Layer 3s bring the App Chain Thesis.... & more complexity to Ethereum

With the launch of Orbit Arbitrum has officially started the Layer3 era for the Ethereum ecosystem.



Expect a couple of big projects to announce a layer3 blockchain soon. My guess is one has got a strong simian vibe.

Steven Goldfeder and Harry Kalodner from Offchain Labs gave a great explanation of the thesis behind Orbit and, of course, the token airdrop on a recent Empire podcast episode.

It's well worth digging into the launch announcement Arbitrim blog post as it goes over their thesis for their token and the layer 3 ecosystem in detail.


Also worth your time is understanding the App Chain Thesis and why people consider it to be the next evolution in blockchain and cryptocurrency.

• Osmosis: The DeFi App-chain from Nansen Research - explains the Osmosis chain and how it works which gives an excellent insight into an app chain in practice
  • The Cosmos LSM: The Catalyst for Interchain DeFi from New Order - worth mentioning here as liquid staking is already an important part of defi and the expansion of this in Cosmos (of which Osmosis is at the heart of) will be vital to the ecosystem's development.
• 2023 Crypto Thesis from New Order - long but information rich report covering the expansion of the execution layers in the major blockchains, which covers app chains and app chain like functionality (e.g., rollups) in other solutions.
• Into The Cosmos: The App-Chain Thesis Enters Ethereum from Liam McDonald on Medium - an overview of the App Chain Thesis as it's playing out on Ethereum (layer3s feature heavily)

Also good to know the conventional wisdom the the App Chain Thesis has largely replaced: 'Fat Protocols'

• Fat Protocols from Union Square Ventures blog - the 2016 post that crystalised the early days of blockchain development and investment. Hotly debated over the years that followed due to it's value flipping thesis (Ethereum didn't exist at the time).
  • Thin Applications from Placeholder Venture Capital - a 2020 follow up by the author above to his much debated Fat Protocols post taking into account the developments since 2016. It isn't the App Chain Thesis necessarily (this was only 2020 after all) but you can see the genus of those ideas here.

Fun fact: There's a persuasive argument that goes: ;Bitcoin is the original app chain'. The reasoning is: It has one sole purpose, to service BTC, and the entire chain is so devoted to this goal there has never been anything able to be sustainably built using the Bitcoin blockchain. Ordinals are arguably a hack. The mechanism that allows for their existence can be stripped out by miners when they're making new blocks as its not part of the core blockchain protocol.


As much as I am excited to see more app-chain products come to the Ethereum ecosystem, not least because of the continuing high gas fees, I have reservations.

A oft stated down side of the Cosmos ecosystem is that it's fragmented by design and lacks the monument of more concentrated ecosystems like Ethereum, Solana or even Bitcoin.


However as Ethereum has sought to address the crazy high gas fees with layer2s, we've seen the leading blockchain technology progressively fragment.



Moving between dapps on different layer 2s (e.g., Lens Protocol on Polygon > Velodrome on Optimism) is already few steps, and it means going through the gas fee hell which is Ethereum:

layer2 > layer1 > layer2

Bridging directly is technically very difficult as each layer 2 needs Ethereum (layer 1) to resolve properly.

It's not so bad right now in Ethereum-land given there's only a few big layer 2s.

However with layer 3s we're going another step up. I think Ethereum really fragments then as separate dapps become essentially islands. There is always a way back down and across to another layer 3 but now it's longer, and still the gas fees of Ethereum await you:

layer 3 > layer 2 > layer 1 > layer 2 > layer 3

That's a lot of jumps.

In Cosmos you can go directly from chain to chain easily and safely as interoperability is part of the protocol. To go from Osmosis to Regen Network it's:

Osmosis > Regen Network

I think this increasing complexity of multiple chains going to be more and more a factor in people's experience of using web3 products.

I don't think users want a completely completely copy paste experience from web2. I think they are smart enough to understand there are trade offs to being more 'hands on' (people understand the DIY vs paying a tradesperson after all) but there is limit to how much friction people will put up with.

For a unique, high value transactions (e.g., blue chip NFTs) maybe but if you want any other sort of high transactions use case (like 99.99% of the internet today) it’s unworkable.