Rong Chen – 08/11/2019
On a quiet evening, I was answering a friend’s question on Wechat, and felt that I should record my answer and make it public. Blockchain technology has recently become a hot topic across the world among blockchain experts, tech enthusiasts, and newcomers alike. It is my hope that this article will clear up a great deal of the ambiguity surrounding the emergence and utility of blockchain and support the Elastos community.
In my opinion, it is inaccurate to describe the blockchain as an official ledger or bookkeeping tool for anyone but miners. The purpose of the blockchain is to establish decentralized trust for the rest of a blockchain’s community. To draw a parallel, in some parts of China, monks are famous both for their wisdom and the discipline they exercise throughout their lives in the processing of cultivating it. Monks spend years – even decades – repeatedly chopping wood and carrying water in monotonous and exhausting routines. Naturally, these physical processes do not enable the cultivation of wisdom in and of themselves; rather, they serve as foundational exercises through which monks are able to look inward and ultimately gain insight into the true meaning of life. Wood-chopping and water-carrying are not exercises that hold any intrinsic value beyond the physical benefits they confer; it is the manner in which the monks practice these activities that enables them to extract wisdom and delve into the profundity of human condition.
In like manner, a blockchain ledger serves as a foundational accounting tool, but that does not adequately describe the profound purpose for which we employ it. We draw on blockchain’s foundational element – its ledger – to establish decentralized trust, the vital glue that holds together our socioeconomic structures.
Alan Turing, considered by many the father of computer science, developed a famous test for gauging artificial intelligence during 1950s which involved two sealed rooms: one containing a human and the other containing a computer. An individual could issue questions through a tiny window to each of the rooms, and if he or she was unable to determine which room contained the human and which contained the computer, Turing declared that the computer possessed intelligence comparable to that of a human being.
To apply the Turing test to blockchains, imagine one room with a computer controlled by a human and the other with a blockchain collectively maintained by a group of miners. Each room attempts to provide some fundamental services required by a society, e.g., issuing citizen IDs, notary publication for national archives, production certificates for individual autonomous cars, currency printing by the mint or central bank, etc. Which room do you believe will provide more trustworthy services? So long as the rooms provide satisfactory services to the people outside, who is to care how many miners are conducting exercises in one of the rooms? It is none of the outsiders’ business to even ask whether the miners are paid in Bitcoin or fiat currency.
Internet users or companies do not really trust one another, but they can resort to public blockchains to settle their disputes, if only for a lack of a better solution. The blockchain’s performance issue regarding TPS (Transactions Per Second) is really a false problem, in my opinion. How often do we go to the census bureau, business bureau, or the public notary in real life? As a standalone technology, the blockchain is not able to complete the computing tasks of the Internet. It is also impossible for blockchains to rely solely on distributed “chains” to fulfill billing and payment needs of today’s banks.
Do not be fooled by Moore’s Law: not all technologies can achieve exponential growth in performance. In the 1960s, the chief designer of the IBM 360 computer, Amdahl, proposed Amdahl’s law, clarifying that distributed software applications could not be efficient. I studied parallel computing at the University of Illinois in 1986, and the experiments I conducted at the time proved that for personal computing (instruction set combination), the eight cores served as a ceiling. Increasing the CPU would only reduce the efficiency and form an inverted U-shaped curve. Independent of the speed of the CPU, as the number of nodes increased, the efficiency would eventually trend to zero.
At that time, I had already decided to change my goal, as I departed from scientific calculation with supercomputers for my life’s journey through operating systems (OS). It may be true that my cognitive ability has a limit because I have not been involved in computer architecture design for 33 years, but even today I still firmly believe in my own experimental conclusions.
Today, the blockchain has so many nodes, and each of them separately records data on any ongoing ledger and must perform a reconciliation of accounting data. Thus, no blockchain can possibly achieve the capabilities of even ONE mobile phone in terms of communication, computation, and storage. This is not a problem that can be solved over time; this is a permanent glass ceiling with respect to performance.
Blockchain is effectively a decentralized computer that is inefficient and cannot be used directly. That is, it is a computer that is not controlled by any organization or individual. It is not necessarily related to mining, which is designed to build decentralized trust. If the countries of the world were to share mutual trust in a singular, central bank, there would be no need to mine.
In the context of the Turing test metaphor, both inside and outside the room, miners use the ledger for internal accounting and to reward one another. To people outside the room, the blockchain provides services based on decentralized trust, such as DIDs and public notary services. Those services are neither cloud computing services nor financial services per se. People should not be confused with what is happening inside the room and outside the room. The term BaaS – Blockchain as a Service – is very illuminating: it reveals that blockchain is a service – NOT a solution.
The peer-to-peer cash concept developed by Satoshi Nakamoto attracted money launderers and Ponzi schemers at first. But the most powerful use case of Satoshi’s idea resides in the construction of a new internet domain – not just its financial landscape – while rewarding miners via profit sharing from real business transactions on the Internet.
Before 2009, all computers could be shutdown by human command, but Satoshi Nakamoto’s Bitcoin blockchain could not be shutdown by anyone. To me, this represents a breakthrough of a magnitude comparable to the first general-purpose computer invented in 1946. Blockchain will play a powerful role in improving the Internet by preventing it from being subject to centralized control.
Blockchain technology can only serve ONE purpose – one which is vitally important: to establish decentralized trust.
A bitcoin market or a few niche markets such as that for ICOs is not sufficient indication that blockchain can be extensively promoted in the mass market of the internet.
Bitcoin has now existed just over a decade, and today the total market value of Bitcoin sits at approximately $150 billion, accounting for 75% of the world’s $200 billion digital asset space, a staggering sum at first glance. However, the market capitalizations of Microsoft, Amazon, and Apple all $1 trillion.
In other words, traditional investment funds and traditional IT companies still have not dipped into the domain of blockchain and digital assets, despite the fact it has become a global buzzword.
Speaking of a substantive issue, how does blockchain technology make money? If no company makes money but NASDAQ still trades their stocks, is NASDAQ not a fund for speculation? Is it not a pyramid scheme? Neither technology nor trust is money, but they both help companies that sell products to consumers to increase profitability. Is there a blockchain company making money today? Not yet. For this very reason, Elastos Foundation has no intention of being or becoming a company.
It is also noteworthy that there are thousands of blockchain-based ICO projects. Even in the event that these projects can all deliver on their promises, it is unrealistic to expect that people will trust such a breadth of decentralized, trustless mechanisms. Generally, people can only recall the top three companies in any specific area, be it e-commerce, search engines, artificial intelligence, or the Internet of Things. Although other companies have advanced technologies and strong business models as well, they remain on the periphery and are eventually acquired by the aforementioned big three.
At its initial establishment, the internet was decentralized. However, due to the lack of technical resources to establish decentralized trust, a lot of scammers and data thieves entered the scene, and so users had to seek protection from large, centralized entities such as Apple. Apple charges a 30% protection fee, whereas some entities in China charge even higher protection fees of up to 80%.
Decentralized trust on the Internet will provide a great service in protecting individuals and their rights to privacy and self-ownership. As far as energy expenditure is concerned, the power consumed by blockchain miners is relatively small relative to the power consumed on the Internet today for the purposes of preventing the malicious activity that blockchain-based infrastructure renders infeasible.
After trust is established on the Internet, the Decentralized Autonomous Organization (DAO) model may gradually affect the organization of future Internet companies, and may even affect the organization of human society as well.
Children born after 2000 are the natives of a virtual domain recalibrated for self-ownership and privacy. “Programmable justice” has another layer of meaning: the justice that blockchain-driven computer programs will achieve.
Perhaps this virtual domain can be called “Cyber Republic;” perhaps this domain can benefit from a currency of its own – that is, one based on decentralized trust.