The Constellation Debate, Part 1 with Ben Coverston (Temporal)

In this deep dive episode of Validated, host Austin engages Ben Coverston from Temporal in a rigorous, data-driven examination of the Constellation proposal—a new framework for Multiple Concurrent Proposers (MCP) on Solana. Ben challenges the assumption that MCP is a clear win for decentralization and censorship resistance, arguing instead that it introduces significant new trade-offs. He highlights how the proposal centralizes power in the hands of the leader, enabling new forms of MEV extraction, transaction delay, and information leakage, especially when the same entity serves as both leader and proposer. Ben contrasts this with alternative solutions like drastically reducing slot times (e.g., to 50ms), which he believes offers comparable censorship resistance with far less complexity and fewer attack vectors. He emphasizes that while MCP is a major research milestone, it lacks empirical motivation and fails to adequately address critical risks like scheduler inconsistency, bandwidth inflation, and the NP-hard nature of block packing under deterministic rules. Ben advocates for open, honest discourse on trade-offs, positioning Temporal not as an opponent of MCP, but as a necessary voice pushing for a more balanced, evidence-based decision-making process ahead of Solana’s SIMD process. The episode also explores Temporal’s broader ecosystem—Humidify, Nozomi, Harmonic—and defends their vertically integrated, proprietary model as a pragmatic response to real-world problems in Solana DeFi. Ben argues that conflicts of interest are inevitable and often beneficial when they solve systemic gaps, and that transparency through composable, audited routers allows users to trust the system even when individual components are closed-source. He concludes that the path forward requires not just innovation, but humility: acknowledging that no solution is perfect, and that real progress comes from rigorously testing assumptions, not just celebrating proposals.