Zuberlin MEV Day

# Zuberlin MEV Day - Preconf.wtf Edition **Location:** MOOS Space, Berlin **Date:** Thursday, June 13, 2024 (dates & times in GMT+2) **Livestream** (starting at 14:00 GMT+2): streameth.org/zuberlin *Curated by [T.I.N.A.](https://x.com/tzhen) & [Kubi](https://x.com/kubimensah) with thanks to the [Zuberlin](https://x.com/ZuBerlinCity) community, stream.eth and PBS Foundation.* ## AM | IRL Researchathon *Note: The morning session for MEV Day will focus on tying the loose ends to the Sequencing Day ([Video](https://streameth.org/zuberlin/videos?page=2)) and Protocol Day ([Videos](https://hackmd.io/Uz0KZmhTS5C7fZ1QKVo-4w)) discussions, through concurrent workshops. The Researchathon Presentation was livestreamed.* How does it work? A set of topics selected from the pool of open questions identified or proposed, will be pre-assigned Navigators based on domain of expertise and interest. The goal is to guide each of the group to work as quickly as possible to either get to a rough spec, surface dependencies and come to a next step or conclusion. To learn more about the researchathon format, check out the 1st [researchathon experiment](https://hackmd.io/Uz0KZmhTS5C7fZ1QKVo-4w) at EthDenver 2023. <details> <summary>10:00-10:30 Team Formation</summary> Navigator for each breakout group will give a short pitch of their particular problem space, why it matters, and build a crew. When the workshop starts, the break out groups will go to one of the 3 workshop areas within the venue. </details> <details> <summary>10:30-12:00 Breakout Workshop</summary> *Note: the list of questions below are primarily proposed by Justin Drake. If you would like to propose your own questions or workshop sessions, please scroll to the bottom of this agenda page and add your own, before the Team Formation starts.* </details> <details> <summary>Proposed Breakout Topics:</summary> - Preconf tip pricing - Navigators: [Barnabé](https://x.com/barnabemonnot) (EF - [RIG](https://efdn.notion.site/Robust-Incentives-Group-RIG-Homepage-802339956f2745a5964d8461c5ccef02)) - Why is a preconf valuable? Selling a *commitment* to a future. Upside: Value of (more) certainty; downside: loss of optionality? What does the tip pay for, exactly? How to think about 1559 base fee adding uncertainty? - The fee market perspective on preconfs: Wait, it's all resource pricing? (h/t John Adler) - Welfare gaps in the online optimisation problem: In hindsight, would the gateway have wanted to offer different prices? How can we formalise this? - What changes should be made to the builder and relay spec - Navigator: [Kubi](https://x.com/kubimensah) ([Titan](https://x.com/titanbuilderxyz)) - Less scary slashing: How do you solve accidental liveness failures? - Navigators: [Sébastien Rannou](https://www.linkedin.com/in/s%C3%A9bastien-rannou-641992233/) ([Kiln](https://www.kiln.fi/)) - Preconf & consensus: leader selection/election - Navigator: [Francesco](https://x.com/fradamt?lang=en) (EF) - Where can TEEs be used to improve the protocol trust guarantees? (collatoral free preconfs) - Navigators: [Moe](https://www.linkedin.com/in/moe-mahhouk-12b005140/?originalSubdomain=de) & [Frieder](https://x.com/fnerdman) ([Flashbots](https://www.flashbots.net/)) - [TEE attestation + collatoral "free" preconfs](https://excalidraw.com/#room=dfe8e0826d63ca5b3f75,L7sp7fru6aJxNcBoX0C2tA) - [TDX slides](https://docs.google.com/presentation/d/1xcwonkSjBkBoRJRFMmcOA6N0IN0LZiscp68qO9MIcyA/edit?usp=sharing) - How do we get from first version of preconfs to synchronous composability? what are challenges? how to solve them? - Navigator: Ellie (Espresso) - Preconfs are solving a UX problem, whereas synchronous composability is a shared sequencer / shared verification problem. The two are separate, but lets explore what synchronous composability means, how we might enforce it, and if it is valuable - Ellie's notes from jam session - **atomic composability** - all sides of a cross chain transaction are either successfully exeucted or none or are executed - **asynchronous atomic composability** - cross chain transactions are not sequenced at the same time (e.g. are sequenced in consecutive blocks) but achieve composability as defined above - **synchronous atomic composability** - cross chain transactions are sequenced at the same time and achieve composability as defined above - What does synchronous composability require? 1. A shared sequencer (could be a single party or a consensus protocol) 2. (Optionally) A verification layer (such as AggLayer) - What does synchronous composability give us compared to asynchronous composability? - Improved latency (don't need to wait for multiple blocks to finalize, faster preconfirmations, improved execution, timely execuction, enables cross-chain flash loans, fast, trust-minimized bridging) - Improved coordination - Better optimistic security in the presence of a verification layer (can slash the shared sequencer if they do not enforce composability) - Better security in the absence of a verification layer - Other notes - Builders will eventually become aggregators of intent solvers - Shared finality layers are related, but orthogonal to synchronous composability. </details> <details> <summary>12:15-13:00 Researchathon Takeaways</summary> Note: Each Navigator shares their group's solution, spec, insights and identify a set of next steps for either continued collaboration, or call to actions.* <summary>Proposed Breakout Topics:</summary> - Preconf tip pricing - Navigators: [Barnabé](https://x.com/barnabemonnot) (EF - [RIG](https://efdn.notion.site/Robust-Incentives-Group-RIG-Homepage-802339956f2745a5964d8461c5ccef02)) - Why is a preconf valuable? Selling a *commitment* to a future. Upside: Value of (more) certainty; downside: loss of optionality? What does the tip pay for, exactly? How to think about 1559 base fee adding uncertainty? - The fee market perspective on preconfs: Wait, it's all resource pricing? (h/t John Adler) - Welfare gaps in the online optimisation problem: In hindsight, would the gateway have wanted to offer different prices? How can we formalise this? *60min Lunch Break* </details> ## PM | Preconf.wtf To learn more about .wtf style unconference, check out [defi.wtf (2019)](https://github.com/carboclan/pm/issues/69), [macro.wtf (2019)](https://github.com/carboclan/pm/blob/master/notes/MacroWTF_Archive.md), [mev.wtf](https://hackmd.io/ivUzk3piQEG8ALzCGbxlag) (2021), [reorg.wtf (2021)](https://hackmd.io/cEw2Z-QcR1yvQ8wAeQZdnQ), [mev.day (2022)](https://flashbots.notion.site/mev-day-836f88806995412dabc1c7bb7ce4e830), [mevconomics.wtf (2023)](https://collective.flashbots.net/t/mevconomics-wtf-on-march-24th-2023/1500), and [censorship.wtf](https://hackmd.io/@timbeiko/censorshipwtf) (2023). Welcome to your antinetworking network event. :) *Note: Remote speaker who need to dial in are marked with * .* ### Chapter 1. MEV & Protocol Design #### `14:30-14:55` 🎙️ Don’t Overload the Proposer ([Slides](https://drive.google.com/file/d/1m8YIdJ_YuKzJsSKz_o4b6Kx1zt4aAd2S/view) | [Video](https://streameth.org/zuberlin/watch?session=666af08807f92b086c2c2e54)) **Speaker:** [Vitalik](https://x.com/VitalikButerin) (EF) **Abstract:** Many ideas being discussed in the MEV space, including L1 preconfirmations, inherently rely on proposers running specific custom logic. This may conflict with the desire to minimize the authority of proposers and maximize protocol-level guarantees such as inclusion lists. > *Related Reading:* *[Don't overload Ethereum's consensus](https://vitalik.eth.limo/general/2023/05/21/dont_overload.html)* #### `14:55-15:20` 🎙️ Overload the Execution Proposer instead ([Slides](https://docs.google.com/presentation/d/1eWxKdykWGFx3OPT7t69R1GaoLYuNSNxTzvQx8M0TGsw/edit?usp=sharing) | [Video](https://streameth.org/zuberlin/watch?session=666af9a007f92b086c2c5117)) **Speaker:** [Barnabé](https://x.com/barnabemonnot) (EF - [RIG](https://efdn.notion.site/Robust-Incentives-Group-RIG-Homepage-802339956f2745a5964d8461c5ccef02)) **Abstract:** We present an iterative roadmap towards increasingly decoupling the role of consensus service provider from execution service provider. Hopefully this is based enough and we can let the preconfers go nuts without wrecking Ethereum :) > *Related Reading: [More pictures about proposers and builders](https://mirror.xyz/barnabe.eth/QJ6W0mmyOwjec-2zuH6lZb0iEI2aYFB9gE-LHWIMzjQ)* #### `15:20-15:50` 🪄 Ethereum R&D in-Protocol vs. Out of Protocol Symbiosis ([Video](https://streameth.org/zuberlin/watch?session=666b027107f92b086c2c7767)) **Panelists:** [Vitalik](https://x.com/VitalikButerin) (EF), [Justin Drake](https://x.com/drakefjustin)* (EF), [Barnabé](https://x.com/barnabemonnot) (EF - [RIG](https://efdn.notion.site/Robust-Incentives-Group-RIG-Homepage-802339956f2745a5964d8461c5ccef02)), [Tim Beiko](https://x.com/TimBeiko)* (EF), [Tomasz](https://x.com/tkstanczak) ([Nethermind](https://www.nethermind.io/) & [Flashbots](https://www.flashbots.net/)) **Host:** [T.I.N.A.](https://x.com/tzhen) ([Flashbots](https://www.flashbots.net/) & [PBS.Foundation](https://pbsfoundation.notion.site/Launching-the-PBS-Foundation-Working-Together-to-Keep-Ethereum-s-Consensus-Layer-Decentralized-4cd7cb83907248b5ae8962d10eef0b70)) > *Related Reading:* > [[Video] Ethereum Protocol Development: Past, Present & Future](https://www.youtube.com/watch?v=NUvLAanHDyg) (Edcon Talk by Tim Beiko) > [Coordination on a decentralized network](https://future.com/what-the-merge-means-for-ethereum-with-danny-ryan/) (from an interview of Danny Ryan) > [What is an Ethereum core developer?](https://hudsonjameson.com/2020-06-22-what-is-an-ethereum-core-developer/) (blogpost by Hudson Jameson) > [Robust Incentives Group (RIG)](https://efdn.notion.site/Robust-Incentives-Group-RIG-Homepage-802339956f2745a5964d8461c5ccef02) Research Areas > [[Video] The Nature and Nurturing of Research: A Modern Synthesis](https://www.youtube.com/watch?v=ByCdXyak3Vs&list=PLhuBigpl7lqu6xWpiXtbEzJQtlMH1tqoG&index=6) (by PL Research) > [Toward an open r&d process for MEV-Boost](https://collective.flashbots.net/t/toward-an-open-research-and-development-process-for-mev-boost/464) (out-of-protocol technical governance) > [PBS.salon](https://collective.flashbots.net/t/mev-week-paris-pbs-salon-agenda-with-slides/2010) --> [PBS.day](https://collective.flashbots.net/t/mev-week-paris-from-zero-sum-to-positive-sum/1963) --> [PBS Guild Proposal](https://collective.flashbots.net/t/pbs-guild-proposal-v3-wip/2223) (Inspired by [Protocol Guild](https://protocol-guild.readthedocs.io/en/latest/)) > -->[PBS Foundation: Working Together to Keep Ethereum’s Consensus Layer Decentralized](https://pbsfoundation.notion.site/Launching-the-PBS-Foundation-Working-Together-to-Keep-Ethereum-s-Consensus-Layer-Decentralized-4cd7cb83907248b5ae8962d10eef0b70) ![image](https://hackmd.io/_uploads/HktXVd_BR.png) ![](https://storage.googleapis.com/ethereum-hackmd/upload_2a3d1e9ee619e0a35d7c5bf5fdbac12b.png) *15min Break* #### `16:00-17:15` 🔥MEV Roast: Are We All Building the Same Damn Thing? *Format: 10min lightning talk + cross-exam from panel + Q&A.* ⚡ **MEV as a thermodynamic engine -- Combustion efficiency and user outcomes** ([Slides](https://docs.google.com/presentation/d/1nJFBQc651l1padximWWvV5RWasv2wiM2axq4keioLTw/edit?usp=drivesdk) | [Video](https://streameth.org/zuberlin/watch?session=666b130b07f92b086c2cb0be)) ⚡ >**Speaker:** [Phil Daian](https://x.com/phildaian) ([Flashbots](https://www.flashbots.net/)) > *Related Content: > *MEV is Fundamental to Permissionless Systems ([Video](https://youtu.be/-3LIHehjzBA?si=2e0RBoCcbuHq5iCM))* > MEV for the Next Trillion ([Video](https://www.youtube.com/watch?v=ACXAzLy3iWY)) > MEV Privacy Trade-offs: Privacy-efficiency Frontier ([Video](https://www.youtube.com/watch?v=ACXAzLy3iWY))* ⚡ **Sequencing Designs: Priority vs. MEV-Boost vs. FIFO** ([Slides](https://docs.google.com/presentation/d/1f6sgPljBip0v2HQ_mt7v-XK56ZovLI0HT1cDx3sruVQ/edit?usp=sharing) | [Video](https://streameth.org/zuberlin/watch?session=666b141c07f92b086c2cb330)) > **Speaker:** [Dan Robinson](https://x.com/danrobinson) ([Paradigm](https://www.paradigm.xyz/)) > *Related Reading: [Priority is All You Need](https://www.paradigm.xyz/2024/06/priority-is-all-you-need)* ⚡ **One-bit-per-attester inclusion lists** > **Speaker:** [Vitalik](https://x.com/VitalikButerin) (EF) > *Related Reading: [One-bit-per-attester inclusion lists](https://ethresear.ch/t/one-bit-per-attester-inclusion-lists/19797)* ⚡ **Who owns the write-lock: Relays vs Gateways vs Sequencers** ([Slides](https://docs.google.com/presentation/d/181tzp3nNrkxHyyPawvWPvwRM4J1uY-5NuasVjH9YuKE/edit#slide=id.g2e5816b839c_0_324) | [Video](https://streameth.org/zuberlin/watch?session=666b14db07f92b086c2cb587)) > **Speaker:** [Josh Bowen](https://x.com/Jskybowen) ([Astria](https://www.astria.org/)) > *Related Content: We Are All Building the Same Thing ([Video](https://www.youtube.com/watch?v=5809YJpbhFg))* **Note:** [Ethereum All Core Devs]((https://github.com/ethereum/pm/issues/1069)) call takes place at this time, you can join the Zuberlin ACD watch party in the garage next door, or stay with us at the main stage. *15min Break* ### Chapter 2. MEV & Preconf Design #### `17:30-17:45` 🎙️Low-collateral Based Preconfs ([Slides](https://docs.google.com/presentation/d/1CC_d6eW_TgkbS6XixAc5YRS9Qwej5TwIiXsgAfHW7YI/edit?usp=sharing) | [Video](https://streameth.org/zuberlin/watch?session=666b1c7307f92b086c2cd78d)) **Speaker:** [Justin Drake](https://x.com/drakefjustin)* (EF) **Abstract:** We show how to enable based preconfirmations with just 1 ETH of collateral. > *Related Reading: 06/10 Zuberlin Based Sequencing Talk ([Slides](https://docs.google.com/presentation/d/1hinNqDFbP4sxd_I4V-q_QEx-DWNdv-umT3eGzw64Mtk/edit) | [Video](https://streameth.org/zuberlin/watch?session=66681113f9b8e98b1ec926f4))* #### `17:45-18:00` 🎙️ The MEV Perspective on Preconfs ([Slides](https://docs.google.com/presentation/d/17xvZZEfKeB_2PAA0tJUASqU8bwg2oNHDVwLc_Xtw9cY/edit#slide=id.g2e57f0f16cb_1_36) | [Video](https://streameth.org/zuberlin/watch?session=666b21a107f92b086c2ce5d3)) **Speaker:** [Hasu](https://x.com/hasufl) ([Flashbots](https://www.flashbots.net/)) **Abstract:** Preconfs create new centralization pressures for MEV supply chain. We establish mental models for reasoning about builder centralization, and discuss a path forward. #### `18:00-18:30` ⛓️ Preconf & MEV Supply Chain ([Video](https://streameth.org/zuberlin/watch?session=666b412207f92b086c2d7e04)) **Panelists:** [Kevin Pang](https://x.com/kevinxpang?lang=en)* (SCP), [Kubi Mensah](https://x.com/kubimensah) ([Titan](https://x.com/titanbuilderxyz)), [Martin Köppelmann](https://x.com/koeppelmann) (Agnostic/Gnosis), [Nathan Worsley](https://x.com/NathanWorsley_)* **Host:** [Hasu](https://x.com/hasufl) ([Flashbots](https://www.flashbots.net/)) #### `18:30-18:45` 🎙️ Restaked Proposer Commitments ([Slides](https://docs.google.com/presentation/d/108r6GUovrZi_qT3RgEl1ZdbpRWHGPh25UDFAPwoixC4/edit) | [Video](https://streameth.org/zuberlin/watch?session=666b415707f92b086c2d7f36)) **Speaker:** [Sreeram Kannan](https://x.com/sreeramkannan) ([Eigenlayer](https://x.com/eigenlayer)) *15min Break* ### Chapter 3. MEV Meets Preconf #### `19:00-19:15` 🎙️ Hardening MEV-Boost ([Slides](https://docs.google.com/presentation/d/1G93jAJeVM7GcwOhYYw-94stgMYm5cMFlCVW-SOLUDcw/edit?usp=sharing) | [Video](https://streameth.org/zuberlin/watch?session=666b415707f92b086c2d7f36) starting 23>) **Speaker**: [Alex Stokes](https://x.com/ralexstokes?ref_src=twsrc%5Egoogle%7Ctwcamp%5Eserp%7Ctwgr%5Eauthor) (EF & MEV-Boost Maintainer) > *Related Material: > Github repo: github.com/flashbots/mev-boost* #### `19:15-19:30` 🎙️ [Introduction to Commit-Boost (Technical Architecture)] ([Slides](https://docs.google.com/presentation/d/1T06pPKcKkU-EdfYkAtXUYQxjfHXkkFBIVRDyWyiEFVs/edit#slide=id.g2731bc99d1b_0_0) | [Video](https://streameth.org/zuberlin/watch?session=666b415707f92b086c2d7f36) starting 43:00) **Speaker**: [Drew Van der Werff](https://x.com/DrewVdW), [Kubi Mensah](https://x.com/kubimensah) ([Titan](https://x.com/titanbuilderxyz)) > *Related Material: > Commit-Boost Github Organization: github.com/Commit-Boost > Commit-Boost Zuberlin Sequencing Day Talk ([Video](https://streameth.org/zuberlin/watch?session=66681afef9b8e98b1ec95fdd) | [Slides](https://docs.google.com/presentation/d/1T06pPKcKkU-EdfYkAtXUYQxjfHXkkFBIVRDyWyiEFVs/edit#slide=id.g2731bc99d1b_0_0)) > Commit-Boost Based Sequencing Call 6.6.24 ([Video](https://youtu.be/jrm4ZUoj9xY?si=90AbCTI5Yb31sDdZ) | [Slides](https://docs.google.com/presentation/d/1a-0rP2knM11g59UmnKn7I7NH8BlFM5wNhczH35sbkSo/edit#slide=id.g2731bc99d1b_0_0))* #### `19:30-20:00` 🧩 [Technial Deep Dive into the Modular Sidecar] ([Video](https://streameth.org/zuberlin/watch?session=666b415707f92b086c2d7f36)) **MEV-Boost Maintainers:** [Alex Stokes](https://x.com/ralexstokes) (EF), [Chris Hager](https://x.com/metachris) (Flashbots), Sarah Liu (Flashbots & Nethermind), [Terence Tsao](https://x.com/terencechain) (Prysmatic) **Commit-Boost Contributors:** [Drew Van der Werff](https://x.com/DrewVdW), [Kubi Mensah](https://x.com/kubimensah) ([Titan](https://x.com/titanbuilderxyz)) ### Chapter 4. Preconf MEV Sauna *Note: The Preconf Sauna, is more than just the title of a presentation on Preconf from Sequencing Day. Some may take the discussion to an actual Preconf Sauna taking place in Zuberlin at 8:00pm. Will this turn into a Roast or a Debate or many deep philosophical conversations deep into the night? We will find out!* #### Proposed Topics - Consensus Debate: [Elijia*](https://x.com/PossibltyResult), [Conor](https://x.com/conormcmenamin9) - Protocol Epistymology: [apriori](https://x.com/apriori0x) - Technical Governance: [T.I.N.A.](https://x.com/tzhen) --- ## Potential Topics - how to make preconfs incentive compatible for proposers? (h/t apriori) - Assume a validator has committed stake to some pre-confirmation system. They can make strictly more value by delaying the confirmations until the last possible moment but staying true to them afterwards. This delay is an undetectable fault since latency isn’t deterministic. And so the rational behavior is to not immediately confirm transaction orderings. - how to decentralize the gateway? And is the gateway just a solver anyways (speculative q)? (h/t apriori)