Reflections on the SNS1 event

[bitbruce]

SNS1 is now 100% controlled by someone and it has lost most of its DAO properties. Of course, it is still logically possible to achieve token decentralization.

NNS and SNS are the main features of IC networks, and this event triggered some thoughts for me.

1. It is worth rethinking whether SNS governance is appropriate in the early stages of a project.
   The project is in the early stage of relatively low market capitalization, it will be easy for someone to take 51% control. Many projects are in such a state in the early stage, and the final result depends on the project controller’s philosophy and development path. When a project is strictly controlled by someone, does it make practical sense to achieve DAO governance on the surface? A project is more reasonable to realize DAO governance after achieving sustainable and stable operation, especially after the code is basically stable.

2. Does the concept of SNS governance of everything need to be restricted.
   This is based on what the purpose of SNS governance is. If the SNS has all the power, then the more likely it is to fail, because there are more paths of governance attack. So it is worth exploring to let SNS have minimal governance authority. For example.
   (1) Making the SNS a controller of the Dapp is only temporary and eventually requires it to give up control over the Dapp, it cannot escalate the wasm of the Dapp and only retains the ability to call certain specified methods on the Dapp at the business layer to implement certain specific governance functions.
   Dapp developers need to be prepared to migrate uncontrolled canister when they encounter bugs, just as ethereum developers do.
   In the end, we still need to pursue the perfect state of “code is law”.
   (2) Whether to allow SNS to change the established Dapp protocol rules, such as the economic model. For example, the current SNS1 can easily be increased.
   The protocol rules of the Dapp have been agreed before the SNS starts, and the SNS is only the executor of the protocol rules. It is unreasonable that an executor can change the protocol rules. Therefore, to restrict the SNS to change the Dapp protocol rules, only allow it to change some specified protocol parameters.

3. The rules of the proposal voting “passed” need to be graded.
   At present, the base for counting votes is the number of weights of neurons with voting rights, the minimum voting rate is 3%, and the rate for “passing” the proposal is 50%. This is reasonable in many cases, but unreasonable in major matters (such as changes to the Dapp protocol), because in many cases someone holding 15% or even a lower percentage of token can control the outcome of the vote, such as the SNS1 event.
   A reasonable approach is to grade the type of proposals, and major decisions at the Dapp protocol level require a higher total voting rate (e.g., 30%) and a higher “yes” vote rate (e.g., 70%), so as to maintain the stability of the Dapp protocol and increase the cost of 51% attacks.

4. Does SNS need to introduce a board of directors.
   As in general corporate governance, not everything is decided by the general meeting of shareholders, and some unimportant daily matters can be decided by a vote of the board. This largely improves the professionalism and efficiency of decision making, especially in the early stages of project development.

[nolyoi]

1. I think the practicality really depends on a case by case basis. I think ideally, a complete dApp would be good. But, how does this prevent anything? They’re still susceptible to the same attacks unless they launch with a large enough market cap to mitigate the attack and that’s not really anything that can be controlled for in many cases imo.

The protocol rules of the Dapp have been agreed before the SNS starts, and the SNS is only the executor of the protocol rules. It is unreasonable that an executor can change the protocol rules. Therefore, to restrict the SNS to change the Dapp protocol rules, only allow it to change some specified protocol parameters.

2. Is this not the case now? I thought the ledger, gov, root, and other core canisters are managed by the SNS-WASM/NNS. Are these able to be changed with UpgradeSnsControlledCanister on NNS launched SNSes as well?

3. Definitely agree. I think this might address a lot of the problems with #1. This is something I’ve had to point out to people multiple times. A lot of people seemed to have a misconception about how this actually works. It’s the neuroned tokens that matter when it comes to voting. What isn’t neuroned has no influence in voting power. Even if there are 2 billion coins, if only 100 are staked, only 51 are needed to take control of the DAO.

[h1teshtr1path1]

1. Most valid point, DAO should have most of the major functionalities and should achieve stable state before going under SNS.
   But Projects undergoing SNS right now are kinda raising funds for further development.

[jwarner]

@bitbruce

SNS1 is now 100% controlled by someone and it has lost most of its DAO properties.

This is only the case because the majority of DAO members were actively selling their stake and no longer wanted to participate in the SNS1 DAO, right? If the majority of DAO members want out of the DAO, and one entity wants to buy them out, I don’t see a problem with that. It’s like a large corporation buying a smaller business that may be faltering, then injecting capital into it to bring it back to life.

Maybe they’ll take it in a new direction, re-distribute the tokens so they no longer have a 51% stake. Use it as a launchpad infrastructure already in place to build something great?

I see a lot of talk about 51% attacks, but this was not that. It was someone who bought tokens that were openly available on the market from people who no longer wanted to participate in the SNS1 DAO. That doesn’t mean it can’t ever go back to a DAO, it has just been reset to what most "DAO"s start out as.

@bitbruce

The project is in the early stage of relatively low market capitalization, it will be easy for someone to take 51% control.

It’s only easy if the majority of the DAO token holders want to sell their stake and someone has the capital to buy it. And if that’s the case, it’s not an attack, it’s willingly handing it over to anyone willing to buy it.

@bitbruce

When a project is strictly controlled by someone, does it make practical sense to achieve DAO governance on the surface?

I think it does. It allows a lot of flexibility in the decision making. Just because I hold 51% stake in something and can override any vote, doesn’t mean I don’t want feedback from the community. And if I genuinely don’t care what happens with a specific proposal, I can abstain and let the other participants decide.

@bitbruce

Does the concept of SNS governance of everything need to be restricted.

I think this would really depend on the type of DAO/dapp/business you’re running. In my opinion, there’s certainly nothing wrong with some restrictions. But that would be up to the DAO/dapp/business founders. There’s an infinite number of ways an SNS could be used in the overall infrastructure of an organization.

@bitbruce

This is reasonable in many cases, but unreasonable in major matters (such as changes to the Dapp protocol), because in many cases someone holding 15% or even a lower percentage of token can control the outcome of the vote, such as the SNS1 event.

In my opinion, no matter how you weigh the votes or grade the proposals. If there are enough people selling their voting power, and someone has enough capital to buy it up, they can swing the vote however they want. Also, as someone said in another thread, a lot of deals like these won’t happen on chain, they’ll happen in offices across the world. “Hey Bob, I really want to push this vote through. Come spend a weekend in the Bahamas so we can talk about it. Bring the family!” Ha ha!

@bitbruce

Does SNS need to introduce a board of directors.

Just make sure your board of directors collectively hold 51% when the token mint happens. Of course, you have to somehow guarantee they aren’t going to sell it.

I think you definitely bring up a lot of good things to consider for anyone seriously considering starting a DAO. There’s a lot of nuances that go into governance, and I’m sure we haven’t thought of most of them because it’s still such a new space. I’m looking forward to watching what happens with SNS1. I’m not a holder of the token, but I’ll definitely pick one up if the price is right.

[ZackDS]

This shat is just great for new users looking at the Dashboard, puts everything about SNS DAO in a trustworthy position.

[bitbruce]

@jwarner

It’s only easy if the majority of the DAO token holders want to sell their stake and someone has the capital to buy it. And if that’s the case, it’s not an attack, it’s willingly handing it over to anyone willing to buy it.

Of course, this is part of the market behavior and can be interpreted as “not an attack”. I think this state is not the goal of SNS. In fact, the controller of SNS1 can do anything now, he can upgrade the wasm of all the cansiter in the dapp, including the SNS1 token, for example, to increase the total supply of SNS1. He doesn’t have to operate through SNS, SNS is just a form for him.

Just like a listed company in the stock market, a company that reaches a certain level of control will be required to launch a tender offer or delisting to all shareholders. The stock exchange is pursuing the goal of a “public company market,” and the SNS is pursuing the goal of a DAO, not a DAO controlled by one person. I am not denying a one-person-controlled project, but rather expressing the view that the DAO pursued by the SNS should not be easily controlled by one person, or at least should raise the economic threshold.
In blockchain, using economic costs solves most of the problems.

Correction : The wasm of SNS1 Ledger is provided by the SNS system, and the controller of Dapp can only choose to upgrade. He can modify Dapp’s canisters code (not including Ledger)

[borovan]

It’s just testament to the IC how flexible this system is.

We (Dragginz) were considering a SNS launch, but it was the first time we saw the whole “blank slate” concept on the SNS-1 landing page… just too tempting. The issue with the SNS sale was that this was always going to be a project that was going to be centralised as far as creative control goes, and we felt we were just adding governance for the sake of it. The facade of decentralised control didn’t really align with the ethos of the IC.

That said, we are going to have rewards for neuron holders, maybe even give them a part to play in the game - create or breed Dragginz that nobody else can. Honestly, fusing game balance with token ownership seems like it’ll take a while to accomplish.

Bear with us, lots more information will be coming next week!

[ZackDS]

@borovan

The issue with the SNS sale was

that you new it would not hold and will not pass. Plain and simple.

[borovan]

Yeah? Well, you know, that’s just like uh, your opinion, man.

[ZackDS]

I was expecting more like Thank you Captain Obvious for stating facts reply but this will do.

Just to be clear I am not mad or angry with you I am just disappointed and little frustrated that I have to redo my charts for the second part of my presentation and explain to new people why they should consider going the old SNS sale route when they can just buy one, and how Sonic failed and a Logan Paul hatching game is on the main page on the IC Dashboard for the “A Service Nervous Systems (SNS) is an advanced form of a DAO” . Investors will eat this up in notime.

[borovan]

I think (hope) that was the last opportunity to buy your way into a SNS sale. I believe that the NNS vote would have passed. Yes this is jumping the gun a little, but crypto was getting really boring and I wanted to spice it up… BAM!

Sorry you have to redo your presentation.

[cyberowl]

What happens to the locked neurons that don’t want to participate anymore? Are they stuck in SNS-1 purgatory.

[borovan]

Good question. I guess I could buy them out or something, or perhaps there’s a SNS governance way to sort it.

I get that not everybody is as excited as I am about a fantasy world with dragons in. The SNS was always going to become ‘something’ though and I can think of far worse outcomes.

[cyberowl]

Oh for sure. I am not against this buyout, just wondering if it was indeed malicious can’t they just drain the other neurons out similar to the treasury. At the same time that would completely change the liquidity at the exchange. I don’t see anything wrong with this but there are some questions we need to ask now like if these canisters held user data, a backup mech would be great to fork to another cluster of canisters if the community disagrees with leadership. Thinking about apps like Distrikt and others.

[infu]

@bitbruce

he can upgrade the wasm of all the cansiter in the dapp, including the SNS1 token, for example, to increase the total supply of SNS1.

Are you sure that can be done? I am not so sure. I think NNS controls SNS canisters like the ledger and governance. Can we get Dfinity to provide clarification on that technical question @lara

[ZackDS]

Looks like Dfinity has left the building.

[chepreghy]

Not at all. But @borovan played by the rules. No serious SNS project will set up a DAO, so that someone can buy it so easily. SNS-1 was meant as a dress rehearsal for the decentralization process, and to reveal what needs ironing out before real teams assign control of their dapp to an SNS. SNS-1 served its purpose, and it looks like, it will have an interesting future, be it not one initially envisioned.

[ZackDS]

Great to see all of you guys were in on it. Need to take a break from this madness and re evaluate the trust and benefit of doubt I gave to IC.

[lara]

Thanks for tagging me!

Indeed, the SNS DAO can only upgrade the SNS canisters to WASMs that have been pre-approved by the NNS but can upgrade the dapp canisters to any WASM.
This is explained in more detail here.

[cyberowl]

I mean worse case scenario is NNS take over right. I think there are bigger problems on the IC and crypto than a buyout takeover. It was a very predictable outcome. Hence why we open source and build data bridges into users personal canisters. As long as the NNS isn’t taken over, canisters can move into other canisters data and value.