Introducing WaterNeuron

Introducing WaterNeuron

What? WaterNeuron is a liquid staking protocol. It offers a more capital-efficient staking method.

How? In regular staking, your capital is locked in neurons for a definite amount of time through the NNS, and only the yield you earn (maturity) is liquid. In liquid staking, your capital and yield are liquid and represented by a single token, in our case, $nICP.

Why? Liquid staking tokens allow users to get a similar yield to the NNS while enjoying fully unlocked capital. Users can then use this capital in various DEFI protocols to:

  • Earn extra yield on DEXes - by putting the token in liquidity pools of various exchanges, you earn a fraction of the fees on the swap.
  • Use as collateral - by locking it in a borrow/lending platform or stablecoin, you can borrow stablecoins against your $nICP, essentially borrowing against your “locked neurons.”

TL;DR When you lock $ICP into the NNS, only your yield (voting rewards, expressed as maturity) is liquid, to unlock your capital you need to wait until your dissolve delay for your capital (locked $ICP) is over. Our solution earns a comparable yield to the NNS while staying fully liquid.

What problems are we solving?

Now that you have a better grasp of the mechanism surrounding this liquid staking token let’s circle back as to why we believe a liquid staking protocol is essential.

  • Capital Efficiency: DeFi is still having a hard time on the IC, partly because any DeFi app using ICP needs to compete against the juicy 8 to 15% that can be generated by locking ICP in a neuron. To that effect, most users seeking safe yield on their token have only one place to go. This locks up a large amount of capital and makes those tokens disappear from the network. On top of that as the NNS incentives large lock-up period for governance reasons most DEFI and governance users get bundled up together.

To fix this, users can now use $nICP instead of $ICP on DEXs or elsewhere to expose themselves to the yields they would get by locking their ICP + the yield they can get from the DeFi app.

  • Governance Decentralization: A significant downside of the IC is the centralization of the voting power. The NNS, as the protocol design intended, is all-powerful. It can vote to upgrade any canister, whether it has been explicitly declared as a controller or not. This problem could disappear if DFINITY didn’t have so much voting power (as most neurons follow DFINITY). WaterNeuron is poised to become an alternative to follow to earn maximum rewards as the known neuron will have a 100% voting rate with educated votes when needed. A counterbalance to DFINITY is crucial to ensure the decentralization of the Internet Computer protocol.
  • $ICP Inflation: The inflation rate on the IC tends to be 5% per year. Token holders who have not locked their tokens see their tokens lose value every year. Liquid staking tokens allow you to hedge against inflation without locking up your tokens.

Protocol Design

The liquid staking token can be minted anytime by swapping $ICP for $nICP. The $ICP swapped will be put together in a 6-month neuron controlled by the WaterNeuron canister.

To make yourself whole again, you can either:

  • Swap it for $ICP on a DEX at a discounted price
  • Redeem it at a protocol level. In this case, we would split your $ICP in your own 6-month neurons that will immediately start dissolving, as pictured below

$nICP thus now represents your $ICP staked in the common 6-month neuron. Now why would you lock it in a 6-months controlled by a cannister when you could do it yourself?

However, it represents more than that, as you have derived from the title, we will do an SNS - all the $ICP received from that SNS will get staked in an 8-year neuron and 90% of the voting rewards will go directly to the 6-months neurons ($nICP holders), as pictured below.

This approach simplifies the staking process, removing the need to select specific trusted neurons to maximize rewards. The protocol votes no by default. The SNS can overwrite the no by proposing to vote yes.

WaterNeuron will launch its SNS DAO with a proposal to start the initial SNS swap. The initial parameters can be found here: docs.waterneuron.fi. We invite you to play with them on the dashboard with the Tokenomics Analyzer.

This SNS is special. 100% of the $ICP deposited will be sent to an 8-year neuron. As explained previously, 90% of this neuron’s rewards will go to the 6-month neuron and will go to the $nICP holder to enhance their yield. The remaining rewards, 10%, will be disbursed to the governance holders who have locked their tokens.

Liquidity

As with any DEFI protocol, early-on liquidity is paramount to success. To that effect, we have two built-in mechanisms to bootstrap the protocol’s flywheel effect.

The first one is enhanced yield for early-on users. If 1 million $ICP gets deposited in SNS, the yield from the 8-year neuron will be substantial. We graphed the APY for $nICP users, from 50’000 to 5’000’000 $ICP locked (0.01% to 1% of the total supply).

The first graph looks at the APY before the protocol hits the $500’000 $ICP total value locked mark. We go from 346,95% yearly return at the 50’000 $ICP TVL mark, 41% at the 500’000 $ICP mark, and from then on, the yield doesn’t go down below 10% until we reach 6’000’000 $ICP tokens (or 1,2% of the total ICP supply) on the protocol:

It then diminishes gradually until we only reach the exact yield of the NNS for a 6-month neuron, 8%, at the 21’800’000 token mark. Afterward, the APY tends to 7.3%; however, this does not take the age bonus of the 8-year neuron into account, which, after four years, will increase the APY by 25%.

Now, let’s get on to the second liquidity bootstrapping mechanism. ⅓ rd of the total amount of $WTN will get airdropped to early user as follows:

This means that until the protocol reaches 40’000 $ICP TVL, every $ICP swapped for $nICP will also receive 64 $WTN, and that is for every token swapped. Between 40’000 and 80’000, it will be 32 $WTN for every token swapped. This airdrop will continue until the 163’840’000th million tokens are swapped—or until the protocol reaches a 2 billion dollar inflow of $ICP (at the current $ICP price of $12.89 per token).

Why would you want the governance token in the first place? Well, 10% of the yield from the 6-month tokens gets disbursed to the governance token holder. As more users leverage liquid tokens, the 6-month neuron and the 10% cut will grow. APY here is calculated given the initial theoretical amount of 1 million $ICP deposited at SNS. The WTN token will have meager inflation.

Wen launch? As soon as canisters can control neurons. The protocol is already built. We have a working version in tesnet. As the implementation changes for the canister-controlled neurons are minimal - a few lines of code need to be removed to accept canister-controlled neurons:

ic/rs/nns/governance/src/governance.rs

Next week, we will do a Twitter Live showcasing the dapp running on a testnet.

As the forum is not ideal for discussions (single-threaded style page) - this post is cross-posted to Taggr where we invite everyone to carry on the discussion and ask any questions they might have.

Your mission, should you choose to accept it:

12 Likes

These two statements seem contradictory in my opinion. Would you please clarify further?

I’m all for decentralizing ICP governance and enabling new entities to get involved by offering educated votes that the community can follow. However, voting NO by default is not educated and it is not a credible contribution to decentralization of ICP. In fact, I think a valid argument could be made that voting NO by default is equivalent to stealing from the NNS. Governance rewards are supposed to incentivize people to make their own educated decisions or to follow someone that will. Voting NO by default is just a mechanism to enrich the people who participate in the WaterNeruon product offering without regard to whether the decision to vote NO is in the long term best interest of the internet computer. I suspect there is more nuance to this proposed mechanism for governance participation. I also know one of the developers of this project is highly credible. Hence, I would really appreciate clarification on how the WaterNeuron can help advance decentralization by offering educated votes.

I understand that the SNS can overwrite by proposing to vote YES, but why would that happen except for the rare occasions when there is a Governance Motion proposal that is highly controversial? What incentivizes the SNS community to try to overwrite a default NO vote? Do you already have ideas in mind that will drive educated participation in governance?

Also, why does the WaterNeuron need to vote NO by default at all? There are known neurons in this ecosystem who are actively and intentionally trying to offer the service of being a reliable and credible contributor to ICP governance. This includes known neurons that are not DFINITY. Why not follow them or a group of them? Today, this is known to be true for Governance, SNS & Neurons Fund, IC-OS Version Election, and System Canister Management proposal topics. You can already guarantee to always vote on these topics if you choose the right known neurons to follow. Even if you choose to follow DFINITY on each of the other technical topics, since they do offer an educated and credible vote on all topics, there is nothing wrong with deciding to follow other known neurons in the future when it becomes obvious that they are offering a decentralized and credible option on those other topics. If you were following other known neurons (instead of voting NO by default), then you would align with the design intent of liquid democracy and you would be respecting the intended purpose of governance rewards.

If you want to create a smart contract that will always guarantee that the WaterNeuron votes, then why not just cast a NO vote only if your Followees have not reached consensus within the last 1 hour of the voting period for the proposal? Better yet, why not just cast a vote based on the majority of the Followees who have voted at that point in time regardless if consensus among Followees have been reached. Then you can guarantee 100% voting participation without having to automatically vote NO on anything.

I know there are already people who vote NO automatically on all proposals. In fact, that’s how the Taggr Network known neuron works when sufficient people don’t vote in the Taggr proposal poll (which happens a lot). It guarantees a vote, but the Taggr voting impact is small and will probably never be very large. By contrast, WaterNeuron is a form of defi that sources its funding directly from ICP governance and could potentially carry a lot of voting power if the project is successful. Personally, I think the mechanism for how the WaterNeuron voting power will be cast needs a lot more explanation and consideration. I have a lot of concern about the idea that WaterNeuron will vote NO by default, which is essentially all that has been presented so far. However, I would have a lot of enthusiasm if WaterNeuron were to follow other known neurons by default while still enabling the SNS to overwrite the Followee votes. It would be great to see WaterNeuon truly offer an option that actually advances decentralization of ICP through educated voting.

6 Likes

Hi wpb, thanks for your interest!

These two statements seem contradictory in my opinion. Would you please clarify further?

Indeed they could seem contradictory but it’s more of a technical detail than anything else. If the DAO didn’t vote yes for a proposal it means that the DAO votes no. Another solution would be to submit two proposals (one two vote yes and another one to vote no) per NNS proposals, but then the WaterNeuron DAO would be flooded with proposals, and some contradictory votes could happen.

Governance rewards are supposed to incentivize people to make their own educated decisions or to follow someone who will.

That’s not true, most people follow Dfinity as they mostly care about the rewards. If we are honest most of the people staking ICP stake it for the rewards and not to participate in the ICP dao.

I understand that the SNS can overwrite by proposing to vote YES, but why would that happen except for the rare occasions when there is a Governance Motion proposal that is highly controversial? What incentivizes the SNS community to try to overwrite a default NO vote? Do you already have ideas in mind that will drive educated participation in governance?

We are building tooling around WaterNeuron. A few examples: a script to verify the releases, we’ll be able to make people submit the hash to a canister which will submit a proposal to vote yes whenever the release is verified, we also will build a canister that will mirror all the proposals from the NNS to WaterNeuron.

why does the WaterNeuron need to vote NO by default at all?

The NO by default is a conservative approach, as we have to choose a default vote we would rather go with NO than YES for the good of the network.

Why not follow them or a group of them?

This is an option that WaterNeuron might explore, like every month elect a known neuron to follow, this would be up to the DAO. And the idea is to build a new voice in the network so following Dfinity is not an option if we want to decentralize governance. Also, WaterNeuron should be composed of some of the best experts in the IC.

If you want to create a smart contract that will always guarantee that the WaterNeuron votes, then why not just cast a NO vote only if your Followees have not reached consensus within the last 1 hour of the voting period for the proposal? Better yet, why not just cast a vote based on the majority of the Followees who have voted at that point in time regardless if consensus among Followees have been reached. Then you can guarantee 100% voting participation without having to automatically vote NO on anything.

I like this idea, we’ll stick with No by default for the sake of simplicity, but this makes a lot of sense and we could upgrade to a better voting mechanism.

To summarize, I think it’s fine to start with a simple NO by default mechanism while keeping in mind that this will change. As there is no oblivious answer, the WaterNeuron DAO will decide whatever mechanism it likes better. It could be electing a known neuron every month, keeping the NO by default, checking the status of the voting on a proposal 1 hour before the end of the voting period etc. It seems like you have some great idea, it would be great if you could be part of the WaterNeuron DAO.

2 Likes

Thank you for your response @WaterNeuron

In my opinion, these ideas are not contradictory. Dfinity is a valid choice on all topics because they care about the network and provide educated and intentional votes. To me, the point is that people make the choice to continue following Dfinity because they believe it is in their best interest. Of course, I think there are other choices too (including some technical topics) if you look at voting records and publicly disclosed manifestos about some of the known neurons. Regardless of who a person chooses to follow, the choice they make is incentivized by the fact that they get governance rewards if they choose Followees that always vote and vote in their best interests. The red flag for me is when a neuron with potentially big voting power just votes NO automatically with a sole purpose of getting voting rewards and without regard to the long term best interest of the internet computer.

This is where I prefer to use the term “let’s be honest”. In my observation, the SNS DAO will follow the lead of the dev team that created the SNS. If you think voting NO by default is a trivial, conservative approach, then it is almost certain that the DAO will never change from that ethos. I’m hoping to convince you to to take a different approach from the beginning by abandoning the idea of routine default voting NO and implementing something more constructive for the ICP ecosystem. I know it is more challenging to build than a default NO, but your team has the skills to provide a greater service than we have ever seen so far from active governance participants.

These are awesome ideas. These are the kinds of ideas that get me excited about what you are planning.

3 Likes

Hey WaterNeuron,

From what I have summarised here:

  • Users receive nICP when they stake in the WaterNeuron protocol.
  • The ICP raised via SNS will be staked in an 8 year neuron to boost the rewards for nICP holders (by adding to the 6 month stake).

Some issues I see with this approach is that from your design: you only get your rewards (in ICP) when you go and withdraw from the WaterNeuron protocol which takes 6 months? So it’s actually 6 months until you can claim your rewards as opposed the NNS’s 7 days?

Also I don’t see plans to provide liquidity on Dexes for the nICP token? So how “liquid” is the nICP token really?

I raise these questions because I hope a valid liquid staking appears on ICP and I understand you may have to make some tradeoff’s in places but have you thought about the tradeoffs?

2 Likes

Hey Jesse,

First of all, we like RakeOff at WaterNeuron and really like the initiative. I can already see RakeOff and WaterNeuron benefit from each other. The two points from your summary are indeed correct.

So it’s actually 6 months until you can claim your rewards as opposed the NNS’s 7 days?

Correct, we’ve chosen to compound the rewards by default, for the sake of simplicity and as it seems like a fair tradeoff. However, we could imaging having another version of the token that doesn’t rebalance and directly distributes rewards to holders.

Also I don’t see plans to provide liquidity on Dexes for the nICP token? So how “liquid” is the nICP token really?

There might be some misunderstanding of what a liquid token is, a liquid token doesn’t mean that you can get in and out without delay, it’s more that you have a liquid token, that you can transfer, sell, etc. of a locked token. There is no point in providing liquidity on DEXs as it should be up to the market to decide what should be the $nICP rate.

2 Likes

Thanks for the kind words about Rakeoff, I like what yous are doing too and I’m just trying to better understand the protocol. Thanks for clarifying in your response.

I still think your design can work and I think it’s similar to what StakedICP done - but there is that tradeoff of waiting until all your ICP is unlocked for your rewards (so in this case 6 months), It’s not the best UX in my opinion.

I know the alternative like distributing rewards directly would be tricky to implement - especially as things scale out.

2 Likes

So let’s pretend I’m a user (which I hope to be :slight_smile: ) and please correct me if I’m wrong:

  • I stake in WaterNeuron and receive nICP.
  • The team is providing no liquidity for nICP so there is a high chance early on I can’t sell my nICP on Dexes.
  • I have to unlock everything and wait 6 months to get my accumulated rewards.

There is an argument here that this is actually less liquid than just staking on the NNS (Which at the minimum gives me my rewards in 7 days). What do you think?

2 Likes

This sounds pretty unlikely because whales could benefit from this protocol by providing liquidity on dexes by providing nICP/ICP and enhancing ever more their yield.

There is an argument here that this is actually less liquid than just staking on the NNS (Which at the minimum gives me my rewards in 7 days). What do you think?

I think this is untrue, first of all, I think we can ignore the rewards in this discussion. Even if you can claim your rewards in 7 days it still takes 6 months to unstake the principal. Again, being liquid just means that you’re able to transfer or trade your nICP, which is not possible on the NNS, hence nICP makes staking liquid.

2 Likes

So, investors in SNS DAO will be funding those who stake their ICP into nICP? Are future investors aware of this?

So, investors in SNS DAO will be funding those who stake their ICP into nICP?

Partly, it’s a number game. The more ICP is converted into nICP the more the SNS participants are winning.

Are future investors aware of this?

Everything should be crystal clear and clearly explained everywhere. That’s why we made a spreadsheet for everyone to be able to play with the numbers easily.

I too think that more decentralization would be very welcome but that this is a harmful way of trying to implement it. Voting according to a majority of the followees seems much more constructive.

2 Likes

If you really, really want a default uninformed vote, why not have two neurons of equal weight, one that votes no and one that votes yes, so that they cancel each other out? This way at least the vote is neutral. But getting eyes, hearts and brains on every proposal brings real value to the ecosystem. Why not really focus on “is this proposal desirable?” and “does this proposal accomplish what it claims to?”. So you need to find a way of getting community sentiment for the first and some hard core programmers and auditors to do the checking for the second. Vote yes when both sentiment and audit pass.

5 Likes

I think delta-neutral voting is a reasonable idea but it doesn’t help ICP’s governance.

We will post a new thread updating the design choices made. But regarding voting, every NNS proposal will be duplicated on the SNS and the protocol will vote on the outcome of the SNS proposal on the NNS proposal on hour before expiration.

2 Likes

Glad to see you reduced the team allocation. :+1:
Are the last epochs going to happen as SNS swaps or custom canisters?

2 Likes

Hopefully SNS Swaps, but if not custom cansiter.

2 Likes

Yes, first decentralized liquid staking protocol on the IC.

For everyone who has been following this thread, we announced the SNS here: WaterNeuron SNS-DAO Launch if you want to continue the conversation.

2 Likes

Is there a reason that liquid staking needs to have voting power at all?
This is intended to be a financial instrument not a “community DAO” correct?

Let’s say dfinity were able to add a function that a liquid staking DAO/neuron were able to declare itself as such, it would void it’s own voting power in return for receiving the full voting rewards as if it were an active voting neuron. The maturity would be the same, which is the purpose of liquid staking, but the voting is void.

Doing this is literally the opposite of decentralisation. It is centralising a large amount of voting power. But the purpose is the ROI yes? So voting power should be irrelevant to a liquid staking platform.

It’s also been said that this is to take voting majority away from dfinity, but now instead of voting no to all, you’ll follow the majority votes on the NNS an hour or so before expiry.

But you’ve also stated that most neurons follow DFINITY. So in effect you’ll be following the majority which you’ve stated already follow dfinity, which essentially means you’ll be following dfinity for voting.

The difference is, people who follow dfinity can revoke that at any time in their neuron. From what I understand of this, this is irreversible. You get the voting majority of hundreds of thousands of ICP after the SNS launch and people cannot chose to revoke who you are voting on. Is this correct?

Would you be opposed to revoking voting power if your earned maturity stayed the same as though you were voting on all proposals? (If you are, why?)

As this would still meet the criteria of your goal, which is to achieve a better ROI on liquid staking, but it would not centralise so much voting power.

Is this possible from dfinity side of things?

2 Likes