cICP - Compounding Stream ICP - Liquid Staking Token

[Lorimer]

@infu

Don’t need to believe it, here is the cICP neuron with locked ICP https://dashboard.internetcomputer.org/neuron/7620974521361381620

The question was about what neurons are designed to do, not how you’ve subverted that design to undo the core purpose of a neuron.

@infu

We are fine if people gift it around by following someone, but we can’t let liquid stakers have control over it?

Yes, because the follower has locked their ICP. If they follow a bad, misaligned, lazy, or otherwise ineffective neuron, they’re only hurting themself if the long-term interests of the IC become compromised or weakened.

At this point we’re going around in circles. Just to be clear, if you’re not going to budge on this I will consider this project a significant threat to the IC (specifically because it undoes the most important aspects of IC governance regarding incentive alignment and long-term stake).

Are you familiar with the context regarding canister controlled neurons?

@bjoernek

It is considered important for neurons to be non-transferable/no-sellable because

• neurons should have an incentive to vote in the long-term interest of the Internet Computer and
• to avoid the possibility of attacks where an attacker acquires tokens only for a short amount of time, votes on a malicious proposal (e.g. transfer tokens) and then sells the neurons again.

Whether you know it or not, I think you’re treading on thin ice. I think you’re making rash decisions that will unnecessarily limit scope for the success of this project - which would be a real shame for everybody involved.

@bjoernek

• Establish a materiality threshold: Set a threshold that triggers mitigation measures if canister-controlled neurons exceed 10% of the total voting power. Neurons controlled by DAOs are exempt from this threshold. The threshold might be adjusted at a later point in time. For example, the threshold might be increased (upon NNS approval) following a materiality analysis of canister-controlled neurons belonging to DAOs, which are not considered to be an issue.
• Delayed implementation of additional measures: Additional disincentives for neuron transfers will be implemented if the materiality threshold is surpassed. While further measures could enhance security, they would also increase complexity for users and require significant implementation effort. The specifics of these mitigation measures can be determined as the situation evolves and will be subject to a separate motion proposal. For instance, these disincentives would reduce the rewards for transferable neurons created after this proposal is passed; to avoid the reduction, non-transferability of neurons could be shown via a proof of knowledge of a cryptographic key controlling the neuron.

I think what you’ve so far described for cICP represents a particularly bad implementation for canister-controlled neurons (with respect to how VP is managed), and I would push to see strict measures implemented to disincentivise these sorts of implementations.

I’ll stop harping on now. I just want to make my position clear up front. As I’ve mentioned before, there’s a lot that I like about this project, and I want to be able to support it. I hope you’ll be able to re-evaluate your position regarding VP fungibility.