Solving the forking problem: Why future blockchains must upgrade themselves

Cryptocurrencies had a difficult start to 2018. Just last month, $60 billion was wiped off the value of the entire cryptocurrency market in 24 hours, and Bitcoin tumbled to $6,800 after reaching its all-time high in December. While the market seems to be recovering now, crypto's volatility and murky regulatory landscape shows that we haven't built a cryptocurrency ecosystem that's ready for mainstream adoption. Blockchain, the technology underlying Bitcoin and other cryptocurrencies, just isn't ready for prime time.

Right now, cryptocurrency transactions can be costly, slow, and tedious to process. The next generation of blockchain platforms must achieve better scalability, faster settlement, and lower fees -- all while retaining decentralization -- in order to succeed. For this to happen, blockchains must be able to upgrade themselves.

Bitcoin is based on a process called Proof-of-Work, which serves as a consensus mechanism within a decentralized network. Its most significant drawback is its lack of scalability. The Bitcoin network generates a block every ten minutes, with a maximum block size of 1MB. This has restricted Bitcoin's transaction speed to no more than seven transactions per second. Because of Bitcoin's inherent forking problem, a transaction can be considered secure only after six confirmations - which takes, at minimum, and hour. Executing a single Bitcoin transaction can take between a few hours to a few days, with transaction fees averaging several dollars and rising with increasing demand.

When you compare this to VisaNet -- which is capable of processing up to 10,000 transactions per second -- it's clear that Bitcoin isn't cutting it. Many argue that raising Bitcoin's block size is the best option, but doing so requires a large percentage of the network to change their software, creating another fork in the blockchain. Bitcoin's inability to handle high-volume transactions at a reasonable speed is a clear hindrance to its mainstream adoption.

As it grows, Ethereum is facing some of the same structural problems. In recent months, Ethereum smart contract execution fees (gas) recently rose to more than five cents from less than one cent in 2016. When CryptoKitties spiked Ethereum's popularity late last year, nearly 30,000 transactions were stuck in limbo as the blockchain struggled to process the backlog. This prompted the Ethereum community to plead with its miners to increase the gas limit by 33 percent just to encourage quicker processing.

These issues with Bitcoin's forking and Ethereum's gas limits are a roadblock to cryptocurrencies' overall growth potential. Future cryptocurrencies must be able to upgrade themselves without a cumbersome process or the involvement of a centralized party.

One way this can be done is through a hybrid approach that combines Proof-of-Work and another consensus mechanism, called Byzantine fault tolerance. Well-known examples of Byzantine fault tolerance-based blockchains include Hyperledger and Stellar. By combining consensus mechanisms, blockchains can allow participants to actively adjust the blockchain's protocol without compromising decentralization or security. With this "committee method", blockchains can separate Proof-of-Work from the transaction verification process. This would reduce confirmation time and allow thousands of low-fee transactions per second, reducing many of the roadblocks faced by major cryptocurrencies today. Each layer of the blockchain could be separated and upgraded in the future, and consensus could be decentralized and permissionless. No trusted authority would be required.

While Bitcoin and Ethereum introduced us to blockchain technology, our understanding of its possibilities and pursuit of new iterations should not stop there. By solving the forking problem and allowing blockchains to upgrade themselves, we can create a comprehensive foundation for the future of blockchain adoption.

Sky Guo is the CEO and Co-Founder of Cypherium. Guo holds extensive infrastructure-level knowledge regarding blockchain consensus, transaction, cryptographic algorithms and messaging mechanisms, and has conducted extensive research on present blockchain implementations and their source code.