Dorax ⚛️

There is almost unanimous agreement that toxic transaction ordering degrades the end user experience.

Multi-millionaires and billionaires keep getting richer and richer while working families get squeezed by a rigged economy. No more. It’s time for a wealth tax — and for our government to start working for working people.

TLDR (from next article coming) - waiting (rev strat) is fully extractive, we should start the walk of shame for all validators running it - batching remains a market-design choice with regime-dependent trade-offs, not a universally fair scheduler - streaming is greatly undervalued. Precisely, Streaming wins the race to speak first; FixedBatch wins the race not to pay too much once speaking first is no longer possible. WaitUpTo is outside any ranking.

Introducing Constellation: a multiple concurrent proposers protocol for Solana. With proposers operating on a 50ms cycle, Constellation gives Solana the fastest protocol enforced economic tick rate of any blockchain.🧵

It's true that 90% of the focus is on validator client, scheduler, and microstructure. The reason is very subtle I think, and that's because despite all evidences no-one is building towards a better Solana. We are still anchored to "revenue driven" mentalities. We do something only if we can earn more. The funny thing is that by doing that we are loosing on other markets. We still focus on FCFS vs FBA when the truth is that an every purpose chain should be FCFS with FBA at application layer. This can be done at app level. FBA or rev strat are still used to create extractive incentives. Do you think a user needs to wait 50ms before executing a simple transfer? No. Do you think an oracle needs to wait 50ms before reflecting on-chain its knowledge? No. I don't see a push towards a better FCFS with FBA at application layer, and this makes everyone still focussed on theoretical work that's not needed anymore. Think about that, what makes HL great for perps? 1. Far more centralized and less distributed than Solana 2. Better control over validators 3. Unique set of rules (scheduler) for every participant under every validator Things like "maker priority" are just fairy tales that one tells oneself to avoid accepting reality. Solana to win needs to have: 1. Control over censoring 2. Unique set of rules for every participant under every validator With these 2 points enforced Solana is already suitable to built a 5x better perp app. N.B.: Pls don't say MCP will solve that, because we need a solution today, not tomorrow. Tomorrow the war is already lost.

talk on solana perps is weirdly mid curve the validator client, the scheduler, and microstructure are 90% of the focus but those, even if executed perfectly (they won't be) are entirely insufficient HL is already the leader to compete, you can't just match what exists, you need to be at least 5x better on a set of dimensions that matter for the end user while also catching up on distribution the non-academic problems are much, much harder to solve for than geeking out on engineering design

talk on solana perps is weirdly mid curve the validator client, the scheduler, and microstructure are 90% of the focus but those, even if executed perfectly (they won't be) are entirely insufficient HL is already the leader to compete, you can't just match what exists, you need to be at least 5x better on a set of dimensions that matter for the end user while also catching up on distribution the non-academic problems are much, much harder to solve for than geeking out on engineering design