Hey you!
Welcome back to “that’s what she said”, the newsletter which is your biweekly dose of all things web3. Last time, we covered Ethereum's origin story — how a teenage Vitalik got frustrated with Bitcoin's limitations, wrote a whitepaper, and launched what became the internet's second-biggest blockchain. We touched on smart contracts, ETH, ERC-20 tokens, and the early stages of the switch from PoW to PoS (if you missed it, go read it now!).
Today, we're going deeper. We're talking about the ICO gold rush that made Ethereum famous (and infamous), the civil war that split the community in two, the full story of how Ethereum ditched mining, and the projects that are actually building the thing right now in 2026.
Grab your Sunday coffee and let's yap!
🫰 The ICO Gold Rush
When we left off, we mentioned that Ethereum's programmability triggered an ICO boom between 2017 and 2018. Let's actually talk about what that meant, because it was chaotic, exciting, and insane.
ICO stands for Initial Coin Offering. Think of it as a crowdfunding campaign, but instead of getting a tote bag or a thank-you card, investors received digital tokens. These tokens supposedly represented future utility in whatever project was being built: access to a platform, governance rights, or some other function. Because ERC-20 made creating a token embarrassingly easy, anyone with a whitepaper and a Telegram group could launch one.
As a result, between 2017 and 2018, over 2,000 token sales raised more than $10 billion combined. In just the first three months of 2018 alone, ICOs raised $6.3 billion (more than all of 2017). Ethereum became the launchpad for an entire new fundraising category.
Some projects that actually mattered:
- EOS: The largest ICO ever. Block.one raised $4 billion over the course of a year-long token sale to fund EOS, a competitor blockchain designed to be faster and more scalable than Ethereum. EOS promised to process millions of transactions per second and eliminate gas fees. It attracted enormous hype and enormous scrutiny. The SEC eventually fined Block.one $24 million for conducting an unregistered securities offering. EOS never lived up to its promises and largely faded from relevance, but it remains a landmark moment in ICO history purely for the sheer scale of capital it raised.
- Telegram (TON): The messaging app you probably use raised $1.7 billion from large private investors in 2018 to build its own blockchain, the Telegram Open Network. The vision was ambitious: a crypto ecosystem baked directly into an app used by hundreds of millions of people — instant payments, decentralised storage, smart contracts, all accessible through a chat interface. The SEC killed it before it launched, classifying TON tokens as unregistered securities and ordering Telegram to return $1.2 billion to investors and pay an $18.5 million penalty. The project was cancelled in 2020. But here's the twist: the open-source code didn't die with it. An independent community of developers picked up where Telegram left off, rebuilt the network, and relaunched it as TON (The Open Network). Telegram eventually came back around. Today, TON is deeply integrated into the Telegram app, powering payments, a built-in wallet, and a growing ecosystem of mini-apps used by millions.
- Tron (TRX): Founded by Justin Sun in 2017, Tron raised around $70 million in its ICO with a promise to build a decentralised internet for entertainment and content sharing — cutting out platforms like YouTube and Spotify as middlemen. Tron was controversial from day one: its whitepaper was found to contain sections plagiarised from other projects, and Justin Sun became one of crypto's most polarising figures, known as much for marketing stunts as for technical output. Despite the criticism, Tron survived and scaled. It became one of the most used blockchains for USDT transfers due to its low fees, and it maintains a large active user base to this day.
- Polkadot (DOT): Founded by Gavin Wood — one of Ethereum's original co-founders and the person who actually wrote its first working implementation — Polkadot raised $145 million in its 2017 ICO. The core idea was interoperability: instead of dozens of isolated blockchains that can't talk to each other, Polkadot would be a "blockchain of blockchains," allowing different networks to communicate and share security. It was technically ambitious and well-funded. Polkadot launched its mainnet in 2020, attracted serious developer interest, and built a functioning ecosystem of connected chains called parachains. It remains one of the more credible Ethereum alternatives in terms of technical depth.
- Filecoin (FIL): One of the most anticipated ICOs of the era, Filecoin raised $257 million in 2017, at the time one of the largest token sales ever. The idea was genuinely novel: a decentralised storage network where anyone with spare hard drive space could rent it out, and anyone who needed storage could pay for it in FIL, without relying on Amazon Web Services or Google Cloud. It took three years to actually launch (the mainnet went live in 2020). Filecoin operates today as one of the largest decentralised storage networks in existence, though mainstream adoption has been slower than the hype suggested.
- Brave Browser (BAT): One of the few genuine ICO success stories. Brave, a privacy-focused web browser, raised $35 million in under 30 seconds in 2017. Its Basic Attention Token (BAT) was designed to flip digital advertising on its head — instead of platforms harvesting your data and keeping the money, Brave would pay users directly in BAT for choosing to view ads. Brave has since grown to over 70 million monthly active users, and BAT remains one of the more functional utility tokens in existence: an ICO-era project that actually did what it said it would do.
- Chainlink (LINK): Raised a relatively modest $32 million in its 2017 ICO — almost underwhelming compared to the nine-figure raises happening around it. Today, it's one of the most important pieces of infrastructure in the entire blockchain space: the dominant oracle network that connects smart contracts to real-world data. Price feeds, weather data, sports results, financial market data — Chainlink delivers it all, on-chain, in a decentralised way.
So why does the ICO boom matter for Web3?
Because it funded the entire first generation of crypto infrastructure. Despite the scams, the failed projects, and the SEC enforcement actions, ICO money paid for protocols and applications that are still running today.
The ICO era also forced crypto to grow up. Regulatory scrutiny pushed the industry toward more structured fundraising mechanisms — security token offerings, private sales, IEOs (exchange-led offerings), and eventually the venture capital-heavy model that dominates today.
⚡ Ethereum Classic
Let me remind you what a hard fork is: it happens when a blockchain changes its rules so drastically that nodes running the old software can no longer communicate with nodes running the new version. From that moment on, the network splits into two separate blockchains — two rulebooks, two histories, and two communities. Usually, one chain becomes dominant while the other slowly fades away or survives as a niche alternative, but sometimes the split is about much more than technology.
The Ethereum/Ethereum Classic fork is one of the most controversial forks in crypto history, not because of a technical upgrade, but because of a philosophical crisis.
In 2016, a project called The DAO raised over $150 million worth of ETH from more than 18,000 investors — one of the largest crowdfunding campaigns in history at the time. The DAO was a decentralised venture fund: investors pooled ETH, voted on how to allocate it, and shared in the returns.
Then, a hacker found a vulnerability in The DAO's smart contract code and used it to drain approximately 3.6 million ETH — worth around $50 million at the time, but worth billions by later standards. To put this in scale: roughly 5% of all ETH in existence was stolen in a single attack.
The community had a choice:
- Option A: Do nothing. The blockchain is immutable. "Code is law", so if the code allowed it, it happened. The investors lose their money, but the principle of an unchangeable ledger is preserved.
- Option B: Hard fork. Roll back the blockchain to before the hack, return the stolen funds, and essentially pretend it never happened.
The Ethereum core developers, led by Vitalik Buterin, chose Option B. A majority of miners and users agreed. The blockchain was forked, the hack was reversed, and the victims got their ETH back, but not everyone agreed with this decision.
A minority of the Ethereum community refused to switch to the new chain. They believed that altering the blockchain (even to fix a theft) was a betrayal of everything blockchain was supposed to stand for. They kept running the original, unedited chain. That chain became Ethereum Classic (ETC).
The philosophical divide is still real today:
- Ethereum (ETH): "The spirit of the law matters more than the letter. The network serves its users, and sometimes humans need to intervene to prevent catastrophic injustice".
- Ethereum Classic (ETC): "Code is law. An immutable ledger is the whole point. Once you start editing history, you've destroyed the thing that makes blockchain different from everything else".
Ethereum Classic is essentially a clone of Ethereum up to block 1,920,000 — the point where the chains diverged. Both support smart contracts and dApps. But ETC has taken a very different path since the split:
- Proof of Work: When Ethereum switched to Proof of Stake in 2022, Ethereum Classic kept its Proof of Work consensus. This attracted a migration of miners who could no longer mine ETH and needed somewhere to point their hardware.
- Fixed supply: Unlike Ethereum, ETC has a hard cap — approximately 210.7 million coins will ever exist. Similar to Bitcoin's model, ETC undergoes a "fifthening" every five million blocks, reducing mining rewards by 20% roughly every two years.
- Smaller ecosystem: ETC's market cap was around $3.78 billion as of mid-2024. Ethereum's ecosystem dwarfs it. Most developers, DeFi protocols, and institutional money flow toward ETH, not ETC.
ETC has a loyal base of idealists and a consistent community of traders, but it has never come close to challenging Ethereum's dominance. The fork was a statement of principle, not a competitive strategy.
💪 Ethereum's Move to PoS
In the first article, we mentioned that Ethereum transitioned from Proof of Work to Proof of Stake. But the full story of how that happened is worth telling in detail, because it took years, involved multiple phases, and represented one of the most complex infrastructure migrations in tech history.
As you remember, Proof of Work (PoW) requires miners to compete by solving energy-intensive mathematical puzzles. The winner adds the next block and earns the reward. It's secure, but it's expensive. At its peak, Ethereum's PoW network consumed as much electricity as a small country, and the computational difficulty of mining placed a hard ceiling on how many transactions the network could process.
Proof of Stake (PoS) replaces miners with validators. Instead of burning electricity, validators lock up ETH as collateral ("staking" it). They're randomly selected to propose new blocks proportional to the amount they've staked. If they behave dishonestly, they lose their stake (slashing). Security comes from economic skin in the game, not raw computing power.
The transition was broken into four phases. Here's how it actually went:
Phase 0: The Beacon Chain (December 2020)
Ethereum couldn't just flip a switch. It needed a safe place to build and test the new system while the old one kept running. That's what Phase 0 was: the launch of the Beacon Chain — a brand new PoS blockchain running in parallel to the existing Ethereum chain, without touching it.
To kickstart it, a deposit contract was established: 16,384 validators each had to deposit 32 ETH into the contract by November 24, 2020. If the threshold was met, the Beacon Chain would launch on December 1. More than 20,000 validators deposited on time.
The Beacon Chain went live on December 1, 2020. It wasn't processing user transactions or replacing the original chain. Its job was narrower and more foundational: coordinate validators, manage staking, and prove that a PoS consensus layer could run stably at scale.
Phase 1: Sharding (2021–2022)
The original vision for Phase 1 was full sharding — splitting the Ethereum blockchain into 64 parallel chains called shards, each processing transactions simultaneously. Instead of one chain doing all the work, you'd have 64 doing it at once, multiplying throughput dramatically.
In practice, Phase 1 evolved significantly as the roadmap was refined. The team realised that Layer 2 rollups — networks that process transactions off-chain and batch them back to Ethereum — were already solving the scaling problem faster than sharding could be built. So rather than rushing full execution sharding, Ethereum shifted focus toward "danksharding": a more targeted version of sharding designed specifically to make rollups cheaper by giving them more space to post their data.
Phase 1.5: The Docking (The Merge — September 2022)
Originally called "the docking", this phase became known simply as The Merge — and it was the main event. After nearly two years of the Beacon Chain running in parallel, Ethereum's original PoW execution layer was connected to the PoS Beacon Chain. The two systems, which had been running side by side, became one. Mining stopped entirely. Validators took over.
The results were immediate and significant:
- Energy: Ethereum's energy consumption dropped by approximately 99.95% overnight. The network that once rivalled countries in electricity usage now runs on standard computer hardware.
- Issuance: The amount of new ETH issued per day dropped by about 90%. Combined with the EIP-1559 fee burn mechanism (which permanently destroys a portion of every transaction fee), Ethereum's supply has become deflationary during periods of high network activity.
- Security model: Attacking a PoS network requires acquiring and staking more than 50% of all staked ETH, and then having that enormous stake destroyed when the attack is detected. Economically, it's essentially prohibitive.
- Scalability foundation: PoS opened the door to the next wave of upgrades, including sharding and further improvements to how rollups interact with the base layer.
The Merge was, technically speaking, a remarkable achievement. A live, $200+ billion network migrated its entire consensus mechanism without downtime, data loss, or a single user noticing their transactions behaved differently. Critics who called it impossible were proved wrong.
Phase 2: The Surge (ongoing)
Phase 2 — now referred to in Ethereum's roadmap as "the Surge" — is where the scaling story continues. The goal is to dramatically increase Ethereum's transaction throughput, from roughly 15–30 transactions per second on mainnet today to potentially 100,000+ when rollups and sharding are fully combined.
The key milestone here is EIP-4844, also known as "proto-danksharding", which launched in March 2024. It introduced a new type of data storage called "blobs" — temporary, cheaper data slots designed specifically for rollups to post their transaction batches. The immediate effect was dramatic: fees on Layer 2 networks like Arbitrum and Optimism dropped by 80–90% almost overnight.
Full danksharding — where Ethereum becomes a highly parallelised data availability layer purpose-built for rollups — is still in development. But the architecture is clear, the direction is set, and each upgrade brings it closer.
🏆 Top Projects in the Ethereum Ecosystem
Ethereum is the infrastructure layer, but the reason it matters is that everything is being built on top of it. Here are the projects shaping the ecosystem right now:
- Uniswap: Launched in 2018 by Hayden Adams — a mechanical engineer with no prior coding experience who learned Solidity from scratch after being laid off — Uniswap is the largest decentralised exchange (DEX) on Ethereum. Traditional exchanges use order books: buyers and sellers list their prices, and a trade happens when they match. Uniswap replaced this entirely with Automated Market Makers (AMMs) — liquidity pools funded by regular users (we'll talk in more detail about this in the upcoming articles). Uniswap has processed hundreds of billions of dollars in trading volume and inspired an entire generation of DeFi protocols. Its governance token, UNI, allows holders to vote on protocol upgrades. It remains the backbone of decentralised token trading on Ethereum.
- Aave: Aave started in 2017 as ETHLend, a peer-to-peer lending protocol. It rebranded and relaunched in 2020 as Aave, transforming into a liquidity pool-based lending platform and becoming one of DeFi's defining projects. The concept is simple: lenders deposit crypto assets into liquidity pools and earn interest. Borrowers post collateral and take out loans against it. Everything is governed by smart contracts. Aave also pioneered "flash loans" — uncollateralised loans that must be borrowed and repaid within a single transaction. They sound bizarre, but they enabled a whole category of sophisticated DeFi strategies. With billions in total value locked (TVL), Aave is one of the most used lending protocols in all of crypto.
- Chainlink: Chainlink runs a decentralised network of node operators that fetch, verify, and deliver external data to smart contracts on Ethereum (and dozens of other chains). Founded in 2017, Chainlink is now the dominant oracle solution in crypto. Major DeFi protocols like Aave, Synthetix, and dozens of others rely on Chainlink price feeds. Without it, a huge portion of DeFi simply wouldn't work.
- Arbitrum & Optimism: Ethereum's mainnet is expensive. During peak usage, a simple token swap can cost $20–100 in gas fees. This is the network's main usability problem, and Layer 2 (L2) solutions are the answer. Both Arbitrum and Optimism are "Optimistic Rollups" — they process transactions off Ethereum's main chain, bundle hundreds of them together, and post compressed batches back to mainnet. The result: the same Ethereum security, but transaction fees are often under a few cents.
Arbitrum, built by Offchain Labs, leads the L2 space with the largest total value locked among all Layer 2 networks. Its developer-friendly environment has attracted major protocols including Uniswap, Aave, and even Robinhood and BlackRock.
Optimism, meanwhile, has become the foundation for the "Superchain" — a vision of multiple interconnected L2 networks all sharing the same underlying technology stack. Coinbase's Base chain, for example, is built on Optimism's OP Stack. - MetaMask: MetaMask is the front door to the entire Ethereum ecosystem. It's how ordinary users interact with Uniswap, Aave, Arbitrum, NFT marketplaces, and every other dApp. For most people, MetaMask is "crypto on Ethereum" in the same way that Chrome is "the internet". MetaMask is evolving from a wallet into a fully-fledged financial platform — with features like its own stablecoin and a card that works at Mastercard merchants worldwide.
Key Takeaways
- The ICO boom of 2017–2018 raised over $10 billion and funded the first generation of Web3 infrastructure, but also attracted massive regulatory scrutiny and an SEC crackdown on unregistered securities offerings.
- Ethereum Classic (ETC) was born from a philosophical dispute after the 2016 DAO hack. It preserved the original, unedited blockchain in principle. Ethereum forked to reverse the hack. Both chains still exist today.
- Ethereum's transition from Proof of Work to Proof of Stake happened in phases: the Beacon Chain launched in December 2020, and The Merge completed the switch in September 2022, reducing energy consumption by 99.95%.
- Layer 2 networks like Arbitrum and Optimism solve Ethereum's gas fee problem by processing transactions off-chain and batching them back to the mainnet, dramatically reducing costs while preserving security.
- The most actively developed Ethereum projects are concentrated in infrastructure: wallets (MetaMask), scaling (Arbitrum, Optimism), oracles (Chainlink), and DeFi primitives (Uniswap, Aave).
Final Thought
The ICO era was messy. The DAO hack was a crisis. The transition to Proof of Stake took years longer than anyone expected. However, Ethereum is still here. It processes over a million transactions a day. It's the foundation of most of global DeFi. It hosts thousands of active dApps. The world's largest asset managers are building on it. Layer 2 networks are making it cheap enough for everyday use.
Ethereum didn't get there cleanly. It got there the way most important things do: through arguments, setbacks, hard choices, and a community stubborn enough to keep building anyway.
If you learnt something new today, pass it on. Share it with your community. Let's spread the knowledge and level up together.
That's a wrap, normies. Next time, we'll talk about stablecoins. Get ready!
Cookies She Left Behind
If you'd love to dig deeper into the topic, I'd also recommend reviewing the following:
- What is an ICO? by Whiteboard Crypto
- Ethereum 2.0 Upgrades Explained by Whiteboard Crypto