I have to disagree with this statement. Voters are executives in the IC in a very direct way. This may or may not be the best model, but to propose otherwise is to offer a different type of governance than the one we have.
I think this is a clever way to try to redefine this situation, but what we actually see is ~175M voting power being removed from the consensus because people would not be able to control it anymore. I find it really strange that if people get to use the voting power (for their benefit), it’s okay, otherwise the voting power has to be removed (for their benefit).
I want to reiterate that we’re talking about 175,000,000 Voting Power / 87,500,000 ICP / 525,000,000$, which certainly isn’t “little to lose”.
I think it’s a failing of proof of stake systems when “decentralization” is a facade for calculated control 
1 Like
We weren’t hired.
We didn’t make the company.
We bought into it, we staked to have a right to vote & increase security - we are shareholders.
This is not a different governance type than we have at all.
1 Like
Shareholders don’t govern a company. They are not compensated for governance. They don’t submit proposals. They can’t overrule an executive decision.
It’s true that the “company” in this case predated its current governance model, but having voting power that directly triggers corporate change is practically the definition of executive authority.
Why are you trying to describe this new innovation in decades-old vocabulary? I’ve seen this a lot recently. Maybe what we’re doing here has no current real-world analogue.
2 Likes
That might be true but it still needs to be internally logical and consistent. My aim was to point out the consequences of this proposal and why those might be problematic.
1 Like
The current design removes voting power and rewards from inactive neurons. Previously, if neurons abstained from voting, their voting power would remain intact, though they would not receive rewards. Either revert to the previous design or redesign the system.
Over time, the NNS will be even more centralized.
2 Likes
Great argument, but one could argue thats a major flaw with some governments. One way to increase power is to remove inactive voting power. But if you cant, you’re forced to reach out… but then you circle back to the same issue for dead votes… which I guess removing inactive vp makes sense.
Guess its a matter of whether you think every vote counts or if you dont have the care you shouldnt get a say.
Some governments require by law you vote. Some governments dont… Which is correct?
The only reason most people would vote atm is for rewards so due diligence is often handed over to folowee. Set it and forget it.
Being inactive is good argument to be made that perhaps your stake or care diminishes, which happens through token dilution.
Hi all,
I would mainly like to comment on some individual point, to ensure that there are no misunderstanding for readers who just have the context of this thread.
1. A neuron’s followees are removed in the current design
As pointed out by Daniel, it is already the case that the followees of a neuron are fully removed if the neuron does not perform any of the required actions (directly vote, confirm following, set following).
2. A neuron can get “back” its voting power at any point in time
A neuron who does not perform any of the required actions gets adjusted voting power. I want to clarify that this neuron gets back the voting power immediately when it perform any of the required actions (votes, sets following, confirms following). I thought this is important to point out as maybe future readers may misunderstand statements like:
but this system takes it a step further by removing their voting power completely rather than just removing their followees
The current design removes voting power and rewards from inactive neurons. Previously, if neurons abstained from voting, their voting power would remain intact, though they would not receive rewards.
So in both designs, no matter if the followees are removed or the voting power is reduced, at the moment when the neuron hasn’t taken any actions for a while
- the neuron does not contribute to the decision (no following / no voting power)
- as a consequence, the neuron does not get any rewards.
In both cases, the neuron can revert this and contribute to decisions / get voting rewards again.
The dynamics to that are a bit different, but I wanted to clarify that this is easily “revertible” in both cases. IMH the “removing following” is actually the more drastic measure than adjusting voting power: for getting voting rewards when the voting power is adjusted, a neuron holder simply has to confirm following. In the case where the followees are removed, a neuron has to explicitly set followers again (or vote directly) to get rewards again.
3. The final design was approved in a motion proposal
Without further context, I think this post can be read as if there were a motion proposal and DFINITY followed up with an implementation that was very different from what was proposed. I would like to provide the context that the final design was discussed in multiple forum discussion with the community and finally voted on by the NNS in this motion proposal. Especially, the motion proposal explicitly mentions the adjusted voting power.
Of course this is not to say that we cannot ever adopt the decision, but I think this is important context.
5 Likes
