I propose not to allow any delegations on the following tracks which should be considered of vital importance:
Root track (could completely change the protocol)
Wish for change track (could produce deep changes in the protocol)
Treasurer track (unlimited spending without any cap, potentially burning the full treasury if the beneficiary address is a burn address)
The idea behind delegation is good, it's the concept that for specific actions you do not want to vote yourself but you rather trust a third party. However for things of extreme importance it is risky to trust a third party because of the well-known agent problem
Just like in your daily life you can issue a POA for a commercial contract but not for extremely important events like for example marriage / divorce / etc
Certain events are so critical that it is important to know that there is no risk that the agent might have a different opinion than the owner
This is not specifically targeted at DV (even if of course their role has to be considered) but it's more about the idea that for certain tracks it is perfectly reasonable to do delegations and for others it's not in order to protect the integrity of the system
SubSquare is proposing to not allow delegations on certain tracks that are considered of vital importance. These tracks include the Root track, Wish for change track, and Treasurer track. Delegation is the concept of trusting a third party for specific actions, but for things of extreme importance, it is risky to trust a third party due to the agent problem. Certain events are so critical that it is important to know that there is no risk that the agent might have a different opinion than the owner. This proposal is not specifically targeted at DV, but it's more about the idea that for certain tracks, it is perfectly reasonable to do delegations and for others, it's not in order to protect the integrity of the system.
Threshold
I propose not to allow any delegations on the following tracks which should be considered of vital importance:
Root track (could completely change the protocol)
Wish for change track (could produce deep changes in the protocol)
Treasurer track (unlimited spending without any cap, potentially burning the full treasury if the beneficiary address is a burn address)
The idea behind delegation is good, it's the concept that for specific actions you do not want to vote yourself but you rather trust a third party. However for things of extreme importance it is risky to trust a third party because of the well-known agent problem
Just like in your daily life you can issue a POA for a commercial contract but not for extremely important events like for example marriage / divorce / etc
Certain events are so critical that it is important to know that there is no risk that the agent might have a different opinion than the owner
This is not specifically targeted at DV (even if of course their role has to be considered) but it's more about the idea that for certain tracks it is perfectly reasonable to do delegations and for others it's not in order to protect the integrity of the system
Hey Giotto,
I am not going to be as sharp as the other responders as I appreciate the underpinning of your argument. Delegators should be careful and aware of the risks when delegating on the aforementioned tracks. However, I believe that at the protocol level, there is sufficient functionality to afford users flexibility around track delegation, (see @BILL 's reply).
As a takeaway from this engagement, I think that tool and wallet developers should include some warning to users if they do provide a "select all track" delegation function. Some may have it already but it is worth noting.
Thank you for your proposal and keep thinking, innovation sees many nays along the way. Consider my nay a soft nay with a handshake.
Regards,
Will | Paradox
Edited