#TermMax @TermMax
There's something strange about spotting an issue inside a vault, only for the built-in safeguard to make you sit tight.
That contradiction is exactly what drew me into @TermMax uneven timelock structure.
Normally, sensitive vault updates follow one path: propose it, sit through the delay, then confirm it. The standard wait is 24 hours, and a Guardian can cancel the pending update anytime during that window.
TermMax doesn't treat every update the same way, though.
Extending the timelock, cutting the performance fee, or dropping a market from the approved list can all happen instantly. Shortening the timelock, hiking the fee, adding a new market, or swapping the Guardian — those all require waiting.
I couldn't stop turning over why that unevenness makes sense. A timelock earns its keep when a curator is asking depositors to accept something unfamiliar. Approving a new market widens where their money can be put at risk. Bumping fees alters the deal they originally agreed to. Cutting the timelock shrinks the heads-up window before future moves.
Those kinds of actions should face resistance.
Now picture an approved market that suddenly turns risky. Forcing its removal to wait — purely because "every parameter change needs a delay" — would flip a protection into a liability.
The underlying logic isn't really about the parameter itself, but about what kind of permission is shifting.
Growing what the vault is allowed to do should move slowly. Shrinking what it's allowed to do should move fast.
I find that split appealing, though real-world cases are probably messier than any clean rule. Pulling a market can lower one risk while shifting concentration into another. "Lowering risk" isn't automatically "free of consequences."
Maybe that's the actual challenge with lopsided timelocks — not whether it's logical to slow down growing risk, but whether risk even has one clear direction once markets are under pressure.
There's something strange about spotting an issue inside a vault, only for the built-in safeguard to make you sit tight.
That contradiction is exactly what drew me into @TermMax uneven timelock structure.
Normally, sensitive vault updates follow one path: propose it, sit through the delay, then confirm it. The standard wait is 24 hours, and a Guardian can cancel the pending update anytime during that window.
TermMax doesn't treat every update the same way, though.
Extending the timelock, cutting the performance fee, or dropping a market from the approved list can all happen instantly. Shortening the timelock, hiking the fee, adding a new market, or swapping the Guardian — those all require waiting.
I couldn't stop turning over why that unevenness makes sense. A timelock earns its keep when a curator is asking depositors to accept something unfamiliar. Approving a new market widens where their money can be put at risk. Bumping fees alters the deal they originally agreed to. Cutting the timelock shrinks the heads-up window before future moves.
Those kinds of actions should face resistance.
Now picture an approved market that suddenly turns risky. Forcing its removal to wait — purely because "every parameter change needs a delay" — would flip a protection into a liability.
The underlying logic isn't really about the parameter itself, but about what kind of permission is shifting.
Growing what the vault is allowed to do should move slowly. Shrinking what it's allowed to do should move fast.
I find that split appealing, though real-world cases are probably messier than any clean rule. Pulling a market can lower one risk while shifting concentration into another. "Lowering risk" isn't automatically "free of consequences."
Maybe that's the actual challenge with lopsided timelocks — not whether it's logical to slow down growing risk, but whether risk even has one clear direction once markets are under pressure.