Internet Computer is the world's first fully adaptive blockchain. The Internet computer network, and the special node machines that host the network, operate under the complete control of the Network Nervous System (NNS).

NNS is a decentralized token governance system that is completely non-accessible. Anyone in the world can submit a proposal to NNS. If the proposal is voted and adopted, it will be executed immediately. The whole process is completely automatic, enabling the network to adapt and develop in real time.

NNS can perform tasks at any time, such as upgrading node machines toUpdate protocols, safely fix applications, adjust economic parameters, or create new subnet blockchains for scaling. It operates within the protocols of Internet computers and can perform the above-mentioned upgrades and modifications without stopping the operation of the blockchain or compromising security.

The NNS network nervous system allows users to use ICP governance tokens to create voting neurons. Anyone can create a neuron, and we expect tens of thousands of neurons to be created after Genesis, and together they will express the will of the community, mediated through algorithms.

A neuron is like a savings account, it sets an exit period, the length of the exit period is used"Delayed dissolution"to configure. The voting rights of neurons, and the voting rewards they can obtain, are related to thenumber of ICPs,"Delayed dissolution"The length of the neuron is proportional to the "age" of existence.

Neurons can vote manually or automatically, which is afollow), rather than delegates, neurons can automatically follow other neurons in a form of fluid democracy.

Neuron holders are placed in the game of encryption economics. The system motivates the behavior of voting "pass or reject" the proposal. Automatically vote in an ideal way, thereby driving the value of ICP to increase in the long run.

secondary title

Governance overview

The purpose of NNS is to allow Internet computer networks to be managed in an open, decentralized and secure manner. It can completely control all permissions of the network.

For example, it can upgrade the protocol and software used by the node machine that carries the network; it can create a new blockchain subnet to achieve expansion; it can split the subnet to balance the network load; it can configure economic parameters, such as gas generation Coin Cycles to ICP conversion ratio; in extreme cases, it can even freeze malicious software containers to protect the network, etc.

The way NNS works is that it accepts proposals, and based on the"Neurons"votes to adopt or reject these proposals. Neurons are also used by participants to submit new proposals. After a proposal is submitted, it is either adopted or rejected, and the process ends almost immediately, or after some delay, depending on how the ensemble of neurons voted.

Each proposal is a specific"Proposal subject"instance, which determines what information it contains. For each topic proposal, NNS has a corresponding system function, which it will call whenever the topic proposal is adopted.

When a proposal is adopted by NNS, it fills the parameters by extracting information from the content of the proposal, thus calling the corresponding system function. Proposals for each theme belong to a specific"or",like"#NodeAdmin"or"#NetworkEconomics", the proposal subject determines the details of how the proposal is processed. In order to prevent users (neurons) from spamming proposals to NNS, if the proposal is rejected, a fee will be charged to the neuron that submitted the proposal.

destroy"destroy") neurons to unlock.

Users are encouraged to create neurons, as they can earn ICP rewards when voting on proposals. The reward comes in the form of the distribution of new ICP minted by NNS. The amount of ICP reward assigned to a neuron is derived from the following factors:

• How much to lock the ICP

• The remaining minimum lock-in period ("Delayed dissolution"）

• How long a neuron exists

• The proportion of votes it participated in, and the sum of the voting activity of all neurons

At any time, each neuron has a currently configured"Delayed dissolution". This determines if the"dissolution mode"Next, how long will it take to dissolve. Once a neuron enters"dissolution mode", its dissolve delay decreases over time, like a kitchen timer, until it reaches zero, at which point its owner can perform a final payment action to unlock the ICP.

For neuron owners who want to get the most value on ICP,"Delayed dissolution"Create a reasonable economic incentive for them. Neuron owners are freely configurable"Delayed dissolution", with a maximum delay of 8 years. Once created, there is no other way to speed up the dissolution except for waiting for the natural passage of time."Delayed dissolution"The higher the NNS, the higher the voting rewards paid by NNS, which encourages users to join the economic game. In this game, a long-term economic incentive mechanism is created, and users will vote for governance according to a very long-term vision.

When governance affairs are complicated, neuron owners may find that they do not have to manually vote directly on every proposal submitted to NNS: First, a large number of proposals will be submitted to NNS, and most neuron owners may not have time to evaluate each proposal; second, neuron owners may lack the necessary expertise to evaluate proposals.

NNS uses a method known as"Liquid Democracy"form to address these challenges. You can set follow rules for neurons, and for any proposal, you can automatically follow the votes by following the votes of a group of neurons. It is also possible to define a universal lazy follow rule, even for proposal topics that have not been set, neurons can automatically follow the votes and get rewards. It is also in their own economic interest to assume that neuron owners have the best interest of the network in managing how their neurons follow other neurons, as they lock up ICP tokens.

It is expected that a significant portion of the total ICP supply will be locked in governance neurons for rewards. This ensures the self-management of internet computers as it makes it impossible for an attacker to gain a large enough stake. Since neuron owners may wish to maximize their rewards by voting on all proposals, most neurons will be actively managed, or configured to follow other neurons so that they can vote automatically.

In practice, once a followed neuron votes on a proposal, a majority of other neurons will also vote because of the follow relationship. This means that the NNS can usually quickly and quickly determine whether a majority of the overall voting power represented by all neurons wants to adopt or reject a proposal, and decide on the proposal accordingly.

secondary title

ICP Token

ICP is a utility token native to the network and plays three key roles in the network:

• Promote network governance
  

ICP tokens can be locked to create neurons, and by voting to participate in network governance, ICP issuance rewards can be obtained.

• Produce cycles for computation
  

ICP can be converted to"cycles", which acts as a gas token to power calculations and is burned when used. NNS will dynamically adjust the ratio of ICP to cycles. This choice is to ensure that network users can always create new cycles at a near-stable actual cost, so that the cost of obtaining gas is stable and predictable.

• reward participants
  

The network mints new ICPs to reward those who undertake important work to enable the network to function, including: providing"voting rewards"; Provided to the service provider running the node machine"secondary title"。

ICP Ledger

account"account", which has two fields (i.e. has two"List"）：

• account identifier
  

bytes, from"control"A unique value derived from the identity of the controller of the account. There are currently two types of controllers: the owner of the key pair; the smart contract (container) that is part of the NNS. The account identifier is derived by concatenating the delimiter of the hash field, the principal ID, and the subaccount (or zero if no subaccount is given).

• balance
  

Positive integer, the minimum unit is one millionth of ICP, and the balance is the amount of ICP held by the account.

send

• send
  

notify

• notify
  

When the recipient of the funds is an account of an NNS container (for example, the account of the governance container), the sender can ask the ledger to notify the receiver of the transfer received by the container. The recipient can then act upon this notification. Two examples of using this capability are creating a neuron and refreshing a neuron's stake. These are described in detail below.

Create neurons

• Create neurons
  

When the controller is the public key holder, they can lock a portion of their balance inside a new neuron. Technically speaking, creating a neuron is done in two stages: first, transfer the ICP to be mortgaged to an account in the governance container (corresponding to a new neuron); then notify the governance tank of the transfer received, and the governance tank will update the records of its internal neurons. In order to transfer these ICPs to a different account, say back to the original account, where they can again be controlled like normal balances, the associated neurons must be completely dissolved. A new neuron that is created is controlled by the private key of the delegator who created it.

• refresh pledge
  

Token Economics

Token Economics

Neurons provide the opportunity to earn rewards by participating in governance. Rewards are distributed to those voters in the form of increased neuron maturity, which eventually produces new neurons that contain additional ICPs. However, the overall economic benefit of acquiring a new neuron also fluctuates with the value of the locked ICP balance.

In order to maximize revenue, neuron owners have a strong incentive to first ensure that their neurons participate in every vote so that they receive the greatest possible voting reward, and second to judge what proposals will best promote The growth of the overall value of the network, and then vote on the proposal.

Dissolution delay

When someone wants to sell a locked ICP balance, they will benefit most if it reaches its maximum possible value at the exact moment in the future when it can be unlocked and sold. Neuron owners will gain the most from the long-term value growth of the network if they vote with a long-term view to maximize the value of the network in the future. For this reason, NNS incentivizes neuron owners to make the dissolution delay as long as possible by distributing larger rewards to neurons.

Since the votes of neurons' owners are more useful in decision-making when their owners have long-term horizons, the NNS also gives more weight to votes of neurons with greater dissolution delays than neurons with dissolution delays of less than six months cannot vote at all.

Of course, since locked balances are transferable, this scheme is less beneficial to the network, as it would allow neuron owners to choose"sell their neurons", even if they have to discount against the unlocked balance.

51% Governance Attack

A key security concern is preventing an attacker from gaining 51% of the voting power, or bringing in unwise voters, to successfully compromise the network. (here"attacker"The term is equally applicable to voters who wish to harm the network, voters who have unintended consequences for bad results, and voters who may simply have over-centralized power. )

secondary title

Calculate voting rewards

Projections suggest that 90% of the total ICP may be locked up in neurons. Regardless of the current level of lock-in, a fixed amount of ICP mint rewards are allocated so that participants will receive greater rewards until the participation rate reaches 90% and the market is able to convince those who are not currently participating to participate.

algorithm

algorithm

A neuron's maturity starts at 0 and increases with voting activity. When the maturity of a neuron grows beyond a certain threshold, then it can generate a new neuron containing additional ICPs and then reset its own maturity to zero.

The number of additional ICPs issued in new neurons was expected to be equal to the ICP locked in parent neurons, factoring in the maturity of the parent neurons. For example, a neuron containing 100 ICPs with a maturity of 10% can generate a new neuron containing 10 additional ICP tokens. The dissolution delay of newly generated neurons is only one day, and the ICP locked inside can be easily retrieved if desired.

There is an approximate equivalence between the maturity of a neuron and the voting rewards it has collected that have not yet been withdrawn by spawning a neuron. (This equivalence relationship is approximate, because maturity only determines how many ICPs will be in the spawned neurons, because"give birth"operation with a degree of uncertainty).

Every 24 hours, we have to calculate how much to increase the maturity of each neuron participating in the vote. We set out to calculate the maximum number of ICPs that could possibly be created and distributed as a reward, which would be reflected in an increase in neuronal maturity. Once we have this amount, we can calculate how much of a relative share of reward each neuron should get when taking into account things like the number of locked ICPs, configured dissolution delays, and age.

We derive the maximum amount of ICP that can be minted and distributed from the current ICP supply and the number of days since genesis. First, this is equal to 10% of the ICP supply divided by the number of days in a year. Over eight years, that number drops to 5%. Note that voting rewards may not be halved in practice as the supply of ICP may grow or fall during this time.

• 10% increase in the first year
  

• 5% issuance rate in eighth year after Genesis, unchanged thereafter

• The rate of issuance is a quadratic function of time

Call the additional issuance time G, and at any time t between G and G+8y, the total reward R(t) at time t is given by the following formula:

R(t) = 0.05 + 0.05[ (G + 8y – t) / 8y ]²

R(t) = Rf + (R0 – Rf)[ (T – t) / (T – G)]²,

where R0 is the initial ratio (10%), Rf is the final ratio (5%), and T is the time when the ratio levels off (G + 8y).