CSL’s Andrew Miller is working to lay the foundation to expand the ability of blockchains. Currently, payments take a couple minutes to process, even though business often moves faster than that. In order to ensure that blockchain infrastructure is ready for the pace of future blockchain business, Miller is laying the groundwork for future advancements in his new project, “Automated Support for Writing High-Assurance Smart Contracts.”
Currently, the programming languages used for cryptocurrencies, online auctions, and distributed government are prone to error. This has resulted in multiple high-profile bugs, such as the CoffeeMiner Attack in January of last year. While the overall impact of these issues was small because of the current limits to online transactions, they could cause a much larger problem when transaction technology improves its ability to handle exchanges more quickly.
“We need to make better programming languages for smart contracts (online transactions),” said Miller, assistant professor of electrical and computer engineering at Illinois. “There is a lot of potential for people to make mistakes even with the simple aspects of these contracts.”This problem will only get worse as programming languages are asked to do increasingly more for cryptocurrency transactions. Right now, every action that takes place within a cryptocurrency blockchain, whether it’s making a payment or playing a game, takes place on a publicly broadcasted channel and is posted in the record. The posting can take a few minutes to complete. Taking a couple minutes per transaction may not be a big deal for most individuals, but it can become cumbersome when dealing with a series of micropayments or a large number of actions. This leads to off-chain payments, which raises new problems.
When payments are taken outside the blockchain, or off-chain, there isn’t a record for any disputes or dispute-resolutions and off-chain actions require expensive methods to take place. Miller, and collaborating researchers from Carnegie Mellon University, are working to build a better programming language that wouldn’t lead users to make off-chain transactions.
“The goal is to build programming languages that can make use of advanced cryptography, like zero knowledge proofs and multiparty computation, but that also helps them to do so correctly,” Miller said. “There are a lot of variations in how this technology could be used in the future.”
This research is funded by the National Science Foundation for three years at $400,000.