Community motion: Emergency unlock for neurons - 神经元紧急解锁

Is that confirm and where can I see the documentation for that, @Severin

found it: Internet Computer Loading

2 Likes

Sounds about right. Also we must agree it’s hard math not for everyone. But if it’s Last updated on May 9, 2023 I would trust it to be right :wink:

From decentralization point of view, this is actually good for decentralization, but it seems that it is a little difficult to implement since you need to consider a lot.

After reviewing the proposal in detail, I think I change my mind, I will support it.

1 Like

If you invested in crypto, it is money that you are willing to lose…

Apart from that, that bet allowed him to develop and improve the blockchain.

Betting 8 years gives back to ICP to have a long-term horizon.

The burn, although it is good, does not have an impact to improve the deflation of the currency.

Ex: %25 burned out of 1000 → 250 ICP
Current Reward % → 0.48

To equip the burnout:
it takes more than 500 days (1.5 years) (17.8 %) (actual)
1.851 days ( 5 years) (%10)

I think the idea is good but the minimum % needs to be at least 50% of the neuron and nobody is going to do it, they prefer to sell it.

I think letting people people unstake at any given time for a flat burn fee will benefit the ecosystem in the long run.

1 Like

Actually, losing the interest that has accrued is a small penalty for a staker who has just locked a neuron but a large penalty for a staker who has kept their neuron for say 7 years of an 8 year lock-up period. If getting a long term neuron is meant to reflect commitment to the long term health of the network, the fast quitter has shown less commitment than the staker who lasted 7 of 8 years and so the fast quitter should have to pay a higher penalty.

If fast quitting is cheap, that allows someone to make a long term neuron, vote against the interest of the chain with correspondingly inflated power, and then quit, paying little.

I think it would be better to make the burn percentage a function of the remaining dissolve delay. I know there are game theorists out there, I’m sure they will be able to come up with a good function if the community accepts this in principle.

5 Likes

I completely agree with your points regarding the penalties for stakers who fast quit and those who hold their neurons for a longer period. Fast quitting can have negative effects on the network’s health and stability, and it is important to ensure that stakers are adequately incentivized to commit to the long-term success of the network.

I agree that game theorists can play an important role in designing a suitable function that can achieve the desired outcomes, and I hope that the community will be receptive to this proposal. Thank you for sharing your thoughts on this important topic.

1 Like

Instead of a flat out burn of the token what would you think if we burned the tokens by buying cycles with the fee. Those cycles could be held by the nns and used to sponsor developers to start up their project and get a working product to market for cheaper.

3 Likes

First of all, thank you very much @blockpunk for the detailed and well thought-out proposal. We appreciate your engagement and clear description of the issue and suggested solution. We can follow the overall motivation that, in very special circumstances, it might be desirable to have an emergency unlock accompanied by an appropriate penalty fee.

Regarding the suggested bidding mechanism, using a market-based approach seems like a clever way to automatically determine the unlock penalty based on supply and demand.

However, we have concerns about the complexity of implementing an interactive bidding scheme. It would require significant effort on the NNS backend and frontend, which could pose security challenges and result in a lengthy implementation timeline.

As an alternative, we propose a simplification for improved security and ease of understanding. We suggest enhancing the “minimum burn penalty” design component by introducing a dissolve delay-dependent unlock penalty function that can be configured by the NNS. Neurons would be able to initiate an emergency unlock with a specified penalty based on this function. The specific shape of the penalty function can be defined separately, but it should have a strong punitive effect (e.g., an 8-year neuron could be subject to a burn penalty of 66%).

The “Emergency Unlock Maximum” design component would still be necessary to prevent flooding the ICP market.

In the future, the mechanism could be expanded to include market-based features if needed. We welcome your feedback and input from the community regarding this proposed simplification.

Disclaimer: This feedback is based on discussions between @THLO and me and does not represent an official position from DFINITY.

4 Likes

Hi all, if anyone has additional thoughts or input, I would greatly appreciate feedback on the proposed simplification mentioned above (using a dissolve delay-dependent unlock penalty function). Please feel free to share your ideas or suggestions.

I don’t think you can throw open the gates without significantly tainting the way people vote. Even a 2/3 haircut may make give people an out that alters how they vote for the good of the network. There has to be some element that makes it so that they can’t be guaranteed an exit. Maybe a max exit per week and a lottery for that ‘space’ each week. A lottery would be more straightforward than a market mechanism.

I’m pretty sure I have the most extreme position on this, but I am pretty sure that a series of marginalized choices lead to a marginalized outcome and I didn’t lock for eight years to end up halfway to the best possible network we could build.

5 Likes

One of the side effects I liked the most about long term staking without possibility of exit is that it leads to long term thinking and long term behavior.

It is so different from anything else and a main reason I am personally so committed to making the IC a success.

Anything that undermines this long term thinking, which I believe this proposal does, is not something I support.

2 Likes

Hi @bjoernek, I think it is impossible to make any accurate predictions on what would happen if this is implemented. Personally, I am neutral and do not have a strong opinion one way or the other.

Since the idea is being entertained, some people are going to be happy, and some people are going to be mad. The easiest thing to do is let the people in the ecosystem decide with a vote.

In my opinion, if people decide to exit the ecosystem by backing out of their commitment, the penalty should be significant, and it should also benefit the people who stay. Using the dissolve delay-dependent unlock penalty function sounds reasonable.

2 Likes

I think there is no need for this, the emergency that started this has passed and so did peoples interest in it. Lost faith in a long term project after only 2 years because of the current price of ICP is not an “emergency” in my opinion.