Should the DAO limit the max VP accounted in a vote from an address?

Whale voting has been a hot topic of discussion for a long time. The conversation happens privately between community members and in public forums such as the #whales-vote channel at DAO Discord and comments on any VP-related governance proposal.

There are two major concerns about limiting VP in a vote:

Preventing sybil attacks such as splitting assets into many accounts to remove the limitations. A roll-out strategy that enables the DAO and members to experiment, learn, and iterate. Since the VP strategy is a cornerstone of the DAO, changes to it should be gradual. Of all the alternatives I heard about, this is the one that I have more confidence in as a first step toward limiting whale power:

Limit the maximum VP accounted from an address to XX% of the threshold of a vote. XX% should be 50% for grant proposals and 80% for any other proposal type.

For example:

POI, Catalysts, Names Bans and Polls [500k VP threshold] => 400k VP Limit. Draft Proposals [1M VP threshold] => 800k VP Limit. Linked Wearables [4M VP threshold] => 3.2M VP Limit. 20k Grant [2M VP threshold] => 1M VP Limit. 80k Grant [4.4M VP threshold] => 2.2M VP Limit. Governance has a 6M threshold => 4.8M VP Limit. The change will:

Set up a simple framework to limit the VP threshold on each proposal type. Level the ground between small and big holders on low threshold votes. Strong impact on Grant proposals, requiring at least two whales to approve the funding. Low impact on governance proposals (that can approve/update these limits). Sybil Attacks The proposed VP limit scheme doesn't prevent an individual to split their assets into many accounts to overcome such limits. However, the risk is mitigated because the splitting operation is operationally expensive for MANA and LAND holders, and impossible for ESTATE and Delegated VP holders. If the DAO recognizes this behavior in the wild, a new proposal should be formulated to address the matter.

Roll Out Strategy Start now with a basic framework and test a strong limit on Grant Proposals since they are the most vulnerable to single-whale voting. Let's keep these systems working for a year and then target to lower the limits based on the results and learnings (e.g. 20% grant vp limit and 50% as a general vp limit).