Announcement on LQ Staking Rewards and Aquafarmers

Dear community,

Due to technical and time constraints, we've run into a regrettable issue regarding correctly and conclusively establishing ownership of Aquafarmer NFTs for the purpose of calculating staking reward boosts when multi-address wallets are used. So as not to delay the release of staking rewards any further, we're going to release unboosted LQ rewards soon and ask the community to guide us towards a solution for paying out the bonus rewards for Aquafarmer holders.

To receive the Aquafarmer boost in the future, you must switch to a single-address wallet setup by 2023-06-06.

Read on for a run down of the situation and our thoughts on a plan to make things right.

Multi Address Wallets and User Identity

Multi-address wallets (such as Eternl in mulitaddres mode) on Cardano were intended to enchance privacy, following in the footsteps of other blockchains. These wallets are also known as "Hierarchy for Deterministic" or "HD" wallets, and were originally introduced in BIP-32 for Bitcoin. They generate new payment public keys and corresponding addresses for each transaction.

Unfortunately, the privacy enhancements work too well.

The Liqwid protocol was originally designed to associate user identity via the payment public key, but with multi-address wallets achieving precise identification has become challenging. We encountered a similar issue with the LQ market participation rewards, where users of multi-address wallets received multiple smaller reward distributions that Liqwid couldn't combine automatically. Without creating a database associating different public keys to the same user -- which erodes privacy and requires high-friction user interaction -- our off-chain algorithms are unable to recognize that different public keys belong to the same user.

How this affects LQ Staking Rewards

Now, we face the same issue with associating aquafarmers with LQ stakes. LQ stakes are UTxOs that carry a hash of the owner’s public key in their datum. With single address wallets, we're guaranteed that we can look up the UTxOs at the address listed and determine whether the wallet holds an Aquafarmer. With multi-address wallets, the Aquafarmer and the LQ stake are usually associated with different keys that, to our algorithms, seem unrelated.

Unfortunately, we haven't found a conclusive and foolproof way to deliver the user experience we originally intended: automatically applying the high-applicable Aquafarmer boost to all stakes that a user owns, while not allowing the boost to be applied to other user's stakes without transferring ownership of the NFT.

What we can do

Consequently, we have a few challenges ahead.

First, we're going to ask users with Aquafarmers to switch to single address wallets. After 2023-06-06, we'll start calculating and rewarding the boosts for aquafarmers only if they're associated to the same address as listed on a stake. This will require unstaking all of the LQ on a multi-address wallet and restaking on the single address wallet, and is the simplest way to start paying out boosts.

Second, we need the help of the community to figure out how to make things right. Although LQ staking went live on April 3rd, we won't be able to fairly distribute boosts for Aquafarmers until 2023-06-06. Moreover, as users switch to single address wallets, their "staking clocks" that determine eligibility for long-term staking APR bonuses will be reset.

Our initial thought is to introduce a temperature check to double the Aquafarmer boost for 44 days (equal to the number of days from April 23rd — when proposal 0 took effect, paying out rewards for short-term staking — to June 6th), and to automatically fast-forward the clock on stakes moved during this period. However, there are obvious policy and technical details that need refinement. For example, how should we handle Aquafarmer holders who already use single-address wallets? How do we re-write our calculations to fast-forward the staking clocks? We believe that arriving at a solution should involve community consensus and a governance vote.