Reevaluating Neuron Control Restrictions

Governance
183

Maybe revoking the PoK? or whatever we decide to use to make them non-transferable. I don’t know if that’s possible or if there is another primitive. I’m imagining something like:

  • Everyone has a PoK.
  • When you wish to make a transfer, you initiate a countdown to revoke the PoK.
  • During this countdown, you are not allowed to vote.
  • Once the PoK is revoked, the transfer is executed.
  • If the recipient is an individual, they will generate a new PoK and continue voting.
  • If the recipient is a marketplace, it will not generate a new PoK in order to remain liquid and will not have the ability to vote.
2 Likes
184

I don’t understand, in which category my neuron will be according to that? I created one of the neurons I have 4 months and another 8 months ago, I don’t really understand what you mean with “ canister controlled”

I created an internet identity, no ledger, just using the seed phrase, it’s mine on that category you mentioned that will be excluded for this disbursement key? You mention before and after “ TECDSA” you really think normal users will know at what point in time this was released and if now we are getting affected by that ? Just leave the neurons created by internet identity or ledger or whatever thing they decide to create it working AS USUAL, just affect the new ones created after the “motion” proposal if you wanna count with our votes.

Anyway as you said there are not many neurons created till
This point in time this neurons will not affect at all your main reasons for doing this proposal,

1 Like
185

Is not about exploits or no exploits, investors never were mentioned on official documentation their neurons couldn’t be sold, and that any dapps created for that purpose will be “exploiting” that. If I were mentioned I couldn’t sell my neuron because of a watermark on it and no one will
Buy it I didn’t lock it indefinitely, also my rewards will be affected, if someone decided to lock for all those years or “indefinitely” of course we are looking for our ROI based on rewards, and now this is affecting my investment plan. I just say one thing it’s better to leave this few neurons unaffected that anyway didn’t represent a huge threat according to what this proposal supposed to be, rather than sacrificing that probably the project will reduce in a huge % the demand for 8year locked neurons 7year 6 year etc… this will scared a lot of potential long term investors if they see what you are causing to us, the first early supportes.

I’m complaining BOTH for, I couldn’t be able to sell when it was NEVER mentioned a year back on the documentation, and the rewards will be reduced if I put non transferability over transferability.

If i want neuron transferability I get WAY less rewards If I want my rewards PROMISED on our initial “DEAL”
With the protocol. And if I choose non transferability I get NORMAL rewards but with the watermark that means I cannot sell the neuron.

186

I believe this display only includes named neurons. I previously mentioned in both private discussions and on forums that I am seriously contemplating transforming one of my neurons associated with my company, which also operates a node, into a named neuron. I propose that all prominent holders ( or “whales” for lack of a better term) should be encouraged to name their neurons and engage more actively in voting.

2 Likes
187

The proposal doesn’t affect your current neurons…just new ones after the import if passed and built.

Canister controlled means a smart contract can control a neuron…lots of cool applications, but smart contracts on the IC can change owners so it can become an isssue.

4 Likes
188

Well, if we can convince Bjorn that it’s worthwhile to add it as an option to existing neurons it might!

1 Like
190

Try to get the proposal approved, where old neurons get affected and all the whales voting against it, good luck

192

If it’s an option for existing neuron holders to add PoK, then it doesn’t need to affect them. They can choose to ignore it and not add PoK.

The value in the option to add PoK gives existing neuron holders the ability to transfer ownership to a canister. But it would be merely an option, not an obligation.

2 Likes
193

Hello Community,

Below is the official view of GeekFactory on the current proposal.

We absolutely support lifting restrictions on canister control over neurons. This will simplify development and unlock new use cases. However, we question whether the idea of enhancing the long-term binding between neuron controllers and neurons is beneficial for the voting behavior, security of the Internet Computer or the future of the ecosystem.

1) Resource Allocation Concerns

Neurons tied to II in the NNS are only a fraction of the voting power. Neurons with substantial voting power are typically managed by alternative measures (DFX, Hardware wallet, Quill) and cannot be transferred. Furthermore, independent voting on Twitter suggests that people are more inclined to lock ICP long-term if there’s a liquidity opportunity even at a significant open market discount.

Given all of the above, why invest in new ways to restrict neuron transferability when the risks are minimal and the community leans towards flexibility? In addition, this proposal is diverting the Foundation’s limited development time and resources away from more critical matters. It is essential to direct all efforts towards projects that offer greater benefits to ICP community.

2) Proposal Segmentation

This proposal incorporates three different ideas into one which makes it hard to make a clear decision. We would like the community to analyze, discuss and vote on these ideas individually.

3) Clarity on PoK Needed

It is not clear how Proof of Knowledge (PoK) will be implemented in this proposal. This could cause unexpected problems down the line. We believe PoK shouldn’t be part of the proposal until everyone understands exactly how it will work.

We hope this will help you to make your own informed decision.

P.S. Responding to accusations of crossing lines, violating unwritten rules, and posing threats to the ecosystem. We always strive to develop solutions that meet the genuine needs of the ICP community, guided by direct feedback and active engagement with our users. We take pride in the overwhelmingly positive responses we’ve received from our community members.

Thank you!

21 Likes
194

If we lose the ability to offer our nuerons on the market, significantly less people will be willing to stake for any amount of time. Having previously created nuerons “grandfather’ed in” isn’t a cope either. Eventually those nuerons will be sold to their “final owner” and essentially ALL nuerons will be affected.

2 Likes
195

My 2cents on this relevant topic.

Divide the proposal in two, one for allowing canisters to control neurons, and the other for giving the concept of “owner” and how that is operated.

  1. Think the first one is consensual and broadly needed.
  2. The second one, definitely needs a lot more R&D time on it, my suggestions:
  • Neurons are already complex, please don’t make them Even more complex. Adding nuances like, is this a PoK neuron, a before or after motion date, an already created disbursement key, will make things significantly worse. Please let’s avoid at all cost going this way. Whatever the change, I feel it should be implemented to all, and keep all neurons equal.
  • The goal is to keep VP under control, maybe we can implement a tiered solution that does a 80:20 effect. Maybe we only need to force the “initial owner lock” system on 20% of the neurons, and we would already get the 80% VP locked as expected. We can allow smaller investors to have the desired flexibility, and selling at a discount on open market. I have the impression bigger neurons / VP need to do KYC, maybe we force the “owner lock” on them?
  • Maybe we could do a snapshot of the current VP and all neurons higher than X, get “owner locked” (or “initial owner recoverable”), and all new neurons higher than Y (targeting DAOs) also automatically get the “owner locked”.
  • That way I feel pretty confident no significant VP (>33% or >51%) would end up in a neuron market.

Thanks for reading and thanks for this hard work. :muscle:

7 Likes
196

I’ve made a number of philosophical and existential comments in this thread, so let me make one concrete and practical one:

The @GeekFactory team are Internet Computer Heros.

They’ve been building amazing tech for the IC and pushing the limits of the tech on a number of fronts.

Products like ID Geek are necessary to move the protocol forward and increase its anti-fragility. As they stated there is no immediate threat posed by their marketplace.

If any of my statements have come across as accusatory, I apologize and hope you’ll give me some grace in trying to solve for far future problems that could threaten the network.

The ID market does enable some users of the platform to knowingly take undue wealth from the rest of the NNS. In the same way, substantial steak knives enable some individuals to walk into their community bank and steal a million dollars, which through second-order effects takes wealth from everyone else if they get away with it. Theft is theft but I’m not blaming the bank for the flawed system.

Just like crypto offers some significant reduction in bank thefts with math(can’t rob banks that don’t exist), if there is a math solution here to make it impossible for people to steal from other members of the NNS, I hope we can have a rational conversation about it.

If our network has moral hazard or hidden volatility in it then we absolutely need teams like GeekFactory building tools like the II marketplace that ferret those problems out and let us build a healthy immune system for the network.

Further to the root, this last bear market was pretty brutal on a lot of us and I imagine the marketplace actually provided far more “good” than “evil” on that front. Sometimes you have to keep a roof over your kid’s head.

I had an interview yesterday with a DAO researcher and engineer from another community that is doing a lot of work in the conviction voting space. He had an amazing quote that hit me pretty hard since this debate has been on my mind.

He was discussing different voting methods and mentioned Velodrome which pioneered time-locked stake voting(in his eyes, he’s unfamiliar with the IC). “When the price started going down, I saw so many friends hurt with nothing they could do. Things started getting nasty and governance broke down. I always wondered why you’d choose this system. It’s like being abusive to your future self. Why would you choose a system that makes you do that to yourself?”

I don’t have a great answer for him. So it’s possible the long-term solution to this problem lies outside the entire scope of the current debate. Thus it certainly isn’t worth hurting anyone or making inappropriate accusations and if I got too close to that line or if I’ve hurt any feelings it certainly wasn’t my intent and again, I apologize.

Keep building.

11 Likes
197

IDGeek is a viable option to sell neurons and it gives the ICP retail investor peace of mind. The option to sell in case of a financial emergency. The only options to possibly remedy restriction of this may be the idea of creating two types of neurons. One where you can sell, and one where you cannot. This way, the investor gets the choice. With the ones you cannot sell, maybe give a bit more of an incentive, such as voting power, or interest earned. Option two would be to introduce an ICP cap. Neurons holding under 1000 ICP can be sold on the market. Anything over that cannot, but grandfather in those that are above that threshold if the neuron was created before this adopted proposal. Again, the point being to grant ICP holders the freedom of choice when investing.

3 Likes
198

Could you elaborate on this?

199

I recently ran this poll on X (Twitter) and it got 31 votes and passed narrowly in favour of canister controlled neurons with a 20% reduction in voting power: x.com

I do think I agree with this point regarding adding new complexity:

I was thinking what if something simple is done that says if a principal is self-authenticated it receives full voting power and if it is a canister principal it receives -20% voting power. This would impact none of the current neurons, have no new complexity and not impact any current dApps (IDGeek).

There is the problem of the tECDSA and HTTP outcall canisters - they would receive full VP as technically they are “self-authenticated”, but I think those canisters are significantly more complex and use much more cycles than a direct canister controlled neuron (they require two HTTP outcalls for every request + tECDSA signing and a lot more code complexity). So I’d say many devs would just opt to create canister controlled neurons directly it would be much safer and easier for devs to code.

I think overall this solution achieves the goals of this proposal while also keeping current stakeholders happy? Any thoughts?

2 Likes
200

See: Reevaluating Neuron Control Restrictions - #52 by skilesare

It’s easy to not see it this way because the maturity minted is so large and a retail investor gets a tiny fraction. It boils down to what we pay stakers for and what we get. If we pay you to stay committed and then you bail after we have paid you, that is generally called by lots of bad names. “But they have to take a reduced fee in the market” doesn’t help because then the buyer of the neuron is getting the benefit of discounted ICP and we all still paid.

201

The issue arises when these long-term commitments are sold, potentially giving the impression that the initial commitment can be easily discarded for personal gain, especially when such sales could result in benefits like discounted rates for the buyer, which seems to distribute the cost among the rest of us.

However, this perspective might miss an important potential upside. Allowing the sale of long-term neurons can actually enhance our ecosystem’s dynamism and resilience. When stakeholders can sell their neurons, we’re not just enabling a financial transaction; we’re potentially trading passive holding for active participation. This is because the sale opens up the opportunity for these neurons to be acquired by individuals who are not just financially invested but are also willing to actively contribute to the ecosystem’s governance and future direction. Thus, what initially appears as a loss in commitment could actually be a gain in engagement and involvement.

Moreover, while it’s valid to be concerned about the immediate financial implications of such sales, including the aspect of discounted ICP for buyers, we should also consider the broader picture. The primary goal is to build an ecosystem that is not only robust but also inclusive and adaptive. Encouraging a more active and engaged community, even if it means navigating the complexities of neuron sales, can lead to a richer, more participatory ecosystem.

In conclusion, while the concerns about the potential for loss in value and commitment due to neuron sales are understandable, it’s also crucial to recognize the opportunities such mechanisms offer. By fostering a marketplace of ideas and engagement, we can ensure that our ecosystem remains vibrant and forward-thinking, driven by stakeholders who are not only invested financially but are also committed to its long-term success and governance.

1 Like
202

Soon we will have AI wanting to participate in the NNS, and that sounds a bit discriminatory towards our future overlords. :slight_smile:

The simplest and cleanest solution is to lift the restriction, and nothing else.

1 Like
203

It doesn’t really solve the issue of a deep neuron market appearing, which would be very much encouraged if this happens. I think the only way to move forward is to address that issue too.

2 Likes
204

Apparently, only a small fraction of the VP is affected.

1 Like