This is a post I've been wanting to write for ten years.
Much of the debate around consensus has been over proof-of-work vs proof-of-stake. But there is an option that is simpler, more robust, and supports in-protocol governance.
contenthub.coil.com
🧵👇
Much of the debate around consensus has been over proof-of-work vs proof-of-stake. But there is an option that is simpler, more robust, and supports in-protocol governance.
contenthub.coil.com
🧵👇
1/ PoW issue - energy usage: True, we can minimize the impact by relying on things like stranded energy.
But a waste is a waste.
There are other computing workloads which could be taking advantage of surplus energy such as training ML models.
But a waste is a waste.
There are other computing workloads which could be taking advantage of surplus energy such as training ML models.
2/ PoS issue - aggregators: Tokens are often held through custodians, exchanges, DeFi contracts. In practice this leads to significant concentration of voting power.
We also should not do governance using PoS - power would beget greater profits which leads to more power.
We also should not do governance using PoS - power would beget greater profits which leads to more power.
3/ To avoid a power-profit feedback loop, Ethereum handles governance off-chain.
But that sucks.
If blockchains are the ultimate in transparency and fairness, shouldn't we be using them to… govern blockchains?
But that sucks.
If blockchains are the ultimate in transparency and fairness, shouldn't we be using them to… govern blockchains?
4/ To recap, PoW is simple, relatively reliable, and expends a ton of energy.
PoS is complex, logically awkward, and plutocratic.
Neither solves the question of governance.
There’s a better way.
PoS is complex, logically awkward, and plutocratic.
Neither solves the question of governance.
There’s a better way.
5/ There is a choice that underlies and supercedes both PoW and PoS. It's when we decide *which* blockchain we want to use and be a part of.
Bitcoin Cash has less hashing power than the most popular version of Bitcoin, yet plenty of people still choose it.
Bitcoin Cash has less hashing power than the most popular version of Bitcoin, yet plenty of people still choose it.
6/ The definition of consensus is voluntary agreement. I pick a set of people and I decide to synchronize with them.
If we have to do that anyway, even with PoW and PoS, why don't we make our consensus algorithm work like that, too?
Enter Proof of Association.
If we have to do that anyway, even with PoW and PoS, why don't we make our consensus algorithm work like that, too?
Enter Proof of Association.
7/ Proof of Association: How It Works
✍️ Write a list of people or entities you’d like to be in consensus with
🤖 Feed it to a software program that scans the network + listens for those on your list
🗳️ When enough of your list vote for a ledger—a quorum—you have consensus
✍️ Write a list of people or entities you’d like to be in consensus with
🤖 Feed it to a software program that scans the network + listens for those on your list
🗳️ When enough of your list vote for a ledger—a quorum—you have consensus
8/ Consensus mechanisms are about how power in a system is defined:
🔋PoW power is defined by energy
💰PoS power is defined by wealth
🫡PoA power is defined by reputation, reliability, and free association
🔋PoW power is defined by energy
💰PoS power is defined by wealth
🫡PoA power is defined by reputation, reliability, and free association
9/ In PoW you don't control how much hashing power a bad actor has. And in PoS you don't control how many tokens a bad actor has.
Your only recourse is to decide not to be on the same network as them through a hard fork, ignoring PoW/PoS.
What then is the point of PoW/PoS?
Your only recourse is to decide not to be on the same network as them through a hard fork, ignoring PoW/PoS.
What then is the point of PoW/PoS?
10/ Proof of Association allows every participant to choose who they listen to. This defines what network they are on. They can:
🫂Choose the most reliable people/entities
👮♀️Remove bad actors
🌍Strengthen or diversify a set of validators across people, orgs, and geographies
🫂Choose the most reliable people/entities
👮♀️Remove bad actors
🌍Strengthen or diversify a set of validators across people, orgs, and geographies
11/ Over time, participants' lists get better and better. More decentralization, more overlap, and higher quality validators.
12/ Best of all - this isn't theoretical. XRP Ledger is a ten year old system which has been using Proof of Association all along.
We'd expect the list of validators to consistently improve over time. And it did.
High decentralization (no entity above 5.7%) + geo/org diversity
We'd expect the list of validators to consistently improve over time. And it did.
High decentralization (no entity above 5.7%) + geo/org diversity
13/ Because PoA consensus represents all stakeholders, governance can happen on-chain. Unlike Bitcoin and Ethereum, XRP Ledger makes decisions about protocol changes via votes by the validators, called amendments.✅
xrpscan.com
xrpscan.com
14/ But this isn't just about XRP Ledger. Proof of Association is a path for different blockchains to:
1. Solve energy waste
2. Solve concentration of power
3. Provide a robust, transparent method of governance
4. Reduce complexity
1. Solve energy waste
2. Solve concentration of power
3. Provide a robust, transparent method of governance
4. Reduce complexity
15/ It's an unfortunate accident of history that much of the blockchain space hasn't paid attention to the tech behind XRPL.
But time has passed & benefits to decentralization that were predicted for PoA have actually materialized.
Take another look.
contenthub.coil.com
But time has passed & benefits to decentralization that were predicted for PoA have actually materialized.
Take another look.
contenthub.coil.com
Loading suggestions...