Module 6 · Ethereum & Shared Applications / 6.5
An unfinished machine, improved in public.
Read Ethereum’s history as a sequence of problems and responses.
A busy bridge teaches its builders where the bottlenecks are. A crowded network does something similar.
Ethereum’s history is easier to follow when each upgrade answers a question: What became useful? What broke? What did people change?
First, people found things to do.
Ethereum launched in 2015 with a shared platform for programs. The DAO crisis in 2016 exposed application risk and disagreement over intervention. The network’s existence did not settle either problem.
In 2017, token fundraising and a collectible game called CryptoKitties brought more activity. Congestion made an important limit visible: a shared chain has limited space, and users compete to use it.
The following years brought lending, stablecoins, and automated exchanges. Applications could connect to each other’s contracts. That made new combinations possible, while allowing a failure in one component to affect others.
The lesson is not that every popular experiment became valuable. Usage revealed both useful capabilities and expensive constraints.
Launch. A public platform for shared programs.
The application chapters, in a little more detail
The 2017 ICO boom made issuing and selling tokens easy; it also enabled weak projects and fraud. A token sale is a financing mechanism, not evidence of a good product.
Maker and Uniswap became early examples of collateral-backed applications and pool-based exchange. In 2020, lending and token incentives drew attention to DeFi. Some activity reflected demand for a service; some reflected rewards for participating.
The 2021 NFT boom made unique token records familiar. It also showed why a token, the associated image, and intellectual-property rights must be examined separately. Market attention is not a substitute for that distinction.
Then, the engine changed.
The Beacon Chain began in December 2020, preparing a proof-of-stake system alongside Ethereum’s existing chain. On September 15, 2022, the Merge joined Ethereum’s execution history to that system.
The balances, contracts, and transaction history continued. The mechanism choosing and confirming blocks changed from proof of work to proof of stake. Ordinary holders did not need to exchange ETH for a new “ETH2” token.
This greatly reduced the network’s energy use. It did not automatically create cheap, unlimited transactions. Consensus and execution capacity are different parts of the design.
The 2023 Shapella upgrade enabled staking withdrawals. The next lesson explains why staking deposits, validator duties, and withdrawal rules belong in the same picture.
The execution history continues. This switch depicts a coordinated protocol upgrade, not an automatic change or required token exchange.
More room, with more layers.
Ethereum increasingly relies on rollups to execute groups of transactions and use the base chain for settlement and data. The 2024 Dencun upgrade introduced blobs: a form of data space designed for this work.
Pectra, activated in May 2025, included changes to account capabilities and validator operations. Fusaka, activated in December 2025, introduced PeerDAS, a way to check data availability by sampling rather than requiring every participating node to download every blob in full.
These changes target different constraints. A cheaper data path for a rollup is not a promise that every application transaction will always be cheap. Fees still depend on demand and the network or service being used.
Roadmaps describe intended work. They are not proof that a feature has shipped. When comparing networks, distinguish a deployed feature, a limited experiment, and a proposal.
Markets and institutions are another timeline
U.S. spot ether exchange-traded products received listing-rule approvals in May 2024 and began trading in July. Those market events did not change Ethereum’s validation rules or endorse every application on it.
Other execution platforms and rollups compete for developers and users. More chains do not automatically mean more useful applications, and Ethereum’s early position does not guarantee a permanent lead.
The idea to keep
History is a record of choices under constraints. Follow the problem, the change, and the new tradeoff—not only the price chart.