DPI MetaGovernance Vote Delegation

IIP: [to be assigned]
Title: DPI MetaGovernance Vote Delegation
Status: Tabled [edit 5/7/21]
Author(s): @gregdocter
Discussions-to: [add if/when added to Gitbook]
Created: 05-05-2021

Simple Summary

‌Delegate DPI voting power from the INDEX multisig to an externally owned account (EOA) controlled by @gregdocter: 0xDA29C3ef220Cce74735E688CC20bDb8FfaD5fd0c


‌Today, DPI metagovernance votes are executed using the INDEX multisig.

This IIP proposes DPI metagovernance votes instead be executed by an EOA controlled by @gregdocter.

This IIP does not:

  • Propose any other changes to the metagovernance process as specified here.
  • Address Snapshot metagovernance (IIP-33).

This proposal merely addresses how the Index Coop executes votes that otherwise would be executed by the multisig.

The outcome of metagovernance votes would remain decided by INDEX holders.


‌Today, each metagovernance vote requires approximately 2 hours of work to coordinate and then execute a vote using the multisig. This work is done by Set team members @setoshi, @asoong, and @gregdocter.

Assuming a conservative 2 votes per week, this amounts to 208 hours (8.7 days) per year spent coordinating and executing metagovernance votes.

Other options, such as changing parameters of the multisig, were considered. However, they do not address the core issue seeking to be resolved here.


This proposed change is stop-gap that is part of a larger meta-governance vision.

As the number of meta-governance votes increases and as more meta-governance capabilities come online, there is an expectation that vote execution will be further decentralized within Index Coop.

This may ultimately take the form of a meta-governance council executing delegated meta-governance votes, protocol ambassadors executing delegated meta-governance votes for their protocols, or some other sort of structure. In the interim, this proposal seeks to eliminate unnecessary operational work while a greater meta-governance framework is being developed.


A simple solution is to delegate DPI voting power to an EOA to enable a community-trusted individual to execute metagovernance votes.

This will reduce time-spent from ~2 hours per vote to ~10 minutes by removing all coordination costs associated with using the multi-sig.

To implement this requires the following:

  1. Primary EOA (@gregdocter): 0xDA29C3ef220Cce74735E688CC20bDb8FfaD5fd0c
  2. Backup EOA (@Cedrick): 0x493F40C55c043E3D19Ff4De413EF27DE545279Eb
  3. INDEX multisig to delegate COMP and AAVE via the Governance module to the Primary EOA
  4. Gas: ETH sent to Primary EOA to cover gas costs
  5. @gregdocter executes metagovernance votes according to the current metagovernance process.
  6. Delegate other protocol voting power as reasonable.

This is intended to be a stopgap solution that lays the path to turning over this power to other trusted community member(s) focused on metagovernance.

The Backup EOA is being designated in case the Primary EOA is unavailable for some unforeseen reason.


  • FOR - ‌Delegate DPI voting power from the multisig to an externally owned account (EOA) controlled by @gregdocter
  • AGAINST - DO NOT delegate DPI voting power from the multisig to an externally owned account (EOA) controlled by @gregdocter


Copyright and related rights waived via CC0 .


Comically easy FOR. Almost surprised it requires a vote.


Very easily voting for, but ideally would like some notification in Discord or other location once a meta-governance vote is executed so folks can see/verify the outcome. Also verifying on Sybil would be nice.

1 Like

It seems to be a no-brainer operationally speaking, but what about the risks associated with going from multisig to EOA? it would be good to have an idea of the worst-case scenarios with potential impacts and mitigation measures.

For example, if @gregdocter or @cedrick is hacked and his account used to vote against the decision of the coop, is it possible to revert the vote?

1 Like

What’s the long term conversation like on this in regards to automating these metagovernance votes? I realize it may be advantageous to have @gregdocter do this now vs the current process, but what about a bot that could do this for us? What about adding this to our Snapshot strategy?

Mostly want to start a conversation about the long term version of this is it doesn’t already exist. If it does exist, can you point me there?

Also good to note that the sooner we can offload this from Greg’s hands and automate it, the better, what with the chance any of us could be hit by a bus tomorrow (not to get morbid :grimacing:) I assume there’s a mechanism to remove that control in such an event, right?


Great proposal! I think this takes us a step in the right direction towards reducing the operational load of metagovernance. My one concern here is with the security of doing this with an EOA. I think a multi-sig with several trusted community members would boost confidence quite a bit. Gnosis-safe already has a fantastic integration with WalletConnect, meaning that we can execute on-chain votes in just a few minutes time, without any of the security risks associated with using an EOA.

When it comes to on-chain governance, we really can’t be too careful, particularlly as our AUM grows, and we expand our metagovernance abilities.


love the feedback/pushback @trx314 and @ncitron, pulling this out to capture the main counterpoints to proposal :point_down:

Ostensibly, the implication of delegating to a multi-sig would be that coordination costs would not be reduced as dramatically.

This seems like a totally-okay tradeoff.

Next steps:

  • Let this specific proposal sit
  • Re-draft according to this ^ feedback to focus on delegating to a multisig

Sounds good @gregdocter!
In general, think it would be a good practice to have a “risks” section for this kind of proposal. My feeling is that a large part of voters do not always have all the technical skills to fully assess by themselves the tradeoffs in terms of security, trustless setup, etc.

1 Like

Excellent work @gregdocter anything we can do to streamline our processes while managing risks should be explored and enacted.