Aura was originally setup to allow each vlAURA to vote with its underlying share of veBAL. This required Aura voters to go to the Balancer Snapshot Space and vote. It also required separate delegation. The result was that much of the veBAL locked in auraBAL did not vote each round. Aura has recently passed an AIP to change how they handle veBAL voting, such that the AURA multisig will vote on snapshot with the veBAL in auraBAL following a snapshot vote of token holders. The governance can be found here:
This BIP proposes that balancer stop counting vlAURA as part of Balancer governance. Otherwise, Aura votes will be counted twice. Once if/when users vote of Balancer Snapshots, and once when the Aura multisig votes with all of veBAL.
Specification
Change the Balancer Snapshot Voting to no longer include the aura-balance-of-vlaura-vebal strategy.
Update the quorum on future votes to be 3 million veBAL.
Extend all Balancer Snapshot votes by an additional 48 hours (120 hours in total) to provide AURA additional time to run a meta-snapshot. This will also ensure that balancer governance ends during the week in all timezones.
– Future snapshots will therefore start on Thursdays and end on Tuesdays.
Timing
AURA is ready to move forward with this new process. Given that it would be unfair for Aura to vote twice with their system owned veBAL, this BIP proposes to run an immediate vote to make a change to the snapshot space before any other votes are performed.
They will make the specified change for all other Balancer snapshot votes from now going forward. The Balancer Snapshot voting strategy should be changed before this, or vlAURA lockers will be able to vote on the balancer snapshot while auraBAL itself is also voting. This creates the possibility of votes being double counted.
The outcome of the vote will then be binary: if a majority of active vlAURA votes “yea” on a proposal, all underlying veBAL owned by Aura will be used to vote “yea” via Balancer Snapshot. The vice-versa is also true.
I have a some major issues with this.
Aura voter apathy should not be Balancer DAO’s problem. If people are not voting, that does not mean you should get to vote for them without them explicitly opting in. I’m sure some native veBAL holders don’t vote; no one else gets their voting power. If users opt-in, great for you! If they don’t, then you shouldn’t get their voting power. Proposed Solution: opt-in system
The winner-take-all nature of this setup gives leveraged power to influential voters of the Aura ecosystem and significantly lowers the cost of a governance attack. Proposed Solution: replace winner-take-all system with percentage-per-vote system that includes abstentions
Less consequential, logistical question
How is vote timing supposed to work for this? Is Aura going to copy over every Balancer DAO vote? Is it going to have the same voting window? If not, how much will it be reduced by?
All opinions my own
PS: Yes, I do understand that from a technical perspective, Aura holds all of that veBAL and can theoretically do whatever the hell they want with it, but this change would dilute native veBAL holder votes.
if this proposal goes to a vote, shall we also consider raising the quorum to 3M veBAL? This would probably help us ensure some decentralisation in the voting process and avoid that any single entity, or delegate, can potentially reach it single handedly.
The way veBAL voting for wrappers is working, the entity that controls the veBAL can decide how it is used/voted with. This has manifested itself as significant wrappers like Tetu delegating 100% of their vote to one person, who also has other veBAL delegated. Based on the current conditions in Balancer governance it is totally reasonable to give active vlAURA voters a larger say in governance. If other governance is passed to regulate how wrappers vote in non-economic balancer snapshots, it makes perfect sense for AURA to adapt its voting to match that.
For now, Aura was based off Convex. As Convex was fighting for a dominant position in the Curve wars so they could maximise ecosystem value, they also had a winner take all approach. As they approached 50% of veCRV, and in co-ordination with Curve, they moved to splitting the vote based on their snapshot in order to avoid 51% situations.
When this all began, it seemed like the vlAURA passthrough voting was a good idea. In the end, everyone else is concentrating and delegating their votes and their wrapper. At the same time, most Aura voters don’t want to take the time to participate in Balancer governance and trying to explain dual delegation creates tricky UX situations that result in too much of Aura’s veBAL sitting on the sidelines when it should be supporting Aura’s interests. Without doing something Aura risks loosing relevance to other players, who seem far less focused on the long term health of the overall ecosystem.
First: Because balancer votes run as yes/no snapshots, there is no way for AURA to vote with all it’s vote weight (no passthrough delegation). Maybe the balancer snapshots could be changed to be some sort of a split your vote/% type model similar to the gauge weight votes with yes and no as options. Then this could be possible.
Second: There are around 10 million vlAURA locked. There’s only about 3 million in LP/for sale. The locked AURA is will distributed between a number of different major stake-holders including:
Gnosis
A couple of Aura Whales
A couple of Bal Whales
Members of the Aura and the Balancer teams
Aura would be quite hard, if not impossible to 51%. Further, AURA currently only controls 25% of veBAL. The largest voting block in the Balancer voting block seems to control more than 40% of veBAL, and as far as Aura can tell controls less than 10% of vlAURA.
All Balancer snapshot posts will be mirrored as soon as they are posted, and end 16 hours before the end of the Balancer snapshot. It would also be useful to propose (either here or in a separate proposal) that Balancer extend their voting window by 24-48 hours to allow more time for active participation both in Balancer, Aura, and other interested protocols. I would also personally be delighted to not have governance ending on weekends
Thank you for your engagement and your feedback and your understanding of the way things are/that this at the moment isn’t really up to balancer governance to deiced. It wasn’t “supposed” to be this way. In the end, there is now 1 delegate who controls more than 40% of veBAL and who’s delegation includes tetuBAL with >5% of veBAL. As a result, Tetu has become a substantial competitor to AURA in the veBAL wrapper space and their delegation is actually larger and more concentrated than all of auraBAL. Aura is doing what it can to be a strong, present and positive force in this ecosystem, while at the same time to try to remain as decentralised a possible.
My opinions are also my own, but I invite future global governance around delegation and passthrough voting for wrappers(and their derivatives), but at the moment there are none, and Aura is just catching up with whats going on around it…
This is a good point. I have found no one who disagrees with it. So. I would humbly request that this governance includes an increase of the Balancer Quorum to 3 million votes, and an extention of the balancer snapshot period by at least 24 hours such that it does not end on the weekend in any timezone.
If someone thinks these 2 topics are different/don’t fit in the same BIP as this request, which really just has to be done and should happen soon, please let me know and I can create a new BIP for the increase of quorum and length of voting.
tl;dr: I support your proposal to remove the vlAURA voting strategy
I don’t think Tetu should be allowed to have that power either, but as you pointed out, that’s how the ve-system is built so my opinion here is irrelevant.
Balancer votes don’t necessarily have to run as yes/no. I see no reason why a weighted-vote system couldn’t be implemented (if not simply activated if the strategy exists).
Yeah, after sleeping on it, I realized there was almost no point in expressing my governance frustration to you over this proposal. Your proposal simply renounces your own voting power that would otherwise be double counted in the future, which is obviously a good thing.
This proposal makes a lot of sense and I also think it’s quite fair.
1/ It suggests more people participate in any BIP and extends the time for voting.
2/ Aura lockers shouldn’t vote twice as it’s not fair.
3/ The proposed change, while making a change in Balancer’s voting system, secures equity and avoids double votes on Aura’s lockers
That proposal prevents double voting for Aura lockers. I can’t see why I wouldn’t be in favor. =)
I would like to bring to attention certain points that appear to be intended to be used as the main argument for moving to the new Aura voting model:
Since the reference clearly concerns the delegation of my votes, I would like to clarify my position so to ease some concerns:
1- My delegation power is limited to 1.5M votes, certainly not 40% of the total veBAL. For reference, Aura alone already controls, with the current model, 1.2M votes. While quickly approaching the number one spot among delegators;
2- the vast majority of the votes delegated to me come from a single tokenholder. Since tetuBAL does not allow (for now) any other option and the entity I represent controls a majority of the tetuBAL pool, we considered an interim solution so to have those votes available for Governance. This is considered extremely important to us as the Balancer Governance has demonstrated a certain degree of hostility and centralisation in the past;
3- The voting preference of the Tetu Council for Balancer Affairs does not always coincide with the desire of the main investor. For this reason it may happen that the Tetu delegation could vote differently from the main stakeholder. A good example has been the vote on the Index Coop’s proposal, where Tetu abstained, while the main investor overridden Tetu’s voting decision. All this is easily verifiable.
This statement seems arbitrary and, unless I’m reading it wrong, assumes this proposal will pass with flying colours, once again raising the suspicion of an impartial voting block that favours certain groups while punishing others.
I’m sure this is not the case, but given past experiences, please allow me to have this impression.
Certainly! First of all, my thinking should be completely irrelevant. However, I also believe that Aura has made an unilateral (for lack of a better word) decision and that this decision must be scrutinized by Balancer Governance. I simply have no idea how the Balancer Governance majority is going to vote on this.
What I do know is that, until their request is fully approved by vote, Aura should avoid any participation in Governance. Hence, Balancer should continue its voting schedule routine as usual, counting on the goodwill of our Aura friends.
In all fairness, Tetu also made a “unilateral” decision on how they delegated their balancer governance. I think we all agree there could be some future governance to think about and better define rules here. For now, I don’t think this is the right thing to get super caught up on.
For the security of the system I think it’s important to not run any other BIPs until this is voted on. Ultimately up to the governance council to decide that though. we will wait and observe their decision.
I worked on the Aura side of the proposal for this with Tritium.
Aura is closely aligned with Balancer, is the largest holder of veBAL, and has over 3,500 unique users. During the current crypto winter engagement is at all-time lows, a system like this protects the protocol without sacrificing decentralization. The current system requires delegating veBAL voting to the Aura delegate or other protocols, which requires gas for users and harms decentralization for both vlAURA/veBAL. On top of delegates being inefficient, there are obvious conflicts of interest that come up where teams avoid votes, this model ensures 100% of votes are used, 100% of the time… once again further decentralizing veBAL.
Using 100% of AURA’s underlying veBAL to vote through snapshots further decentralizes both protocols and protects from systems like tetuBAL which are largely controlled by sole entities. For obvious reasons, Balancer holders and Aura Holders are largely the same groups. Not leveraging the full power of Aura’s veBAL makes governance much less predictable and at risk of being manipulated.
Keep in mind this is strictly for governance proposals as Aura has its own Hidden Hand bribe marketplace.
I find this statement slightly offensive, as I am a long-time holder of $dxTETU, which at the time of writing controls 536,060 veBAL, and I am in no way being “controlled” by any third party.
I would much rather we extend the current Aura system to Tetu. We could then ask StakeDAO to implement the same, and any future wrappers as well.
In fact, when Tetu was discussing this delegation, we also flagged this issue:
In other words - I was happy to punt on it then, but if it’s an issue, let’s respond by fixing Tetu, instead of giving any one actor more unilateral power over Balancer.
You can find it offensive, but that doesn’t mean it isn’t true.
I would say fixing the issue of decentralization with Tetu is important as one holder controls the majority of supply (Humpty). With that being said, the above proposal is still a step in the right direction for both AURA/BAL as it decentralizes the protocols by maximizing veBAL votes on governance proposals. Extending the current system does not solve the issue, this proposal does.
Humpy controls the majority of the tetuBAL supply; he does not hold much $dxTETU and as per the specifications of tetuBAL, he does not control governance powers for his veBAL while they are in the LP.
As I said before, the delegation to Andrea was simply a quick fix. Personally, I was under the impression that Balancer even preferred it that way. If they do not, let’s remove that delegation and implement the same Snapshot Strategy that Aura is currently using.
This proposal isn’t to discuss Tetu, it’s to make vlAura voting more efficient.
Personally, I don’t think that Aura and Tetu strategies have to match and I think that is an entirely different governance discussion. I’d recommend making another forum post in governance if you want to go into detail on Tetu for the time being.
I would also like Tetu to move a vote of locked dxTETU holders to make decisions around the future of Balancer governance and the kind of ecosystem we are building together upon. It seems like their interests are most aligned with the long term success of Tetu and the veBAL ecosystem Tetu is building upon.
500k votes and 2.5m votes being “centralized” are very different topics.
Saying that “this isn’t about Tetu” isn’t fair either. The changes that Aura has made were very directly in response to TetuBAL, and this is the first chance that we are getting to discuss them, since I was not going to wade in on Aura’s internal governance discussions.
Perhaps if we can agree on this, then we can move forward with my suggested course of action rather than what is already underway in this proposal. I understand that this might be inconvenient for Aura as they have already voted on this change, but ultimately it might be in the best interests of the Balancer protocol.
Aura has the freedom to change its governance practices as it wish and this change has been decided. This vote is a simple formality to ensure that there is no double voting. I added a quorum increase and an extension of the voting period, both of which make governance stronger, upon requests of various parties and with no contention. This particular issue needs to go to vote this week, and AURA will vote using the new format as determined by its governance starting next week.
It makes tons of sense to figure out some additional governance around frameworks for wrappers and how to protect the integrity of Balancer Governance and veBAL economics as more parties start to build in this space. This is probably an ongoing conversation, but I don’t think this particular forum post is the right place to have it. Feel free to DM me on discord (I’m in the Tetu one) or start a discussion thread somewhere here.
Tetu’s Balancer Snapshot delegation makes it part of a larger voting bloc, that is larger than both auraBAL and tetuBAL put together.
Aura built off of the Convex playbook. The passthrough voting was an experimental deviation.
Governance has turned out to be more contentious and concentrated than expected. I suggest we study how Convex handled this, and when they changed from winner takes all to splitting their entire veCRV stash based on the ratio of their internal snapshot in thinking about any framework we might put in place.
I very much support working in the future to develop governance standards for all veBAL wrappers, but in the case of Balancer Snapshot Governance which shapes the direction and future of the protocol, I personally don’t think the passthrough system where 1 vlXtoken votes with it’s portion of veBAL that xDAO controls is the right thing for the ecosystem at the present time.
