In various discussions over the last two months, we’ve noticed that some people have trouble wrapping their head around Saito’s consensus mechanism. Although our whitepaper tries to outline the solution as simply as possible (and in only four pages!), since understanding it requires conceptualizing blockchain security as an economic problem rather than as a mere technical challenge, we wanted to put this blog post together to help people make the adjustment.
So if you are having difficulty grasping Saito, start by remembering that the reason Proof-of-Work and Proof-of-Stake are considered secure is that they meet the following criteria:
* make it difficult to produce blocks, while…
* measuring the amount of work that nodes do, and…
* paying nodes in proportion to the amount of work done.
In Bitcoin these properties are guaranteed by SHA256. The difficulty of hashing ensures that it is expensive to rewrite the blockchain, while the randomness of hashing ensures that payment is made in proportion to work done. Proof-of-stake has similar properties.
Scaling a blockchain means paying much higher costs for network operations. The first problem this creates is measuring who is paying for those costs. How can you do this in an open access network where you don’t even know who is participating? The second is more subtle: not all work done in a network is of equal value. Should we pay nodes for storing a full copy of the blockchain, for instance, even if no-one ever downloads it from them?
In the absence of objective information about who is providing storage and bandwidth and how valuable their contributions are, the only safe way for proof-of-work and proof-of-stake blockchains to cover these costs is to pick a subset of privileged nodes and compensate them while monitoring their performance. This is a fallback approach to scaling, but it is not a good one as it erodes the open access properties of the blockchain.
But now we get to Saito, which solves both problems. In fact, Saito is the first and only consensus mechanism that:
* makes it difficult to produce blocks, while…
* measuring the amount of value that nodes provide the network, and…
* rewarding nodes in proportion to the amount of value provided
How does it do this? Adding cryptographic signatures to transactions lets Saito identify nodes that provide valuable bandwidth (i.e. route valuable transactions for the network). The burn fee likewise measures the value provided by block-producing nodes. Saito knows how to measure value, which allows us to reward it without being cheated. And so all participants in the network are rewarded in proportion to the value they contribute.
The result is a strengthened blockchain. In addition to improved security in situations of high bandwidth and storage consumption, Saito eliminates sybil attacks, lowers transaction fees, and ensures that the payments made to each type of participant ( router / block-producer / miner ) are set according to the market value of their contribution to the network. And if you missed this the first time around? Read the whitepaper again and you should see it now!