What's the hype around decentralization all about? Why is it important? Why do we want decentralization, when centralized financial institutions seem to be working just fine?
The answers to these questions range from political, ideological views all the way to software engineering challenges, network security, fault tolerance and other technical aspects. In this article we discuss some of the main points about distributed ledgers, the challenges posed by decentralization and why Bitcoin is such a relevant technology. Decentralization faces several challenges and, in the specific field of cryptocurrencies, one well known problem had to be overcome in order to build a viable crypto.
It was this problem which Bitcoin solved and revolutionized the world of decentralized digital currency.
Suppose you command an ancient army and you have 11 generals. You have surrounded the enemy territory and now you must decide on whether to attack or not. Your main problem is knowing that you have traitors among the 11 generals. How do you coordinate an attack such that all generals obey the command and no single general is able to betray your orders?
This problem was discussed in an academic paper by Leslie Lamport et al titled "The Byzantine Generals Problem". Lamport based the research on previously existing computing problems such as Dijsktra's dining philosophers problem (which studies computing concurrency) and the well known Chinese generals' problem to adapt the idea for computer science. In the Bizantine Generals version, computers are generals and the links between computers are communications channels. The order to "attack" or "retreat" would come via the link to these computers.
In short, the main contribution derived from this academic work to Bitcoin (and most other cryptocurrencies) is that whatever the majority of generals decide will be considered the consensus and therefore will be what all generals must do. This concept of following the agreement of the majority is analog to the length of the Bitcoin blockchain. The longest blockchain that has been agreed upon is the legitimate one. You can have as many blockchains as there are nodes, each with its own view, but only one blockchain must be considered the legitimate one, otherwise it'd be a chaotic mess.
Reaching a consensus in a distributed environment is, thus, the biggest challenge. It is this problem which Bitcoin solved through the apparently simple solution just discussed in the previous paragraph: the longest chain wins. But there's a problem.
We mentioned that in our fictitious army we have 11 generals. Assume that 5 of them have decided to attack, while the other 5 have decided to retreat. Now, here's the problem: the 11th general is corrupt and is likely to betray our army. Whatever he decides becomes the majority, because he holds the 11th vote. One vote now gets to decide over 10 other generals! This describes a version of what's called the 50% + 1 attack and unfortunately it can, in theory, be performed against Bitcoin. As long as both "halves" are honest, this will not be a problem. But now there's a new problem to consider!
If you have a total of 11 units of fictitious hashing power (in practice it'd be hashes/second) mining Bitcoin and 6 units are held by anyone, then this entity holds what we know as 50% + 1 mining power. Now, this is the real danger. This entity could, in theory, inject a fraudulent transaction into their blockchain, "mine" it while violating the chain (the fraudulent transaction would appear out of nowhere) and move on having the longest chain. The other miners would search for consensus and would find this chain to be the longest! The Bitcoin system would have been defrauded.
This attack is the sum of all fears in the world of cryptocurrencies. The only reason it does not happen today is the sheer amount of distributed hashing power available on the Bitcoin network. There are so many people mining BTC today, with so much hashpower (and proportional amount of money invested) involved, which makes it unfeasible for anyone to gain 50% + 1 advantage. In this video, Mr. Andreas Antonopoulos gives a satirical (yet accurate) example of how expensive and futile it'd be to try and defraud a single Bitcoin block (a span of 10 minutes) :
The last step to create a cryptocurrency is to apply the problem we've just solved to digital cash.
Blockchain is a generic name for a specific kind of database and, by itself, it is not revolutionary. In fact, common relational database systems can be used to implement a blockchain data structure within them.
The real innovation introduced by Bitcoin is the solution to the Byzantine General Problem which we just discussed.
So, what happens when the longest blockchain consensus just reached via the solution of this classic problem involves money? What if the object of the consensus were a financial transaction ledger? As you can see, the moment we turn the blockchain into a transaction ledger it instantly becomes decentralized digital cash. If you cannot defraud the longest chain consensus, and the longest chain contains a transaction ledger, you get a cryptocurrency
So, all this makes for a very interesting academic debate, but why, exactly do we want decentralization? Why is it important?
Well, there's an elephant in the room in the current worldwide financial systems. Central banks control the creation of currency and this system has been abused to the point where eight people now hold as much wealth as half of humankind combined. How did this happen?
It's a mix of a centralized banking system, unfair taxation and corruption.
During the early 2000's Japan's economy had stagnated. There was no inflation, there was no deflation, there was no growth and it seemed like Japan had finally reached the highest evolutionary step in capitalism - the perfect economy. But that's not how the system works. Our system requires inflation and constant growth, otherwise it collapses.
It is not enough for companies to produce amazing profits and pay big fat dividends, they must continue to grow indefinitely. Therefore, the "perfect" Japanese economy could not stay in equilibrium, something had to be done. The solution to this was to print more money.
Everyone then realized something curious : as long as this money remained in the financial system, and did not reach the masses, inflation would not happen. They could pump the financial markets to astronomic levels and as long as this money remained in the financial sector and did not leak onto main street, it would not cause inflation. The name of this process is Quantitative Easing (QE).
When the Western market collapsed in 2008, QE was the way the FED and ECB found to kickstart the Western markets back into life. But the result is nothing short of a social disaster.
Banks have been printing money freely and The People cannot do the same. Regular folks work 9 to 5 for a relatively fixed wage, while central banks print trillions of dollars to be pumped into the financial sector only. This money never reaches the middle class. The result is absurd concentration of wealth. AKA centralization.
Cryptocurrencies challenge this notion by introducing a financial instrument that is never political, always technical and transparent, money is produced at a well known rate, everyone can verify the creation and destruction of cash and, above all, it is 100% decentralized. In cryptocurrency, banks cannot simply print money for themselves like they do now. If any mining entity or anyone else for that matter found a way to defraud the system, it'd be recorded on the blockchain and everyone would be able to see the fraud. It's impossible to create money out of ether and spend it freely, which is what banks do today.
The entire problem with the present day financial system is centralization. This is why we need to decentralize and give the power over their own money back to the people. We've reached an absurd point where money is 100% virtual and a handful of people are deciding their own virtual wealth, with inexistent money being printed for their own benefit, while the rest of humankind is running in the hamster wheel waiting for next month's pay.
The main challenge of having a decentralized system is to agree upon a consensus. The system must be made in such a way that it is immutable, impossible to defraud (in Bitcoin this is guaranteed by strong cryptography) and there must be a way for the decentralized ledger to be agreed upon by the majority of network participants. As soon as any of these principles is violated, the system will no longer be trustworthy.
We need decentralization because the current centralized financial system is unfair. Another stunning statistic illustrates this: 42 people hold more wealth than 3.7 billion others combined. Centralized banking has failed us - and is now a new crash waiting to happen. Financial authorities cannot keep the markets working by printing more money, creating more debt and by withholding all this wealth within the financial system. The centralized system simply does not work and we'll live through constant boom/bust cycles which have plagued the financial systems for decades. It has been 10 years now since the 2008 subprime crisis and another large crisis looms in the horizon, with QE having injected tens of trillions of U$ into the financial sector while none of this money reached 99% of the world's population.
Cryptocurrencies promote financial decentralization. While there is still a lot to do to make cryptos scalable enough to become the future of financial systems, working towards decentralization is the main goal and must be the focus of future financial systems.