A basic principle of the blockchain is to distribute (in fact to duplicate) a register (the recording of data) between several actors. The fact of having several identical copies and control mechanisms makes it possible to dispense with a central body (the trusted third party). Instead, it is the community that validates the transactions together and agrees (the consensus ) on a reference version of the registry.

More precisely, it is a member of the community (called "minor" in some blockchains) who does the work and then submits it to the validation of the other members. Imagine ten people who do not know each other in a room: we choose one - by an algorithm known to everyone - who checks the transactions and then writes them in the register. Once she's finished, she submits her work for proofreading to the other nine.

Kết quả hình ảnh cho công nghệ

Choose the "minor"

This totally decentralized method immediately raises a question: who will, and how, choose the member (the blockchain network node) to validate the current transactions and replace the trusted third party?

This is where the famous "proofs of" come into play. They make it possible to determine which person in the room will be in charge of writing the transactions in the shared register.

The most famous and widespread "evidence" in public blockchains (including the Bitcoin blockchain) is "Proof of Work".

Contrary to what one reads a lot, the PoW does not consist in verifying that a transaction is valid (ie to verify that the debtor indeed has the funds by analyzing the preceding lines of the register). This verification is part of the "consensus" but it requires little computing resources and can be done by any computer on the market.

PoW is another part of the validation, which makes it often confused with the verification step. But, unlike this one, PoW is VOLUNTARILY very greedy in resource.

Bitcoin consensus (validation) = transaction verification (simple) + PoW (complicated)

"Transaction validation is an important step in the operation of Blockchains since it is the process that leads to the writing of transactions in the distributed register . It must be safe and reliable, "explains Damien Lecan of Space Elephant . "What better than to make it extremely complex and difficult to discourage fraudsters or those who would harm? ".

The aim of PoW is therefore to make writing in the register (the creation of a block) artificiallycomplex and costly in energy.

"This is the goal of PoW: to have a transaction that is safe, extremely difficult and complex, that allows transactions to be validated and stored in blocks," says the CTO.
A lottery more than a puzzle

We often read that PoW is a "complex mathematical puzzle". Why not. But the metaphor magnifies the essential.

Recall that a block of the blockchain consists of several transactions selected by the minor, metadata such as the time stamp, the signature of the minor, the hash of the previous block and the hash of all this information - hash which, himself, will be taken up in the next block (at each block N, we make a hash of hash N-1, hence the notion of string).

"Concretely, the PoW Bitcoin consists of calculating the" hash "of the block that the minor wishes to validate [...]. This "hash" is calculated from the data of the block themselves and a random ("nuncio") that varies completely the "hash" of this block among 2 32 possibilities, "says Damien Lecan. "The complexity comes from the fact that this" hash "must be below a certain value (the" target "). Concretely, this is equivalent to finding one that starts with several "0", which is extremely unlikely. To achieve this, we must calculate many "hashes" by varying the "nuncio" to find a compatible with the requirements.

Finding the right "nuncio" is like doing a brute force contest - more than solving a puzzle.

A puzzle induces the idea that there are clues, forms to observe and methods better than others to solve them. While the only way to increase your chances of finding a hash before the other minors - and therefore to be chosen, and therefore to be paid - is to increase your computing power. But even being the most powerful, a minor is not sure of being the first to find the right nuncio.

"PoW is a paid lottery," slice the CTO. "It's about being the first to calculate a number among a phenomenal quantity, in a short time and that satisfies a given criterion. Having more power allows you to generate hashes faster and increase your chances of finding a result first. But there is a strong random component in finding the right "hash". [...] By analogy, you can buy a lot of Lotto tickets to increase your chances of winning in each draw ", but without being sure to win the jackpot.

The defects of its qualities

PoW generates a huge energy expenditure (compared to that of whole countries). It should be understood that when creating a block, all the nodes (N) having chosen to be "minor" of the blockchain turn their farms of servers and specialized graphics cards to full power by launching in the brute force contest. A block is the result of the work of a winning minor AND that of the N-1 losing miners. Hence a huge waste of resources.

Another flaw, PoW slows the creation of blocks. The slowdown was also the first objective of this technique, which was initially used in a different context, to slow down the sending of emails and thus fight against the sending of spam.

"This slowness is relative: an average block every 10 minutes for Bitcoin," nevertheless tempers Damien Lecan. "Its creator estimated at the time, in 2009, that it was a good parameterization from a functional and technical point of view".

In the context of Bitcoin, the complexity (and therefore the "target" discussed previously) is therefore regularly re-evaluated according to this speed of creation of the blocks. If the power of the miners increases, the complexity adapts accordingly to slow them down and maintain the historical delay of 10 minutes between two blocks.

A stroke of genius

Despite these two flaws, PoW is a real stroke of genius. "It has demonstrated that the principle of Blockchains could work on a global scale," says Damien Lecan.

It is indeed it that made it possible to make the attempts of massive falsification too expensive, while creating a motivation for the minors (by paying the winner) which made them virtuous (the one who owns bitcoins has no interest to make drop his course by attempting to falsify his register).

Certainly, this idea of ​​PoW in the heart of blockchain "1.0" shows today its limits. But it has been running for almost ten years, so far flawless.

"Technical solutions exist, some demonstrated in other blockchains or prototypes, and could push the limits of Bitcoin". Sufficient? "Nobody can really know," admits the CTO of Space Elephant. "There will surely be new limits, then new technological innovations."

Blockchain 1.0, Blockchain 2.0, Blockchain 3.0

Rather than improving the Proof of Work, other "proofs of" are now implemented in the second generation of blockchains. We can cite the Proof of Authority (the miner turns from node to node, block after block), the proof of stakes (the nodes that have the most assets have priority), or even the notary service ( as in Corda ). But there are dozens of others.

These alternative consensus are used, for example, in private blockchains - where actors know each other upstream and where the benefits of PoW have little meaning in view of its constraints.

And already the blockchain 3.0 begins to point the end of its link. "The next blockchains should embark on a governance" on-chain ", ie integrated natively in the operation of the blockchain itself," predicts Damien Lecan thinking including French Tezos. "But we are waiting for them to really break through." Patience is the mother of virtue. Especially with young technologies.