Emin Gün Sirer: Real Potential of Blockchain is in Development of New Methods of Data Settlement
Emin Gün Sirer is one of the most prominent and respected figures in blockchain development. An associate professor of computer science at Cornell University, co-director of IC3 and co-founder of @bloxrouteLabs, he has produced some of the most influential research papers into Bitcoin.
Binary District Journal caught up with him at the Genesis London blockchain conference, where he led a discussion on a range of topics including decentralisation in Bitcoin and Ethereum, the use of hardware to settle more cryptocurrency transactions, and issues with the scalability of Bitcoin.
Building Blocks to Scale Bitcoin?
The Bitcoin network is being underutilised, Sirer believes. This is because the current one megabyte limit on blocks is insufficient.
For many years, the block size debate has split the Bitcoin community into two: supporters of the 1MB block size and supporters of a larger block size to increase the capacity of the Bitcoin network.
The argument in favour of a 1MB block size is that if the block size is increased, it may interfere with the decentralisation of Bitcoin as it could prevent nodes from being synchronised efficiently. In Bitcoin, and most public blockchain networks, each node is required to go through every single transaction that is settled. If nodes are required to verify a larger number of transactions due to bigger blocks, node synchronisation could be affected.
The supporters of a larger block size, however, have argued that the exponential improvement in the chip manufacturing industry and the rapid decline in costs for memory have made node synchronisation a non-issue, given that node operators can purchase large memory at lower costs. A few years ago, 500GB hardware exceeded hundreds of dollars because hardware with large memory was scarce. Today, 1TB hardware memory can be purchased for less than a hundred dollars.
“The costs of running a node is a much, much smaller problem. Everything says that the blocks can be increased”
“Indeed the costs of storage have dropped immensely, and so all of that tells us a very large story that says the resources are much cheaper,” Sirer says. “Because of the bandwidth increase, the orphan issue is much less of a problem than it used to be. Because of the drop in costs, the costs of running a node is a much, much smaller problem. Everything says that the blocks can be increased.”
During his presentation at the Genesis London blockchain conference, Sirer also revealed the findings of a Cornell University study in decentralisation in Bitcoin and Ethereum. The study disclosed that an increase in the capacity of Bitcoin nodes by a factor of 1.7x does not have any impact on the decentralisation and node synchronisation of Bitcoin. Sirer pointed out that because these findings were released in 2017, the factor of improvement could have risen substantially as of 2018.
“That's what the data says,” Sirer tells us. “It's not something I think, it's just something. The data says that if people were happy with the amount of decentralisation in 2016, then the network at the moment is much, much, much better, by about a factor of 1.7x, and we haven't taken advantage of that available extra resource that came online.”
Also at Gensis, Sirer’s colleague at Cornell, researcher Iddo Bentov, discussed the possibility of utilising hardware systems such as Intel’s SGX to process Bitcoin and other cryptocurrency transactions at a large capacity. The basic concept of hardware-based off-chain cryptocurrency payments is using peer-to-peer channels to send payments with lower fees, like a micropayment system.
BloXroute, a company advised by Sirer, is working to commercialise hardware-based cryptocurrency settlement systems as a second layer scaling solution. Sirer explained that a hardware-based system could increase the transaction capacity of any public blockchain like Bitcoin and Ethereum to several thousand transactions per second.
“The network layer of blockchain has been ignored by many, and that's where a lot of the performance lies”
“The vision for this company is to act as the infrastructure provider for other people who want to build more exciting, different consensus layers, and smart contract layers on top,” Sirer says.“The network layer of blockchain has been ignored by many, and that's where a lot of the performance lies.”
“A lot of the current protocols that we see – certainly the Bitcoin protocol, certainly Ethereum as it stands right now – do not take advantage of what could be done much better at the network layer. We're going to elevate that with bloXroute, and by doing so, we unleash many, many higher transactions per second for the blockchain itself that runs on top.”
During his presentation, Sirer further emphasised that studies show that fewer Ethereum nodes are connected to institutions in comparison to Bitcoin, suggesting that the Ethereum network is actually more decentralised than Bitcoin. Given Ethereum’s focus on Sharding and Plasma, two scaling solutions that are expected to optimise the way nodes verify transactions, Sirer and other researchers at Cornell and bloXroute are interested in Ethereum.
“The data shows that the nodes are both in the latency space,” says Sirer. “And also geographically more distributed round the world. Ethereum nodes tend to come from all sorts of places, smaller networks, and homegrown entities, as opposed to Bitcoin nodes, which tend to be located in data centres. Our study found that the majority of Bitcoin nodes, 56%, are in data centres.”
Real Potential of Blockchain
According to Sirer, the real potential of blockchain technology is in the development of new processes, systems, and methods of data settlement, rather than placing existing infrastructures and assets on top of the blockchain.
Several blockchain projects are currently focused on placing assets like gold and real estate registries on the blockchain. While such activities could be useful, it is difficult to consider them as innovations because those type of projects do not improve the applicability, accessibility, and adoption of the blockchain. “There is so much interest in this area because a lot of people realise that this is indeed going to change the way we do things,” Sirer says, anticipating a muddled and somewhat flooded scenario as a result of its popularity.
“The real potential of blockchain technology is in the development of new processes, systems, and methods of data settlement”
“What will happen is we will see a chaotic world of people pushing all sorts of things,” he says. “It will have an enormous component of scams, and people pushing things that will not work. It will have some component of people pushing stalled, regular products, some not so exciting ideas, except with blockchain.”
Technologies like off-chain protocol settlement and solutions that can optimise the way blocks are mined, nodes verify transactions, and data is transferred, can revolutionise the implementation of blockchain technology.
Right now, the supporters of the 1MB block size seem to be winning the battle. Bitcoin, which has kept its 1MB block size, maintains a $155 billion market capitalisation after incorporating segregated witness (SegWit2x) technology. This follows a vote in mid July 2017, where mining pools and companies representing roughly 80-90 percent of the Bitcoin computing power voted in favour of SegWit2x, which reduces the amount of data verified in each block by up to 65% by removing signature data and having it attached in an extended block. However, Investopedia reports that talk of doubling the block size to 2MB in November has ramped up and is expected, which would go some way to improving Bitcoin’s scalability.
Bitcoin Cash, in comparison, was formed after a hard fork from Bitcoin over concerns about the transparency of SegWit2x and whether the cryptocurrency was following the vision of Satoshi Nakamoto, the anonymous developer behind Bitcoin. It increased the block size from 1MB to 8MB to accelerate the verification process. Its market capitalisation lags behind at a far from insubstantial $27 billion, but its adjustable level of difficulty, which ensures that the chain’s survival and transaction verification speed, has raised concerns about its security regardless of the number of miners supporting it.
Illustrations by Kseniya Forbender
To contact the editor responsible for this story:
Margarita Khartanovich at [email protected]
- Artificial Intelligence Isn’t Ready to Take Over From Doctors and Nurses, Just Yet
- Machine Learning Vs. Analysts: Will AI Eventually Replace Data Scientists?
- Why The Danger of Deepfakes Is No Danger At All
- Sarcastic Robots? How Deep Convolutional Neural Networks Are Making AI Worryingly Human
- Traffic Lights in The Sky: Flying Cars Can Appear Sooner Than You Think