#dusk $DUSK @Dusk
Two lines from the Dusk staking docs, a few paragraphs apart. One: there is no protocol waiting period after a successful unstaking transaction. The other: soft penalties can move part of active stake into locked stake. So which is it — free to walk, or capital that can be pinned in place?
Both. And the difference is where the actual risk lives.
Entry and exit are not symmetric. Minimum direct stake is 1,000 DUSK; the docs set a floor, not a ceiling. Exit is immediate — full unstake, or partial as long as what remains still clears 1,000. Entry is not. A new stake becomes eligible at the start of the epoch after the next boundary. An epoch is 2,160 blocks, so depending on where your transaction lands, activation runs roughly one to two epochs — about 6 to 12 hours.
Read that as behaviour, not documentation. Leaving costs a fee. Returning costs half a day of earning nothing and counting for nothing. Under volatility, that asymmetry quietly pushes the validator set to thin out faster than it refills.
The lock that does exist is a penalty, not a schedule. Soft penalties cover failed participation: suspend eligibility, move part of active stake into locked stake. It's still yours. It just does nothing. Hard penalties cover provably invalid consensus behaviour — invalid votes, signing conflicting proposals — and can burn stake. Hence the one operator rule I'd bolt to the rack: never run the same consensus key on two active nodes.
Fair is fair. Separating downtime from provable misbehaviour is good design, and reserving the burn for the latter is defensible. My question is narrower: is "no lockup" an honest headline when an uptime failure can immobilise capital?
So — delegators and pool users, which risk are you actually pricing? Slashing, or the quieter one: stake sitting locked while a node sits suspended?
Two lines from the Dusk staking docs, a few paragraphs apart. One: there is no protocol waiting period after a successful unstaking transaction. The other: soft penalties can move part of active stake into locked stake. So which is it — free to walk, or capital that can be pinned in place?
Both. And the difference is where the actual risk lives.
Entry and exit are not symmetric. Minimum direct stake is 1,000 DUSK; the docs set a floor, not a ceiling. Exit is immediate — full unstake, or partial as long as what remains still clears 1,000. Entry is not. A new stake becomes eligible at the start of the epoch after the next boundary. An epoch is 2,160 blocks, so depending on where your transaction lands, activation runs roughly one to two epochs — about 6 to 12 hours.
Read that as behaviour, not documentation. Leaving costs a fee. Returning costs half a day of earning nothing and counting for nothing. Under volatility, that asymmetry quietly pushes the validator set to thin out faster than it refills.
The lock that does exist is a penalty, not a schedule. Soft penalties cover failed participation: suspend eligibility, move part of active stake into locked stake. It's still yours. It just does nothing. Hard penalties cover provably invalid consensus behaviour — invalid votes, signing conflicting proposals — and can burn stake. Hence the one operator rule I'd bolt to the rack: never run the same consensus key on two active nodes.
Fair is fair. Separating downtime from provable misbehaviour is good design, and reserving the burn for the latter is defensible. My question is narrower: is "no lockup" an honest headline when an uptime failure can immobilise capital?
So — delegators and pool users, which risk are you actually pricing? Slashing, or the quieter one: stake sitting locked while a node sits suspended?