We can certainly measure the activity of delegates as part of this.
Perhaps we could set this up as an actual experiment. Enable it for only half of addresses at random, and after a while, measure the percentage of tokens delegated to active delegates from each set of addresses.
I’m definitely on board with this, especially if it’s first run as a small scale experiment.
However, @cory.eth makes a good point about edge cases, especially with regards to many holders voting at the same time, bumping up gas significantly, or people voting during regular high gas periods.
Is there a risk that the fund would be drained if even a few people would vote when the gas is like 3ETH at some points, or is that a preventable scenario?
I love the idea of running this like an actual experiment! Excited to see the draft proposal.
How about doing a poll to understand how many users want to change their delegation (either to themselves or to someone else) but are not able to do so because of the transaction cost?
+1 for running a small scale experiment
Maybe there could be some form of gas price limit? I remember unstoppable domains used to offer claiming (minting) credits which would only mint if gas was under 60GWEI (cant remember the exact number). Or a limit on redelegation tx per block so not so many tx get fed to the network at once?
I think as long as its clear how things work, people should not mind having to wait for gas to be under a certain price or to join a queue of redelegation transactions waiting to be sent to the mempool in order to get a free redelegation.
Also +1 for running a bounded small-scale experiment.
One other comment. More generally, I don’t see why the redelegation process would need to be nearly so often as monthly. Annually would be fine by me, with the first one obviously coming/starting some short time after ENS Governance determines/votes to do such a thing. But after that, annually, or semi-annually, cooked into the code would be more than sufficient.
If a reason emerges to do it “out of band”, sooner than the code-bounded period, then that could be initiated by ENS Governance in the oddball / unforeseen case.
I fully support this proposal. I would even go more extreme and say that once a year is enough. Most elections you only get to vote once every four years, and I think it would be nice to have a yearly event in which we encourage people to reconsider their delegates. Of course they can use the “once a year” anytime they want, and can redelegate paying their own gas as much as they want–but imagine we had a “get out and vote” campaign every year, during the airdrop anniversary?
I consider delegation part of voting and voting to be a public good, so is within the scope of the DAO to subsidize.