We’ve heard this time and again, especially from folks in the financial sector: “blockchain is here to stay, Bitcoin not so much”. Is this really so?
We disagree, and in this article we discuss the origins of Bitcoin and how the problem it solved only makes sense when there’s a reward for participants to join the system. In order to deliver this reward securely, you will necessarily need to develop some system to verify the transfer of value among untrusted parties. We will show how this leads to Satoshi Nakamoto’s breakthrough in cryptocurrencies.
We also argue that a centralized blockchain doesn’t really solve any innovative problems and could be substituted by a secure RDBMS or other NOSQL database management system. But first, to better understand the issue, we need to go back to the basics of why Bitcoin is special: What problems did Bitcoin solve that weren’t solved before? What, exactly, is new about Bitcoin?
Eureka! The Breakthrough Introduced By Bitcoin
Bitcoin’s main innovation was to solve the Bizantine General’s problem for digital payments. This is a classic military strategy problem that basically outlines the issues faced when you need to work with a large group of people that you don’t necessarily trust. You may trust a general close to you, but it is difficult to trust 200 generals spread out in thousands of miles, each being either honest or a traitor (the honest/traitor duality is a simplification of much more complex human relations). In fact, there is no definitive solution to this problem as traitors continue to exist in every field of human endeavor – and every so often a large spying scandal explodes in the media.
Bitcoin solved this for digital money through a very clever game that employs proof-of-work and self-adjusting difficulty depending on how many players and resources are detected. We will not go into the details of Bitcoin mining, as there is plenty of great content about that on the WWW. But, suffice to say, that due to this system the cost to defraud the system is countless times higher than the potential profits. Thus, there’s no ROI upon defrauding the Bitcoin network. As Andreas Antonopoulos puts it in the following video, defrauding Bitcoin in today’s context is a futile excercise (even by nation-states with nearly unlimited resources):
This is accomplished through its brilliant distributed proof-of-work mining system. We now take this idea for granted and hundreds of coins have forked this system from Bitcoin and created their own version of the consensus mechanism, but until Satoshi Nakamoto implemented it, nobody had solved the distributed ledger consensus problem before.
Miners spend hundreds of thousands of dollars every minute verifying each and every byte in new transactions waiting to be committed into the blockchain. Defrauding the system would require the attacker to hoodwink each and every one of these miners and full nodes – a feat until now proven intractable. The longest chain in the blockchain wins, because the longest chain has been accepted by most miners. The longest blockchain is, therefore, the consensus!
It is a simple concept in hindsight, but one that took decades of cryptographic research to be put together in the right way.
In fact, the ACM published a very interesting article which shows all the historic academic research that went into each component of Bitcoin. What Satoshi Nakamoto and the other Bitcoin pioneers (including the late Hal Finney who many believe is the Satoshi Nakamoto) achieved was to put all these cryptographic pieces together in a way which has stood the test of time. Their consensus mechanism through proof of work is plain simply brilliant and has withstood some of the most sophisticated and expensive attacks ever devised against any other known technology.
Suppose, for instance, that a large bank wants to deploy a blockchain internally. No mining is required, because the main problem solved by Bitcoin does not exist in this case, which is the distrust among unknown parties in a financial system. Everyone in a bank has signed a contract, has been hired and will be legally responsible for anything that happens within their sphere of work within the bank. Thus, you don’t need a “consensus” mechanism inside a financial institution with trusted participants.
When you run a blockchain within a trusted environment, it is nothing but a secure and immutable database and not at all related to Bitcoin. In fact, most RDBMS already incorporate enough security features that you could simply deploy a traditional database and simply audit any modifications. Gigantic business operations function perfectly using large scale databases such as DB2 and Oracle – this is definitely not the niche for blockchains.
Centralized blockchains that run in trusted environments may well survive without Bitcoin. But they would not really bring anything new to the IT environment.
Decentralization Is The Key
Decentralization, therefore, is where Bitcoin really shines through. Being able to transfer money securely between untrusted parties, in a 100% secure and fast manner, paying extremely low fees is the true genius behind Bitcoin!
But is there a use for this technology that does not involve the transfer of value? Let’s focus for a moment on the main piece of the Bitcoin puzzle: mining.
There is only one incentive for miners to continue to spend hundreds of thousands of U$ in electricity and mining hardware costs: the token earned as a reward mining. Without the token that we have come to know as “Bitcoin” no one would be willing to verify the transactions for free. What else could be the motivation for otherwise free mining? Ideology? A sense of community? These may work for small communities, but we’ve seen such P2P tech in the past and it never
Therefore it is very difficult to conceive of a distributed consensus system based on proof-of-work that does not offer incentives to its participants. In fact we may have just reverse engineered Satoshi Nakamoto’s original thought process. What everyone wanted to use proof-of-work for in the early 1990’s was to make spam and other network floods very expensive. By attaching a cost to each message in a network, be it instant chats or email, one could introduce a deliberate obstacle. This obstacle would be very small, unnoticeable by the end user, but would become very expensive for spammers which need to send billions of messages to convert a tiny amount into sales.
But then, who would pay for this proof of work? What would the incentive be for a network of voluntaries to check proof of work headers in emails, if it weren’t for some kind of real value that could be exchanged for money? By following this reasoning, the only way to reward participants for mining transactions and verifying proof of work results is to pay them somehow. For this, you need a distributed ledger, so you need a way to reward participants in a proof of work network somehow and you need this to also be decentralized and 100% secure. The solution to this problem is now what we’ve come to know as Bitcoin.
As you start to think about how to solve the spam problem and other proof of work related cryptographic schemes, you will necessarily come to the conclusion that participants must be rewarded when they contribute honest data to the network. This simple concept leads to Bitcoin’s implementation of a distributed ledger that everyone can trust without having to actually trust any of the participants.
A centralized blockchain is possible to implement without any of Bitcoin’s innovations, but it will not solve the same problem. In fact, a centralized blockchain is simply a immutable database. To run a decentralized system, you need to make sure the cost of defrauding the system is higher than the values traded within the system. This requires proof of work or some other staking system, which necessarily leads us to the necessity of a value token. We now call this token “a Bitcoin”.
Therefore a decentralized network that does not offer a reward token will probably not survive very long or will not develop critical mass. For example, there is no incentive to share data on the Bit Torrent P2P network. Sharers either do it for ideological reasons or they spread virus or other malware along with their share. There’s always the need for an incentive and decentralized blockchains’ only incentive is to pay miners some form of trusted token that carries actual value.