I have seen the problem, we should reconsider the role of node providers, they are also investors, they also need to take risks, not full-time staff guaranteed 100% profitability, we hold ICPs have no obligation to support them!
1 Like
You guys have been talking about this for days. How will a route be decided for one or multiple of these solutions?
I see a bunch of great quick patches and overall restructure of the chain solutions but when will someone submit a proposal? Then if dfinity is against everything being discussed here then thereās no shot in fixing these underlying problems.
Or are you guys just waiting for Dom to show up and give a response on his opinions cuz thatās unlikely.
While itās not likely Dom gives his opinion on this topic, itād be great to have some degree of feedback regarding solutions from the foundation, given theyāve designed these mechanisms.
With that being said, Iām personally waiting for a consensus to be reached on the forums before creating a proposal.
2 Likes
I like the following
voting rewards being locked to the neuron is fully dissolved idea.
Increase of transaction fees
Either remove unused node providers or have a fixed rate of icp that can be earner or decrease the amount of rewards they receive.
According to this thread node providers receive basically a 2x every month guaranteed so maybe decrease the rewards or a fixed monthly pool where the node providers get paid out based on how much they were used that month.
Voting rewards being locked til fully dissolved will remove a lot of sell pressure.
1 Like
I understand that dfinity wants their service to be cheap but weāre at a point where users are losing hope. If thereās no buyers and everybody loses faith in this project due to being so much in the red then everybody loses apart from obviously the VCs whoāve already sold.
I really like the idea of adjusting rewards based on usage. Iāve been mulling over a cool space ship metaphor to understand this better, and this is what Iāve come up with.
If we liken our network to an advanced starship hurtling through the cosmic void, we quickly realize that its survival depends on a harmonious balance of systems: propulsion, navigation, life support, and so forth. Each subsystem functions in an inter-dependent cycle of energy conversion, resource allocation, and dynamic responsiveness.
Propulsion: The Inflation Dilemma
Letās start with the crux of our problem: inflation, or in our starship terms, the āengine thrust.ā We aim to reach optimal velocity without overheating the engines or expending too much fuel. The challenge is to modulate thrust (token minting) so as not to lead to thermal runawayāakin to the risk of our engines exploding or propelling us into an uncontrolled trajectory.
The central issue weāre facing now is the looming risk of inflationāor, borrowing from our metaphor, thermal runaway. Just as thermal runaway can damage a system by allowing it to become too hot, excessive token creation can dilute value, reducing the āenergyā (value) each token can contribute to work. To counteract this, we could implement safeguards akin to thermal cutoffs, where the token minting rate is dynamically adjusted based on system demand, much like how a heat engine adjusts its fuel intake.
Navigational Systems, The Shipās Officers, and Bridge Crew: Governance
To mitigate unintended consequences while altering the core engine thermodynamics, we need robust feedback mechanisms. If weāre dealing with an intricate vessel, the feedback loops act as governors or regulators that adjust the system parameters to prevent failures. In a similar vein, governance decisions should be tied to real-time metrics, offering a āliveā view of how changes in node rewards or transaction fees affect the overall āthermal equilibrium.ā
The question of node provider costs, isnāt just about minimizing outlays. The reports ought to include information on how much āthrustā (computational value) each āengineā (node) needs to align our vessel on the desired trajectory. This forms a crucial part of our thermodynamic calculations. These node providers should be remunerated in a manner thatās directly proportional to the computation need from them. While all nodes should be compensated enough to at least break even (keep the engine warm), nodes that arenāt needed to perform as much work shouldnāt be getting the same amount of āfuelā as nodes that are running at maximum capacity.
To achieve this, automated reports need to be generated, and a team needs to be assembled whose job it is to conduct continuous in-depth analysis of the financial state of the network and then assemble periodic (say, weekly, bi-weekly, and/or monthly) reports for the voters to read and base their decisions on. Small changes can be automated, similar to how the network automatically adjusts cycles cost based on XDR, while larger changes can be deliberated upon and put to a proper vote.
Life Support and Mess Deck: User and Developer Onboarding
The life support system is essential for keeping the crew alive, and a welcoming environment on the mess deck is crucial for crew morale. This isnāt just about keeping the air breathable (simplifying onboarding with better documentation and tutorials); itās also about making sure that the onboard environment cultivates innovation and growth (through educational programs, grants, and hackathons).
Shields: Transaction Fees and Service Costs
Just as the starship has defensive shields to guard against cosmic threats, the transaction fees and service costs act as protective measures, creating an economic boundary that filters out āspamā transactions or low-value contributions. These shields canāt be too strong, or else they could obstruct important supplies or allies from coming onboard (deter new users and developers). Too weak, and you leave your starship (the network), vulnerable to all kinds of cosmic mischief. Hence, the shieldās parameters need to be a subject of ongoing governance, fine-tuned based on active threat assessments, user experience, and the networkās current ābattle conditions.ā
The shields also act as an absorber of excess energy production (inflation), so more power can be routed to the shields (up transaction and service costs) when more energy needs to be dissipated, as long as weāre mindful of the side effects and balance the energy accordingly.
Communication Nexus: Inter-chain Alliances and Strategic Partnerships
Think of strategic partnerships as forming alliances with other civilizations or trading outposts. These alliances could offer critical supplies, advanced technologies, or military support. In our case, alliances with other blockchain platforms will provide new users, developers, and increased computational usage, while affording our partners advantages as well.
Summary
To sum it up, these changes should be made
- Scale node rewards based on usage, with a minimum so that they at least break even.
- Establish automated feedback mechanisms to make adjustments, and a team to assemble reports for the voters to base decisions on.
- Increase transaction fees and service costs dynamically when inflation is too high, while being mindful of the consequences of doing so.
- Improve documentation and make tutorials to help onboard new users and developers. Hold more frequent events like hackathons and boot camps.
- Start reaching out to other blockchains and popular projects on those chains to increase demand.
4 Likes
We canāt just lock node rewards in a neuron. I agree we should do that, but we have to cap node rewards too. If not, excessive amounts of voting power is transferred to Node Providers and instead of having run away inflation result in the price dumping, we get run away inflation resulting in everyone whoās not a node provider being diluted.
And the minute we impose a cap on the amount of ICP node providers can receive, we have to impose some sort of system for tracking and repaying the missed value that node providers havenāt received as a result of the cap.
I do agree, itās time to reach a consensus and begin drafting a proposal. Dfinity has demonstrated no appetite for participating in this discussion. Itās time to act.
2 Likes
We may need their input, their ānoā vote has killed decent proposals before
2 Likes
At the beginning, I invested ICP for profit. With the deepening of understanding, the NNS is the most exciting thing I have ever seenļ¼you can make real voting!I think this is one of the important ways to govern future societyļ¼especially for us who live in a dictatorship country, this is a hope, just like my usernamećI hope the NNS will become stronger and stronger.
4 Likes
As Iāve been following this thread it donāt think weāve identified a working solution that will evolve to consensus, though correct me if Iām wrong @Accumulating.icp; and my deepest thanks for introducing this topic with due amount of research and sensitivity.
The general sticking points for proposals so far is that messing with NP rewards is a bad idea, and no-one knows where money to make up the difference would come from. Iād like to introduce a more drastic, though simple solution that addresses the issue at a different ārootā.
By @Accumulating.icp estimates, inflationary sell pressure at current levels comes from:
Monthly N.P Rewards | 583,577
Monthly N.P Rewards Sent to Exchanges | 275,287.49
Monthly Disbursed ICP | 1,091,778
NNS neurons contributing 2-4X current NP rewards tells me itās closer to the ārootā, and we know from @Kyle_Langham that neuron dispersion will continue at ~1M+ ICP/Month for about the next year, and then drastically decrease. My interpretation of this is that we just need to get āover the humpā without having node rewards do irreversible damage to ICPās Total Supply.
Therefore my proposal is an NNS upgrade the locks dissolving neurons with a short dissolve delay while ICP is below a certain price (Iād recommend $5/ICP). Rewards continue normally, but dissolving neurons with <12month dissolution become locked until ICP rises above $5 again. I think maturity is insignificant enough to remain unchanged, but we can decide together if the community supports this idea.
Itās not the teamās place to initiate this, but I think there is finally legitimate precedent to stop the bleeding, and not for ātoken priceā, but so I can build with the peace of knowing this network will exist in a year. It will most anger the people that it best serves, and it will take all the fuel out of the fire that ICPās alleged āenemiesā might be looking to use towards something more permanent at this very moment.
Caveat: This is an absolutely horrible idea IF thereās still a significant amount of liquid ICP held by $0.03 round seed investors or other VCs. My impression was that they pretty much all got out, but I donāt know if this can be investigated/calculated.
7 Likes
Againā¦ this is debt restructuring.
New nodes should be warned that under a certain threshold there is a cap on $icp received. Cause they are investors!!!
Think of AWS or other cloud providerā¦ their equipment is part of the equity!! Thus, node providers should share losses!! Any other solution takes us to zero.
4 Likes
We are an extremely leveraged company right now and we are drowning on interest payments!!
4 Likes
The changes you mention here arenāt necessarily right or wrong. But we need to come up with changes that we can submit in a proposal and vote on in an effort to resolve the specific issue of run-away inflation as a result of Node Provider rewards specifically. The rest of the categories of focus that you mention will each require separate deliberation.
Here is a summary of necessary changes that Iām proposing in an effort to resolve the very specific issue of run-away inflation as a result of node rewards payments during times in which the price of ICP is depressed:
1.) a fixed fiat value is assigned as the amount due to be paid to node providers (we can discuss what we think this value should be. If it is less than what is currently paid to node providers, so be it.). Node Provider rewards are dispersed in ICP.
2.)There is a cap placed on the amount of ICP that may be dispersed as payment to node providers. (We can discuss what number would be appropriate as a cap)
3.) Any value that hasnāt been dispersed to node providers is tracked as 0% interest debt owed to the node providers by the network. This debt is later serviced when the price of ICP has risen enough to permit the NNS to be able to service the debt while staying within the ICP cap that we decide on.
This solution was discounted as a mere ādebt restructuringā and not a solution. I will now make my case as to why this is the better option amongst the other options proposed.
1.) It addresses the issue of run-away inflation as a result of Node Provider payment obligations. It does so by placing a cap on the maximum amount of ICP that is to be dispersed as payment to node providers.
2.) It addresses the issue of neuron stakers being diluted by rewards paid out to node providers. It does so, again, by placing a cap on the maximum amount of ICP that is to be dispersed as payment to node providers.
3.) It transfers a reasonable amount of risk from the NNS over to node providers. It does so by allowing the NNS to receive services from node providers, undisrupted, even when the value of ICP is too low to cover full costs to Node Providers while staying within the rewards cap. The NNS tracks the value owed to Node Providers as accounts payable. This is done at a 0% interest rate and with no hard deadline in which the NNS is obligated to repay node providers. Thus, affording the NNS time to allow the price of ICP to recover.
4.) It takes into account the fixed overhead overhead costs of node providers. It ensures that they are sufficiently compensated in a manner that preserves economic incentive for them to continue operations.
5.) It forces node providers to become investors in the NNS. Not by locking up their rewards into a neuron, leaving them unable to pay their overhead costs in the short term. Instead it mandates that node providers invest in the form of providing a 0% interest line of credit to the NNS in times of economic strain.
Iām open to other options. But if anyone does decide to propose another option, it should fulfill the following requirements:
1.) take into account the operation expenses of node providers in a way that doesnāt remove all economic incentive for Node Providers to contribute to the network.
2.) eliminate the risk of run-away inflation as a result of node provider rewards dispersals.
3.) eliminate the risk of all other neuron stakers being diluted by rewards paid out to node providers.
Before anyone comments (again) with āit doesnāt solve the root of the problemā, I want to reiterate that the only way to solve the root of the problem is to increase network activity and that is the job of the developers. What we need to achieve here is a system that allows developers to continue doing their work during periods of economic strain. The solution Iām proposing is meant to solve the issue of the ecosystem collapsing as a result of short term price depression. It is not intended to solve the overarching issue of insufficient network activity. Again the issue of network activity is to be resolved by developers in the ecosystem. Our one and only goal in this discussion needs to be to create a system that allows developers to keep developing during times of price depression, while preventing the NNS from collapsing.
6 Likes
Could you link to the product youāre working on (if itās for this)? Iām having trouble threading together all your replyās.
And just to clarify, my reasoning for not focusing specifically on the NPs is because their contribution is consequential to network stability/integrity, while angry neurons canāt do anything detrimental. Thereās already caps on NPs, but the reason they might fail is a direct result of NNS dispersion (I think), which can be easily tapered. But Iām out of my depth when trying to understand the position of node providers and think your solution is the best in this thread IF we found the perfect parameters and implementation (which seems much harder). Would love those NPs to jump back in here.
2 Likes
I completely agree that their contribution is vital. A cap on node rewards is also necessary. Thereās no silver bullet to this issue. Thereās going to need to be some give and take from both parties involved (the parties being the NNS and the Node providers). I think the the solution Iām proposing does the best job at acknowledging the nuance here.
Hereās a link to the post i made where i reference the product that Iām building to serve as a long term, sustainable solution to this issue: Node Provider Inflation Spiral - #54 by Jesse
To summarize, Iāve built an app that doesnāt require coding experience in order to own, operate, update or transfer. The plan is to give every consumer within the ICP ecosystem their own app and by doing that, it increases network activity by leaps and bounds. This removes the bottleneck in network activity from being tied to the number of devs in the ecosystem, to being tied to the number of users in the ecosystem.
To release this app will take me more time. Iām months away from an alpha launch and probably a year away from the product being refined enough to the point where itās ready to be distributed en masse.
There is reason to believe that network activity will increase enough to resolve this issue, but we need time to work.
4 Likes
I like this solution. I would vote for this.
2 Likes
Before we begin to backslide: If there are no further counter proposals, Iām going to start a new thread for the purpose of discussing the parameters of the solution posed in post that Iām currently replying to.
5 Likes
Thank you for your dedication.
2 Likes
I donāt think the amount of conversation on this topic is excessive - itās a complex problem with a lot of noisy data. Moreover, getting a reasonable, well-thought-out proposal is going to take some time, and putting something forward that makes us feel good but takes us in the wrong direction, could be devastating. One of the lessons that my business career drilled into me was to be really careful before messing around with incentives - you can really create a lot of ill will with stakeholders that is hard to recover from (years).
At this point, I am not sure we even understand what is driving ICPās recent price declines or that we understand who our investors are, and what causes them to buy, sell, or hold ICP relative to other assets. There are over 250 million tokens floating around unstaked and it looks like 5 to 10 million tokens trade on a typical day (200 million exchanges a month). Kyle seems to have the best data and at current levels, it is far from compelling in terms of explaining recent price behavior. Maybe but maybe not. (BTW USD have been inflating at 5 to 10).
I think that we can agree that we do know certain things and that certain actions might be helpful:
a.A The whole idea of staking is to give control of the network to those with a long-term interest and to reward that behavior. We have a strong preference for compounding rather than liquidation of rewards. Therefore a staking structure that incentivizes compounding vs. disbursing rewards would be better. Maybe full reward if you compound and half if you pull out rewards before maturity.
b. We need to burn more. It seems that the transaction fees could be increased and made variable based on the size of the transaction. Not clear that this will change pricing behavior but seems like a limited downside. Could be a baby step in the right direction.
c If ICP price drops too low, the model breaks down. There is some price level/rate of inflation where it causes a spiral. One way to deal with this would be to cap the maximum inflation of the network (e.g., 12 percent/year. declining over time or $x share) with a waterfall of actions that would kick in that prevent inflation from going above that - it could be either be deferrals or a shadow tracker per DavidF where the reward is paid out when either price hits a certain threshold or at some point the reward is unlocked at a future date.
d. Put a freeze on hardware/node providers for the time being.
But we need to be very careful - if we just clamp down and start changing incentives, we may generate exactly the opposite behavior that we want. For example, what if you change maturity and everyone hits the dissolve button? Or loses confidence in the network and stops investing and sells loose tokens? Or if you defer node rewards and then the node holders decide to dump 100 percent of their rewards and liquidate current holdings? Easy to get wrong.
8 Likes
The others points you mentioned are all valid. Each of them are intricate topics in and of themselves which will require extensive deliberation separately. The issue we currently need to zero in on is the one youāve listed that Iāve quoted here. This is the most pressing issue at the moment as it is the one with the greatest likelihood of causing a network collapse under current conditions.
I completely agree that incentives need to be thoroughly considered. This is why Iām opposed to the idea of forcing Node Providers to receive their payments as staked neurons. This could easily disincentivize node providers from contributing to the network as it would make covering operations in the short term prohibitively expensive for various node providers around the world. Node providers should maintain the autonomy to be able to assess and balance risks in a manner thats specific to their businessesā geographical and political conditions. Requiring node providers to accept their compensation as staked neurons would strip them of that autonomy.
The proposition I made is done with two goals in mind:
1.) Decentralize risk. Currently, risk is centralized to the NNS. If the price of ICP drops below a certain threshold, the result is an inflationary collapse of the entire NNS. If this happens, everyone is screwed, all at once. By placing a cap on inflation generated for node reward dispersals while tracking the balance owed to node providers and later paying that balance down as the price of $ICP rises, node providers are allowed the autonomy that they need in order to balance their own respective balance sheets in accordance to their specific geographical and political environment. On the other hand, if we force node providers to receive their rewards as staked neurons, it raises the already high barrier to entry for new nodes wanting to contribute to the network; anyone wanting to be a node provider would have to have enough money to run an entire data center at a deficit for however long it takes for their first neuron to unlock. This is a huge barrier to entry to becoming a node provider and Iām firmly against imposing it.
2.) Maintain sufficient incentives for node providers to continue to contribute to the network. Iāve considered the perspective of node providers and came to the conclusion that anyone whose willing to buy specific hardware to contribute to the network has at least some level of trust in the network. By implementing a system in which the node providers provide services to the network on credit during times of macroeconomic strain we presume that node providers trust the network enough to issue a 0% interest line a credit. This presumption is not all that farfetched and I assert that it is the least likely to negatively impact the incentivization of both current and future node providers.
I do believe its important for us not to rush a solution, but it is equally important that we do not find ourselves in an āanalysis paralysisā scenario at such a precarious time for the network. As you said, we donāt understand who our investors are or what motivates them to divest. Iām of the position that our investors are dangerously close to divesting if this issue is not contained ASAP. Weāve already seen the price of $ICP decouple from similar networks like Filecoin. My guess is this is the result of some investors losing trust in the network deciding to divest as a result. An inflationary death spiral could happen at any moment and it could happen fast. If you have a better proposal, by all means do let us know, but if not, I think its important that we begin working out the details on the proposals that Iāve made.
3 Likes