Conversion of maturity – if this becomes a feature of the proposal – would be a manual event. What would the difference between automatically doing it, and accumulating the maturity as compounding in the first place?
It is very likely not going to be a part of the SNS in its first release.
Yes, you will always be able to retrieve “liquid maturity” as soon as it is represents more than 1 ICP in value. In order to preserve this freedom, you would NOT enable the compounding maturity flag on your neuron.
2 Likes
Indeed, the auto-compounded maturity, which treats maturity as a kind of virtual stake, would be entirely opt-in. The current maturity scheme, as it stands, would still remain.
6 Likes
Hmm you are definitely right that it definitely would cause a lot of additional anxiety at the point of disbursal. And we are supposed to use it in the ICP ecosystem. This proposal is certainly a headscratcher !!!
3 Likes
If anything we want to incentivise spending ICP in the ecosystem as much as locking up ICP via staking. The ICP ecosystem will not grow with out increased “spending” The suggested changes do not address the issue of “spending” ICP in any way so I think they need to be reworked with that in mind.
3 Likes
To all on this thread: please see Request for feedback: Compounding Maturity Proposal and let your ideas be known!
2 Likes
Something that should worry everyone at this point is inflation more than tax advantages, Kyle Langham estimates that a price around $3 is very dangerous because it could produce an inflation death spiral. If we’re entering a bear market and BTC has a few more dumps to take we’re going to be very close to that price range. For the sake of the whole network a dynamic APY or some other solution would have to be implemented in that case anyway, even if a death spiral doesn’t happen inflation will get out of control. Maybe a reduced ICP reward can translate into community fund rewards, let’s say we get larger community project token allocations or through some other advantage that will mean a big reward once the network gets out of the hole it fell in.
If the concern is the arbitrariness of the APY we can just make APY a function of price, less price translates to less APY, as price “normalizes” so will APY go back up to where it would be.
4 Likes
APY is not the main reason why the $3 price could mark the beginning of a death spiral in Langham’s analysis. Because the total amount of ICP produced through APY is not a function of ICP’s price. The problem lies with node rewards which are paid in fiat-equivalent ICP. Therefore, the lower the ICP price, the more ICP needs to be minting to pay node providers. If the network keeps expanding and price keeps falling, it could lead to an inflationary spiral that can only be stopped by curtailing the number of nodes, which itself would be a huge setback.
I do agree that long-term inflation is an issue and it would be good if Dfinity finds ways to curb it.
4 Likes
Death spiral would be stopped with a well funded and diversified Dfinity Treasury. Each month, additional ICP could be minted and converted to stable coins. This stable coin treasury should be used to be node providers if ICP price goes below a danger threshold. Stable coin emergency treasury should increase or decrease as a function of the risk to reach the threshold. mmm… actually this could be a governance proposal.
5 Likes
Also important to consider that the more nodes we have the more cycles we can potentially burn. Cycle burn should offset node provider rewards, so I don’t see where the inflationary death spiral would happen if nodes are always burning more cycles than they’re receiving in equivalent ICP. Of course we have to switched over to the fully automatic renumeration yet. Does anyone have insights on this part of the inflation equation?
3 Likes
I may be wrong, but I believe there is a long list of nodes that have been on standby since genesis. I don’t think node providers would invest in equipment to be on standby without getting paid. Therefore, I suspect deploying new nodes will not cost more than what is already getting paid out. This is my assumption. I don’t know this as a fact.
1 Like
I understand. Thank you
Great question. I heard about the IDL nodes. One of the team members confirmed we have ~800 nodes on IDL, just waiting to join the network: see:@Luis-comments.
Question: Are those 800 IDL nodes being rewarded?
(in case Yes: Onboarding those new nodes will be fine. In case ‘no’ + ‘ICP price goes <5 $USD’ Onboarding new nodes might be a problem regarding inflationary pressure along 2022-2023.)
I hope someone could clarify about that. @Luis
Why/how would they be rewarded if they arent on the network and contributing to the protocol?
Hi @JaMarco , that is a relevant point, and it is why I am curios about that. To know if the IDL nodes are being rewarded or not.
For nodes to be IDL -and ‘ready to connect to the network’- it would mean that:
i) the nodes are already bought (a large sum of money is vested in hardware that is currently IDL.)
ii) If the nodes are ready to join the network, it means that they are deployed in a Data Center. Those nodes providers are already paying stack rent/fees and probably already paying for secure/high-speed connections made available to them. And still, they are IDL.
If it is the case that Dfinity decides to keep them IDL, them -maybe- there might be a compensation.
1 Like
Yes, all nodes from the initial set of node providers are rewarded from the point in time where the node provider proved that they are racked and stacked. For the initial set of nodes this was the only way to convince node providers to invest. Just as a reminder, the thousands of interested node providers came after launch. Before launch you needed to be very brave to invest in something that sounds undoable and doesn’t exist yet 
We will know better at the end of this year, but it already looks like we estimated the number of needed nodes for the first 18 month very well. It gave us enough capacity and time to develop the operational maturity that is needed to grow into the millions of nodes (if needed).
2 Likes
Operational maturity to support millions of nodes is definitely one angle. However the most important angle is : what would we do with millions of nodes?
Currently we are not using even 10% of installed capacity (all of 500 nodes). So our plan was to grow a 1000 fold in compute capacity over the next couple of years. Currently we ARE NOT EVEN CLOSE in demanding a 1000 fold increase in capacity. So why are we not actively changing our plan in response to the business?
Edit: i.e. virtually all such hosting contracts have an contingency built in…“if we are unable to use these nodes in a timely fashion because of business reasons, we reserve the right to terminate such nodes…”…if you do, investigate whether we need to execute that clause.
3 Likes
Well the problem is nobody uses the network. You dont need many nodes. You dont need many blocks per second. I am not seeing any kind of development on the IC that would suggest that dapps, projects are launching on the chain. like zero. The only number that maters is how much ICP gets currently burned. vs. currently minted. the divergence between these numbers is astronomical.
So before making complicated solutions like a treasury or what not, just get the fucking number up of apps and people that use this ghost chain. After that decentralizing the nodes is a must too.
Besides that I disagree on Dominics proposal regarding governance staking and tax. totaly disagree.
He should also stop shitting on other projects (its partially a reason why nobody uses the IC) and deliver the product we have been waiting for for years.
1 Like