Coil incentives

While the single-head Gummiworm protocol can be self-hosted and thus comes with its own incentive structure, the Coil Network provides an attractive opportunity for new heads to benefit from the collectivized validation of a diverse and trustworthy set of coil peers.

In such a network, however, a carefully balanced incentive model is required. The model needs to create the conditions for the following set of goals:

• Head operators should be disincentivized from dishonest behavior or poor infrastructure maintenance that leads to a liveness failure and evacuation
• Head operators should have predictable costs for long term forecasting
• Head operators should have complete control and claim to the fees within the L2 as a result
• The protocol should represent yield for coil operators
• Align coil operators with the health of the protocol over the specific self-interest of any one head
• Consistent revenue for a treasury, which can invest in further expanding the protocol
• Coil operation should be sybil resistant
• Head operators should be unable to select their own coil peers directly

To achieve this, we propose the following loose tokenomics and incentives model, though full implementation would require a much more rigorous analysis of the game theory involved.

First, to disincentivize dishonest behavior among head peers, we require a substantial bond be posted by each head (sourced from the peers in whatever ratio they see fit). In the event that consensus breaks down, this bond is used as compensatory damages for the participants who are inconvenienced by the evacuation. This creates a strong disincentive to attempt to censor transactions, forge ledger state transitions, or run a low quality network that eventually evacuates.

Second, as part of the process of opening a head, a per month (or some other suitable time period), per coil peer cost is negotiated. At the beginning of each month, this amount is paid by the head operator out of the equity compartment. In principle, this rent can be in any token, but in practice, we anticipate it will be in stablecoins. This rent is then either distributed continuously to the coil throughout the month, or in a lump sum at the end of the month. This creates stable, predictable costs for the head peers and predictable revenue for the coil peers.

A small percentage of this rent accrues directly to the Sundae treasury.

In order to align coil peers with the long term health of the Sundae and Gummiworm protocols, we institute 3 measures:

• First, we require the bond placed by head peers, and an additional pledge posted by coil peers, to be in SUNDAE tokens
• Second, the coil peer can choose for anywhere between 20% and 100% of their revenue to be received in SUNDAE. This portion is swapped through SundaeSwap, and is exempted from the sundae treasury cut, resulting in higher overall revenue for the coil peers.
• Third, the revenue received becomes a part of the coil peers pledge. While the coil peer is active for any head, and for a waiting period after, they can only withdraw a portion of their bond per unit time. The withdrawal rate will grow asymptotically for long-running reliable coil peers to reward reliable service. This ensures that a coil peers alignment with the Sundae protocol grows over time.

Overall this creates disincentives for misbehavior; reliable costs for the head peers; attractive revenue for the coil peers; and Sundae protocol revenue to fund a treasury for future developments.