Where did this craze come from? Among other factors, the acceleration of investments/acquisitions of blockchain companies has acted as the main catalyst, including theR3CEV consortium, which alone accounts for 60% of financial institutions and a market capitalization of $600 billion. In the Fintech and Insurtech industry, 2015 was the “year of blockchain,” sparking a stir in the press, as evidenced by the cover of The Economist in October 2015. Even with regard to regulation, the climate is favorable thanks to the tax exemption on the use of digital currency in the European Union.

Birth of a technology       
The idea of blockchain was born with Bitcoin, known to the general public as a currency but in fact, primarily a set of technologies that form the backbone of a computing protocol. Bitcoin has as much in common with a TCP/IP protocol (aka Internet) than with a currency.

In theory, a blockchain is a duplicate record, shared between users and validators. The information flowing through this platform is grouped in blocks, each of which represent a page of the book. Blocks are automatically stacked chronologically in order to timestamp information in their order of arrival. During the creation of a new block, validators secure and lock the contained information, including the reference to the previous block. It becomes impossible to falsify a block without changing all other blocks secured to the “chain”. Therefore, the blockchain turns into a great audit tool and allows at the same time to establish trust between the actors in the system. However, in practice, all blockchains aren’t tamper-proof. How can we be sure that time-stamped data in a blockchain will never be distorted? To understand this, we need to look back at the context in which the Bitcoin protocol, mother of all blockchains, appeared.

For decades, great books of accounts were computerized and maintained by private entities such as banks and financial institutions. This centralization was necessary in the fight against fraud and “double spending” which was reflected, basically, by the success of several payments of a same amount at once. A closed system allows to discard any manipulation of operations, for example, the addition of gigantic sums on the accounts. Such a concentration of power can seem risky, but a strict regulation governs these activities to ensure compliance and stability.

The blockchain technology was developed against the backdrop of the 2007 financial crisis, at a time when trust in financial institutions was at its lowest point, with phenomena of banking panic and virtual freeze of interbank transactions. The idea of a public ledger account safeguarding the decentralization of power resurfaced in this specific context. However, the question about the vulnerability of public records vis-à-vis fraud remains unresolved. How can we ensure the transparency of all movements and protect write access in a database open to all?

In November 2008, a developer known as Satoshi Nakamoto offered a solution by publishing a white paper titled Bitcoin: A peer-to-peer electronic payment system in the Cryptography mailing list. This document described the general operation of a time-stamped server (renamed “blockchain” by developer and activist Hal Finney) with a write access privilege that requires “proof-of-work”.

Seven years later, the proof-of-work security was used to guarantee the immutability of transaction data in the Bitcoin blockchain.

The Bitcoin blockchain
The challenge of proof-of-work is comparable to the game of Sudoku in which testing is difficult but verification, very easy. This operation (also called “mining”) is based on consensus rules as well as validators (also called “miners”) scattered around the globe. The main function of mining is to ensure the safety of a blockchain, that is to say, of a public database!

Mining consists in creating a lottery in which the more minors (validators) develop computing power, the more they get tickets to win the competition. Therefore, winners are paid in account units, also called “bitcoins” (without capital letter). The Bitcoin protocol security level increases in proportion to the number of contributors. The combination between a unit of account, a principle of mining and blockchains builds a decentralized governance through a consensus of quasi-anonymous actors.

Therefore, the safety of blockchains is guaranteed as long as a majority of honest miners control together more computing power than another group of miners that wish to carry out an attack.

All blockchains that use proof-of-work may be vulnerable to Goldfinger attacks. These attacks consist in bringing together a computing power equivalent to just over 50% of the total power. As a consequence, the attacker is free to enable/disable certain transactions during 10 minutes in the Bitcoin network. The purpose of these attacks is to undermine the confidence of network users. Today, Bitcoin has become the most powerful network in the world with over 1,200 Peta Hash per second (PH/s) which means that even if a 51% attack is theoretically possible, in would be very expensive (approximately 5 billion dollars for the Bitcoin blockchain during 10 minutes).

Fortunately, Bitcoin is an open source project. A simple software patch, followed by a restarting of the network nodes, would be enough to solve the problem (see Contingency plans). The source code for Bitcoin is auditable by everyone provided that you understand programming languages such as C++. Thus, everyone can take part, edit, compile and even rewrite the code to create their own blockchain. In the Bitcoin ecosystem, there are thousands of new blockchains, called “Altcoins” for “Alternatives corners.”

The Cambrian explosion of blockchains
We are now living a Cambrian explosion that is giving birth to many bockchains, thus allowing large scale experimentation. There are now many types of blockchains, with sets of properties that vary depending on the project. Today, anybody can create their own system: the requirements for starting are accessible and simple.

As we have seen, originally, a blockchain is a public database. In order words, this flow of information is visible to everyone. It is a tremendous step forward in the fight against corruption, but from the perspective of a private entity, these benefits are perceived as disadvantages.

Actors such as IBM or Digital Asset Holdings are working on private blockchains that require a permission for taking part in the consensus. Despite the decentralized micro-provision/macro-centralized of private blockchains, some features of public blockchains are maintained: a strong architecture because the database is distributed; thepermanence of the system because data is unchangeable; the uniqueness because there is only one system, one single general ledger.

These features offer significant advantages such as the reduction of the systemic risk related to the insolvency of a participant, the governance of off-chain assets provided by smart contracts or distributed virtual machines (Turing Completeness) which will optimize services or reinvent them from scratch.

For example, companies such as banks will use this technology to minimize maintenance costs. The development of private blockchains takes off in a particular context of ongoing upgrade of computer systems that are decades old. The R&D departments of all major banking groups are highly interested in blockchains and develop, alone or through consortium such as R3CEV, prototypes that allow to respect the laws in force in the respective countries (for example, KYC/AML regulations – Know Your Customer andAnti-Money Laundering).

By contrast, public blockchains such as the Bitcoin protocol can be used to ensure data privacy through the use of asymmetric cryptography: financial information and user identities are protected because the user is the only one to remains in control of disclosure. This could perhaps give ideas to states, while citizens are exposed to the risk of their data being hacked by national institutions or private companies, as recalled by the hacking of Sony Pictures Entertainment in 2014 and that of the US public service in 2015.

In summary, a distinction can be made between:

– Public blockchains whose general ledger is open. These blockchains can be secured by using a proof-of-work mechanism that requires a certain amount of computing power to obtain the privilege of write access to the blockchain. However, new alternative mechanisms such as proof-of-stake, are currently being tested. These require a number of digital assets to generate a new block. The proof-of-transaction and proof-of-blockexperiments point at a future where several security mechanisms could be arranged in parallel or in series.

– Private blockchains whose general ledger is closed, also known as “permissioned blockchains”. Today, these blockchains aren’t implemented yet, but it is easy to imagine that they will use either proof-of-stake, or a known and identified validator, as is the case for IBM’s Open Blockchain. These are B2B-oriented products that will be dedicated to cases of very specific use such as syndicated credit, trade finance, supply chain provenance in physical blockchain, compensation or repo. These blockchains are clearly aimed at the industry and do not affect end-users.

Hybridization of blockchains
But that’s not all. Private blockchains are also experiencing a fragmentation while their environment is only just emerging. Richard Brown, CTO within the R3CEV consortium, said that the term “blockchain” very often covers a wide variety of ideas. He noticed that many of his partners and colleagues discussed completely different matters while they thought they were speaking about the same thing. Rather than speaking of “blockchain” indiscriminately, he suggests using the term “shared and replicated registers”: “shared” because multiple actors can read or write on different parts of the register; “replicated” because everyone can get a copy of the register if necessary, rather than relying on a central authority.

However, it would be unfair to present all shared and replicated registers as private blockchains. Most of these records are merely updates of bank computer records that simply adopt some aspects of the blockchain technology. These shared and replicated registers, that we started developing two years ago, are very different from original blockchains, as they don’t take advantage of the full blockchain potential. This could very well be a transitional step in the long process digital revolution in the banking, financial and insurance sectors.

In the world of public blockchains, the most interesting developments are occurring at the level of transaction scripts and smart contracts. The Ethereum smart contracts project is a good example, but also projects such as Streamium, that enable the viewing of streaming videos, show how Bitcoin could be used to create models of interaction that are simply not possible on financial platforms, today.

As we can see, these two worlds of public and private blockchains could lead us to the same place: a world based on shared databases, operating autonomously.