What finally clicked for me about Moonlight and Phoenix on DuskVM is that different state models don’t need different finality models.
Moonlight arrives with a public account trail: balances, sender, receiver, amount, nonce.
Phoenix arrives through encrypted notes, shielded outputs, and nullifiers.
I kept assuming those two shapes would somehow require two different ways to reach finality after DuskVM.
But that’s not really DuskDS’s job.
Moonlight can stay account-shaped. Phoenix can stay note-shaped. DuskVM can accept both without forcing one to become the other, while Dusk L1 still gives the resulting state a deterministic finality boundary.
So the mistake was assuming that different state representations need different endings.
They don’t.
Dusk can let Moonlight and Phoenix remain fundamentally different underneath while still giving both the same answer to the one question that matters at the end:
when is this state final?
#dusk $DUSK @Dusk
Moonlight arrives with a public account trail: balances, sender, receiver, amount, nonce.
Phoenix arrives through encrypted notes, shielded outputs, and nullifiers.
I kept assuming those two shapes would somehow require two different ways to reach finality after DuskVM.
But that’s not really DuskDS’s job.
Moonlight can stay account-shaped. Phoenix can stay note-shaped. DuskVM can accept both without forcing one to become the other, while Dusk L1 still gives the resulting state a deterministic finality boundary.
So the mistake was assuming that different state representations need different endings.
They don’t.
Dusk can let Moonlight and Phoenix remain fundamentally different underneath while still giving both the same answer to the one question that matters at the end:
when is this state final?
#dusk $DUSK @Dusk