In the beginning there was Blockchain
And it was good.
Blockchain innovated in such different ways, pointing to a less centralized environment, and it pointed out the direction to go.
Blockchain as it name says, is just a chain of blocks, each block containing transactions with a coin base transaction at the top to reward the miners, that's people borrowing their computing power to the network, which made the necessity for a trusted authority obsolete, creating a trustless network.
Each block also contains a Merkle Tree Root, which allows for transaction validation.
That's amazing and everything that came after that as well.
But it also brought a few issues:
- Hard to scale, and we all saw that.
- Transaction fees, discouraging microtransactions.
- Computing power becoming more centralized than we expect.
For a while, we've seen different projects arising, but all of them with a slightly different version of the blockchain, yet the blockchain was still there... until now.
From now on I'm gonna summarize a bit of my understanding of blockchainless protocols. If there's any misunderstood feel free to point it out in the comments section.
The Tangle
IOTA introduced a different approach, it completely replaces the current Blockchain with a graph, when you see their visual representation it is easy to understand why it was named Tangle.
Instead of blocks, there are transactions, which they call sites. Instead of a chain, there's a direct acyclical graph.
To a node emit a transaction it must approve 2 other transactions.
A transaction can be indirected approved. If transaction C approves B and B approves A, it's considered that C approves A.
So as the transaction input rate increases, also the computing power increases, in theory solving the scalability ghost.
It also solves the problem with fees, there's no need for fees in this scenario, the cost to emit a transaction is not paid by fees, it's paid by approving other transactions instead.
Nodes also can calculate statistics about the network, making it possible to identify lazy nodes, for example.
So it also addresses the centralization problem, having a computing power better distributed.
IOTA dedicated a good part of its white paper to describe possible attacks and ways to mitigate it, explaining about the algorithms to select tips to approve and ways to optimize it.
Basically, each transaction A contains a set of properties:
weight: Sum the weight each transaction that approved A.
score: Sum the weight of each transaction by A
height: Longest path from A to Genesis (The first transaction)
depth: Longest path from A to a tip (An unapproved transaction)
With that in mind, it's possible to improve the tip selection algorithm and alleviate attacks.
I'm very impressed by the technology behind IOTA and IMHO it has great potential to set a new standard.
Bitlattice
I also recently heard about Bitlattice which seems to follow a similar idea of a blockchainless distributed ledger.
I'll definitely keep an eye on it.
Conclusions
At least, in theory, this new structures seems to be fixing the most crucial problems we have on the blockchain now.
A no-fees, scalable and decentralized solution.
From now on I think more projects will look into different approaches than Blockchain.
No, Blockchain is not dead, but it's nice to see new possibilities coming out and people thinking outside the box blockchain to solve the concerns we face now.
Nevertheless, the future is bright, whether it's in a blockchain, in a tangle, or any other data structure that we come with.