How do you solve a problem like blockchain?

Fausto Jori and Matthew Coward of Reply talk through approaches to overcoming some of blockchain's shortcomings

In a recent article Computing pointed out that while blockchain is a very interesting innovation it is not the solution to all of the problems being placed at its feet. Even for the purpose for which it was designed, supporting the crypto-currency bitcoin, issues have emerged that make it clear that we are still very much in the era of Blockchain 1.0.

Shortly after the article was published we were contacted by Fausto Jori and Matthew Coward of Reply, a company that has created a competence centre of experts and academics in order to create blockchain "accellerators" in the areas of IoT, finance, property, voting and ticketing. They offered to talk through the work being done to strengthen the weak links in the chain.

Computing: Transactions on a blockchain are stored, in theory, forever. Will the encryption algorithms remain sufficiently robust to protect these records for all time, or will public blockchains become a gigantic honeypot, a juicy target for states and hackers?

MC: That should certainly be a consideration. Let's not forget that on the bitcoin blockchain you can see the contents of anyone's wallet, so you should be very careful about what you put into a public blockchain. Those sort of considerations may send you towards a private blockchain. Fintech companies wouldn't want to put their clients' data into the public blockchain.

Because it's behind a firewall you have traditional network security, however the downside of [private blockchains] is that you have less mining power to secure that network. Bitcoin has one-third of a nuclear power station driving it, which gives an indication of the amount of power required to hack the bitcoin network. You'd need half a power station to even be able to think about hacking the bitcoin network.

But doesn't a private blockchain defeat the object, i.e. by making a Sybil attack possible and requiring a central authority?

MC: You can have a private blockchain that is run between companies that trust each other and still have a consensus-driven application where if you have one bad node it doesn't matter because so long as you have a majority of honest nodes they'll quickly outvote you. We see these more and more as a way to get around the issue of what information do you put in the public blockchain

FJ: We're going to see many examples of private blockchains connected to the public blockchains, but there's always going to be compromises. This is where you need experts.

Since the transactions occurring on blockchains can be reverse engineered, isn't there a danger in using the technology for applications such as voting?

MC: Again, you need to be very careful about the information you put in a public blockchain. Is it encrypted? Can you abstract the information in any way? It's all about engineering the right solution, but it's not trivial.

The bitcoin blockchain has been described as the worst database in the world, only suitable for storing and transacting very small amounts of information, bloating to enormous proportions, capable of just three transactions per second with a 10 minute commit time.

FJ: There are very different blockchains for different purposes and you need to pick the blockchain that suits the particular application.

MC: The bitcoin blockchain was never designed to host big applicatons. Ethereum is much quicker and it has a 15 second block time. The next generation blockchains like Ethereum enable you to write Turing complete programs on the blockchain. True, it doesn't have a huge amount of processing power but you can code what you want. Extrapolating, you can have an application running in the blockchain and even if a node is attacked it doesn't matter.

Do these newer blockchains require the same amount of energy to run? Each bitcoin transaction uses enough energy to power a house for a day

MC: The amount of power you put into the system is equivalent to the security you get out, and relates to the amount of energy you need to attack it. Ethereum uses a different protocol but it's still proof of work. It has fewer miners on it than bitcoin but it's still very secure. How do you address the enormous energy consumption? Well, there are other protocols based on proof of stake, which use less power than proof of work.

What do you see as the most promising areas?

FJ: We chose three different technologies based on Ethereum, bitcoin and MultiChain. We started with fintech and now we are looking at manufacturing, telco, media and retail.

MC: The value is moving from currencies to smart contracts and the Internet of Things.

Want to know how IoT can help your enterprise? Come to Computing's Internet of Things Business Summit 2016 - it's free for qualifying end users