deployed, with top tokens having market caps in the tens of billions.

ERC-721, authored in January 2018, did for non-fungible tokens what ERC-20 did for fungible ones. Each ERC-721 token is unique, making them perfect for digital art, domain names, game assets, and anything where uniqueness matters. ERC-721 launched the NFT phenomenon that peaked in 2021-2022 when OpenSea processed over $3 billion in monthly volume.

The EIP Lifecycle

Getting an EIP from concept to implementation is neither quick nor easy. You write up your proposal and submit a pull request to the EIP GitHub repository. EIP editors review formatting, assign a number, and if acceptable, it enters Draft status.

Your EIP is now public and the Ethereum community can comment across GitHub, Ethereum Magicians forum, AllCoreDevs calls, and Discord. This review period is where proposals either gain traction or quietly fade.

If your EIP survives with positive support, you work with client developers to implement it. Once implemented and tested on testnets, Core EIPs enter Considered for Inclusion status. If included in a hard fork and successfully activated on mainnet, the EIP moves to Final status.

Many EIPs enter Stagnant status after 6+ months of inactivity. This can take years - EIP-1559 was first proposed in April 2019 but didn't activate until August 2021.

The Politics of Protocol Development

Every EIP creates winners and losers. EIP-1559's fee burning was economically beneficial for ETH holders but reduced revenue for miners. Some mining pools opposed it vocally.

Account abstraction (EIP-4337 and related proposals) has been discussed since 2016 but took until 2023 to get working implementation, not because the technology was impossible, but because there were multiple competing visions with different priorities.

Client diversity adds complexity. Ethereum maintains multiple client implementations to prevent single points of failure. This means Core EIPs need specification clear enough that different teams can implement them identically.

While Ethereum claims decentralization, a relatively small group of core protocol developers exerts significant influence over which EIPs get prioritized. This makes practical sense but can create bottlenecks.

Notable EIPs That Changed Everything

EIP-4844 (Proto-Danksharding), successfully implemented in early 2024, introduced "blob-carrying transactions" - optimized for rollup data availability. Layer 2 rollups post transaction data to Ethereum for security, but this was expensive. EIP-4844 created cheaper temporary storage, reducing L2 costs by 5-10x. This doesn't generate hype headlines but genuinely improves ecosystem economics.

EIP-1167 (Minimal Proxy Contract) standardized a way to deploy cheap proxy contracts, reducing deployment gas costs by over 95% in many cases. This enabled new patterns in DeFi and NFTs, but end users never interact with it directly.

The Account Abstraction Journey

Account abstraction deserves special attention because its journey illustrates the challenge of implementing complex changes to mature protocols. Vitalik has talked about it since Ethereum's early days - eliminate the distinction between externally owned accounts and smart contract accounts. Currently, every transaction must originate from an EOA, creating UX problems.

This vision spawned numerous attempts: EIP-86 (2017), EIP-2938 (2020), EIP-3074 (2020), EIP-4337 (2021). EIP-4337 took a clever compromise: implement account abstraction without protocol changes using a separate "UserOperation" mempool and "bundler" infrastructure. By early 2025, it has real adoption, with major wallets implementing it.

Account abstraction's decade-long journey shows both Ethereum's careful approach and the challenge of evolving a protocol with billions of dollars of existing infrastructure.

The EIP Process in 2025

The EIP process itself continues evolving. Early discussions happened primarily on GitHub and Ethereum Magicians. By 2025, governance is more fragmented across Discord, Twitter, and dedicated calls. This has pros (broader participation) and cons (harder to track discussions).

Core development has become more professionalized. Most core developers are funded through grants, foundations, or companies. This brings stability but some worry about centralization of influence. AllCoreDevs calls have become the de facto decision-making body.

Why EIPs Matter

The EIP process represents one of the most successful experiments in open-source governance for critical infrastructure. It has coordinated thousands of developers to build and evolve a protocol securing hundreds of billions in value, all without central authority.

This matters beyond cryptocurrency. As more infrastructure becomes software-based and globally distributed, we need governance mechanisms that can coordinate large-scale technical decisions without centralized control. The EIP process demonstrates that decentralized communities can execute complex changes with clear procedures, technical rigor, and community buy-in.

Most importantly, EIPs embody the principle that people affected by a system should have a voice in how it evolves. The EIP process gives every Ethereum user a mechanism to participate in shaping the protocol's future. Every merged EIP represents thousands of hours of thought, debate, and implementation by people who believe decentralized infrastructure can be both functional and fair.

References

1. "EIP-1: EIP Purpose and Guidelines" - Ethereum Improvement Proposals, https://eips.ethereum.org/EIPS/eip-1
2. "The History and Impact of EIP-1559" - Consensys Research, 2024, https://consensys.net/blog/ethereum-eip-1559-retrospective/
3. "ERC-20 Token Standard" - Ethereum.org Documentation, https://ethereum.org/en/developers/docs/standards/tokens/erc-20/
4. "The Road to Account Abstraction: EIP-4337 and Beyond" - Ethereum Foundation Blog, 2023, https://blog.ethereum.org/2023/01/account-abstraction-roadmap
5. "Ethereum's Governance: How EIPs Shape the Protocol" - Bankless Research, 2024, https://bankless.com/ethereum-governance-eips-explained
6. "EIP-4844: Proto-Danksharding Implementation and Impact" - L2Beat Research, 2024, https://l2beat.com/research/eip-4844-analysis
7. "NFT Standards: From ERC-721 to ERC-6551" - OpenSea Developer Resources, 2024, https://docs.opensea.io/docs/nft-standards-overview
8. "The Politics of Protocol Development" - Ethereum Research Forum, 2024, https://ethresear.ch/t/protocol-governance-challenges/
9. "Ultra Sound Money: ETH Supply Analysis Post-EIP-1559" - Ultrasound.money, 2025, https://ultrasound.money
10. "Comparing Blockchain Governance Models" - Messari Research, 2024, https://messari.io/report/blockchain-governance-comparison