Could there be some merit to, instead of having the voting power decay or increase over time, simply have it stay fixed? So if we were to lock up our YFI for a year, that gYFI always has 365 days of voting power. I don’t really like the idea that my YFI will have more or less power based on when I lock it up– what we should be rewarding here is the fact that the YFI is being protected and locked into governance to be used for its intended purpose.
If voting power stays fixed over the course of the vote lock, then that person would have the exact same voting power in all of the proposals they vote on, and are equally rewarded for it every time. If it were a decaying function– then theoretically someone with enough capital could buy up enough YFI, vote-lock it immediately before the vote, and have a much higher relative power than anyone who has been consistently and reliably locked up and voting in the protocol.
Because of this, I also think making the gYFI non-transferrable is the ideal solution. I think this is the only way it works with equal-weighting votes across the whole period. If gYFI is transferrable, then I think that would bring in a whole separate set of issues if someone wants to redeem their YFI but has lent out their gYFI.
The only issue I see with all of this will be for users who are currently lending their YFI out (for instance, what the YFI vault is doing). If we allow locking of YFI, and a borrower stakes this borrowed YFI– it is stuck, and it’s very likely that lenders will not be able to withdraw their YFI and a good chance that borrowers may get liquidated due to the inevitable high interest rates at high utilization (I’m thinking of CREAM specifically here). Realistically, this may be a net positive, as it will discourage borrowing/lending of YFI since the vote-locking will be an option that lenders will need to be wary of.
Overall, I am 100% for this proposal, but I believe that vote-locking should be weighted equally across the course of the lockup and that gYFI should not be transferrable.
I like the concept but not all the details. x/365 seems too high and could discourage newcommers interested in governance. x/60 to as much as x/100 seems sufficient. I would cap at 1. I also recommed starting everyone “fresh” rather than making this retroactive.
Is there a way to determine how long YFI has been staked in the governance pool and use that instead of issuing a new token? No need for a lock as YFI in governance is staked and would lose voting and reward power if unstaked and transfered.
I recommend voting weight and reward weight be calculated the same way so that share of rewards increase as voting weight increases.
For a bit of analysis of people attempting to front-run the NuCypher staking contract, check out the Gauntlet report. We spent some time analyzing something similar to a decaying vote, albeit for 30 day staking periods.
Great perspective. I would keep it at a sliding scale - 10% bonus when you cross 3 months, sliding up to a 50% bonus at 36 months. But capped at 36 months.
I think this is a great idea because it improves security. The security of our governance needs to be taken seriously, we cannot afford to have YFI hijacked by malicious actors. We need to prepare ourselves for potential attacks on governance, and time weighted voting power is a step in the right direction.
I agree with others that having the gYFI be transferable defeats the purpose. Having increased voting weight based on time, but no new token, seems best.
@zeframlou’s link on conviction voting is interesting. Am I right in understanding that it removes time limits for proposals, and instead just passes proposals when they meet some threshold?
I definitely “for” this proposal.
By adding the ability to lock a token by increasing the weight of your vote, we will give more rights to those who want to be with the project for a long time (also I would like the reward in government locks to be not much lower than the reward in the vault, so as not to create a “punishment for the desire to vote”, but I don’t understand how to achieve this).
Also, I do not quite understand why most of the topics are against gYFI be transferrable. I proceed from the assumption that if a system can be bypassed for some malicious activity, then it will always be bypassed, and therefore it is not worth putting unnecessary obstacles to ordinary people using this system. In my opinion, trading in blocked management tokens is easy to bypass simply by selling your ethereum account on OTC. At the same time, we simply stop our money lego and limit the space for fantasies of what can be built above.
And yes, users of the government system should be aware that if they sell their gYFI tokens, then the buyer can use them for his own purposes, which may not correspond to the goals of the seller. Therefore, a) the seller must treat these tokens responsibly and understand what value they carry;
b) see who he/she is selling to and what goals the buyer has.
The transferability of such tokens will also provide an opportunity to leave the system for people who, for some reason, are disappointed in it and do not want to take direct participation anymore (and this will always be the case, since not all decisions have unequivocal support). Without the opportunity to leave, we just get people who will try to harm the long-term perspective of the project as much as possible with their votes, as they will be upset by some decision made in the system.
I am generally AGAINST. I do not think the question has been addressed: what is the function of YFI? We want to minimize its function from what used to be governance to now earnable governance. This is like setting up a boys club inside a boys club where 1 wallet like CRV takes 51% and never gives it up due to literally no one being able to oust them
If anything, I think this solidifies YFI (or in this case, gYFI), as being for governance. Having a weighted model that rewards locking up the token for governance ensures that those who are truly holding YFI for governance (and not for speculation/dividends) are the ones who have the biggest voice in governance decisions.
I’m not sure about your argument comparing to Curve– token distribution for YFI was much more equitable compared to curve– it’s not like Andre gave himself a big chunk of the supply; he was earning it from farming just like everyone else.
My point is it further centralizes power to the few. The original purpose of YFI was to govern itself, not to generate a different token for those who specialize into it
I don’t think yYFI holders should be able to vote. They are using their YFI as collateral (and thereby putting it at some risk). They should not also be able to benefit from Yearn income by doing so. One, or the other.
But also, I think gYFI should not be transferrable. Otherwise, it sort of negates the point, no?
With the launch of YFI on AAVE, this proposal gains much more substantial pressure to be implemented in one form or another before permanent damage can be done to the protocol and it’s reputation by people leveraging AAVE to manipulate votes.
Agreed. The other discussion on Aave is here. (Linking for context.)
Do people want to start work on putting this into a YIP? I’m still not in favor of making gYFI transferable, but I think we can start by considering (compiled from across the two threads):
additional multiplier on votes based on time locked up
additional rewards for stakers based on time locked up
increase lock-up time for voting to be longer, or release the lock-up in stages over time
changing the voting system to be something that favors long-termists like conviction voting
I was against the idea of a gYFI at first. I recognied the potential problems with YFI being used on other farming platforms or as collateral. I believed that increasing vote share and reward share for time in the governance pool (up to some cap) would be beneficial but not enough given the potential yield avalable elsewhere – and I still do.
I am now more interested in separating governance from financial interest in the project. Yes, I know it sounds odd. Here is an idea: give every YFI holder a gYFI token and assign the gYFI the governance rights and the YFI token the financial rights. Make everything tranferable but institute the time based benefits for voting and for rewards. We could require significant gYFI locks for voting power. YFI could still be used o other platforms or as collateral without putting governance at risk.
is automatically solved by rewarding stakers proportional to gYFI held/staked– if you lock up for longer, you get relatively more gYFI for the same amount of YFI compared to those who are not locking up theirs for as long.
I think perhaps we should do a soft vote on vote.yearn.finance. My proposal would be to do the following poll:
gYFI is not transferrable, and gYFI is minted based on committed time-lock, where gYFI = YFI * x/365. So locking up 1 YFI for 1 year (or 365 days) mints you 1 gYFI. This YFI is now locked for the year, and the YFI becomes unlocked after 1 year. Redeeming the gYFI for the YFI burns the gYFI. Voting weight stays the same throughout the entire duration.
Same as 1, but gYFI is transferrable.
gYFI is not transferrable. gYFI functions similarly to veCRV– in that as expiration of the lock approaches, vote weighting decreases.
Same as 3, but gYFI is transferrable.
My biggest problem with the gYFI being transferrable is I just don’t see how the time lock would really work effectively– as in, how you would know when the original underlying YFI could actually be unlocked. But as always– I do not write smart contracts so I’m not sure of all the ways something like this could be implemented.
I would also really appreciate if someone could explain why they think decaying voting power is a good idea– it seems we would want those who commit to a long lockup (invested in governance) to have an equal say in all matters throughout the duration of their vote-lock.
I am against if it’s linear, but if weight is distributed using a logarithmic curve I’d be in favor. Holding for the first and second week should weigh much more than holding for the 51st and 52nd week.