OIP-4 Implementation Update #1 — Delegate Reputation Score (DRS) Details and Adjustments

General Discussions
1

As committed during the OIP-4 discussions and vote, we want to ensure that implementation remains transparent and accountable to the community. This post serves as the first public update on the implementation of OIP-4: Delegate Compensation and Delegate Reputation Score Integration for stOBOL.

We encourage all ongoing discussion to remain in the original OIP-4 thread to consolidate feedback and ensure context remains accessible. This post is strictly a factual update on how the proposal is being implemented, what has changed since the vote, and what the community can expect next.

We welcome input on this implementation, but to keep the discussion productive, we ask that feedback be:

  • Structured and justified — Please include the reasoning and/or data behind your concern.
  • Solution-oriented — If you raise a problem, we’d love to hear any ideas on how to address it.

We welcome feedback that helps improve the implementation, not to re-open a debate on whether OIP-4 should be enacted. That time has passed and OIP#4 was validated by vote.

Final Calculation Details

You can view the full formula and logic for the Delegate Reputation Score (DRS) here :backhand_index_pointing_right: DRS v1 Implementation Spec

Adjustments Since Vote

1. Scoring Period Adjustment

Originally, “active delegate” classification required a DRS ≥ 65 calculated over the past 3 governance cycles. We are making a small change and this now evaluates the average score over the last 5 proposals.

Why: This ensures the score reflects actual delegate behavior across real decisions, rather than an arbitrary time window. Since the number of proposals can vary significantly across cycles, using a fixed number of proposals offers more consistent evaluation.

2. Retroactive Bootstrapping

To avoid penalizing delegates for not anticipating scoring rules, the system will apply a simplified scoring logic retroactively to the five most recent proposals prior to DRS activation.

  • Each of these past proposals will carry a 20% weight.
  • If a delegate voted on 4 or more of the last 5 proposals, they will begin with a DRS of 80+ and qualify as active.
  • Starting from the next proposal, the full DRS formula kicks in. So progressively, the score will be based on 4 past + 1 new, then 3 past + 2 new, etc., until it’s entirely derived from live tracked behavior.

3. New Delegate Bootstrapping

New delegates should not need to wait five proposals to qualify and start to be compensated. For new entries:

  • The DRS is calculated over the number of proposals they were eligible to vote on.
  • If they’ve participated actively from their first eligible proposal, their DRS will be high enough to qualify immediately.
  • However, if early participation is poor, their score will drop quickly.

In UI: A visual tag will indicate delegates still within their “evaluation window” so token holders can assess their track record accordingly.

4. Update Cadence

The DRS will be updated once per governance cycle (every 3 weeks).

Why:

  • Allows complete view of voting impact, rationale submissions, and forum engagement across a full proposal lifecycle.
  • Prevents volatile mid-cycle updates and reduces delegation churn.
  • Promotes clearer decision-making by showing fully-formed scores.

Distribution Curve for Delegate Rewards - Request for Feedback

OIP#4 specified that once a delegate is classified as “active,” their share of rewards from the Delegate Rewards Pool would be proportional to the square root of their delegated voting power.

This approach was selected to:

  • Avoid hyper-concentration of rewards toward the largest delegates,
  • Support smaller delegates who maintain high DRS and act diligently,
  • Encourage a more pluralistic, non-linear distribution of influence.

However, following internal reviews and community conversations, some valid concerns were raised:

  • Square root-based distribution, while fairer in theory, may introduce attack vectors for Sybil behavior or farming incentives,
  • A linear distribution (1:1 with delegated VP) would be simpler and safer — but might overly reward the largest holders and discourage long-tail participation.

What we’re asking

We’re inviting you to share feedback on which model to adopt:

  • Should we keep the square root model, aiming for long-term fairness and decentralization, with a commitment to monitor and adjust if abuse occurs?
  • Or should we adopt a linear model from the beginning, prioritizing simplicity and minimizing Sybil risks?

This is a reversible decision (a “two-way door”) and can be updated later via governance. But since implementation will begin shortly, we’re asking for input by 2025-06-18T16:00:00Z .

Please respond in the OIP#4 thread with your thoughts. All feedback welcome.

Next Steps

We are now in the build and integration phase with Curia and Tally. Once live, the Association will monitor the system and assess:

  • If certain weightings (e.g. participation vs. rationale) need adjusting
  • If additional components should be integrated
  • If more granular or qualitative governance metrics should be introduced

We’ll report back regularly on progress and welcome suggestions from the community. We want to thank everyone again for the thoughtful feedback that shaped this design.

This phase is about building quality infrastructure together. Let’s keep the conversation focused, open, and collaborative.