Proof of Worth
Socially useful “mining” using DAOs and Ulex
Bitcoin’s Proof of Work consensus algorithm (POW) is controversial. The power consumed by bitcoin mining is starting to match that of small countries, and continues to grow. While I fall into the camp that maintains that the main feature of POW is to provide unrivaled security, I can understand the discomfort it engenders, especially when projecting the long-term bitcoin price and the subsequent effect this will have on demand for electricity. In addition to the high energy consumption, critics are also concerned that the actual work done is of no use beyond securing the network. The hashes POW produces are essentially dice rolls. Perhaps this is a missed opportunity.
In response to the perceived wastefulness inherent in POW, a number of initiatives were started to create a socially useful consensus algorithm. Primecoin, a cryptocurrency designed to “mine” for prime numbers, was one of the earliest examples in this trend. However, market adoption never really took off. It’s almost impossible to determine why something fails in the open market but one of the reasons I’ve seen offered for Primecoin’s demise is that the consensus algorithm can only discover a certain class of prime numbers that aren’t particularly useful or difficult to calculate off-chain. Algorithms exist for discovering primes which are more suited to the needs of science but are not possible to implement as a blockchain consensus algorithm. In a sense, Primecoin was more of a hopeful proof of concept than a useful alternative to POW coins.
When I first started coming to grips with the Bitcoin protocol, I asked myself “Why isn’t there a vast array of different coins, each mining something useful for humanity?”
The answer comes back to the purpose of the mining algorithms chosen: consensus. The work that is being proven is such that miners can churn away offline, figuring out the best result without disrupting the network. This keeps them busy and prevents the network of nodes from being spammed with false solutions provided by would-be miners.
Once a solution is discovered and offered up by a miner, the nodes which check the solution can do so in a matter of milliseconds. The asymmetry of POW is that it takes very long to do the work while taking almost no time to prove the work has been done. This is perfect for a network which is trying to achieve Byzantine Fault Tolerance, a necessary ingredient for a functioning distributed ledger.
In POW the hashing power committed by a miner is proportional to how quickly that miner is likely to discover the solution.
Since the first miner to propose a correct solution is awarded the block minting right, POW is also proof of mining investment. This brings us to one of the other wasteful aspects of POW mining: the sunk hardware costs of mining equipment as well as their gradual depreciation. For cryptocurrencies that are designed to be more ASIC resistant, miners not only face depreciation costs but obsolescence costs as well. In effect, cryptocurrency POW mining has sparked a Red Queen’s arms race among miners. The whole purpose of the sunk physical investments is to incur a resource cost on miners that translates to improved security on the network.
In response to the criticisms of POW, competitors attempted to find a socially useful mining algorithm. Many of us in the space are familiar with the most popular alternative: Proof of Stake(POS); the long awaited successor to POW in the Ethereum network.
Proof of Stake skips POW’s energy-hungry arms race by requiring block validators stake cryptocurrency rather than burn CPU cycles. Under POS, the more currency a candidate stakes, the greater the likelihood that they will be the winning validator. Validators are required to vote on winning blocks to achieve consensus and their voting power is proportional to the amount they’ve staked. Losing voters have their tokens burnt to penalize minority attacks on the network and to encourage unanimous consensus. Analogous to POWs feature of a cheap means of validating past work done, proving stake under a POS system is also trivially quick.
So in both POS and POW, the winning validator is likely to be the one who has invested the most resources. POS simulates the economic cost of investing in mining equipment and burning electricity without wasting real world resources. POS creates the same game theory dance using smart contracts without requiring physical resources to be wasted.
Redefining the Problem
We now know that if we propose an alternative consensus algorithm, the work should be done outside of the blockchain and the proof that it has been done should be easy to verify. The reason it has to be easy to prove is so the network can mint blocks in a regular interval and remain synchronized. Unstated in all of this is the implicit constraint that time plays in binding all the pieces together. What if we removed time as a constraint? Then the process of proving no longer has to be quick, it simply has to be objectively possible and low cost. For instance, if Bitcoin minted blocks once a year then the work being done could be such that it takes up to a year to verify objectively. In this case, the possibilities for consensus algorithms increase dramatically.
If traditional charities are committees for good, Proof of Worth is a factory for good.
Second Layer Mining
Ethereum gives us all the tools we need to simulate cryptocurrencies. Up until now these tools have been mostly used to produce static, capped tokens, but there’s no reason you can’t simulate a traditional POW coin that runs on top of Ethereum. In fact, the ethereum.org site uses this as an example smart contract tutorial.
Suppose that instead, we set a new objective standard that is of use to society. This standard can be “mined” and a Decentralised Autonomous Organisation (DAO) will oversee whether it meets the minimum requirements to be considered a proof. The person who submits the proof is then issued new coins in a coinbase transaction. Proof of Worth is a scheme that seeks to hybridize the best of POW and POS while bringing in a new layer of governance in the form of Ulex in order to transform subjective inputs into objective states in order to achieve a socially useful outcome.
Below I will build out a fictional example which will be used to extend the concept.
Thoughtcoin: A decentralized peer-to-peer electronic cash based on Proof-of-Good-White-Paper consensus algorithm
Thoughtcoin is a second layer cryptocurrency that runs on Ethereum. It has a supply schedule identical to Bitcoin and ends with a total supply of 21 million. Instead of approximately 10 minutes, block generation times are 50 hours. The governance of Thoughtcoins and the calibration of its parameters will be overseen by a DAO. Because Thoughtcoin runs on Ethereum, coin transfer time is decoupled from block minting time — users need not wait 50 hours for transfers to be confirmed. The purpose of the block time is to give the governance DAO sufficient time to evaluate the white papers submitted.
The Thoughtcoin DAO will specify in detail exactly what is and isn’t a white paper in a list of rules hosted on IPFS. When miners submit a white paper, they must stake a certain amount of Thoughtcoin. Miners have two hours after the previous block in which to submit their papers, after which a 24-hour period begins in which papers can be disputed using a decentralized dispute resolution mechanism such as Ulex. The disputes are subject only to the rules of the DAO on what constitutes a valid white paper.
After the dispute window has passed, holders of Thoughtcoin are able to stake Thoughtcoin in the DAO to vote on the paper they thought was most enlightening. The white paper that receives the biggest stake by the end of the block epoch is considered the winning paper. 50% of the losing voters’ Thoughtcoin is transferred to the winning voters. In this way, a Schelling Point is created around the best paper. Voters can change their vote while the block has not been sealed.
10% of the block reward is distributed to the winning voters so that there is a reward for them even when 100% of voters concur. The winning white paper author is then issued the rest of the block reward in addition to all of the transaction fees since the last block. The winning paper author is analogous to the winning miner in Bitcoin.
The coin above creates its own incentives by acting as the voting ticket for future white papers. Initially the creator of Thoughtcoin will be the only person to possess the coin but (assuming he or she creates faucets or manually disperses it) over time, more individuals will be able to participate in the mining and voting. Eventually a market for the coin will emerge that represents the economic value of good white papers.
Thoughtcoin could probably do with more design iterations. For instance, we might want to reward some authors whose white papers come in second or lower because they are still of value to society. In addition, the community will need to decide if losing resubmissions are accepted in later blocks. My intuition is that they should be so that good authors aren’t penalized for coincidentally sharing a block with better authors. In addition, the governance can be more sophisticated than a simple voting DAO. In the event that the “worth” being mined is for a particular community rather than humanity at large, it might be more appropriate to use a nuanced, community-oriented governance structure such as a DISC.
The point of the Thoughtcoin example is just to illustrate the minimum moving parts necessary to make Proof of Worth a reality using existing technology.
By creating a self-perpetuating subsidy for good white papers, we can create an almost barrier-free funding mechanism for innovators and a source of good quality research for the internet at large by aligning incentives using the same forces of self-interest that keep Bitcoin alive. We might envision a coin based on Proof-Of-Authentic-Charity that rewards users who meaningfully improve the world without gaming the system. Perhaps Proof-Of-I-Ate-Plants-For-A-Month for a coin that tries to encourage the world to move away from animal consumption or Proof-Of-Volunteering-At-A-Retirement-Home.
So long as the block minting times are long enough and the rules listed in the DAO are valid in a decentralized court and so long as truth can be agreed upon, a worthy cause can be mapped to a consensus algorithm that incentivizes its own maintenance.
Technological Constraints and the importance of Ulex
Until a functioning decentralized court system such as Ulex is on the Ethereum blockchain, Proof of Worth coins are impossible. Ulex is the link that will enable the transformation of subjective measures (such as Proof-of-Helping-Dogs-In-Need — for a real Dogecoin?) to objective blockchain states that enable a fully functioning cryptocurrency. Ulex also acts as an anti-spam filter because all submissions require deposits that are subject to adjudication.
I anticipate two main objections to Proof Of Worth second layer tokens running on Ethereum. The first concerns the current consensus algorithm of Ethereum. At the time of writing, Ethereum is still secured by the Proof of Work algorithm. If you believe that it is wasteful and should be replaced then Proof of Worth tokens do nothing to achieve this. If anything, they increase the demand for the Ethereum platform and therefore the difficulty of the underlying mining algorithm. In suggesting this, I am making the assumption that at some point in the future, Ethereum will successfully move to Proof of Stake. In the short run this objection holds.
The second objection I anticipate sounds something like this:
Tokens created on Ethereum do not require an additional layer of consensus because transactions are secured by the Ethereum miners. Are you not orchestrating a big song and dance for nothing? Wouldn’t it be easier to just create a standard charity?
It is true that the purpose of Proof of Worth is not to achieve consensus. Rather I’m fascinated by how the alignment of incentives achieved by it motivates self interested individuals to invest great resources into continually achieving a collective goal. While Bitcoin mining is an always-online, tireless engine for generating consensus, Proof of Worth inherits consensus generation from the underlying Ethereum network; its purpose is as a non-stop engine for generating any behaviour that humans value. Block generation is like a production line of social good. If traditional charities are committees for good, Proof of Worth is a factory for good.
Another aspect of Bitcoin I wish to emulate is how the token is imputed value thanks to its function in securing mining. Since Proof of Worth generated coins are the same tokens required to vote on the worth of miners’ offerings, it automatically answers the question of “what are these tokens good for?” If you’re interested in the emergent, collective organization that cryptocurrency mining creates and the implications for society, I encourage you to see this article.
Proving Proof of Worth
We will soon have the tools we need to build cryptocurrencies with ‘mining’ processes that deeply integrate subjective human valuation. A Proof of Worth approach — if it works the way we hope, and we’ll soon find out — sidesteps the problem of public distaste towards POW mining that’s (rightly or wrongly) perceived as wasteful. We will be able to create a new breed of tokens for human advancement that can be tightly focused on specific human problems and opportunities.
The examples I gave were off the top of my head but the possibilities are almost without limit. If you think of a good candidate cause for a Proof of Worth coin, please mention it in the comments. One suggestion I’ve heard is a Proof-Of-Scientific-Replication for a coin that encourages published science to be replicated, something for which the current incentives are not well aligned.