Words of caution: We need to formalize the governance system

Proposals Protocol and Governance
#1

Hey all, I’m sure you agree that this past week has been incredibly exciting. It was particularly exciting for me, not just because of all the YFI (pronounced waifu) I farmed, but also because of the creation of the community governance system. I’ve been researching and building decentralized governance systems since late 2017 when I started building betoken.fund, a crypto hedge fund dApp 100% owned, governed, and managed by a decentralized & meritocratic community. I’ve always thought it’s a no-brainer to govern DeFi protocols through DAOs, so I’m stoked that it’s finally becoming mainstream.

However, it’s far less exciting for me to see yearn use a naive token-weighted voting system, as it is seriously flawed. I’ve written an extensive critique of it: Why Voting Tokens Are F**king Horrible, And 4 Ways to Fix Them, which is pretty long so here’s the TL;DR:

  • Token-based governance systems assume that the price of the token is strongly correlated with the success of the underlying protocol, which is just not true, due to market manipulation, complexity of the protocol, and the possibility to short-sell the governance token.
  • Flat organizations (which the yearn governance system currently is) are bad, because not having a formally defined power structure would only lead to the tyranny of structurelessness, where informal power structures form spontaneously from preexisting social relations between members, producing leaders whose power cannot be stripped and preventing talented people without “good friends” from obtaining power.
    • If this sounds far-fetched, just keep in mind what happened to the French revolution and the Russian revolution.

To make yearn a protocol that can last for decades (if not centuries), we’re in a unique position to thoughtfully design a governance system that does not explode in our faces. In my opinion, what we need to do is:

  • Formalize the governance process. It’s one of the first things that needs to happen, and I’m glad to see it’s already spontaneously happening with YIPs and RFCs, but there’s still a long way to go. For instance, there’s no protocol for debating a YIP, or for what a YIP needs to contain.
    • Proposals like voteLock are going in the right direction IMO
  • Establish an executive council. We need clear designations of power & responsibilities, and having a council of people working for yearn full time is a good idea. For instance, who should be in charge of the yearn Twitter account? Who should be the forum mods? Who should filter out malicious proposals? Who should have the right to manage the GitHub repos? We need to be able to answer questions like this unequivocally, and having an elected executive council is IMO the best way to do it.
  • Share fees/YFI with the council/the dev team/people analyzing proposals. To have a healthy ecosystem, we need to establish clear internal incentives for people to contribute to the protocol in areas we value. We should reward the dev team, reward those who provide valuable analysis of the pros and cons of proposals, and reward those who help the governance system running smoothly. Giving YFI rewards seem to make the most sense, since it will be imbued with governance power once YIP 10 passes. YIP 14 seems to be going in the right direction, but I think it makes more sense to reward contributors with not just money, but also governance power.

Would love some feedback & more ideas! There’s a lot of work to be done to build ourselves a solid governance system.

43 Likes
Too many cooks in the kitchen
Too many cooks in the kitchen
Too many cooks in the kitchen
#2

@zeframlou completely agree, especially about creating a platform to effective debate YIPs

We are creating a social governance platform called The Ether for projects like Yearn that is solely devoted to providing a place where protocol changes can be effectively debated and statistics on the debates are provided to all users so they are more informed when voting on YIPs.

The platform will be useful in helping users formulate opinions on how they should vote. We will be going live very shortly.

10 Likes
#3

Maybe one practical step forward would be to evaluate the existing DAO platforms: Aragon, DAOstack, Moloch, OpenLaw and Colony.

There are several interesting differences between then and their governance systems. The common ground is that all of them offer voting systems with blockchain type censorship resistance and transparency, and the opportunity for a community to manage assets together.

7 Likes
#4

Highly agree @zeframlou.

I think there are past examples which we can learn from
DXdao has been doing a tremendous job at structuring the governance system, worker guidelines, and the vision of the DAO, and this was all done pretty much fully decentralized.

Aligning voter incentives with the success of the protocol.
Example of necdao and DeversiFi: necdao is a dao that governs a portion of the exchange’s profits. Nec tokens are earned from trading action in the dex and they can be used for a trading fee discount. In addition you can stake (lock) the tokens to get voting power in the NecDAO, which holds a portion of Deversifi revenue.

In our case staking (locking) YFI tokens for a year aligns the interest of the voting power with the success of the protocol.

Happy to chat more about this.

3 Likes
#5

Establish an executive council. We need clear designations of power & responsibilities, and having a council of people working for yearn full time is a good idea. For instance, who should be in charge of the yearn Twitter account? Who should be the forum mods? Who should filter out malicious proposals? Who should have the right to manage the GitHub repos? We need to be able to answer questions like this unequivocally, and having an elected executive council is IMO the best way to do it.

This is a really good idea. We need to have some clarity on this. I would support a proposal to codify some of these executive roles.

2 Likes
#6

This topic deserves more attention. I too feel that it is important to establish an executive team to help keep this group focused on the right issues.

4 Likes
#7

However, it’s far less exciting for me to see yearn use a naive token-weighted voting system, as it is seriously flawed.

It sure does. Idea was\is to switch to something more reasonable.

Formalize the governance process.

We are doing our best to roll out understandable workflow to YIPs. It looks better but still far from ideal.

Establish an executive council.

I am doing my best to delegate as much stuff as I can without falling into “good friends” trap you described. This is one of the reasons I bump people trust level here, add other telegram modes I didn’t know before and ask people to participate in dev with PR on GitHub.

Share fees/YFI with the council/the dev team/people analyzing proposals.

Like this one on both sides. As an entry level, would be cool to have monetary reward. As a next step, would be great to be involved in governance process.

@zeframlou:

  • Do you think it would make sense to migrate the whole system to another working DAO platform? Which? Would you be willing to make research on that?
  • Is there any specific role you would like to have in the system? :slight_smile:

Thanks for valuable input.

7 Likes
#8

That sounds awesome! Debating has been done rather informally for most DAOs right now, it would be wonderful to have a platform offering a more clear & formalized process. Not much info on the website though, where can I learn more?

1 Like
#9

Yeah dxDAO and necDAO have been hugely successful, though to be honest I haven’t been following their forums. It was especially impressive for dxDAO to launch multiple well-developed products like omen and mesa.

Their success might be partially attributable to DAOstack, who have been spot on from the very beginning about the importance of voter attention & community management. If we were to move to an existing DAO platform I would prefer DAOstack over Aragon/Moloch honestly.

Feel free to ping me on Discord! My handle is @zeframL

1 Like
#10

You can check out our YouTube videos (especially drip #1) to get a better idea.

We will also be releasing more information in the next 48 hours in anticipation for our discussion at the Ethereum Magicians Summer Session at 9am EST on Wednesday.

I will make sure to post a link to that information here once it is released.

Finally, you can join our Discord

1 Like
#11

Can I ask what is your role in YFI? I see that you are a mod. How else are you involved? As developer? Just wondering.

#12

Voluntarily bootstrapping eco-system :sweat_smile:

8 Likes
#13

I think there’s no rush to migrate YFI governance to an established DAO framework, as currently things are not working too terribly. The framework is only the tech, the governance process itself is what’s important.

Establishing an executive council is the thing to do ASAP in my opinion, as having clear designations of responsibilities & power is important especially in the early stages of a protocol. The concrete next steps are:

  1. Decide on the roles of the council members (community mod, developer, governance wizard…), and specify the power & responsibilities of each role
  2. Decide on the process of choosing the council members
  3. Decide on the form that the council entity will take (off-chain & informal/Aragon DAO/DAOstack DAO/MolochDAO)

These look like three potential YFIs, though probably should write up an RFC first.

Among the DAO frameworks, I’d say DAOstack is the best if we want to use something out-of-box, and Aragon is the best if we want really customized stuff (which would of course require work).

I’d love to help design the YFI governance system, and I can do some smart contract dev if that’s needed.

8 Likes
#15

I don’t think we should rush to do this actually, almost all of the things you mentioned (i.e., who is a mod, who is in charge of twitter) can be voted on an decided in the current format. Now that YFI holders are the only holders there is no reason to create a seemingly small executive council that will merely concentrate power into a few people who can potentially exert influence over a larger majority by vote delegation. As for malicious proposals, that is why there is a quorum - uninteresting/bad proposals won’t meet quorum - and any that do with easily be voted off since now only YFI holders can vote. Intentionally voting for malicious proposals would be suicide, which is why YIP was implemented.

I see an executive council or the sort as a way to extract unnecessary fees from the protocol in the form of payments in order to “manage” the things you mentioned why can be easily managed in the current format. If it is deemed in the future it is unfeasible we can re-evaluate, but creating something right away is premature and likely will be an additional cost at the expense of YFI holders.

Vote delegation or the creation of a council will also enable an oligarchy to more easily form, which is against the whole point of decentralization.

5 Likes
#16

It might be sensible for YFI holders to vote to approve a “constitution” of sorts for YFI/yearn/The Project. This will set out an appropriate governance mechanism which include procedural and substantive safeguards for YFI holders. For example: i) ordinary matters that only require a simple majority vote; ii) super-majority matters requiring 75% vote; iii) a mechanism to vote in or remove an an executive council; iv) scope of the executive council’s powers, etc (it is a little more detailed/nuanced than that, but you get the idea).

The executive council is important because YFI holders’ votes mean nothing unless they are implemented. The fear that a small executive council might exert a disproportionate influence over the project is balanced by, for example, including the power for YFI holders to vote to remove the executive council.

To draw an analogy to a company (which is not entirely appropriate, I know), YFI holders are the shareholders who appoint the directors. The directors, in turn, run the company and make day-to-day management/execution decisions. If at any time a shareholder wants to remove their nominated director, there is a mechanism to do so in the company’s constitutional documents.

4 Likes
#17

Great idea let’s do it!

1 Like
#18

Copying over my response to someone on Discord with a similar opinion:

It’s not that we will create people with autocratic power, it’s that they will always emerge in any system, and where we can make a difference is whether we recognize those positions of power and formalize them, so that we can constrain their power and define ways the system owners (in our case YFI holders) can punish/oust them if they become malicious.

The response from @Neo is also pretty good, I like the analogy with shareholder-director relationships. Just because YFI holders are giving power to an executive council doesn’t mean that the council will become the de facto rulers of the protocol, as long as the governance power is held by YFI holders and the council members are elected by the YFI holders. YFI holders are the owners, and the council executes the will of the owners.

7 Likes
#19

How do we reorganize the council, how often are terms, will there be 6 month term voting? I think we should let the system run as it is now and evaluate that. We can always implement this in the near future.

#20

I like the idea of accessibility for all! Requiring an expensive amount or excessive amount of YFI to govern the space, could shift power dynamics to be unequal. My question is what safety measures are being proposed to make sure a counsil that is given fees doesn’t become the powering party of YFI? I’m conserned that a board with power over proposals and or other things might have too much sway.

#21

There would have to be a way to submit a no confidence vote and force board members out.

1 Like