4.4 Recursive Appeal System
Statistically, random sampling of small samples (only 3 arbitrators in the first round) can lead to erroneous rulings due to probability bias or targeted bribery. OmniPact does not assume that a single round of rulings is absolutely correct, but rather provides a mechanism to correct errors.
We model the dispute resolution process as a multi-round game. Each round of appeal expands the jury size, thus, according to the law of large numbers, increasing the probability that the verdict approximates the objective truth .
4.4.1 Exponential Cost Escalation Model
To prevent malicious users from launching "dust attacks" or stalling through endless appeals, appeals must come at a high financial cost.
1.Jury Scaling
The number of arbitrators required for each round of appeal, $N$, grows exponentially.
设$N_k$ represents the number of arbitrators in the $k$th round (initially $k=0, N_0=3$):
Round 0: 3 Jurors
Round 1: 7 Jurors
Round 2: 15 Jurors
Round 3: 31 Jurors ...
This increase ensures that the weight of a new round of voting always exceeds the sum of all previous rounds, thus guaranteeing that a new ruling can override an old one.
2. Appeal Fee Calculation
The party initiating the appeal (Appellant) must pay enough to cover the costs of a new round of arbitrators.
: Standard service fee for a single arbitrator.
: Agreement tax or destruction fee (to prevent zero-cost attacks).
3. War of Attrition
This is essentially a chicken game of "who is more convinced that they are right".
The honest party is convinced that as the number of people increases, the truth will be seen by the majority, and is therefore willing to pay the high appeal fees upfront because they expect to receive a refund and a reward if they ultimately win the case.
The lying party: Knowing that the cost of bribery will increase exponentially with the number of people involved (bribing 31 people is much harder than bribing 3), the attacker's expected value quickly becomes negative as appeal costs skyrocket.
When $C_{appeal} > Amount$, a rational attacker will drop the appeal.
4.4.2 The Supreme Court: Finality Enforcement
In theory, appeals can proceed indefinitely, but to ensure commercial efficiency, there must be a finality. When the cost of appeal exceeds the amount in dispute or triggers an ecological security threshold, the case will be transferred to the OmniPact Supreme Court.
1. Triggers
Economic ceiling: When.
Round limit: The game is still undecided after reaching Round 5 (63 arbitrators).
2. Composition of the final jury
The final review panel is no longer composed of randomly selected ordinary nodes, but rather of "veteran nodes" consisting of the top 1% of Stakers or DAO Elected Delegates.
Real-name/semi-real-name: Veteran nodes are usually linked to real-world reputation (such as well-known Web3 organizations or auditing firms).
Public voting: The final appeal will no longer use a commit-reveal process, but will instead be conducted through a public, named vote. This means their verdict will be directly subject to moral scrutiny by the entire community.
3. Finality & The Right to Fork
The jury's ruling is final and directly modifies the financial status of the OES contract.
What if the jury is corrupt?
This is the ultimate defense for decentralized protocols. If the final jury makes an obviously wrong ruling (e.g., the whole world saw the seller send an empty packet, but the jury ruled in favor of the seller), the community will initiate a Social Consensus Fork.
Honest community members will deploy a new version of the OmniPact contract.
In this version, the staked tokens of corrupt jury nodes will be removed (slashed).
The market will sell off old chain tokens and buy new chain tokens.
This ultimate deterrent of the "potential fork" forces the final jury (even though they own 51% of the tokens) to refrain from wrongdoing, as doing so would cause the value of their tokens to drop to zero.
Last updated