Blockchain has been the technology of the past year. Bitcoin, the most well-known blockchain-based cryptocurrency, saw its trading price skyrocket to more than $19,000 in late 2017. Since then, investors have poured money into cryptocurrencies and anything remotely related to blockchain technology — an optimistic choice, especially since the price of Bitcoin had dropped to $7,000 as of March 2018.
Advocates have declared that decentralized blockchain-based currencies could become the future of monetary transactions, as governments and banks are removed from the process. Meanwhile, critics have warned that the decentralized nature of cryptocurrencies make them far too risky, and that governments will inevitably crack down on unregulated transactions.
While political and economic concerns have understandably dominated discussion of blockchain, the technical aspects of the technology have gone largely ignored. Many in the public think that blockchain “just works,” but this assumption is both inaccurate and potentially dangerous. It is important to recognize that blockchain is, in fact, a technology, and its very real technical challenges will need to be addressed in order to realize its full potential.
The Building Blocks
Understanding the technical problems with blockchain requires first understanding how it works — easier said than done. At a simplified level, a blockchain can be thought of as a digital ledger of transactions that is distributed across the internet, with each page of that book representing a “block.” Whenever someone wants to send cryptocurrency, he or she broadcasts a transaction across the internet, along with a unique digital signature indicating that the transaction is legitimate.
Broadcasted transactions are not immediately added to the blockchain. Instead, new transactions are only added when a new block is added. New blocks are created by users in a computationally intensive process called ‘mining.’ When a new block is mined, the ‘miner’ aggregates broadcasted transactions, adds those transactions to the block, and broadcasts the block to the world. This system of relying on computing power to create new blocks is known as ‘proof-of-work.’ Users then add the new block to their own digital transaction book, creating a decentralized way of tracking transactions around the world.
Each new block contains a signature, or ‘hash,’ of the previous block, ‘chaining’ every block to the block before it — hence the term ‘blockchain.’ A valid blockchain must have a valid signature on every block, and if any of the previous blocks is tampered with, the signatures of all subsequent blocks are invalidated. This creates a problem if different blocks are created and broadcast at similar times, each bearing the same signature as the previous block, but for different transactions, since it is unclear which block should be accepted. Most blockchains therefore use a protocol where the longest blockchain is considered to be the ‘official’ chain.
The main problem blockchainers around the world have been dealing with is scaling blockchain technologies for a global audience. Like pages in a book, each block in the blockchain has a specific size. For Bitcoin, the size of the block is limited to 1 megabyte, which holds around 2,400 transactions. Since the Bitcoin protocol limits new block creation to, on average, once every 10 minutes, there are only an average of three Bitcoin transactions per second worldwide. For comparison, Visa and Mastercard together process nearly 2,000 transactions per second.
As a result of the block size issue, transactions may take hours or even days before they are officially recorded on the Bitcoin blockchain, making it difficult to confirm that payments for goods or services are actually made. Furthermore, since the demand for space on each new block far outstrips supply, miners have begun to demand transaction fees for users. In December 2017, the average cost of a single transaction was $34. Transaction fees have since fallen to less than $1, temporarily addressing the problem, but transaction fees will almost certainly grow as Bitcoin expands.
The Bitcoin community has tried to address the problem. In mid-2016, there was a hard fork of Bitcoin, essentially switching some users to a new, incompatible transaction book known as Bitcoin Cash. Bitcoin Cash expanded the block size to eight megabytes instead of one, facilitating eight times as many transactions every 10 minutes as the original Bitcoin, now known as Bitcoin Core. However, even with the expanded block size, Bitcoin Cash can still only handle around 61 transactions per second.
Bitcoin Core implemented a different fix, known as Segregated Witness, or SegWit for short. SegWit makes each transaction smaller rather than making the block bigger, and allows for the use of the Lightning Network, an ‘off-chain’ network that enables faster transactions by relying on intermediary transactions between networks of users. These transactions are considered off-chain because they are only committed to the actual blockchain at a later point in time.
While initially promising, these solutions are far from perfect. Emin Gun Sirer, associate professor of Computer Science at Cornell, contended in an interview with the HPR that both approaches are short term measures. For Bitcoin Cash, the larger block sizes require more network space to transmit each block. According to Sirer, “If [Bitcoin Cash] ever gets into a situation where the number of transactions is higher than the network can deliver, then they will run into a wall.” SegWit, on the other hand, can at best double the on-chain transaction rate. Lightning Network faces its own scalability problems. As Sirer noted, “The capacity of the network depends on the financial relationships that people have. The transactions are limited by channel capacity.” Although the Lightning Network allows for faster transactions, the size of those transactions is limited. If solutions to scalability are themselves not scalable, then the issue is only delayed, not solved.
Because blockchain was originally designed as a method for financial transaction, blockchain networks are built with security as a focus. Though several security problems remain, Sirer explained that the main security threats to blockchain are not threats to the network, but threats to individual users. A user’s identity on the blockchain is based entirely on their digital signature, technically known as a private key.
Problems arise when the private key is accessed by malicious actors or when the private key is lost. Unlike with a centralized banking system, there is no insurance or method of retrieval for a stolen or lost private key. This could certainly be considered user error, but if blockchain is expected to become widespread, then it will have to account for users who are less technically savvy and who are potentially vulnerable to malicious actors.
Beyond the individual level, there is also the security challenge posed by exchanges. Converting between cash and cryptocurrency is most often done through cryptocurrency exchanges, which are especially vulnerable to malicious attackers. In January of this year, over $600 million was stolen from the Coincheck exchange. In 2011, Mt. Gox, one the largest Bitcoin exchanges at the time, was the victim of a $460 million theft. Digital security is a problem with traditional financial institutions as well, but the decentralized, anonymous, and irreversible nature of cryptocurrency transactions makes them much more vulnerable.
Finally, there is also the possibility of unintentional bad transactions through forks, where two different blocks are simultaneously created. Silvio Micali, a computer science professor at MIT and Turing Award winner, explained to the HPR that the transaction ambiguity created by forks of the blockchain means vendors will be uncertain about whether a payment has actually been recorded. This problem will be especially prevalent if blockchains try to address the scalability issue by decreasing the time between creation of new blocks. Doing so would increase the probability of two blocks being created at similar times, growing the number of forks in the blockchain, and increasing the ambiguity users face.
Perhaps the greatest technical problem with Bitcoin is not a technical one at all, but a human problem. As with any new technology, it is inevitable that technical bugs will be discovered and that changes to the underlying technology will be necessary to fix those bugs. However, as Steve Wilson, Vice President and Principal Analyst at Constellation Research, argued in an interview with the HPR, the distributed governance structure of blockchain technologies makes any change incredibly difficult to implement. “If there’s a bug in the Bitcoin, these guys even argue about whether it’s a bug or a feature.”
In addition to difficulties in modifying the implementation of the blockchain, there is also the growing power of miners, who decide which transactions are added to each new block. According to Micali, “The miners are an external force that get to control who gets to spend the money. They cannot spend my money, but they can prevent me from spending it.” In a system with a large number of individual miners, this is less of a problem, but that is not how most blockchains work in practice. Since creating new blocks is computationally intensive, control over the blockchain will gradually shift towards miners with the most computing power. Already, Micali noted, “The blockchain of Bitcoin is controlled by three mine pools in China, and the Ethereum blockchain is controlled by two minepools in China.” The increasingly centralized power of miners presents a direct challenge to blockchain’s promise of a completely decentralized and distributed system of transactions.
New Chains on the Block
The problems facing blockchain today are numerous, but they do not diminish its promise. Already, individuals around the world are developing creative solutions to the problems of today’s blockchains. Two such individuals are professors Sirer and Micali.
Sirer has worked as the chief scientist at bloXroute, a technology venture that helps make existing cryptocurrencies like Bitcoin more scalable by enabling fast information propagation across the blockchain network. Meanwhile, Micali has led the development of Algorand, an implementation of blockchain that relies on proof-of-stake rather than proof-of-work. Instead of delegating new block creation to the miner with the most computing power, Algorand randomly selects an individual to create each new block. Then, 1000 other users are randomly selected and vote to approve or reject the new block. According to Micali, this results in a blockchain that is highly scalable, secure, and efficient. Sirer agreed with the use of this method: “2019 will be the year of proof-of-stake.”
Micali and Sirer are far from the only ones working to advance blockchain technology. There are hundreds, perhaps thousands, of developers and experts around the world trying to make blockchain technology into the truly revolutionary technology it has the potential to be. Changes to blockchain technology will almost certainly meet resistance, but change is often necessary for advancement. As Wilson contended, “The tragedy of Bitcoin is that it was really a proof of concept.” Sirer agrees: “Bitcoin is technologically stagnant.” If blockchain is to advance, new technologies that solve real problems will have to be embraced by blockchain users.
The most important step towards solving the problems with blockchain, however, is realizing that such problems exist. Focusing purely on the regulatory and economic issues of blockchain not only ignores the flaws of the underlying technology; it minimizes blockchain’s potential. Wilson said that blockchain can be used for much more than cryptocurrency, and it is already being used for interbank settlement, financial products, shipping manifests, and more. The problems of blockchain cannot be ignored; they must be understood and fixed. Blockchain may be the future, but the future of blockchain is far from certain.