Ethereum just published the most detailed upgrade plan in its history. Seven upgrades. Five goals. One massive rebuild.

If you are wondering who the idiot this guide is written for….it’s me.

Ethereum researcher Justin Drake published what he calls the "Strawmap," a proposed schedule of seven major upgrades stretching through 2029. Vitalik Buterin, Ethereum's co-founder, called it "very important" and described the cumulative effect as a "Ship of Theseus" rebuild of Ethereum's core.

That metaphor is worth understanding.

The Ship of Theseus is an ancient Greek thought experiment: if you replace every plank of a ship one at a time, and eventually every single piece has been swapped out, is it still the same ship?

That's what the Strawmap proposes for Ethereum.

By 2029, every major part of the system gets replaced. But at no point is there a planned "stop-the-world rewrite." The aim is backward-compatible upgrades that keep the chain live while planks are swapped, though each upgrade still requires operators to update their software, and edge cases can change. It’s a complete rebuild disguised as incremental upgrades. To be strictly precise, while the consensus and execution logic are being rebuilt, the state (user balances, contract storage, and history) is preserved across all forks. The "ship" is rebuilt while it is still carrying its cargo. All aboard!

"Why not just start over from scratch?" Because you can't reboot without losing the thing that makes Ethereum valuable: the apps already running on it, the money already flowing through it, the trust already built. You have to replace the planks while the ship is sailing.

The name "Strawmap" is a mashup of "strawman" and "roadmap." A strawman is a preliminary proposal that's put out there knowing it's imperfect, specifically to get people poking holes in it. So this isn't a promise. It's a starting point for debate. But it is the first time Ethereum's builders have spelled out a structured, time-bound upgrade path with clear performance targets.

The people working on this are some of the best cryptographers and computer scientists on the planet. And it's all open source. No licensing fee. No vendor contract. No enterprise sales team. Any company, any developer, any country can build on it. The upgrades that JPMorgan will benefit from are the same ones available to a 3-person startup in Sao Paulo.

Imagine if a global consortium of world-class engineers were rebuilding the internet's financial plumbing from scratch and you could just... plug in.

How Ethereum Actually Works (60-Second Version)

Before we talk about where it's going, here's what it is today.

Ethereum is basically a shared global computer. Instead of one company running a server, thousands of independent operators around the world each run a copy of the same software.

These operators verify transactions independently. A subset of them, called validators, also stake their own money (ETH) as a security deposit. If a validator tries to cheat, they lose it. Every 12 seconds, validators agree on which transactions happened and in what order. That 12-second window is called a "slot." Every 32 slots (about 6.4 minutes) forms an "epoch."

True finality, the point where a transaction becomes irreversible, takes roughly 13 to 15 minutes, depending on when your transaction lands in the cycle

Ethereum processes on the order of 15 to 30 transactions per second, depending on how complex each transaction is. For comparison, Visa's network can handle over 65,000 per second. That gap is why most Ethereum apps today run on "Layer 2" networks, separate systems that batch up lots of transactions and then post a summary back to Ethereum's ground floor for security.

The system that gets all those operators to agree is called a "consensus mechanism." Ethereum's current one works and is battle-tested, but it was designed for an earlier era and limits what the network can do.

The Strawmap aims to fix all of that. One upgrade at a time.

The Strawmap's Five Major Goals

The roadmap organizes everything around five goals. Ethereum already works. Billions of dollars move through it daily. But it has real limits on what can be built on top of it. These five targets are about removing those limits.

1. Fast L1: Finality in Seconds

When you send a transaction on Ethereum today, you wait roughly 13 to 15 minutes before it's truly final, meaning irreversible, done, no take-backs.

The fix: Replace the engine that gets all those operators to agree. The goal is to achieve finality in a single round of voting within each slot. Minimmit is one leading candidate in the research, a protocol designed for ultra-fast consensus, but the exact design is still being refined. What matters is the target: finality within a single slot. Then the slot times themselves get compressed: the proposed path is 12 seconds → 8 → 6 → 4 → 3 → 2.

Finality isn't just about speed; it's about certainty. Think about a wire transfer. The time between "sent" and "settled" is the window where things can still go wrong.

If you're moving a million-dollar payment, or settling a bond trade, or closing a real estate deal on a blockchain, those 13 minutes of uncertainty are a problem. Shrink it to seconds and you've fundamentally changed what this network can do. Not just for crypto-native apps, but for anything that moves value.

2. Gigagas: 300x Bigger

Ethereum's main network handles about 15–30 transactions per second. That's a bottleneck.

The fix: The Strawmap targets 1 gigagas per second of execution capacity, which works out to roughly 10,000 TPS for typical transactions (the exact number depends on how complex each transaction is, since different operations consume different amounts of gas). The key idea is a technology called "zero-knowledge proofs" (ZK proofs).

Here's the simplest way to think about it: right now, every operator on the network has to redo every single calculation to check that it's correct. It's like having every employee in a company independently redo every colleague's math. Safe? Yes. Wildly inefficient? Also yes.

ZK proofs let you check a compact mathematical receipt that proves the math was done correctly. Same trust, a fraction of the work.

The software that generates these proofs is still too slow. Current versions take minutes to hours for complex work. 

Getting that down to seconds, a roughly 1,000x improvement, is an active research problem, not just an engineering challenge. Teams like RISC Zero and Succinct are making rapid progress, but this is still at the frontier.

A 10,000 TPS mainnet with fast finality means simpler, fewer moving parts. Fewer things that can go wrong.

3. Teragas L2: 10 Million TPS Across the Express Lanes

You still need Layer 2 networks for truly massive volume (and customisation). Today, L2s are limited by the amount of data Ethereum's main network can handle for them.

The fix: A technique called "Data Availability Sampling" (DAS). Instead of every operator downloading all the data to verify it exists, they each check random samples and use math to verify the complete dataset is intact. Think of it like checking that a 500-page book is actually on the shelf by randomly flipping to 20 different pages; if they're all there, you can be statistically certain the rest is too.

PeerDAS shipped in the Fusaka upgrade, which laid the groundwork for everything the Strawmap builds on. Scaling from there to the full target means iterative expansion: more data capacity each fork, stress-testing network stability at every step.

10 million transactions per second across the L2 ecosystem opens doors to what is currently impossible on any blockchain. Think global supply chains where every product and shipment has a digital token. Or millions of connected devices generating verifiable data. Or micropayment systems handling fractions of a cent. These workloads are too big for any existing network. At 10 million TPS, they fit with room to spare.

4. Post-Quantum L1: Preparing for Quantum Computers

Ethereum's security relies on math problems that are extremely hard for today's computers to solve. This applies across the system, both the signatures that individual users make when sending transactions and the signatures that validators use to reach consensus. Quantum computers, if they ever become powerful enough, could crack both, potentially allowing someone to forge transactions or steal funds.

The fix: Migrate to new cryptographic methods (hash-based schemes) that are believed to be resistant to quantum attacks. This is a later-stage upgrade because it touches nearly everything in the system, and the new methods use much larger data (kilobytes instead of bytes) which changes the economics of block sizes, bandwidth, and storage across the entire network.

Quantum attacks on today's cryptography are likely years to decades away. But if you're building infrastructure meant to last, infrastructure that might hold trillions of dollars in value, "we'll figure it out later" isn't a real answer.

5. Private L1: Making Transactions Confidential

Everything on Ethereum is public by default. Unless you're using a privacy app like Railgun or a privacy-focused L2 like ZKsync or Aztec, every transaction, every amount, every counterparty is visible to anyone.

The fix: Build confidential transfers directly into Ethereum's core. The technical goal is to let the network verify that a transaction is valid, that the sender has the funds, and that the math checks out, without revealing the actual details. You can prove "this is a legitimate $50,000 payment" without revealing who paid whom, or what for.

Today there are workarounds. EY and StarkWare announced Nightfall on Starknet in February 2026, bringing privacy-preserving transactions to a Layer 2 environment. But workarounds add complexity and cost. Building privacy into the foundation removes the need for middleware entirely.

The Seven Forks (Upgrades)

The Strawmap proposes seven upgrades on a roughly six-month cadence, starting with Glamsterdam. Each upgrade is deliberately scoped to change only one or two major things at a time, because if something breaks, you need to know exactly what caused it.

Hegotá follows with further structural improvements. The remaining forks (I* through M*) extend through 2029, progressively rolling out faster consensus, ZK proofs, expanded data availability, quantum-resistant cryptography, and privacy features.

Why Does This Take Until 2029?

Because some of these problems are genuinely unsolved.

Replacing the consensus mechanism is the hardest. Imagine replacing a plane's engines mid-flight while thousands of co-pilots all have to agree on each change. Every change needs months of testing and formal verification. And the push to get cycle times below 4 seconds eventually hits a physics problem: it takes about 200 milliseconds for a signal to travel across the globe and back. At some point, you're fighting the speed of light.

Making ZK provers fast enough is the other frontier problem. The gap between the current speed (minutes) and the target (seconds) is roughly 1,000x. This requires both mathematical breakthroughs and purpose-built hardware.

Scaling data availability is hard but more tractable. The math works. The challenge is doing it carefully on a live network holding hundreds of billions in value.

Post-quantum migration is an operational nightmare because the new signatures are so much larger that they change the economics of everything.

Native privacy is politically sensitive on top of being technically hard. Regulators worry that privacy tools enable money laundering. Engineers have to build something that's private enough to be useful but transparent enough to satisfy compliance requirements, and it also has to be quantum-resistant.

And these can't all happen at once. Some upgrades depend on others. You can't scale to 10,000 TPS without mature ZK proofs. You can't scale L2s without working on data availability. Those dependency chains determine the timeline.

Three and a half years is actually aggressive for what's being attempted.

2029?

First, there's a wildcard. The Strawmap explicitly notes that "the current draft assumes human-first development. AI-driven development and formal verification could significantly compress schedules."

In February 2026, a developer named YQ bet Vitalik that one person could use AI agents to code an entire Ethereum system targeting the 2030+ roadmap. Within weeks, he shipped ETH2030: an experimental Go execution client claiming roughly 713,000 lines of code with all 65 Strawmap items implemented, labeled as running on test networks and mainnet.

Is it production-ready? No. As Vitalik noted, there are almost certainly critical bugs throughout, and in some cases, probably stub implementations where the AI didn't even attempt the full version. But Vitalik's response is worth reading carefully: "Six months ago, even this was far outside the realm of possibility, and what matters is where the trend is going... People should be open to the possibility (not certainty! possibility) that the Ethereum roadmap will finish much faster than people expect, at a much higher standard of security than people expect."

Vitalik's key insight is that the right way to use AI isn't just to go faster. It's to take half the gains in speed and half in security: more testing, more mathematical verification, more independent implementations of the same thing.

The Lean Ethereum effort is working on machine-checked formal verification for parts of the cryptographic and proof stack. Bug-free code, long considered an idealistic fantasy, might actually become a basic expectation.

The Strawmap is a coordination document, not a promise. Its targets are ambitious, its timelines are aspirational, and its execution depends on hundreds of independent contributors.

But the question isn't really whether every target gets hit on schedule. It's whether you want to build on the platform with this trajectory, or compete against it.

And the fact that all of this, the research, the breakthroughs, the cryptographic migrations, is happening in the open, for free, available to anyone... that's the part of this story that deserves way more attention than it gets.

Disclaimer:

1. This article is reprinted from [[Snapcrackle](https://x.com/Snapcrackle/status/2029174289379811594)]. All copyrights belong to the original author [*Snapcrackle*]. If there are objections to this reprint, please contact the Gate Learn team, and they will handle it promptly.

2. Liability Disclaimer: The views and opinions expressed in this article are solely those of the author and do not constitute any investment advice.

3. Translations of the article into other languages are done by the Gate Learn team. Unless mentioned, copying, distributing, or plagiarizing the translated articles is prohibited.

**