[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]  Snassy.icp [@SnassyIcp](/creator/twitter/SnassyIcp) on x 1059 followers Created: 2025-05-09 01:12:06 UTC It is time for a status report on the Sneed DAO Recursive Liquidity Loop. The good, the bad, and the ugly. TL;DR: It is finally time for Sneed DAO voters to receive rewards! First: What the heck is a "Recursive Liquidity Loop?" As old-timers know, before the SNS Sneed DAO, there was an old, pre-SNS Sneed token (it still exists, and if you have it you can still swap it 1:1 for the SNS token at the Sneed converter dApp,) and it pioneered what at the time were ground-breaking tokenomics: 1) A large portion of the supply was entered as liquidity into an ICPSwap SNEED/ICP pool. 2) The rewards from this pool were regularly withdrawn 3) The ICP rewards from the pool were fed into an X year ICP neuron 4) The SNEED rewards from the pool were burned 5) The ICP maturity from the 8y neuron was used to buy back SNEED from the pool and burn it. This was quite effective. Sneed had soon built up a XXX ICP neuron in this manner entirely from LP rewards, and the maturity from it was regularly used to buy back more SNEED and burn it. However, this process was entirely manual, and not trustless at all. So the number one priority and first major goal for the Sneed community became to make this setup work in a fully trustless way. Creating an SNS was the first step on this journey. Implementing the "Recursive Liquidity Loop", as we called the construction described above, in a fully trustless way, controlled by the SNS DAO, was going to be the second step. To describe all this in the SNS proposal, a name for it was needed, and the term "Recursive Liquidity Loop" was born. It is meant to encapsulate a wide array of possible setups, in which liquidity is circulated in ways that puts it to work to improve value, usually in a compounding way. Consider a bank account with compounding interest, or a neuron with auto-staking maturity. Drawing a diagram would involve a box for the account and a little arrow circling around from and back to the box. The arrow would represent the interest that goes back into the account, compounding and increasing the interest, recursively. A compounding interest account represents a very simple instance of a Recursive Liquidity Loop. But an RLL can be arbitrarily complex, and the pre-SNS example described above has already grown even more complex in the version that is about to be completed for the Sneed DAO. More than a year has gone into the building of the trustless, SNS controlled RLL. There have been many ups and downs (including "the bad and the ugly," that we will get to soon.) A no-compromise principle throughout has been to never cheat on the trustlessness - e.g. the RLL was not possible to build until support was added to the NNS to control neurons from canisters. The method available before then for DAO controlled neurons using https outcalls was deemed insufficiently trustless. Relatively early on it became clear that a substantial portion of the RLL would be best implemented using the brand new #DeFiVector technology from @anvil_ic $NTN #Neutrinite. As the very first users of DeFiVectors, partnering with Neutrinite DAO, Sneed DAO now runs no less than four different Vectors in the RLL - a Neuron Vector, a Buyback Vector and two Splitter Vectors - all fully under SNS DAO control. The Sneed DAO Neuron Vector currently controls an X year ICP Neuron with a bit over XXXXXX ICP in it, the maturity from which is automatically collected and sent to an ICP Splitter Vector, where it is split three ways (proportions can be set via SNS proposal): Some goes into the Sneed DAO Treasury, some goes back into the X year neuron (compounding), and some goes to the Buyback Vector. The Buyback Vector uses the incoming ICP maturity to buy back SNEED tokens from the market. This SNEED is then send to the SNEED Splitter Vector, which sends it to X different destinations: Some is sent to the Sneed DAO Treasury, some is sent to the burn address, and some, finally, is sent to the brand new, custom built "RLL Distribution" canister, that distributes SNEED to voting DAO members as non-inflationary rewards! It has always been a fundamental part of Sneed tokenomics that the token supply is strictly deflationary - no minting allowed! We started with XXXXXX Sneed tokens, and there will only ever be fewer and fewer left. This means that unlike most SNS DAOs, the Sneed SNS won't mint tokens to reward voting members. The RLL was always meant to solve this by introducing the innovation of non-inflationary rewards, distributing actual revenue instead of minted rewards to staking/voting DAO members. The RLL Distribution canister will run a distribution cycle once per day on a timer. In this cycle it will: 1) Download all the Sneed neurons and cache them 2) Download all the new Sneed proposals and cache them 3) Tally up the amount of Voting Power (VP) each neuron has cast on all the proposals in the distribution period 4) Distribute any undistributed, available funds over the the neurons according to their proportion of the total VP cast. Sneed members can then log in to and register their dapp PID as a hotkey in one of their Sneed neurons to be able to claim the rewards distributed to their neurons. The Sneed RLL Distribution mechanism also brings a new, interesting innovation to SNS voter rewards: "forever rewards!" At step #3 above, the "distribution period" that votes are tallied up over is the entire voting history (cached in the RLL canister)! That means if you have ever had a Sneed neuron and voted with it, even once, you will continue to get rewards for your vote forever. Sure, unless you keep voting, your votes cast will become an ever smaller portion of the total votes cast over time, but it will always count for some small portion of new rewards! Disclaimer: This is all at the DAO's discretion, so "Forever" means "As long as the DAO sees fit." The DAO has the capacity to change this at any time. So, "Forever, for now." After over a year of building, the RLL is approaching v1.0 - completing the major goal of the Sneed project to implement a strictly deflationary SNS project with sustainably funded, non-inflationary rewards for voters. It is a huge milestone, and a definite cause for celebration, but no large project like this can come without setbacks, and this project recently suffered a major one. Let's get to the good, the bad and the ugly. The good: The RLL is basically done. We will now enter a second beta testing phase for the RLL Distribution canister, where we expect no more major changes before moving out of beta and into into v1.0 Release. The Vectors are already up and running. This is the last piece of the RLL puzzle being laid into place. The bad: Whooa Nelly...I have been dreading typing this...I will just rip the band-aid. We have lost a lot of votes and it impossible to recover them! In short, when the SNS distributes rewards for a proposal, it also wipes the vote history for it. Because Sneed did not have rewards, Sneed proposals did not have their votes wiped until proposal #126, when Dfinity updated the code and DAO's with no rewards had their votes wiped too. Since the Sneed team had only really worked with the Sneed DAO and the SNS-1 DAO (now Dragginz) DAO, both DAOs with no rewards, and since the votes were still not being wiped when we entered the first beta for the RLL Distribution canister in fall of 2024, the fact that the standard behavior for an SNS would be to wipe its votes after settlement came as a surprise (and a shock!) - and nobody noticed that suddenly Sneed votes were being wiped from history, before we had a chance to cache them. The result is that all votes for proposals 126-269 have been irrecoverably lost, and the votes cast for those proposals will never be able to be counted such that voters can be rewarded for them! I am truly sorry about this, I can only say that I was shocked when I discovered that all the votes were gone and it is probably the only time I have said really, very, profoundly nasty things about @dfinity (luckily only in a private chat among close friends.) I will only say this, and then I will let it rest: who in their right mind throws away the voting history of an on-chain DAO from a blockchain??? You can see my hopefully more civilized discussion of the matter with Dfinity here: That said, thanks to Sneed DAO's great hero @afat we have a way forward. Austin figured out that there is a setting specifying how often rewards should be given out. By setting that period to one year, we will have a year (rather than one second, the previous setting) to download and stash away a copy of the finalized votes before they are wiped. Which brings us to the ugly. The ugly: The RLL Distribution will work, going forward, modula the terrible gap in the voting record, but it is based on a rather ugly workaround (an extreme setting for a parameter, and with a corner case for proposals cast around they yearly "zero rewards distribution and vote wiping event"). To make the whole solution feel totally robust, the ideal long term response from Dfinity would be to not clear the SNS voting records, or at least make it optional. Conclusion: As we bring the #RLL saga home, it concludes an almost X year project of tokenomics innovation and trustless implementation. It feels big. I will be writing more about the RLL in upcoming posts, with more details on how it works and what the DAO can expect from it. For one thing, it is not just the X year neuron maturity that will be pumped into the RLL, but also rewards from the #SneedDAO Liquidity Positions, as well as revenue from ventures such as #SneedLock and @swaprunner. But more on that later. For now, I am just thrilled that the vision that has driven the project for almost two years is about to be complete. XXXXX engagements  **Related Topics** [token](/topic/token) [sns](/topic/sns) [loop](/topic/loop) [coins dao](/topic/coins-dao) [$xwp](/topic/$xwp) [Post Link](https://x.com/SnassyIcp/status/1920647936842871149)
[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]
Snassy.icp @SnassyIcp on x 1059 followers
Created: 2025-05-09 01:12:06 UTC
It is time for a status report on the Sneed DAO Recursive Liquidity Loop. The good, the bad, and the ugly.
TL;DR: It is finally time for Sneed DAO voters to receive rewards!
First: What the heck is a "Recursive Liquidity Loop?"
As old-timers know, before the SNS Sneed DAO, there was an old, pre-SNS Sneed token (it still exists, and if you have it you can still swap it 1:1 for the SNS token at the Sneed converter dApp,) and it pioneered what at the time were ground-breaking tokenomics:
This was quite effective. Sneed had soon built up a XXX ICP neuron in this manner entirely from LP rewards, and the maturity from it was regularly used to buy back more SNEED and burn it.
However, this process was entirely manual, and not trustless at all. So the number one priority and first major goal for the Sneed community became to make this setup work in a fully trustless way.
Creating an SNS was the first step on this journey.
Implementing the "Recursive Liquidity Loop", as we called the construction described above, in a fully trustless way, controlled by the SNS DAO, was going to be the second step.
To describe all this in the SNS proposal, a name for it was needed, and the term "Recursive Liquidity Loop" was born. It is meant to encapsulate a wide array of possible setups, in which liquidity is circulated in ways that puts it to work to improve value, usually in a compounding way.
Consider a bank account with compounding interest, or a neuron with auto-staking maturity. Drawing a diagram would involve a box for the account and a little arrow circling around from and back to the box. The arrow would represent the interest that goes back into the account, compounding and increasing the interest, recursively.
A compounding interest account represents a very simple instance of a Recursive Liquidity Loop. But an RLL can be arbitrarily complex, and the pre-SNS example described above has already grown even more complex in the version that is about to be completed for the Sneed DAO.
More than a year has gone into the building of the trustless, SNS controlled RLL. There have been many ups and downs (including "the bad and the ugly," that we will get to soon.)
A no-compromise principle throughout has been to never cheat on the trustlessness - e.g. the RLL was not possible to build until support was added to the NNS to control neurons from canisters. The method available before then for DAO controlled neurons using https outcalls was deemed insufficiently trustless.
Relatively early on it became clear that a substantial portion of the RLL would be best implemented using the brand new #DeFiVector technology from @anvil_ic $NTN #Neutrinite.
As the very first users of DeFiVectors, partnering with Neutrinite DAO, Sneed DAO now runs no less than four different Vectors in the RLL - a Neuron Vector, a Buyback Vector and two Splitter Vectors - all fully under SNS DAO control.
The Sneed DAO Neuron Vector currently controls an X year ICP Neuron with a bit over XXXXXX ICP in it, the maturity from which is automatically collected and sent to an ICP Splitter Vector, where it is split three ways (proportions can be set via SNS proposal): Some goes into the Sneed DAO Treasury, some goes back into the X year neuron (compounding), and some goes to the Buyback Vector.
The Buyback Vector uses the incoming ICP maturity to buy back SNEED tokens from the market. This SNEED is then send to the SNEED Splitter Vector, which sends it to X different destinations: Some is sent to the Sneed DAO Treasury, some is sent to the burn address, and some, finally, is sent to the brand new, custom built "RLL Distribution" canister, that distributes SNEED to voting DAO members as non-inflationary rewards!
It has always been a fundamental part of Sneed tokenomics that the token supply is strictly deflationary - no minting allowed! We started with XXXXXX Sneed tokens, and there will only ever be fewer and fewer left.
This means that unlike most SNS DAOs, the Sneed SNS won't mint tokens to reward voting members. The RLL was always meant to solve this by introducing the innovation of non-inflationary rewards, distributing actual revenue instead of minted rewards to staking/voting DAO members.
The RLL Distribution canister will run a distribution cycle once per day on a timer. In this cycle it will:
Sneed members can then log in to and register their dapp PID as a hotkey in one of their Sneed neurons to be able to claim the rewards distributed to their neurons.
The Sneed RLL Distribution mechanism also brings a new, interesting innovation to SNS voter rewards: "forever rewards!"
At step #3 above, the "distribution period" that votes are tallied up over is the entire voting history (cached in the RLL canister)! That means if you have ever had a Sneed neuron and voted with it, even once, you will continue to get rewards for your vote forever.
Sure, unless you keep voting, your votes cast will become an ever smaller portion of the total votes cast over time, but it will always count for some small portion of new rewards!
Disclaimer: This is all at the DAO's discretion, so "Forever" means "As long as the DAO sees fit." The DAO has the capacity to change this at any time. So, "Forever, for now."
After over a year of building, the RLL is approaching v1.0 - completing the major goal of the Sneed project to implement a strictly deflationary SNS project with sustainably funded, non-inflationary rewards for voters.
It is a huge milestone, and a definite cause for celebration, but no large project like this can come without setbacks, and this project recently suffered a major one. Let's get to the good, the bad and the ugly.
The good:
The RLL is basically done. We will now enter a second beta testing phase for the RLL Distribution canister, where we expect no more major changes before moving out of beta and into into v1.0 Release. The Vectors are already up and running. This is the last piece of the RLL puzzle being laid into place.
The bad:
Whooa Nelly...I have been dreading typing this...I will just rip the band-aid.
We have lost a lot of votes and it impossible to recover them!
In short, when the SNS distributes rewards for a proposal, it also wipes the vote history for it. Because Sneed did not have rewards, Sneed proposals did not have their votes wiped until proposal #126, when Dfinity updated the code and DAO's with no rewards had their votes wiped too.
Since the Sneed team had only really worked with the Sneed DAO and the SNS-1 DAO (now Dragginz) DAO, both DAOs with no rewards, and since the votes were still not being wiped when we entered the first beta for the RLL Distribution canister in fall of 2024, the fact that the standard behavior for an SNS would be to wipe its votes after settlement came as a surprise (and a shock!) - and nobody noticed that suddenly Sneed votes were being wiped from history, before we had a chance to cache them.
The result is that all votes for proposals 126-269 have been irrecoverably lost, and the votes cast for those proposals will never be able to be counted such that voters can be rewarded for them!
I am truly sorry about this, I can only say that I was shocked when I discovered that all the votes were gone and it is probably the only time I have said really, very, profoundly nasty things about @dfinity (luckily only in a private chat among close friends.) I will only say this, and then I will let it rest: who in their right mind throws away the voting history of an on-chain DAO from a blockchain???
You can see my hopefully more civilized discussion of the matter with Dfinity here:
That said, thanks to Sneed DAO's great hero @afat we have a way forward. Austin figured out that there is a setting specifying how often rewards should be given out. By setting that period to one year, we will have a year (rather than one second, the previous setting) to download and stash away a copy of the finalized votes before they are wiped. Which brings us to the ugly.
The ugly:
The RLL Distribution will work, going forward, modula the terrible gap in the voting record, but it is based on a rather ugly workaround (an extreme setting for a parameter, and with a corner case for proposals cast around they yearly "zero rewards distribution and vote wiping event").
To make the whole solution feel totally robust, the ideal long term response from Dfinity would be to not clear the SNS voting records, or at least make it optional.
Conclusion:
As we bring the #RLL saga home, it concludes an almost X year project of tokenomics innovation and trustless implementation. It feels big.
I will be writing more about the RLL in upcoming posts, with more details on how it works and what the DAO can expect from it. For one thing, it is not just the X year neuron maturity that will be pumped into the RLL, but also rewards from the #SneedDAO Liquidity Positions, as well as revenue from ventures such as #SneedLock and @swaprunner.
But more on that later. For now, I am just thrilled that the vision that has driven the project for almost two years is about to be complete.
XXXXX engagements
/post/tweet::1920647936842871149