Synopsis
Invest $75,000 in the initial cycle of the Bananapus juicebox. Bananapus is an offshoot community from JuiceboxDAO's contract devs to experiment with more high-risk token ops that could make Juicebox projects harder to kill and easier to grow. Its main focus is trying to implement a particular integrative solution to cross-chain Juicebox-based organizations, onchain governance based on staked positions, and token distribution strategies.
Motivation
The next big problems JuiceboxDAO needs to solve are:
1) replacing multisig ownership with a great onchain governance UX.
2) becoming accessible on L2 EVMs.
3) distributing JBX in a manor which incentivizes network propagation.
These are all high risk for JBX to experiment directly with. JuiceboxDAO has spent the better part of its past year's budget on risk management -- contract dev, testing, versioning, fixing, UX for it all. See here. In an effort to cut down on risk management expenses going forward, the DAO should instead fund and take part in another community that'll attempt to implement solutions to these delicate needs. Bananapus will take on those risks and capitalize on opportunities that may come from being a first mover, with the thesis that these needs might be best solved in a single mechanism that can be shipped in scoped chunks. JuiceboxDAO can later adopt what it likes into JBX if proven safe and useful.
Bananapus will be initially governed onchain through the NANA ERC-20 token and a Governor contract. It will then:
1. build a Juicebox treasury that uses a NANA payment terminal, issuing artful NFT gov tokens as the 721Delegate-like NANA-denominated tiers -- essentially staking and unstaking: ERC-20 paid in, NFT minted out. NFT redeemed, ERC-20 reclaimed -- and move to NFT governance. See here for ideas of how this concept may be exposed to users in web clients.
2. create a component that distributes a reserved rate allocation (or any inbound coin) to be claimable by holders of a subset of 721Delegate-like tier holders. These claimed rewards can be vested, and fortified if the recipient unstakes before a certain future date.
3. use these components together to create a way for project token stakers on L1 to not only receive distributions from other L1 projects, but also those of projects running on L2 instances of the Juicebox protocol, and vice versa. There are a few implementation ideas in discussion.
These are all high risk and somewhat pie-in-the-sky, but I think they can work. Each component can be of generalizable high use for projects.
I think JuiceboxDAO should be steady in how it relates with this experiment:
1. invest initially because the people involved -- jango, drgorilla, 0xba5ed, viraz, hopefully others soon -- have proven trustworthy and the treasury design is aligned with JBX interests. In other words, scoop some NANA and consider participating in Bananapus governance of its juicebox. Fund development inso doing. Invest more over time if it's working.
2. acknowledge that Bananapus has potential to be a DAO that manages a meaningful responsibility if it figures out how to effectively deploy the Juicebox protocol on other EVMs. Acknowledge that this may not reflect directly on JBX unless JBX continues being a useful community to create a vibrant ecosystem, in which case NANA will always have incentive and desire to route NANA to be claimed by JBX stakers, else it'll be forked and outcompeted.
3. most likely Bananapus will find some issue along the way and spend too much time debugging or re-architecting, in which case JuiceboxDAO might not want to continue committing its resources to this particular experiment. Now that the protocol's versioning seems steady, it makes sense to reduce ongoing payouts to developers who attempt to build solutions to the next big problems directly onto JBX, and instead fund offshoots that offload most risk for others to take.
4. If Bananapus succeeds in proving out its thesis, JBX can choose to either a) continue aligning with NANA and promote its growth, or b) fork the playbook and compete with NANA directly.
The Bananapus treasury will run in 6 week funding cycles, with 0 payouts, 0 discount rate, 100% redemption rate, a 7-day ballot, and a 30% reserved rate => 10% to bananapus.eth (the onchain Governor), and 20% to dao.jbx.eth. We will try to get to a point where these reserved chunks can soon be claimable directly by NANA and JBX stakers. The treasury will issue 1m NANA / ETH.
Our goal will be to deploy on an L2 with a 1st Cycle synced with mainnet Bananapus's 2nd Cycle in 6 weeks time. The syncing isn’t necessary, but it makes operations cognitively simpler for governing token holders.
Bananapus's 1st L2 Cycle will be similar: synced 6 week cycles. 0 payouts, 0 discount rate, 100% redemption rate, 7-day ballot, 30% reserved => 10% to the L2 Bananapus Governor, 10% to L1 Bananapus Governor, and 10% to ecosystem DAOs of the L2 being used. The treasury will issue 1m opNANA[polyNANA,zkNANA] / ETH. It will use a 5% protocol fee. These numbers are all subject to change.
Our goal by the end of mainnet Cycle 2 will be to have moved governance from the NANA ERC-20 to staked NANA NFTs using a new 721StakingDelegate component that's very similar to the now-familiar 721TieredDelegate.
Bananapus will be highly experimental and will have no guarantees.
I'm also excited for Bananapus to build collaborations with Peel, Nance and WAGMI once we get some bits of tooling off the ground.
Specification
Send $75,000 of ETH to the Bananapus treasury, to be deployed by Saturday, April 15 2023. This will mint NANA for JuiceboxDAO both as a financial contributor and reserved rate recipient.
Give JuiceboxDAO the power to exercise redemption of its NANA at any time without a new proposal if and only if there is a proposal to NANA governance that proposes a payout from the Bananapus treasury that suggests a takeover. The redemption value of JuiceboxDAO’s NANA should be higher than its initial $75k investment if this is the case. See below for the math.
Also send 20 million JBX to bananapus.eth if JuiceboxDAO ever chooses to adopt a NANA-based component for JBX.
_____________
[Math]
Assuming 30% reserved rate, with 20% going to dao.jbx.eth, 1m NANA / ETH issuance rate, 50% governance quorum.
action 1:
~40 ETH paid by dao.jbx.eth => 40m NANA issued, 28m NANA to dao.jbx.eth, 12m reserved which includes 8m to dao.jbx.eth.
resulting state 1:
dao.jbx.eth now has 36m NANA of a total 40m NANA.
**action 2:**Imagine an exploiter (exploit.eth) trying to capture the Bananapus treasury by going after over 50% of tokens.
X ETH paid by exploit.eth => Xm NANA issued, 0.7Xm NANA to exploit.eth, 0.3Xm reserved which includes 0.2Xm to dao.jbx.eth. Where X guarantees 50.01% of circulating NANA to exploit.eth.
resulting state 2:
Total NANA supply of (40+X)m.
dao.jbx.eth now has (36+0.2X)m NANA.
exploit.eth now has 0.7Xm NANA. Set Y = 0.7Xm.
solve for Y, where Y = 50.01% of (40+X)m:
Y = 0.5001(40 + X)m
Y = 20.004m + 0.5001Xm
replace Y and solve for X:
0.7Xm = 20.004m + 0.5001Xm
0.7X = 20.004 + 0.5001X
0.1999X = 20.004
X = ~100.07
Given the treasuries new balance of 140.07 ETH and a 100% redemption rate, dao.jbx.eth could redeem its 56.014m NANA out of total 140.07m NANA (~40%) for 56 ETH.
A 7 day reconfiguration delay would give the DAO 7 days to execute this retrieval of funds before the NANA treasury can be rugged.
Rationale
Easy and transparent.
Risks
Bananapus will likely not work. At most the DAO is risking $75,000 of ETH.
Timeline
Next funding cycle