Audio of the Week: "Inside the Modular Mullet Era of Crypto Scaling" on Expansion

Every Friday, I'll write about a podcast episode form my play list that week.

This week’s audio is an episode of Expansion from June 26, 2024. This new Blockworks podcast on the modular ecosystem launched last quarter, and I’m catching up on episodes. The podcast is relevant for understanding the trade-offs between monolithic and modular systems and the value of abstracting away complexity.

The conversation features Olaf Carlson-Wee (Founder & CEO of Polychain Capital) and Nick White (COO at Celestia Labs) discussing the shift from monolithic to modular blockchain architecture.

It begins with the host asking Olaf to recall his quote from an Epicenter episode seven years ago. Indeed, we interviewed Olaf in 2017, just two years after founding Polychain. This was around the time Olaf was featured on the cover of Forbes and was aggressively investing in ICOs.

When you're building a traditional web application, these days the most premier way to do this is a service-oriented architecture. Where the architecture of your app isn't on one monolithic code base, but is it rather on different modular parts that each serve a very precise function, but that can all interact. This means that if one modular piece fails, maybe just one part of your sites api goes down, but you web interface stays up for all of your users. So this is the benefit of a service oriented architecture. you don't get catastrophic system wide failure. An individual piece can get upgraded or fail, without bringing down the whole system.

Olaf goes on to describe his time at Coinbase (where he was the first employee) and observing how systems were architected. It made sense to him that blockchains would take a similar trajectory for scaling. This shows how early Olaf was thinking about the future of blockchains, making bets on protocols like Cosmos and Polkadot—both early attempts at modularizing the stack.

Later, when comparing the security of an L1 like Bitcoin to that of an L2 rollup, Olaf analogizes the different security models to the different amounts of money we deal with.

I might carry around $100 in a wallet on my person. At my house, I might have a few thousand dollars in a safe. I might have tens of thousands of dollars in my bank account. When we're thinking about dollars its quite intuitive. You wouldn't want to carry around the entire balance of your bank account in your pocket. Nor would you want to have to go to your bank every time you wanted to use a vending machine.

Indeed, security should match the monetary value in the transaction's context. When making a transaction on the Bitcoin blockchain or L1 Ethereum, one buys into an incredible amount of security, regardless of the amount.

Olaf’s analogy makes intuitive sense until you realize that we do carry the entire balance of our bank account in our pocket, in the form of debit cards. Technology has abstracted away the security model, making it possible and safe to transact any amount with a single device. It’s important to recognize that this comes at a cost, through payment networks, interbank settlement systems, and insurance. Yet, as a society, we accept this cost for the sake of convenience.

In most places, you don’t have to worry about whether a merchant supports your bank or card issuer. A single payment terminal will essentially work with any bank in the world. This isn’t the case in some countries, where you might see a handful of payment terminals on a merchant’s counter. These places have not yet accepted the cost of absolute abstraction, but I believe they will eventually arrive there.

I don’t disagree with Olaf’s analogy. In fact, I think it’s helpful for understanding the different security assumptions one should have about L1s and rollups. However, it also highlights the importance of abstraction as a crucial tool for improving user experience and making these technologies easier to adopt.