There’s been a lot of noise lately ETH L2 vs alt L1. So here’s a messy, unfiltered brain dump on scalability, tradeoffs, and how I see the current state of things: Let’s start here: Ethereum is slow. It’s expensive. It’s not ideal for most applications. And honestly, it hasn’t been for a while. That’s why we have alt L1s Solana, Avalanche, Polygon PoS, and others. They exist because Ethereum couldn’t scale fast enough. And to their credit, they’re fast. They’re cheap. But none of them get that way without sacrificing something. Every L1 that claims to be “better” than Ethereum is really just making a different tradeoff. Usually it comes down to running fewer validators, or skipping full verification by every node. Sometimes, it means requiring higher-end hardware so fewer people can even run a node. That leads to lower decentralization, even if the chain feels smooth and fast. Does that make them bad? Not necessarily. If you’re building a game or a social app, do you really need Ethereum-level decentralization? Probably not. A lot of apps simply don’t need “bank-grade” security. They need cheap fees, fast confirmations, and easy user flows. So in that sense, alt L1s are totally valid. They trade a bit of decentralization and security for way more usability. And for many use cases, that’s the right call. Now let’s talk about L2s. L2s aim to get the best of both worlds: Security and decentralization from Ethereum plus scalability. The whole idea is to process transactions off-chain (cheap + fast), then roll them up and verify them on Ethereum (secure + decentralized). Take something like Polygon zkEVM it batches a bunch of transactions, generates a ZK proof, and sends that proof to Ethereum for validation. For users, it feels snappy and affordable. For developers, it’s just EVM same tools, same wallets. And under the hood, Ethereum is doing the heavy lifting, verifying that the batch is legit. If the proof doesn’t check out, it doesn’t get finalized. What’s powerful here is that ZK tech doesn’t rely on social trust—it’s math. If Ethereum doesn’t approve the proof, it’s game over for that batch. That’s what makes ZK rollups so promising: You get Ethereum-grade security You keep EVM compatibility You gain real scalability But here’s the catch: none of this is perfect yet. All the big L2s Arbitrum, Optimism, Base, Scroll, Polygon zkEVM they all still have training wheels. They’re upgradeable via multisig. There are centralized sequencers. Some have trusted provers. Why? Because the tech is still evolving. We’re not at the “set it and forget it” phase. There are bugs to fix, features to ship, edges to smooth out. Most teams have roadmaps to reduce these powers phasing out centralized control, making sequencers decentralized, limiting upgradability to on-chain provable bugs. But we’re not fully there yet. So yes, L2s have risks. Just like alt L1s have risks. No magic chain. No free lunch. Everything has tradeoffs. The key is knowing which tradeoffs make sense for your use case. That’s what the scalability conversation should be about not tribalism.