NXT: A Cybernetics Perspective V 0.1: Proof of X

31 December 2013

NXT: A Cybernetics Perspective V 0.1 Proof of X Abstract Systems (like NXT) can be analysed to determine what they are capable of, beyond their intended application. Every system has some accidental or unintended features. The following is some preliminary research about the constraints and possibilities of the fundamental functionality of the NXT system.

Economic Measures: Proof of Stake, Proof of Work, Proof of X Protocol Level Proof of Stake (POS) is the fundamental activity which secures the NXT network. It imposes a positive economic incentive on a user to engage in forging and to otherwise increase their holdings by additional investment in NXT tokens (coins). POS is implemented at the protocol level of the NXT system. Analogously, Proof of Work (POW) is the fundamental activity which secures the Bitcoin network. It imposes a positive economic incentive on a user to engage in mining and to invest in ever more powerful mining hardware. POW is implemented at the protocol level of the Bitcoin system. Of course, a user does not have to engage in forging or mining in their respective systems, despite the positive economic incentive to do so (particularly if they deem the incentive to be too small or the associated investment to be too great). Ideally however, from a security perspective, everyone should engage in supporting the network. The positive economic incentive is exacerbated with an increase in price of the tokens/coins in both systems. This can be a seemingly derogatory situation in the early stage price discovery of the tokens of a new system. Both the POS and POW schemes have a pressure motivating users toward one course of action only (at least at the Protocol Level). I.e. there is no incentive to directly motivate a user to “not-forge” or to “not-mine”. As we will see in the “Antagonistic Pairs” analogy later on, this would be like a human arm having only a bicep muscle and no tricep – so the functionality and control of the arm is hindered.

Application Level Other schemes beyond POS can be implemented at the application level of the NXT system. These other schemes, Proof of X (POX, where X is some variable), need not be utilised exclusively to secure a network. POX can be utilised for some other representative purpose. E.g. POX could be utilised for proportional representative voting, allocation of dividends, application of user ratings in various categories, lottery reward distribution, macro-economic influence of token distribution, manipulation or stabilisation of token price etc. Such POX schemes will have different levels of accountability and trustworthiness (discussed below) which will affect their practicality, application, scope and usefulness.

1

31 December 2013

What is X? A facetious answer is that X can be anything. The following non-exhaustive list might highlight some of the more obvious or relevant substitutions for X with regard to the NXT system.          

Proof of Work Proof of Stake Proof of Burn Proof of Isolation Proof of Excellence Proof of Contribution Proof of Support Proof of Charity Proof of Raspberry-Pi-Deployment Proof of …

An important factor of what X is, is whether or not X can be verified. And how unambiguous and trustless can the verification be. Can Proof of Work be verified? Yes, for certain types of work. Can Proof of Support be verified? Yes, for certain types of Support (e.g. total uptime supporting the security of the NXT network, regardless of block reward probability). A Proof of X scheme should be expensive to the user involved (e.g. expense of time and/or money), but it should also be cheaply and easily verifiable by every other user on the network.

Strong POX The following two POX schemes are important in the NXT system because they are unambiguously algorithmically verifiable. Verifiable “Proof of Burn” Proof of Burn (POB) is a scheme where a user destroys coins (or equivalently renders them un-usable or inaccessible). In the NXT system, one could burn tokens/coins by sending them to a verifiably inaccessible account. Thus, NXT coins/tokens can be entirely irreversibly removed from circulation. This must be implemented carefully as the action cannot be undone (or if it can be undone then the security of the entire NXT system will have already been undermined). A verifiably inaccessible account can be generated in a number of ways. A simple way would be to gather entropic data from multiple users and combine it to form a new account number. E.g. asking 10 NXT users for single random numeric digits might result in receiving the numbers 2,4,2,6,7,2,8,1,7,5. Now simply send some NXT coins/tokens to account number 242678175 and they can be considered “verifiably burned”. Verifiable “Proof of Isolation” Proof of Isolation can be defined as verifiably removing NXT coins/tokens from the potential forging pool. This can be done by demarcating them into Silos. Another way to conceptualize Proof of Isolation is as Proof of “Burn-of-Stake”. The stake is effectively being waivered or burned. A Silo is simply an account/address which contains the isolated coins/tokens. Proof of Isolation can affect the “AVG_BASE_TARGET” value/calculation by making it more probable for every other user to forge a block.

2

31 December 2013

There can be different types of Silo: Strong Silo This is an account with 1 input and 24 hours (1440 blocks) or less elapsed time since that input and no outputs. It can be verified that the account cannot forge any block for at least a 24 hour period after it received its input. This is a feature implemented at the protocol level of the NXT system. The coinage contained in the Silo can be quantified as the number of coins in the silo multiplied by 1 day. 10 Million NXT in the Silo is 10 Million coin-days. To lengthen the isolation period of a certain amount of funds they could be moved from one Silo to another within the 24 hour (1440 blocks) time period. This would incur transaction fees each time the funds are moved or cleared. The coin-age of funds moving between Silos could be given as: Coin-age = (Number of NXT x Number of Days) - Transaction Fees Weak Silo This is an account with 1 input and more than 24 hours (1440 blocks) elapsed time since that input and enough funds such that “a block should statistically have been forged by now” and no outputs. It can only be verified that it has not forged a block (even though it should have), it cannot be verified that it is truly isolated from forging.

The single input constraints of both the Strong and the Weak Silos could be relaxed. However, accounts which have no outputs (outgoing transactions) are not made known to the NXT blockchain until they make their first outgoing transaction. This could make it computationally intensive to search the entire blockchain for multiple inputs to the same address. An outgoing transaction could be sent to the entity who requires Proof of Isolation from the Silo account, to identify the Silo and consequently to make the Silo known to the blockchain allowing for additional inputs to compound the isolated funds and presumably increase the earned reward (economic or otherwise).

3

31 December 2013

Weak POX The cases for the following POX schemes are not as strong as the previous two. However, they highlight some use cases of the NXT system at the extremes of its constraints. Proof of Time-Locked Cold Storage Time-locked Cold Storage Accounts could be implemented within the NXT system in a number of ways. For example, to lock an investment away for a certain period of time the passphrase for the account in which it is stored could be chosen in such a way that it would take an approximately known length of time to brute force the passphrase by someone who does not already know the passphrase. Proof of Work Consequently, for those nostalgic for mining, a POW scheme could be constructed to determine the passphrase for and accordingly unlock the time-locked cold storage accounts. The criteria of the passphrase would determine the passphrase mining strategy and ultimately the type of POW scheme employed. Proof of Excellence A true Proof of Excellence scheme ought to involve producing some quality output which was previously unknown to or previously unachievable/unattainable by others. This is difficult to implement and to verify, however. An approximation of Proof of Excellence could be implemented in the form of Proof of “Pre-Defined-Excellence”. Scheme 1: In this case an actor could solve a difficult problem and store a bounty in an account with a checksum of the solution as the passphrase to the account. Then the actor could set a challenge for another user to replicate the solution of the problem to retrieve the bounty from the account. Scheme 2: In a related case, the user could perform increasingly excellent work to determine one passphrase after another in a chain of tasks. This could involve designing new algorithms to solve problems in ever more intricate or more elegant ways. Temporal elements could be added to these aforementioned schemes to achieve a truer Proof of Excellence implementation. For example, in Scheme 1, the reward could diminish over time. In Scheme 2, a race between competitors could be organised where the victor earns the entirety or proportion of the reward. A key point to note here is that, in the case of Pre-Defined-Excellence, the solution and its corresponding checksum are what unlock the account containing the bounty. The method of solution might be wildly different to the initial reference solution. Consequently, new and potentially better approaches to solving real world problems could be discovered.

4

31 December 2013

Economic Measure Antagonistic Pairs The following is a useful (but imperfect) analogy to highlight the practicality of having opposing economic incentives. In muscle physiology, an antagonistic pair occurs when a set of two muscles have opposing effects. An example is the bicep and tricep pair of muscles of the upper arm region of many animals (e.g. humans). An "antagonist" is a muscle that acts in opposition to the specific movement generated by the agonist and is responsible for controlling the motion, slowing it down and returning a limb to its initial position.

An example of the potential outcome of an antagonistic approach in the Bitcoin system: Suppose there are just two miners mining/supporting the Bitcoin network. Miner 1 is mining at a rate of 1GH/s and Miner 2 is mining at a rate of 4GH/s. For every 5 blocks, statistically Miner 2 should mine 4 of the blocks and receive 4 of the block rewards and Miner 1 should mine 1 of the blocks and receive 1 of the block rewards. (This is CASE 0). A simple agreement between all miners (in this scenario, just two) to reduce their mining rates by one million fold to 1KH/s and 4KH/s respectively would statistically result in the exact same block reward being yielded by both parties. (This is due to the difficulty adjustment of the Bitcoin system.) Unfortunately a simple agreement like this requires trusting individuals. This would be especially problematic when many more than 2 parties are involved. Some means of verifying that work is not being carried out at full capacity would be required here. This could be termed Proof of Inaction. CASE 1: Proof of Inaction could periodically determine the miner’s maximum output and then calculate their nominal output as a percentage of their maximum output. Miner’s could then be rewarded for mining up to some upper bound determined by a scaling factor (say, 1 millionth) of their maximum output. While this hypothetical scheme does not mitigate against the high investment cost of mining hardware, it does greatly reduce the nominal running costs of supporting the network. This scheme does not necessarily encourage more users to participate in mining. CASE 2: Proof of Inaction could periodically determine the miner’s maximum output and then calculate their output as a percentage of their waivered output. Miners could then be rewarded for “not-mining” down to some lower bound determined by an absolute mining power – say 1KH/s. I.e. the miner’s share is not determined from 0H/s up to their maximum output, but from their maximum output down to 1KH/s. While this hypothetical scheme does not mitigate against the high investment cost of mining hardware either, it does greatly reduce the nominal running costs of supporting the network AND this scheme may encourage more users to participate in mining as the barrier for entry is small. Miner 1 Output

Miner 1 Ratio

Miner 1%

Miner 2 Output

Miner 2 Ratio

Miner 2%

Total Network Power

Pure POW Case 0

1GH/s

20%

4GH/s

5GH/s

1KH/s

20%

4KH/s

4GH/s / 5GH/s 4KH/s / 5KH/s

80%

Proof of Inaction Case 1

1GH/s / 5GH/s 1KH/s / 5KH/s

80%

5KH/s

Proof of Inaction Case 2

1KH/s

≈0.99GH/s / ≈4.99GH/s

≈20%

1KH/s

≈3.99GH/s / ≈4.99GH/s

≈80%

2KH/s

5

31 December 2013

The following Economic Measure Antagonistic Pairs have been identified in the NXT system: Pair 1 1. Proof of Generation 2. Proof of Burn

(POG) (POB)

Proof of Burn is the Antagonist to the Proof of Generation’s Agonist. Every complete system should be prepared for the possibility of its own demise. At the time of demise the system should fail gracefully. In the case of the NXT system, all NXT tokens could be verifiably burned to reduce the effective usefulness of the system towards zero. It would be prohibitively expensive to burn all the NXT tokens as this would most likely require one individual or organisation to purchase all of them. Pair 2 1. Proof of Stake 2. Proof of Isolation

(POS) (POI) or alternatively Proof of “Burn-of-Stake”

Proof of Isolation is the Antagonist to the Proof of Stake’s Agonist. A hybrid POS/POI scheme could be implemented in the NXT system with little or no change to the system. Incentives (economic or otherwise) can be devised to reward those actors who isolate NXT tokens from the potential forging pool. The purpose of a POS/POI hybrid scheme is not to prevent forging or to stifle decentralisation. Instead it is an attempt to avoid the pure POS scheme running out of control. Free market behaviour would allow users to perform their optimal strategy or ratio of forging and isolating to maximise their return. Pair 3 1. Proof of Work 2. Proof of Cold Storage

((POW) with the possible inclusion of Proof of Excellence (POE)) (POCS)

Proof of Cold Storage is the Antagonist to the Proof of Work’s Agonist, in this particular scenario. As outlined earlier, this pair of schemes could enable practical real world problems to be solved in new and exciting ways.

6

31 December 2013

Implications The implications of additional Proof of X schemes being employed in the NXT system are far reaching and may be intractable: only time will tell how they develop. Some of the implications are as follows:   

It is possible to manipulate the total number of tokens in circulation It is possible to incentivise against forging (by offering a better alternative) It is possible to incentivise for isolating

Proof of Burn – Possible Applications:    

Destruction of the NXT network by burning all the NXT tokens Incapacitation of the NXT network by burning a lot of NXT tokens or a number of specific tokens (particularly when fungibility is reduced with the introduction of the coin coloring and asset definition features) Migration from one blockchain to another Proportional representation for a wide range of applications

Proof of Isolation – Possible Applications:   

Reduce the security risk associated with a large number of tokens at stake in a hot wallet Reduce the upward pressure on the Proof of Stake scheme Proportional representation for a wide range of applications

7

31 December 2013

Summary Ultimately the wider NXT economy and its actors will decide what to do with any Proof of X scheme (if anything) at the application level (or indeed at the protocol level if there is sufficient motivation to build-in another fundamental POX). It would be interesting to embed the Proof of Isolation scheme deeper into the protocol level to successfully counterbalance the current Proof of Stake economic measure monopoly. Exploring the available economic measure antagonistic pairs may greatly enhance the longevity of the NXT system. It is a testament to the NXT project, developers and community that the NXT system can inadvertently harbour such immense extensibility and flexibility in addition to all the deliberate and powerful features it inherently possesses. If Bitcoin has been described as decentralised programmable money, NXT then can be described as something akin to a decentralised programmable economy.

8