DXD Voting power - Guilds & 2.0

dxgov 2.0

DXgov Quick Summary

The priorities of DXgov so far:

The team at DXgov has been focusing on guilds for SWAPR governance as well as major groundwork in preparation for governance 2.0. Since we already needed a new application to replace alchemy as well as an interface for guilds we took this opportunity to build an application which could handle both systems, saving us resources as well as turning into a product to bring revenue to DXdao. The product (Project DAVI) is designed as a global governance app for all DXdao governance solutions with a focus on user experience and accessible autonomy.

When 2.0? Everything done until now

So Wen Gov 2.0? I think it’s important not to understate the scale that the governance 2.0 transition is going to take. Completely changing the governance logic for our DAO which all of us depend on is no small task. You can read more about the reasons behind the Governance 2.0 development here

Up until now, we have rewritten schemes and permissions to be far simpler, more flexible and more secure. The next step which we have an audit scheduled for in September will complete all work prior to the actual voting machine logic change, a major step towards 2.0.

We will deploy a new version of DXdao in parallel alongside the existing one (rolling deployments goerli→Gnosis Chain→Arbitrum→Mainnet) so that we can begin a slow transition of funds and operations over. Once this process is complete and the new voting machine is developed and audited, we can upgrade the voting machine to begin using 2.0.

However, this does NOT mean the process is near complete. The transition over to the new architecture will take time, especially on mainnet. And the actual work to write the new voting machine will take not only developer resources but also a lot of oversight and detailed technical discussion and specification, not to mention actually switching to that governance model.

So TLDR, when 2.0? Likely early-mid 2023 is a realistic estimate given all the work above and assuming no major blockers.

Does the community want a DXD Guild?

Guilds are being built to deliver governance to DXdao’s spinout products such as Swapr, with this effort being prioritized due to the potential impact on product fee generation. Guilds could also be used to provide DXD holders a voice…

The interesting part comes when we consider the guild being given some reputation in DXdao. Since the guild has autonomy it can exercise this reputation and vote in DXdao proposals. This would allow DXD holders to sway votes and most importantly voice their opinions in public view of DXdao governance (visible as dxd.guilds.eth).

To be clear this is not a replacement for Governance 2.0 and that is still the goal, but this would be short term autonomy and a voice long before 2.0 arrives in production.

This idea of having a DXD guild vote in DXdao governance is not without its challenges.

Having reputation and technically being able to use it is not the same as having accessible autonomy, a key feature we have been building for Project DAVI.

Creating a proposal to vote on a proposal especially considering time constraints is no simple task. We have an aim of any DAO being able to vote in any other Project DAVI DAO. Without going into details, the application would mirror proposals and then put them on chain dynamically. The tricky part here is timing and it isn’t likely something that would work with short fast proposals, but long important proposals would be absolutely possible.

So some resources would be needed to develop a nice user experience for this in DAVI. This work WILL be done at some point, the question is when. Is there enough demand for this that we can take some time in DXgov to develop it before 2.0 and potentially add a small delay to the delivery of 2.0 or would the DXD community rather wait until 2.0 in 2023 to get any voice in governance?

I do think it is important to highlight that the resources for this feature are purely frontend, no contract work is needed. Whereas gov 2.0 is far more involved on the contract and coordination side of resources.

How we would handle a DXD guild

I would propose a number of something like 4% since this cap is in place for any given address in DXdao.

To protect from a small number of DXD holders controlling this 4% I would suggest a gradual increase with the goal of 4% REP in the guild once there is a decentralised distribution of members with a relatively high quorum.

This should help prevent instances of a couple of DXD holders having large amounts of REP or placing a low price on a large chunk of REP. This is of course something gov 2.0 has measures in place for but we have to manually prevent that here. We should even place some guidelines around slashing REP if guild voting power distribution started getting too centralised. And of course, this solution is still not perfect, e.g. 70-30 vote still disregards the 30% voting against winning action when viewed from DXdao governance perspective. This is something that longer-term gov 2.0 addresses.

Example of what this DXD guild voting behaviour would look like:

  • DXD Guild is created, and it controls an avatar like its own sovereign DAO
  • DXdao rewards the avatar with 4% rep through a contributor proposal (gradually)
  • DXvote proposals showing up in DXD guild, and DXD holders voting
  • When DXD guild concludes a vote on a proposal, it shows up in DXvote as a 4% Rep Yes or No vote on the DXvote proposal

While DXgov’s primary goal is to deliver best in class governance to DXdao and its products, there is also potential that DXgov’s work will generate additional traction and revenue in the future. Balancing this has been tough but for the most part, they are symbiotic, helping each other grow. By solving DXdao’s problems with which we have first hand experience, we believe DXgov tools will become some of the best “real DAO” solutions in the space.

Dynamic roadmap of how DXgov is evolving

It’s over to you now, we can see the value in this from both a DXD holder and Project DAVI product perspective but if it’s decided that having a 4% say in proposals is not going to be exercised then we can postpone this feature of DAVI and the DXD guild as a whole. Please let us know what you think.

13 Likes

Completely support this initiative, but think 10% is probably a better figure. With Gov 2.0, the effective influence will be even higher, so I think 10% allows a better transition period from 0 influence to the influence Gov 2.0 brings.

3 Likes

I think the reason why we said 4% is cause it is the maximum to be issued to a single contributor. And yes 10% sounds like a way more reasonable number of rep to be issued to DXD holders to be used in the DXD guild.

3 Likes

Is it a good idea if we give 10 to 12% but split across 3 different DXD holders (as delegates) giving a 4% REP each. It kinda creates diversity and create more engagement from the DXD holders.

If the guild votes 60% in favour and 40% against - does that result in 10% REP (assuming the guild has 10% REP) voting in favour for the DAO proposal? Or 6% in favour and 4% against?

How are proposals setup in Guilds? I guess the voting time for Guilds has to be less than the actual proposal - so that the Guild can still vote on the proposal?. So if I submit a proposal on Gnosis Chain (where I believe Guilds will be deployed first?) my proposal takes 1 day to boost and then has 4 voting days. So is every proposal mirrored in the Guild? I say mirrored but it would actually have to be re-submitted with an on-chain transaction - right? So perhaps it makes sense to only post them in the Guild after they boost? Will every single proposal also be in the Guild (doubling governance overhead?) or just specific ones? If only certain proposals, who decides?

Is there a minimum quorum for DXD guild? Or if no one were to vote, would 0.0001 DXD be sufficient to post a 4% or even 10% rep vote – assuming no one else votes in the guild?

2 Likes

All good questions

If the guild votes 60% in favour and 40% against - does that result in 10% REP (assuming the guild has 10% REP) voting in favour for the DAO proposal? Or 6% in favour and 4% against?

No, as highlighted in the original post this is the downside of voting like this, it will represent the group as a whole. This is what Gov 2.0 addresses and is obviously a better solution, guilds for this are not perfect.

How are proposals setup in Guilds?

Some of the details here still need ironed out in terms of UX. Guilds will likely first be deployed on Gnosis chain but there is no reason mainnet could not be done shortly after or even at the same time if it had demand.

The UI to make mirroring smooth is the part that needs development so I wont go over it too much here. But yes, essentially the UI would choose the best time to request/automate an on chain mirror and until that point it would stay as signatures off chain. Every single proposal would be in guilds that work with the time limits but only ones with attention would of course be voted on and put on chain.

Anyone could of course manually make a proposal to vote in DXdao before we implement the mirroring feature, it would just require more overhead of working out the best time to vote.

Is there a minimum quorum for DXD guild?

Absolutely, all guilds require quorum due to the lack of holographic consensus, the number has yet to be decided and suggestions are welcome.

but think 10% is probably a better figure

I agree, although I do also still think this number should slowly ramp up just to be safe as the DXD guild grows, we don’t know what sort of activity it will get and one person with 10% would be bad. The 4% number was fairly arbitrary as it is the current rule for one address although a new rule makes sense to represent a collection of people.

2 Likes

only 4% or 10% seems really low?

why not 20%?

I think this is better seen as a short term compromise. In theory maybe 50% makes sense but that large influence is what gov 2.0 is meant to address.
The issue with this solution is that it lacks the checks and balances of gov 2.0 which would make 50% very dangerous.
The guild would represent DXD holders as a whole even if they were divided but with the power of the whole, this is what dlabs highlighted above.
But yes, 4% was certainly low, and I actually think if we saw high usage of the guild’s power then a higher amount could be justified over 10%.

5 Likes

Also just a general comment to remind everyone that REP is inflationary. So if you mine someone 4% REP and they don’t earn regular REP - it will reduce overtime.

Moreover, if you mint 50% REP - after minting it’ll only be 33% REP in the DAO.

3 Likes

Having had some more time to think on this topic, and with more recent happenings in the DAO, I think a relatively high REP allocation to the DXD guild is appropriate given the importance of upcoming decisions.

I agree with @EvanVanNess 's initial pre-edit sentiment that 50% seems a logical number in principle, but also recognize the issues that presents in itself.

I’m drawn to a 33% allocation (for clarity, this is 33% post-dilution, so whatever amount is required to end up at 33%), as it gives DXD the significant influence that it deserves in these decisions (a meaningless, token-of-appreciation allocation that doesn’t actually give DXD a significant influence would signal a failure to recognize the issues presented in recent weeks), but meanwhile doesn’t give DXD holders the power to pass proposals by themselves against the interests of the DAO, or REP holders - which is the only real argument I can fathom against even a 50% allocation (and a fair argument, by the way!).

If there is disagreement on this kind of range, based upon DXD holders being given too much power to act against others’ best interests, I’d invite challenges to consider the ‘flip side’ alternative of a lower allocation allowing actions against DXD holders’ best interests in a similar manner.

A 33% allocation still presents a huge hurdle to pass anything detrimental, while recognizing the issues being discussed recently surrounding the importance of DXD, which I think and hope are more front-of-mind now amongst REP holders.

3 Likes

yeah, 50% was somewhat hyperbole to make the ethical point that DXD holders deserve to be full partners. The DAO’s treasury comes from DXD holders, after all.

the fact that it’s bloc voting means it probably should be in the 20-30% range.

2 Likes

I think the Governance 2.0 model is a good starting point for this conversation. The WG wrestled with a lot of these issues on how to incorporate DXD into governance (a DXD guild was even considered). In the agreed-upon model, DXD/REP voting power is split 50/50 albeit with some voting boost/decays. Technical development of Gov 2.0 has taken longer than expected, and the proposed DXD Guild would be a stopgap fix to get DXD holders’ opinion into governance as soon as possible, which I think is extremely important.

My concern with a number like 33% REP to the DXD Guild is that according to @ross that REP could not be split up to reflect the breakdown of DXD holders. If a vote in the DXD guild is 51-49%, the Guild would still vote with all 33%.

And unlike the Governance 2.0 model, which capped the amount of DXD an address could vote to 1,500 (plus required the address to have REP), the DXD guild would not have a cap on the amount of DXD, so this would essentially mean 1-2 large DXD holders could vote with 33% REP.

The other thing that Gov 2.0 had is it gave more voting power to DXD holders that lock up their DXD into the future (similar to the veCRV model). I think something like this could be done in Guilds? So, you’d stake your DXD for 6 months or something to get voting power? I also wonder if there is a way to include DXD Swapr LP’s in governance (Caney Fork is a Swapr DXD/ETH LP).

2 Likes

This is a characteristic of the holographic consensus model used by REP also in that a 52-48 vote results in a full passing of a proposal, not ‘52% of funding is approved’ or similar; but actually in this case has the additional check and measure that it’s only a minority vote within the wider REP voting system. So, even if 1-2 large DXD holders exercised a majority vote (and so they should be able to, if their relative stake in the DAO is sufficiently large and others have had ample opportunity to vote and not taken it, signalling their own neutrality), then REP holders still have more than enough veto power. Sure, a relative split would arguably work better, but I think we should be mindful of @KeenanL 's recent alternative restructuring proposal, and take a moment to assess whether we’re seriously addressing the first point made after the preface:

It’s clear that these big decisions related to the future of the DAO are coming up fast, and I think giving DXD a relatively small influence in those key decisions that is effectively checking a box in name only would be unjust, after everyone seems unanimous in the messaging that DXD’s influence in these decisions has been zero for far too long, and that a legitimate push to address that is needed.

I’d also be mindful and cautious about adding further technical scope to Guilds, when as above, these upcoming pivotal conversations are imminent. If it’s part of the current implementation, then great, but I think pushing Guilds out further for a nice-to-have would be a mistake. The vast majority of DXD held has not changed hands for long periods of time, so the implication is that a large percentage of DXD is held by long-term holders anyway in a backward looking sense, so is a fair proxy for long-term holding in general. Again, forward looking is more important, of course; but just circling back to the ‘the solution doesn’t have to be perfect if it’s an improvement’ idea, especially for something interim.

1 Like

Agreed with some of the concerns here from the two responses above.
The couple of large holders controlling the allocated REP was an issue highlighted originally and the solution was to have some manual limits in place to control this so that we would require a certain number of members before minting higher amounts of REP. Its not a great solution but open to additional ideas that don’t require new contracts.

On Chris’s idea with locking up DXD similar to gov 2.0, we actually require this as things stand now. We will eventually upgrade the DXD token to be able to do snapshot functionality without locking, but as of now the technical limitation of the token contract means we have to lock up DXD for the guild.
What we get to decide is for how long? Given that this is a short term solution I would suggest it be for a short period to avoid the case of people wanting to move to gov 2.0 but not being able to because there DXD is locked in a guild.
The other issue with a long locking time is that it could turn people away and further cause more centralisation.

And just an update on things.
I think its certainly obvious there is demand for a DXD guild so we will be going ahead on that front, the only remaining issue is of how much REP is minted to the guild. This will of course go through DXdao governance so will be decided separately to deploying the guild.

We highlighted that some development would be needed for a mirroring interface to make this process run smoothly and that is still the long term plan and remains a core and important feature of DAVI. But we can also support the DXD guild voting autonomously upon launch, governors would have to manually make a proposal to vote but the UI for this will be simple with how DAVI is architected.

4 Likes

Glad to hear @ross that these features are largely already a part of Guilds from the get-go. Would be curious to hear your own thoughts on what % REP is the right move?

It’ll shock no one to hear, but I’d agree with @hughesconnor and @EvanVanNess that the percentage needs to be sufficiently high. Doubly so given the important decisions being made at the moment without our input. 33% seems sufficiently high for a starting point, yet sufficiently low given the concerns already highlighted.

That one, two or any number of DXD holders would have significant say in DAO governance, that’s a factor of token voting which both guilds and gov 2.0 will introduce in some form or another. We’ve been using non-transferable REP only voting, and while it has some benefits over token voting; it’s got some pretty big drawbacks that have led to the current situation.

The reality is that due to two years with poor value model and lack of governance over the past (plus significant agent-principle risk from REP holders), DXD has grown more centralized as many holders have thrown in the towel and exited. And few new entrants have looked at DXD and its drawbacks and have considered it worth buying even well below book.

The way to address that is to give the token significant value and meaningful governance rights; which gives new holders a reason not be deterred by past poor performance (both market and DAO performance).

Water things down now with a too small a REP % and you’ll have more of our already limited of holders decide to exit, and end up with a far more centralized DXD base when Gov 2.0 is ready to implement.

Let’s not ignore the fact that if anyone is concerned DXD is currently too decentralized, they’re perfectly able to buy DXD way below NAV value now. They’ll increase decentralisation of ownership, improve market price and increase their own say in governance.

2 Likes

Wondering if there was a specific governance decision to build guilds, if yes, can you please refer me to it?

I understand this is the way to manage spin out products, but I was under the impression the Gov 2.0 will be the basis DXdao and spinout products.

My opinion is that we should focus only on delivering the results of the working group and governance decisions below, anything else is a distraction.

Gov 2.0 will likely take another 6-12 months from this point. Waiting that long to address the imbalance of influence and power between REP and DXD holders would be a failure of the DAO to take recent issues raised seriously. In any case; as above, guilds is practically ready now, so the distraction you speak of is already a sunk cost.

2 Likes

I’m sorry that things take a long time to develop in general, and with additional layers of difficulty when developing smart contracts on solidity.

Rushing side initiatives and half baked ideas is what got us here in the first place.

If the task is not detrimental to the mid/long term goals of DXdao, we should cease to pursue it.

You’re missing the point and strawmanning - Gov 2.0 is clearly a priority and the best solution we have, nobody is arguing against that. But Guilds is essentially completed already, and a clear improvement to the status quo. The work required from this point to implement is minimal (I get the impression it’s essentially deciding a REP %age and then deploying) vs. the benefits. To call for it to be ceased at the final hurdle just as DXD is set to gain influence, after it has been mentioned countless times on governance, product strategy, and other calls over the course of 2022 (and even earlier - the first mention I can find was in a post dated August 2020, 25 months ago, which you participated in), seems odd.

I think calling Guilds half-baked is pretty disrespectful to all the work that’s gone into it; it’s a fair, useful, and necessary interim solution between the status quo and Gov 2.0, adapted from development work we are advertising as a product to other DAOs and organizations - I find it odd to call a version of our own external facing products ‘half baked’. It’s also like asking why did we release Swapr v1 if we knew Swapr v2 would be better? It’s a vast improvement over the current situation, and Gov 2.0 is another 6-12 months away as outlined in the original post; in which time, if the status quo persists, more DXD holders leave out of frustration. This outcome ‘doesn’t align with DXdao’s long term goals’ of being a positively valued organization, nor having a decentralized holder base. Much of these issues come from the inability to give DXD holders a voice in the past.

To reiterate @KeenanL 's latest post… The culture of saying no to any solution that isn’t perfect is a real issue here, and I hope something that will be addressed rather than just paid lip service.

The task looks to be 1) 95+% completed, and 2) extremely beneficial to DXdao’s long term goals, in that it finally gives power and influence to DXD holders, and does something to stop the decay of the organization’s value as determined by the market, and also stop the centralization of its holder base.

Another 6-12 month delay to giving DXD holders any rights whatsoever is the absolute worst possible (in)action we could take…

3 Likes

My personal view on the amount of REP being minted to the DXD guild is that something in the range of 20-35% makes a lot of sense. If we look at a certain recent proposal on gnosis chain with high turnout we can see what DXD guild’s impact would be (bearing in mind that minting something like 20-30% would also dilute current REP holdings). Regardless of how the DXD holders would have voted here it is clear to see that we have a lot of active rep and to make an impact as stakeholders the REP amount should be at least 10%.

However, in saying this I believe the best way for us to go about minting that rep is to gradually propose increasing it. DXgov will deploy this guild with the settings best suited for this type of mirror proposal voting and then start the guild off with a proposal to mint 4% REP.

Initially, we will do this on gnosis chain since it is where governance is most active, which has its own issues as the speed of proposals happening is also far faster there. We have never done this before and so a lot will be learnt, we will also write a short guide on how this way of voting can best be used. For example, if a proposal is passing via boosting and the guild would not have time to vote on it then governors could downstake the vote to draw out a larger discussion on the proposal.

It would be good to get feedback on mainnet deployment demand as the proposal time makes more sense with mirrored proposals but the initial DXD guild will require locking of DXD and gas costs should be considered.
With the token working group also ongoing it will be important to decide how we feel about locking up tokens for power. With upgrades planned we would have the option of adding snapshot functionality directly to the DXD token, meaning no locking for the guild.

After we mint the initial rep and ensure everything is working as expected it is over the guild and what they want to propose. If it wanted to try minting 30% in one chunk then it could, however, I would advise a slow increase from the new guild and propose member and decentralisation goals also.

I think it would also make sense to include in each REP mint that this REP amount will be revisited or outright burned upon Gov 2.0 to not interfere with the balance we struck previously for 2.0.

2 Likes