Bitcoin and the law of centralization
The first time I went online it was over a 2,400-bits-per-second dial-up modem. At that time, there was no internet in India and the nearest I could get to something that resembled it was CiX, a bulletin board service operated by the unforgettable Atul Chitnis. Since only one person could log on at a given time, everything on CiX took place asynchronously. You dialled up, downloaded whatever you had to and quickly logged off so that you freed the modem up for the others waiting to log on. In those early days of the internet, we spent more time offline than connected. It was a tiny walled garden but given we had no options, it was the only way we could get online.
When the Indian government finally allowed its citizens access to the internet, everything opened up. We suddenly had access to all that the world wide web had to offer. Our only constraint was knowing where to look. It was everything that I had imagined the internet would be—a decentralized universe of diverse knowledge so vast that it was always going to be impossible to completely cover.
But before I knew it, things started to get organized. Google came along and offered a completely different kind of search—one that kind of took the fun out of old-school exploring. Social media companies learned our likes and dislikes and started serving us content they knew we’d like directly into our news feed.
E-commerce companies began to offer us the opportunity to buy anything we could dream of so that we could shop the planet from the comfort of our armchair.
With that things began to swing back to how they were right at the very beginning. Today, even though we have the illusion of choice, most of us rarely use that choice to venture beyond the few sites we use. We can go anywhere we want to but choose to partake of the internet as it is delivered to us by the few sites we trust will serve us the content we like.
We don’t realize it, but our internet experience today is not that dissimilar from what it was for when our only option was CiX.
It is a law of nature that no matter how decentralized a service is to start with, left to itself, things eventually tend towards centralization. This results in power being concentrated in the hands of a few putting so much pressure on market economics that law and regulation is forced to intervene.
The decentralized technology that is all the rage today is bitcoin. It offers an alternative to centralized financial systems that are dependant on a central banker to maintain and manage the central ledger on which every single debit and credit transaction that takes place within the system is recorded. It does this by storing a log of every transaction on a blockchain—a distributed ledger, which isn’t stored in any central server. Every new bitcoin transaction must be added to the blockchain in order to be verified as valid. Once added, it becomes an irrefutable part of the chain of transactions that stretches back to the founding of the chain.
Anyone can add a transaction to the blockchain. All you need to do is solve a really hard cryptographic puzzle and the block of transactions you are working on gets appended to the chain. It is this democracy of effort that allows the system to operate without the need for a central banker to intermediate all transactions.
Over time and as the computational effort required to solve these cryptographic functions increased, the bitcoin miners who solved these cryptographic problems started organizing themselves in pools. These mining pools were loosely organized coalitions working together to achieve greater mining efficiencies.
The trouble is that the moment a mining pool controls more than 50% of bitcoin’s total cryptographic hashing output, that pool can independently determine what transactions get added to the blockchain and what cannot—effectively defeating the decentralized design of the blockchain. This is the single greatest flaw in the design of bitcoin—one that shook confidence in the entire system when, in 2014, the mining pool GHash controlled more than 50% of the total computational power on the network for over 12 hours at a stretch.
At the time, in order to defuse the situation, some miners voluntarily left the pool immediately, restoring faith in the system. But with bitcoin at its current stratospheric valuations, even the threat of a 51% attack could dramatically destabilize the cryptocurrency. Earlier this year, over a threatened fork in the bitcoin protocol, fears of a 51% attack were rekindled. Even though that fork never came to pass, perhaps the time has come to question whether the decentralized design of bitcoin can survive on its own or whether it needs some help.
Rahul Matthan is a partner at Trilegal. Ex Machina is a column on technology, law and everything in between. His Twitter handle is @matthan