Went into the reward distribution this week expecting one number and found four recipients.
Block rewards on Dusk are split between the block generator with an additional portion based on credits. The development fund. The validation committee. And the ratification committee.
That last pair is what caught me.
Because Succinct Attestation runs proposal then validation then ratification. Three steps. And the reward structure pays the validation committee and the ratification committee separately rather than treating them as one group doing consensus work.
Which tells you something the architecture pages do not spell out. Those two committees are genuinely distinct roles with separate incentives rather than the same participants voting twice.
Paying them separately means each one has its own reason to do its own job properly. Ratification exists to check the validation outcome. If both were paid as a single pool the incentive to actually scrutinise the earlier step gets weaker.
The generator being paid partly on credits rather than a flat amount is a second signal. Behaviour over time affects reward rather than just showing up for a round.
My caveat is that reward design is easy to admire on paper and hard to evaluate without watching it under stress. Incentives that look sound in a document sometimes produce strange behaviour once real value depends on them.
does splitting rewards across consensus roles strengthen the design or just complicate it
@Dusk #dusk $DUSK
Block rewards on Dusk are split between the block generator with an additional portion based on credits. The development fund. The validation committee. And the ratification committee.
That last pair is what caught me.
Because Succinct Attestation runs proposal then validation then ratification. Three steps. And the reward structure pays the validation committee and the ratification committee separately rather than treating them as one group doing consensus work.
Which tells you something the architecture pages do not spell out. Those two committees are genuinely distinct roles with separate incentives rather than the same participants voting twice.
Paying them separately means each one has its own reason to do its own job properly. Ratification exists to check the validation outcome. If both were paid as a single pool the incentive to actually scrutinise the earlier step gets weaker.
The generator being paid partly on credits rather than a flat amount is a second signal. Behaviour over time affects reward rather than just showing up for a round.
My caveat is that reward design is easy to admire on paper and hard to evaluate without watching it under stress. Incentives that look sound in a document sometimes produce strange behaviour once real value depends on them.
does splitting rewards across consensus roles strengthen the design or just complicate it
@Dusk #dusk $DUSK