The Simple Story Everyone Believed
The original concept for sharding was beautifully simple and easy to grasp. Imagine the Ethereum network is a single, congested highway with bumper-to-bumper traffic. Transaction fees (gas) are the high tolls you pay for the privilege of inching forward.
Sharding, in this analogy, was the plan to build 64 new lanes parallel to the main highway. By splitting the blockchain into many smaller chains, or 'shards,' the network could process transactions in parallel, dramatically increasing capacity and lowering fees. Each shard would handle its own transactions and smart contracts, spreading the load. This was 'execution sharding,' and it’s the version of the story that stuck in many investors' minds: more lanes, more speed, lower tolls, directly on the main Ethereum highway.
The Big Pivot to Data Blobs
The problem is, that's not the plan anymore. The Ethereum developer community realized building and securing 64 independent execution chains was incredibly complex. At the same time, another scaling solution gained momentum: Layer 2 rollups. These are separate blockchains (like Arbitrum and Optimism) that process transactions off the main chain and then post a compressed summary back to Ethereum. This inspired a pivot. Instead of building more lanes for the main highway, what if Ethereum could become the world's most efficient and secure provider of 'paved ground' for these other highways to build on? This led to a new, rollup-centric roadmap. The plan shifted from 'execution sharding' to 'data sharding,' specifically through an upgrade called Proto-Danksharding (EIP-4844) that went live in March 2024. This introduced 'blobs'—a new, cheaper way for rollups to post their transaction data to Ethereum. This data is only stored temporarily, just long enough to be verified, which is a key cost-saving innovation.
Why It's All About the Rollups
The core misunderstanding is thinking Ethereum itself is supposed to get dramatically faster and cheaper for the average user. With the new plan, it won't—at least not directly. The goal of Proto-Danksharding and the future full Danksharding is to make it incredibly cheap for Layer 2 rollups to operate. Over 90% of a rollup's cost came from posting data to the main Ethereum chain. By creating a separate, low-cost data market with blobs, EIP-4844 slashed those costs. The savings are then passed on to users of the Layer 2 networks. So, while a transaction on Ethereum's base layer might still be expensive during peak times, a transaction on a rollup that uses it for security becomes much, much cheaper. The new strategy is to scale Ethereum by enabling a thriving ecosystem of Layer 2 solutions, not by trying to cram all activity onto the base layer.
The New Investment Thesis
This changes the investment thesis for Ethereum. It's no longer about becoming the world's fastest 'supercomputer' that processes every transaction. Instead, Ethereum is positioning itself as the foundational settlement and data availability layer for the entire crypto economy. It's like the transition from a central mainframe to the internet protocol (TCP/IP) that enabled countless applications to be built on top. Ethereum's value now derives from being the ultimate source of security and truth for hundreds of other chains. Its job is not to do everything, but to provide the one thing that everything else needs: verifiable, secure, and (now) cheap data availability. Investors who still expect sharding to bring super-low gas fees directly to Ethereum's mainnet are missing the pivot. The real action and the direct beneficiaries of these scaling upgrades are the Layer 2 solutions and the applications built upon them.
