What is a blockchain?

Thanks to Tim Berners-Lee, the World Wide Web has literally changed our daily lives, revolutionizing our learning and the amount of information available.

Unfortunately, the (exponential) abundance of this data has revealed in recent years a problem around the trust granted to this information and to the institutions that generate/control it.

How can we trust information?

The giants of the tech industry have gradually destroyed the trust of a large portion of us regarding the origin, collection, and sharing of our data.

What if the Blockchain were the beginning of a solution?

The short history of blockchain

Even though the beginnings of blockchain come a bit later, a 1982 thesis by cryptographer David Chaum proposes a protocol similar to blockchain.

His thesis is titled: "Computer Systems Established, Maintained, and Trusted by Mutually Suspicious Groups"

"Blockchain" was actually the subject of study for the first time in 1991, by two researchers: Stuart Haber and W. Scott Stornetta.

They wanted to set up a system where timestamped documents (that mention the time at which it was created) would neither be falsified nor obsolete.

In 1992, the system was improved by incorporating the concept of a Merkle tree, which allows several documents to be combined into a single block.

The first blockchain was invented in 2008 and unveiled to the public on January 3, 2009, by a person whose real identity is unknown but who appears under the persona of Satoshi Nakamoto.

This implementation of blockchain was the core of Bitcoin

What is a blockchain?

To put it simply, a blockchain is a database.

That is to say, you can add information to it, which, once recorded (in our chain of blocks), is impossible to modify or delete.

The data can be:

• An employment contract
• An acknowledgment of debt
• Courses
• Articles
• Scores in a game
• Etc...

In reality there is no technical limit on the type of data to be stored; the whole nuance lies in the way this data is stored.

The difference with a classic database?

A blockchain has two main characteristics that differentiate it from a classic database.

1) The data structure

A blockchain groups information into groups called Blocks.

Each block contains three parts:

• The data: The data is stored immutably (impossible to modify) in the blockchain's ledger and contains all the raw transaction data.
• The hash: The unique identifier representing (mathematically the current content of the block, in a reduced form). The concept of hashing is explained in this other blog article!
• The hash of the previous block: To form a chain, each block includes the hash of the previous block. This increases its security, because if there is a modification, the hash would be different and would automatically reject the fraud attempt. The first block has as its previous hash, the original block value "00000".

The hashing algorithm used for Bitcoin, for example, is the SHA-256 algorithm.

2) Data storage

Classic databases have a centralized structure, roughly like this:

Storage takes place at a central point, on a main server, with other servers used for redundancy or data sharing (sharding), but all managed by a common node.

This makes it a vulnerable system because there is a single point of failure

The blockchain, on the other hand, is a decentralized database, which is rather represented as a spider's web:

A decentralized database operates in a distributed manner and stores its information on a network of storage nodes (commonly called "nodes").

Each node has a complete and synchronized copy of the ledger (the ledger is the record of every operation/transaction carried out on a blockchain).

In a distributed architecture, if one node stops working, another one takes over.

Note that there are dozens of different distributed systems, protocols, and tools, but the latter are often based on centralized technologies, and decentralization happens at the logical application level (like PeerTube or Mastodon, for example).

Only blockchain works directly at the data level, and its value also lies in the trust and verification of data, in order to prevent any falsification.

How does blockchain prevent fraud?

What allows a blockchain to prevent the falsification of its data is the validation system through a consensus algorithm.

Proof of Work

Like Proof of Work (PoW), there are different consensus algorithms, and each blockchain will decide on the process used to reach an agreement on the current state of the ledger.

If the concept of Proof of Work is unfamiliar to you, we've dedicated an article on Code-Garage to it.

Used on Bitcoin and Ethereum, for example, PoW puts blockchain nodes (called miners) in competition with a single objective: solving complex cryptographic problems.

This (computational) effort proves that one is legitimately entitled to help validate blocks. Once solved, the block becomes valid and is added to the blockchain, and the miners are rewarded with cryptocurrencies.

By linking to the previous block, this new block helps create a chain of blocks (hence the name "blockchain")

While this consensus method has the advantage of being simple to implement, there is however a major flaw (among others, such as its energy consumption) in this algorithm:

If a person or group were to hold 51% of the active nodes (those of the miners, as opposed to users who are passive), they would then have control over the node network.

They would therefore be able to falsify the history of transactions present on the blockchain.

Proof of Stake

There is another consensus algorithm called Proof of Stake (PoS), created as an alternative to "Proof of Work".

This mechanism consists of putting nodes in competition to validate blocks, as with miners, but basing the ability to validate them on the amount of cryptocurrency already acquired in the past.

That is to say, the more you hold thanks to your commitment to a blockchain, the greater your chance of mining the next block, and therefore obtaining new cryptocurrencies as a reward.

Here too, a major flaw exists: validators with the largest stakes can have uncontrollable influence.

In conclusion, there are still many possibilities for creating new consensus processes to solve existing problems. Everything remains to be done in this technical ecosystem!