Provider Incentives

The Capacity model incentivizes participants that are interested in publishing protocol-related data on the blockchain to hold FRQCY. The principal use case for Capacity is Providers acting on behalf of coinless users who have delegated certain permissions to the Providers to act on the users' behalf.

Capacity balances two core goals related to scalability. The first goal is providing a low transaction cost mechanism for Message publishing. Given the volume of Messages generated in social networks, Providers must be extremely cost-conscious about Message publishing costs. The second goal is minimizing spam in a system designed to minimize Message publishing costs. As the cost of Message publishing decreases, the viability of spam Messages increases proportionally. The design of Capacity attempts to balance these two goals.

Capacity Model

While all transaction fees on Frequency can be paid for with FRQCY, certain types of transactions can use Capacity which strikes a balance between Message publishing cost and spam prevention using three core mechanisms: staking, renewal, and unstaking.

Capacity Staking

Capacity is generated by staking FRQCY. There are several mechanics and restrictions involved in Capacity Staking.

The amount of Capacity generated is proportional to the number of FRQCY staked, with the specific ratio being set via Frequency Governance. While the ratio can change over time, changes to the ratio only affect staking done after the change; existing stakes maintain the ratio in effect at the time the stake was made. This incentivizes stakers to maintain stakes with ratios perceived as favorable, generally stabilizing net investment in the network. The roadmap anticipates that the staking ratio will migrate from the current model where it is set by governance to a future model that adjusts the ratio algorithmically based on chain metrics. The initial Capacity Staking Ratio for Frequency is 50:1; 50 FRQCY must be staked for each Capacity unit generated.

When staking for Capacity, the staking account must designate a specific Provider to receive the Capacity. While any FRQCY holder can participate in staking, Capacity can only be spent by a Provider. Capacity Staking is specifically targeted at managing costs for Providers by giving them a mechanism with low price volatility for publishing Messages. This restriction ensures that Capacity Staking focuses on the Message publishing use case, versus other potential derivative possibilities and incentives.

Capacity can only be spent by the Provider who receives it. There is no transfer mechanism for Capacity: only the Provider who receives the Capacity can use it on eligible transactions as an alternative to token fees. This focuses the utility of Capacity on minimizing transaction cost while avoiding a secondary market in Capacity speculation.

Capacity Renewal

Capacity renews on a periodic basis. Despite the staking ratio for Capacity requiring multiple FRQCY to be staked for each unit of Capacity generated, any Provider with recurring Capacity-eligible transactions will quickly find using Capacity more economically efficient than spending tokens.

Capacity Staking has a Capacity Epoch, which is defined as a fixed number of blocks. The number of tokens staked by a given Provider determines that Provider's Capacity Allocation, which may be increased by Provider Boosting. Providers can spend up to their full Capacity Allocation on transaction fees each Capacity Epoch. At the beginning of each Capacity Epoch, the Provider's Capacity Allocation is renewed, allowing the same amount of Capacity to be spent again in the next Capacity Epoch. The initial Capacity Epoch for Frequency is 7,200 blocks (a duration of approximately 24 hours, given Frequency's 12-second block time target). Capacity incentivizes sending Messages by reducing transaction costs. In a social networking context, the utility of the network increases as users communicate more, so Capacity acts to increase overall network utility over time by reducing the transaction cost of Message publishing.

To illustrate the utility of staking for a Provider, given the initial Capacity Epoch length of 7,200 blocks, and the initial 50:1 Frequency to Capacity Staking Ratio, consider two Providers with 50 FRQCY, Provider 1 staking their tokens, and spending 1 Capacity on transaction fees each Capacity Epoch, and Provider 2 spending 1 FRQCY on transaction fees each Capacity Epoch. After approximately 50 days, Provider 2 would be at fund exhaustion, while Provider 1 could continue to spend 1 Capacity each Capacity Epoch indefinitely, preserving the optionality to unstake all staked tokens at some time in the future.

Unused Capacity does not roll over to the next Capacity Epoch. Any unused Capacity remaining in a Provider's Capacity Allocation at the end of a Capacity Epoch is lost. When Capacity renews at the beginning of a Capacity Epoch, it can never be more than 100% of the Provider's current Capacity Allocation. This makes Capacity usage more analogous to time-bound resources such as wireless spectrum or bandwidth than to accumulation-bound resources such as a bank balance. The combination of favorable staking ratios, the fact that the staking ratios are locked at the time of stake, and the fact that Capacity cannot be transferred might incentivize Providers to hold excessive stakes at favorable rates. However, the fact that Capacity cannot be rolled over incentivizes behavior that optimizes Capacity Staking for peak use cases, but not much beyond that, since such stakes provide no extra economic value for the encumbrance of staking.

Capacity Unstaking

FRQCY staked to generate Capacity can be unstaked. However, unstaked tokens are not immediately available, and remain locked for a period of time after unstaking before being otherwise usable by the staker. This delay is referred to as the Thaw Period.

The Frequency Thaw Period is a key anti-spam measure. In a model such as Capacity where tokens can be staked instead of spent to pay transaction fees, there is a risk of creating an environment where there is zero marginal cost for sending spam. Requiring a non-trivial staking ratio and combining that with a significant Thaw Period does not directly increase the cost of spam, but puts the value of the staked FRQCY the spammer acquired at risk, as the effects of any large-scale abusive behavior theoretically have time to be reflected in whatever mechanism the spammer used to acquire FRQCY before the unstaked tokens can "thaw" and become usable again by a bad actor. The initial Thaw Period for unstaking Capacity is 30 Capacity Epochs.

Requiring bad actors to acquire FRQCY to send "free" Messages, but making them wait for the repercussions of their behavior to be reflected in the value of their FRQCY holdings does not eliminate bad behavior. However, it does limit the economically sustainable volume that can be sent without impacting the value of the bad actor’s FRQCY holdings to a greater extent than the value of the Message.

Capacity immediately stops being generated when unstaked, but the staked tokens are only unlocked at the end of the Thaw Period. This lowers the volatility of Capacity Staking Ratio changes. By default, if the staking ratio goes up (requiring more FRQCY per unit of Capacity), existing stakeholders have a strong incentive to retain their existing stake and benefit from the better ratio. However, if the staking ratio goes down (requiring fewer FRQCY per unit of Capacity) there could be strong incentive for large numbers of stakers to unstake tokens and restake at the lower rate, freeing up excess tokens for other use cases or increasing the amount of Capacity they can generate. This could have drastic effects on freely circulating token supply and pricing. The Thaw Period mitigates some of this potential volatility as staking ratio rates decline by requiring those who contemplate unstaking to not only react to the new lower ratio, but also predict that the ratio will still be lower at the end of the Thaw Period, while losing the benefit of generating Capacity during the Thaw Period. The combination of the risk of the ratio going back up and the loss of utility during the Thaw Period further helps balance Frequency's Economic Model.