How to have a Proof of Stake reward system while leveraging the philosophical merits offered by Proof of Work (showerthought)
Disclosure: This is merely a showerthought I've had. This is not a protocol proposal for Nano. This is also not a protocol proposal for a separate cryptocurrency or unrelated ecosystem.
If you want to know my motivation behind this thought, it was in designing a second layer on top of Nano (which does not affect the core protocol or require any software changes) to perform IPFS-like and Ethereum-like functions using the native Nano coin. A full elaboration on this idea is not presented here, but essentially, it requires creating a new blockchain model that only mints it's coins when a Nano is locked up in return, and vice versa. Tokens, smart contracts, and distributed data storage would be the goal of such a scheme.
Preface: In the debate between Proof of Work versus Proof of Stake, the POW side of the argument focuses on morality and economic incentives, while the POS argument focuses more on pragmaticism and efficiency (also, the morality of environmentalism). The moral downfalls of POS however do not apply to Nano, because people running nano nodes to not make money doing so. If they did, then the following would apply:
1) There is no significant capital investment risk for POS (which encourages a planning element)
2) There is no capital depreciation for POS (which motivates timely action)
3) There is no significant power bill for POS (which mandates profit analysis or else monetary loss)
All three of these things is what makes POW, on the surface, a morally superior concept (aside from the environmentalism argument). There is a problem with POW however, all these things demands external resources which would otherwise be unneccesary, increasing the costs of usage on the blockchain. Increasing these costs just feeds the power companies at the users' collective expense.
The summarized disadvantage of POS is that it greater enables "the rich getting richer" than what POW would otherwise do. The loose rationale for this, is a lot more than simply merited, virtuous, and intelligent individuals are rich; many rich individuals are parasites, politicians, they receive large inheritances, they win their money through luck, accident, or by stealing. We don't want to enable thieves, cronyism, or the apathy of rich kids.
If we could simulate the characteristics of Proof of Work that make POW a mechanism by which only the wealthy whom has merit and intellect can accumulate wealth, then we would have effectively solved the moral problems of POS and the resource-wasteful problems of POW at the same time.
The Three Step Plan for a Meritocratic POS system
Step 1: Reward all transaction fees to not anyone, but to only the holdings of the participants who 1) Run a node with all of the necessary specifications and 2) only to the nodes which stake their coins in a Locked Capital Contract
There is an inherent incentive to staking a larger amount of money in normal POS systems, because this means you get a larger share of the reward. However, in this system, there is a counter-incentive at play, the more you stake, the more dedication is required in order to retrieve your old funds.
This is because the Locked Capital Contract doesn't allow you to retrieve your staked coins until you earn an equal amount of coins. So if you stake $1000, then you cannot retrieve your $1000 until you earn a new $1000. Instead of receiving it all at once, you can design this LC contract to give you, for example, 1 dollar for every dollar you earn (but at some point not less, to avoid unnecessary transactional traffic).
This essentially creates a virtual analogy to having to invest in additional capital equipment, which would have otherwise performed POW and given the monetary proceeds to the power companies and therefore likely governments.
Step 2: Fix a relative and dynamic fee to the staked funds within the Locked Capital Contract. Due to the problems with trying to invent a trustless "central clock" in the software engineering universe, your measurement of relative time can instead be transaction count. What you could do, is say, any node who operates slower than the average node gets a "virtual capital depreciation fee" on their staked funds.
A few different points to unpack:
'Work time' is just a conceptual placeholder for how fast and efficient your node is in reality.
Pinning an amount to this "virtual capital depreciation fee", or vcd fee for short, I've found to be rather difficult. So instead I propose a relative "virtual capital depreciation fee". (Given that 100% of the average 'work time' would cause no increase or decrease in your capital investment) having for example 99% of the average 'work time' could cause you to lose 1% of your (compound/continually recalculated) capital investment every arbitrary number of transactions. If you make it every 1 transaction you will get a fairly unstable and hypercompetitive network, so to increase it's attractivity, you could make it every 1000 transactions. 1000 is just an arbitrary number.
Where do these vcd fees go? It goes to the nodes that are outperforming the average node. By counting up the net "efficiency values" of all the overperforming nodes, you can derive a percentage of how much of that "efficiency value" belongs to you. If it's 1%, then you receive 1% of the collective lost funds of the underperforming nodes.
A way that maybe we can measure 'work time' is by each node judging every other node by how fast they send information. Perhaps every inter-node data transfer starts with sending a hash of the data, and then the receiving node times how long it takes to receive the rest of the data, based on elapsed transactions. I'm sure there's other ways we could measure other angles of 'work time' as well.
Step 3: Create a virtual power bill. This virtual power bill can't be percentage based, because that wouldn't motivate anything, and that doesn't reflect how reality works. Instead it has to be a fixed rate. You could say that every node must suffer 1 Satoshi per transaction processed, but I think there is room to argue that even this would contain an element of arbitrariness. So how do we put this part of the mechanism on a dynamic market like we have the rest?
In reality, the power companies set the rates for usage of electricity. So who in our blockchain analogy could be looked at as "the power company"? Well, it's the users that give life to the ecosystem. And it's the coin holdings that secure the ecosystem. But if we merely give the power to set these prices to anyone, then it will inevitably result in the wealthy pricing the poorer out. So instead, we could make a stipulation; all coins are able to vote for a price on the "virtual power bill", except for any funds that are currently staked in a Locked Capital Contract. This forces the rich to either seek more money themselves and take on more risk, or to sacrifice more of that in order to engage in pricing out the poor.
Additionally, this money should not be funneled back into the rich. If the users are the power company, then the users get the money. Of course this money will never be anything close to what they spend, but it's a nice way to recycle the wealth in the ecosystem without wasting it.
But how should we give the money to the users? Wouldn't the rich take most of this? Yes they would. So that's why instead of basing it on coin holdings, it should be based on fee expenditure. Basically, After having participated in this POS blockchain ecosystem, you get a partial refund for the fees you've spent. Not more, and you're not stiffed in order to unfairly benefit anyone else. It's just simply distributed based on relative expenditures in the form of fees.
Since the "virtual power bill" does in fact raise the price of the fees, giving this virtual power bill back to the users will "equal out" the increase in the expensiveness of the fees to use the network, resulting in no economic inefficiency (aside from a few negligible extra transactions in the ecosystem, which can definitely be mitigated by grouping transactions together).
Conclusion: By simulating virtual Capital Investment, Capital Depreciation, and a Power Bill, we might be able to bestow upon Proof of Stake the meritocratic values found within Proof of Work. I have no knowledge that this idea would certainly work, it's just an idea I had. The biggest potential problem i see, is there may be a practical or engineering challenge with trying to measure how fast and efficient each node is in a fully distributed system.
Feel free to critique my idea here or offer any kind of problems you think someone trying to create this might run into.