Currently used most of the time: Proofs of Work and Stake

Bitcoin consensus was found first via proof of work (PoW) = one CPU, one vote, or so the intention seemed to be. As we know, this didn’t last. Now, BTC mining is decreasingly profitable and relegated to massive operations near cheap electricity. Furthermore, environmental concerns over the energy expenditure of PoW chains (BTC, ETH as of time of writing) are now becoming politically prevalent.

Proof of Stake (PoS) was a leap forward in transaction validation efficiency and environmental impact. Its Achilles heel is that it incentivizes token hoarding, i.e. the rewards of validating transactions favors those who were able to purchase dirt-cheap tokens early on (VCs).

So, while these systems are not the biggest problem in crypto now, the demand for increasingly efficient and equitable consensus may grow with overall blockchain adoption, perhaps even to the point of outgrowing PoS chains.

There are several other proposed consensus mechanisms

Each offers different tradeoffs on the classic blockchain trilemma - the challenge of maximizing decentralization, security, and scalability all at once (and/or the belief that only two can be optimized at a time). However, there is usually an aspect of ponzinomics that presents an advantage which disincentivizes the ‘public’ from jumping in after VCs have established their positions in a project.

These mechanisms are proofs of [+ a TLDR description based on my cursory understanding]:

1. Activity [PoW + PoS -→ Faces same criticisms as both of these]
2. Authority [Trusted validators are, VCs, partners -→ considered highly centralized]
3. Burn [Miners burn tokens to earn right to validate -→ quantity burned = probability of selection as validator]
4. Contribution [Both a token security deposit and past validation activity affect probability of being selected as a validator -→ niche, evidently useful for on-chain validation of actions initiated off-chain (iExec)]
5. History [Each validator encodes passage of time to properly order transaction data quickly, without the entire network needing to be perfectly in-sync -→ Ex. Solana, which still maintains some decentralization]
6. Importance [PoS, where both tokens staked and other metrics like number of transfers inform each validators selection odds -→ “weighted PoS”]
7. Space / Capacity /Storage / Replication / Spacetime [Disk space is used to store potential mining solutions or user-uploaded data, etc. -→ quantity of space allocated = probability of having matching solution or otherwise being selected as validator based on other data storage (Subspace, Filecoin, Storj)]

If you’d like to understand how all of these work in more detail, check out:

However, as you may have noticed, none of these designs have had a breakthrough adoption event thus far. During peak 2021 bull market, projects such as Filecoin made some waves to be fair, however, their consensuses were focused on the service they provided, decentralized storage, and don’t seem viable for general-purpose consensus.

IMO, Subspace has a chance to produce the best blockchain trilemma optimization yet

Creating equitable incentives for the average person to participate in consensus and ensure reasonable decentralization requires incentives including low entry cost, predictable rewards, some profit, no punishment for node downtime, etc. Proof of Capacity, specifically Subspace, seems to be balancing equitable incentives well, implementing a consensus protocol called Proof of Archival Storage. The project’s whitepaper describes the consensus mechanism as follows:

“Consensus in Subspace is based on proofs of replicated storage of the history of the blockchain itself. Farmers store the history collectively, many times over, with each farmer storing as many replicas as their disk space allows. Consensus and computation are then decoupled, such that farmers only propose an ordering for transactions, while staked executor nodes maintain the state and compute transitions.”

In all honesty, I’m still trying to wrap my head around the nuances of this consensus mechanism (see link below for more, just might write a follow up on this as I continue to read), but am gradually piecing together that this incentive system encourages broad participation, seems to save on compute costs by removing the need to compute solution hashes at the time of transaction, and has the makings of being competitively secure and fast.

Final notes

This is not endorsed by Subspace, I have no ulterior motive to promote them, I just think it’s an awesome project and potentially the best optimization of the blockchain trilemma thus far. I’m going to participate in farming to the extent I can and hopefully learn enough to be a critical thinker toward it’s improvement or replacement.

References:

Below: Subspace Network Whitepaper -→