Use proposers to run an auction for proof production. This proposal is compatible with both PBS and Fernet.
Context
Monero has a decentralized mining pool, which uses a second blockchain to coordinate work. This proposal defines a fair structure under which such work pools may be created. Highly-available decentralized networks are expected to have the cheapest access to capital and compute, assuring liveness without adding complexity to the core protocol.
Details
(init) provers deposite bond collateral
builder broadcasts bond requirements and proof data to provers
a bond consists of a proposal portion, and an MEV portion
provers with sufficient bond broadcast bids
builder signs and broadcasts the winning bid
proposers attest to timely availability of the winning bid
prover submits proof to L1
If a prover fails to submit a proof in time, the builder avoids slashing by submitting the signed winning bid.
The provers proposal bond portion is burnt, and the MEV portion transfered to the builder.
I’d kindly ask you to please adhere to the submission format for it to be considered a valid proposal.
Title: A concise, descriptive title for your proposal
Summary: A brief, easy to understand summary of your proposal (about 300 words)
Comparisons: Explain what makes this solution unique and different from alternative solutions
Details: Explain the prover coordionation protocol, its components (including parameters), and its functionality
Questions: Any outstanding questions
Additionally some type of diagram (&/or visual) would likely be an efficient way to get on the same page, prior to diving into some of the specific details and questions. If you could update when you get a chance, that’d be great!
Hey @cooper-aztecLabs!
Your feedback is noted, and I have added section headers.
I am here to have fun. I communicate succinctly and in good faith. I believe my ideas are rational and well thought out. This is enough. Let us focus instead on designing a robust privacy network.
To better facilitate 3rd party prover network support, prover deposits may optionally be any quantities of any assets. If a builder accepts these bids, their bond remains at stake, and they receive the provers deposit if they are slashed.
Have you thought about the common occurance (“Winner’s Curse”) where in bidding markets like this, winners tend to overbid and thus create value loss over time? How would you counteract this?
Winner’s curse applies in common value auctions (the value of the item is the same but imprecisely known to all bidders). This is a private value auction, where bidders have different values because their compute overhead is different (and known ahead of time).
An English auction (open first-price ascending, this proposal) is pareto-optimal (and equivalent to a Vickrey auction (sealed second-price)), incentivising bidders to bid their true value (less the information value of bids).
Edit: There does appear to be some uncertainty regarding L1 fees.
I propose a gas rebate (funded by base-fee), which refunds L1 fees for submitting proofs.