Publications

Publications

Leonardo and the blockchain


Published on May 2, 2019 5 minutes of reading On May 2, 2019, the first blockchain and media industry research in Canada is published. The result of extensive work, with the support of Telefilm Canada and the Canada Media Fund. To be the author and co-initiator, I thank these institutions for their trust and support, as well as the Pôle Médias HEC Montréal for their much appreciated collaboration. But May 2 is very special this year, as an extraordinary anniversary of the death of Leonardo da Vinci. The "Divine Artist", as his first biographer Giorgio Vasari had already named in the sixteenth century, died just 500 years ago today. Da Vinci is one of the greatest "explorers" that our world has known, by its intellectual curiosity, combining sciences and different forms of art, including painting. A genius is said, as fascinating as mysterious. The Mona Lisa still smiles. So to publish this research report on the anniversary of the death of "the humanist with total knowledge" made me ask myself the question: what would he, Leonardo, have done with and for the blockchain? Would he invent it? Perhaps. But, certainly, he would have been interested in it and would have known how to use it. Here are some illustrations, anachronistic of course, related to both the work and the hectic life of da Vinci, to materialize the still too often abstract applications of the blockchain. After all, with Leonardo, in addition to artistic expression, it was engineering and scientific accuracy that mattered. So, speaking of blockchain, what would Leonardo have done? Da Vinci is the questioning of established knowledge and dogmas. A true spirit of disruption. Although in my opinion the blockchain is more a slow factor of disruption (spread over years), put in the context of the fifteenth and sixteenth centuries, in the short time, his spirit and innovative methods have been able to have, or to seek, a disruptive impact. Whether in science or arts. Da Vinci did not only open the way for his successors, he was also inspired by the work of others, predecessors or contemporaries. The difference is that he is the first to have truly gathered information and data systematically. He "revalorized the observation". This observation, combined with experimentation, was his mode of learning. The principle of proof of concept specific to the experimentation of any technology. And especially the blockchain, still in the experimental era, and whose practical applications raise as much hope and enthusiasm as skepticism. Many of his observations and findings have been written in a variety of documents, most of which have been disseminated over the years throughout Europe. The trail of many of these documents has been lost. Leonardo could have saved them on a blockchain and authenticated the origin. On the one hand counterfeits or unauthorized appropriations would have been limited because they would be more easily identifiable and, on the other hand, the traceability of these writings would have been greater and more reliable, ensuring a wider dissemination of his work. In fact, it would seem that the precision of his researches and achievements, and here speak in particular of his anatomical drawings, "was such, that if they had been published, the knowledge on the human body would have been preceded by a century". Could Leonardo have benefited from the blockchain as a disintermediation system to publish himself? After all, with the example of Bitcoin, which does not involve a banking intermediary, the blockchain is the promise of a new distribution by which the creator regains control. The audience, too. Notary son, Leonardo had to understand the role of the intermediary. Indeed, the blockchain, by its mechanism of consensus, is mainly to authenticate and notarize the transactions in an open ledger. But Leonardo seemed to have an imagination far more fertile than his ability to materialize his ideas. Very little of his scientific research and even his artistic works (about twenty of them, it seems) have actually become realities. Could he, having recorded his works on the blockchain, grant to others, craftsmen, engineers or artists, licenses for the realization of these, to follow the progress and to be remunerated on the derivative productions? All thanks to the transparency that allows a chronological distributed ledger as the blockchain? And then, da Vinci was as self-taught and versatile as methodical, to have come to build a true "universal knowledge". It would seem difficult from now on to reach such a level of knowledge "in our situation of compartmentalized disciplines". In fact, in the continuity of the previous idea, the blockchain could have supported this universal spirit and it could do it now: it would be a great tool of collaboration, where the contributions of each and everyone would be authenticated, validated, authorized, followed and recognized. Blockchain is meant to be the interconnection of knowledge, combined with respect for intellectual property, for the common good. A true cooperative and smart contract. Yes, but, and as we have seen, the blockchain is experimental. Like the works of Leonardo. And this one always had to run after the money. His knowledge, his resourcefulness and his power of seduction allowed him to benefit from the patronage of the Medici, Sforza and Kings Louis XII and Francis I. But perhaps Leonardo could have imagined the tokenization of his research and his works? In other words, thanks to the transparency and traceability offered by the blockchain, as well as the cryptography and smart contracts with which it works, allow Leonardo, the creator, to issue digital tokens. Tokens to digitize assets and value them. First for their financing, then for their dissemination. Imagine, your ancestor having invested a few crowns in The Mona Lisa? A digital right on an asset, authentically and steadfastly registered on a decentralized database? What would be the value of these tokens from these few crowns now? So speaking of value and traceability, at the time of writing these lines, note that the most expensive painting ever sold, Salvator Mundi, originally scheduled to be exhibited at the Louvre Museum of Abu Dhabi, would have disappeared. On the one hand doubts already existed about the origin of the work (would it be the most talented of its apprentices who would have painted it finally?). But also, the opacity about the identity of the buyer and especially about where the painting would be raises questions ... could a system of traceability and authentication based on a distributed ledger have been useful? ... In any case and to conclude, if da Vinci had in mind a "ceaseless quest to grasp the universal laws that govern mankind and the universe", it is quite possible that a parallel could to be drawn. By helping to reinvent relationships between individuals and between organizations, such as some of the principles on which industries, institutions, and perhaps societies are based, it is conceivable that some of the noblest applications of the blockchain may be consciously or unconsciously in a comparable quest. The idea of ​​this whimsical text, and some of its parts (especially those in quotation marks) are inspired by a series of articles published in the Quebec newspaper Le Devoir on April 27, 2019. This is the work of journalists Isabelle Paré, Jean-François Nadeau and Stéphane Baillargeon, as well as the results of the interviews they conducted with specialists. Thanks to them for this inspiration. One last note: it is also fun to know that the actor and producer Leonardo DiCaprio should play in 2019 the role of Leonardo da Vinci. He who, through his production company, is already working on a biographical film about Nolan Bushnell, founder of the game company Atari. Project whose bet is to largely base its model of financing and promotion on the emission of digital tokens. Another story to follow.




A blockchain is a distributed ledger technology


This article is complementary to the research "Blockchain technology and the Canadian media industry", a project undertaken by Telefilm Canada, the Canada Media Fund, and Badel Media, in collaboration with the Pôle médias HEC Montréal. Published on May 2, 2019 10 minutes of reading Defining the blockchain Defining the blockchain in a simple way is never easy and each subject specialist will approach it in a different way and in a different context. But to sum it up, some fundamentals have to be presented. It is often they that allow the reader to visualize the applications and the scope of this technology. Let's say right now that the blockchain consists of computer protocols, mostly open source, and is part of the Distributed Ledger Technologies (DLT) family. Ledgers have historically been records of information about transactions, usually the transfer of an asset, its use or its ownership from one person or organization to another. Their computer version, a DLT, is not a new concept and several approaches and technologies emerge, complementary or competing with each other. What particularly distinguishes the blockchain from the other DLTs is the notion of consensus. In other words, the establishment of a mechanism from a set of rules accepted by network participants, and necessary for its proper functioning. Let's stress again that distribution is at the heart of the blockchain: distribution of information, processes, roles and responsibilities, in a common vision, a common goal. Consensus is the essential element at the base of the governance of a decentralized system that wants to no longer be based on a central supervisory authority. Key features If we decide to limit the description to its simplest, we can define the blockchain as a database. A database that has the particularity of being distributed, that is to say, in a way, replicated and synchronized on the different computer nodes of a network. This distribution has the following benefits: Greater transparency of data, through the multiplication of access points; which facilitates their authentication but especially their follow-up, ie their traceability; Immutability of these data: already secured by the cryptography techniques used on a blockchain, the data entered into a distributed ledger will be all the more immutable as each "copy" of this register must be consistent with the others. Basic uses of the blockchain All applications of the blockchain relate to at least one of these four uses: The cadastre, or the chronological register of transactions between authenticated entities; The transfer of value that induces disintermediation as a potential impact; Automation, via smart contracts, or coded contracts, which have the benefit of greater efficiency of administrative and business processes; Digitization of assets with the promise of creating new forms of values ​​and business models. As part of this introduction to the basic operation of technology, we focus on the first two points: cadastre and value transfer. Basic elements of operation It should also be noted that the operation described primarily concerns public blockchains of the bitcoin blockchain type. Private and hybrid or consortium blockchains (eg linked to an organization or business sector) will have types of consensus that are not generally based on the same principles. They will therefore be much less "decentralized" in the true sense of the term because these blockchains are more about distributing roles and processes, increasing traceability and efficiency, establishing new types of relationships between stakeholders, with a common goal of responding to a sectoral issue. Let's summarize here the basic elements of public blockchains: Data is encrypted and "sealed" in blocks, once a validation and authentication process is completed via the consensus mechanism accepted by network members. This mechanism is also based on cryptography principles. For the bitcoin blockchain (started in January 2009 and the first business application of blockchain technology), this consensus is called Proof-of-Work. The blocks are connected to each other by a unique identifier. In the end, these interconnected blocks constitute a chain. The longer a chain is, the more it becomes secure and resistant to modification. Attempting to modify the data in a block would amount to compromising the integrity of the chain itself. Any modification is traceable and the possibility doubtful, given the lack of equivalency between data in the affected block and that of the other blocks in the chain. A Blockchain is theoretically unalterable because taking control of one involves taking control of 51% of a network’s nodes. Once again using Bitcoin as an example, this type of action would require extremely powerful computing resources, therefore making it difficult to hack such a system. Despite the environmental impacts (electricity consumption, etc.), it is this requirement for computing resources that ensures the system’s reliability. Consensus, governance, and trust Distributed ledger technology introduces a type of collective accounting, which is theoretically verifiable by anyone in real-time. Given the absence of a central control authority, the system itself, with its decentralization and its consensus mechanism, becomes the trust protocol between participants, which is certainly its greatest innovation from a technological and governance perspective. Mining is the centerpiece of the system’s confidence. This mining operation, characterized by computer resources that are made available to a network through incentives and compensation, allows the verification of new transactions and records them in the register following approval by the consensus protocol in place in the network. For example, the Bitcoin blockchain confirms transactions following a competitive resolution mechanism involving mathematical calculations therefore known as Proof-of-work. Mining based on Proof-of-work Every computer connected to a blockchain can, in theory, become a “miner”. This is because, in principle, mining (in its current form as proof of work) requires ever greater computer resources. Generally speaking, these resources are of the ASIC type (Application-Specific Integrated Circuit), in other words, systems developed for a specific application. In this function, the miners choose a group of transactions from a pool of all the transactions awaiting verification by the network. The rewards granted to miners in exchange for processing play an important role in determining the choice and therefore the priority of the transactions to process. This is because the economy of mining is based logically on creating a net profit in each operation (compensation for transaction processing versus the cost of computing resources). Bitcoin’s miners are compensated through a combination of newly issued virtual currencies and transaction fees or commissions. As an illustration of this idea, at the moment, the first miner to solve the mathematical problem related to the verification of a block of transactions that are ready to be approved is rewarded, with 12.5 new bitcoins. This can extend to the number of bitcoins that are or will be injected into the network, up to a maximum of 21 million (at the end of August 2018, a little more than 17 million were in circulation and the last bitcoin should be issued around the year 2141, or 132 years after its creation). This is based on a rate that decreases sufficiently over a long enough period of time to attain 21 million, where the “extraction process” also becomes the essential activity behind the issuing of new bitcoins. Inside a block The competitive process of the Bitcoin blockchain is configured in such a way that every 10 minutes a new block is formed. To respect this rate, the calculation to be performed is subject to an adaptable level of difficulty that depends on the calculating power of the entire network. Additionally, the rate can vary completely, depending on the consensus protocol used by a blockchain other than that used by Bitcoin. But before another block can be created two steps must occur: The miners must first solve a specific mathematical problem: a cryptographic hashing function that seeks to randomly identify a value (a nonce) whose juxtaposition with the data of a previous block produces a hash, which must be under a certain threshold (the mathematical problem to be solved). Put simply, this hash is also the unique digital signature of a block and is a unique string of characters (256 in the Bitcoin blockchain). When data in a block changes, a new hash is generated. Hence, a problem of correspondence that can be easily detected over a decentralized network whose nodes possess the same number of copies of the ledger. The hash is a sort of fixed-size, standardized image that makes it practically impossible to find an output value based on an input value. Therefore, the original block of a chain cannot be modified. Unchanging, this initial block is known as the genesis block. Once a miner has solved the problem, it must broadcast the block to all of the other computers in the network who then verify it. Only blocks containing unanimously approved transactions are added to the chain. Blocks (of a variable maximum size, but for example no larger than 1 MB in the Bitcoin blockchain) are integrated into a chain in chronological order. Every approved and broadcast block is characterized by its signature (Hash), its Timestamp, and a data field describing the transactions. The immutability of a blockchain is supported by the fact that it is copied in its entirety to the nodes on the network (the total size of the Bitcoin blockchain at the end of August 2018 was approximately 180 GB).





download-as-pdf.png
download-as-pdf.png

Contact

manuel@badelmedia.com   |   514-607-3640

Montréal, QC, CANADA

  • Blanc LinkedIn Icône
  • Blanc Facebook Icône