Since 2016, Blockdaemon’s Blockchain R&D Engineer Chris Spannos has been involved in researching Ethereum scaling for mass adoption. Recently, Chris shared his journey and his work on optimizing proposer-builder separation (PBS) for Ethereum Layer 2 solutions. We sat down with him to discuss his work and its impact on Ethereum's future.
- How did you first get involved with blockchain technology, and what drew you to Ethereum specifically?
Yeah, so I got involved in 2016. I had a longstanding preexisting interest in decentralized distributed systems, and I discovered smart contracts, which really impressed and excited me. That was around the time of the Ethereum DAO hack, where a hacker figured out how to exploit smart contracts and siphon out what was then the equivalent of about 60 million U.S. dollars in ETH. The Ethereum community embarked upon this incredible decentralized coordination journey to recover and fork the chain. Observing that whole process as a bystander was hugely impressive and inspired me. I wanted to get involved, so that's how I got started, and I was sucked in from there.
- In 2018, you received a grant from the Ethereum Foundation. What was the focus of your project, and how did it impact onboarding new developers?
After the hack, I started getting involved in the research forums and decided I wanted to contribute to some of the scaling efforts. To my amazement, the Ethereum Foundation awarded me a hackternship, which is a grant-funded project to onboard developers. Basically, I interviewed various client developer teams - at that time, the Prysm implementation with Raul Jordan's team, which is now under Offchain Labs over at Arbitrum. I interviewed them and a lot of other client developer teams all around scaling and mass adoption. So that was a pretty large research and discovery project focused on scaling and mass adoption, which introduced me to early scaling protocols such as Plasma, kind of an early precursor to Optimism's chain, as well as Starknet and some other early rollup protocols. It introduced me to problems which continue to this day and is why I'm continuing to do this research and work in this field.
- You interviewed client developer teams and early Layer 2 projects like Starknet and Plasma. What key insights did you gain from those conversations?
At that time, I gained key insights into the crucial differences between optimistic and zero-knowledge rollups, their different approaches, and how they would evolve over time. With Starknet, which uses zero-knowledge proofs to validate every state transaction, you can basically trust the chain inherently. With optimistic approaches like Plasma, there were exit games where if any fraudulent behavior was detected, it had to be reported, and users would try to exit the chain. Two very different approaches. Additionally, there was a differing trajectory on the Ethereum roadmap at the time, which was evolving around what was called sharding. It's interesting because, as Vitalik Buterin recently pointed out, these paradigms, which were once considered opposing, are actually similar in that they both deal with data sharding and rollups. The same problems of scaling and decentralization still exist, and it's really about who runs these systems and who's responsible for them.
- What was the "Scaling Today" project, and how did it contribute to the community's understanding of scalability issues?
The Scaling Today project was an output of the Ethereum Foundation grant and my interviews with various developer teams. At that time, we discussed many issues around what it was like to build clients. For example, with Prysm and Raul Jordan's team, the sharding spec was moving very fast, often going through rapid iterations. Some issues that arose were around the need for documentation. At that point, the tooling in the Ethereum ecosystem was very immature, as were many of the protocols in development. There was very little documentation, and it was very hard for developers from a web2 background to figure out how to use the tooling or even get involved and help build. So my project, Scaling Today, was an attempt to highlight those issues, surface them, and publish them in the hopes they would be addressed. Looking back today, we can see the tooling has really evolved - tools like Hardhat and Foundry for deploying smart contracts, and an incredible amount of tooling for static analysis and security. It's really come a long way. I'd like to think that somehow my research contributed a very small fraction to that, but really there are so many inspiring and amazing developers in the Ethereum ecosystem who are just driven, and it's great to see all the work that everyone's done.
- As a Blockchain R&D Engineer at Blockdaemon, what challenges in MEV, interoperability, and Layer 2 scaling have you encountered?
Yeah, so MEV, which stands for maximal extractable value, is the ability for actors to have control over what transactions are included in a block as well as the order of those transactions to extract profit from the network or users. It includes arbitrage between exchanges - centralized or decentralized - as well as latency games and timing games on blocks. As one of the early R&D engineers at Blockdaemon, I was tasked with helping explore MEV in different ecosystems, exploring it in the Ethereum ecosystem with MEV Boost and also in the Cosmos ecosystem. There's a lot of interesting technology that's emerged around MEV in Solana, the integration of Skip into various Cosmos protocols, and evolving efforts around how to more equitably distribute MEV throughout Ethereum and other ecosystems. So it's been great to work on these things. There are still various issues that need to be resolved in how to more equitably distribute MEV.
Interoperability and cross-chain communications are really critical across Layer 1 and Layer 2 and even Layer 3 chains. That's working with various protocols from Layer Zero and EVM ecosystems for cross-chain liquidity pools and DeFi applications to the inter-blockchain communications protocol, IBC relayer, that has come out of the Cosmos ecosystem but is applicable across various protocols.
As for Layer 2 scaling, as L2s emerged, one of the things that has become apparent is there are issues around centralized sequencers, and there's more demand for transparency around the sequencers because that is where blocks get constructed, transactions are ordered, and then executed. So there are some issues around MEV there as well as sequencer fees and how those fees are distributed, which is a very timely issue right now. I helped do early investigations into Espresso’s decentralized sequencer that Blockdaemon spun up - an early prototype of their sequencer - and did some write-ups, internal presentations, and helped onboard Espresso so that we have now participated in their testnet, as we've announced. It's very exciting news, and there's more to come on that front. The Blockdaemon node ops team that's been working on that has been doing fantastic work, and it's such an exciting project. So I've been very lucky to have an opportunity to help in the early onboarding of that initiative.
- Vitalik Buterin mentioned a shift from Layer 1 scaling to a rollup-centric model. How do you see this transition affecting Ethereum's scalability and decentralization?
The rollup-centric model is one where we have the Ethereum Layer 1 chain, and then we have all these Layer 2 chains emerging - protocols like the optimistic protocol from OP Labs, zkSync, Starknet, and many other different kinds of protocols. If you look at data on websites like L2beat, you can see that on the L1 chain, transaction throughput is about 12 transactions per second, whereas on many rollup L2 chains, it's much higher - anywhere from 50, 75, to 150 transactions per second, or even more. The idea of freeing up that bottleneck in transaction throughput and enabling mass adoption for everyday retail usage and large-scale institutional usage is all related to scaling up Ethereum. The rollup-centric paradigm or framework is critical for scaling Ethereum. As Vitalik mentioned, it aims to achieve the same goals as the sharded roadmap, but they shifted from sharding to this rollup-centric roadmap. Many of the same principles apply, such as having sharded data and provers and how to decentralize all of this. The main goal is not only to increase transaction throughput, which is critical for mass adoption and usability, but also to keep the protocol evolving in a way that maintains decentralization, transparency, and censorship resistance.
- Can you explain what Proposer-Builder Separation is and why it's crucial for Ethereum's future?
Proposer-builder separation (PBS) is an approach that evolved after The Merge, Ethereum's shift from proof-of-work to proof-of-stake in September 2022. Before PBS (both in the Proof-of-Work (PoW) and early Proof-of-Stake (PoS) phases), validators (or miners in PoW) were responsible for all tasks in block production: proposing, building, and validating blocks. Validators would collect transactions, order them, and create a new block for the chain. This bundling of roles allowed for direct control over which transactions were included, their order, and how block rewards were distributed—giving rise to challenges like MEV exploitation.
PBS separates these tasks, building different infrastructure and protocols around them. This separation is crucial because it addresses issues like toxic MEV, where actors can control which transactions are included in a block and their order to extract profits from the network or users. By separating block proposing and block building, PBS aims to decentralize these processes and minimize MEV exploitation, helping to ensure that profits are distributed more fairly and that the network remains decentralized and secure.
- Your current research focuses on optimizing PBS in Ethereum Layer 2s. What are the main objectives of this project?
My current research focuses on optimizing proposer-builder separation (PBS) in Ethereum Layer 2s. The main objectives of this project are to help the ecosystem understand how PBS might be introduced across Layer 2s and what effects this might have on centralization, censorship, and adoption of different approaches to sequencers. I'm exploring questions like how decentralized sequencers can lead to distributed block building or partial block auctions, which aim to decentralize the node setup. My approach is to build a rubric that looks at centralized, shared, and decentralized sequencers, examining aspects like leader selection in sequencing, token delegation and distribution for fair block building, and distributed proposers or multiple proposer setups. This rubric will help inform my analysis of different L2 protocols such as Starknet, Linea, Base, Taiko, and others, to understand how they're implementing PBS or ways they might implement it.
- How do you believe your research will enhance fairness, efficiency, and decentralization in Layer 2 solutions?
I believe my research will enhance fairness, efficiency, and decentralization in Layer 2 solutions by uncovering areas where fair block building may be a solution that some L2s can adopt. Maybe there's a way in which what they're doing can be tweaked a little, such as implementing proposers and multi-proposer setups or pre-commitments. There are various initiatives out there addressing different pieces of the puzzle from different directions. By getting as much of a bird's-eye view of this puzzle as possible before it's put together, I hope to see how we might connect the pieces, what makes sense and what might not. With relatively few people working on this, I aim to contribute to the effort and help improve PBS approaches and equitable MEV distribution in Layer 2 solutions.
- What role has the PBS Foundation played in supporting your work, and how has mentorship from Simon Brown influenced your research direction?
The PBS Foundation has been great. It's such an inspiration to have them take on board my application for research and support it, providing guidance from the team. Simon Brown has been amazing. He's with Linea at Consensys and has done research around some of these questions on PBS and L2s, and different ways they're thinking about approaching PBS at Linea. They've published in the ETH Research forums. He's been a brilliant mentor, posing questions I hadn't thought about and helping me think about things that open new directions. I really appreciate their support.
- Looking ahead, what do you see as the most pressing challenges and opportunities in scaling Ethereum for mass adoption?
Looking ahead, I think the most pressing challenges and opportunities in scaling Ethereum for mass adoption involve both technical and coordination aspects. There are technical challenges such as pre confirmations, inclusion lists, execution tickets, multi-setup approaches, and different approaches to blockchains. These are highly challenging areas, both in terms of engineering and the economic incentive mechanisms used, like auction mechanisms. Coordination within the Ethereum community is equally important. Aligning various stakeholders - developers, researchers, node operators, and users - is essential to implement these technical solutions effectively. There's also an opportunity to enhance user experience, security, and accessibility, making it easier for people to interact with Ethereum. By addressing these challenges collaboratively, we can build scalable, decentralized solutions that will pave the way for mass adoption.
To learn more about Chris’s contributions to Ethereum, read his recent article, “My Journey Scaling Ethereum: Introduction to My Work with the PBS Foundation”. Follow him on Mirror at spannos.eth.
