The Lee Center

JASON HICKEY

Networked systems can be a major tool in a crisis, but, on a daily basis, we will still use them to purchase products using a credit card, bid and buy from a Web auction, and contract for services. All these activities require a balance between ease of access and security (against viruses, hackers, etc).

Lee Center member Jason Hickey, Assistant Professor of Computer Science, says mathematics will help with day-to-day security. The goal of good programming, he says, is to make it impossible for a computer programmer to write something that could damage a program in someone’s computer—whether intentionally or accidentally.

Increasingly, bad code is more than just a nuisance; it could literally be a matter of life or death. Automobiles and passenger planes now routinely use computers and software to control fuel injection, smog control devices, air bags, and engine checks, for example. “Obviously, [the airline industry] can’t afford a failure, so currently they must spend millions of dollars to have every line of code checked.”

But the coding language in use today is poorly written and susceptible to hackers, says Hickey. “Our effort is to come up with a set of high-level mathematical rules where a computer programmer is constrained in what he or she can write,” he says.

Hickey is developing “automated theorem provers,” a kind of toolkit containing a set of mathematical rules or equations. A programmer would load into this toolkit the set of equations that will be the foundation for a new piece of coding. The toolkit would do one of two things—check the newly written code to see if the equations are correct or search the Internet to see if such a proof, or line of code, already exists that can be used to fulfill the same objective. Once checked, the toolkit would issue a certificate guaranteeing that a particular piece of software is error free.

In this way, “malicious math” couldn’t be hidden within the new code as long as the automated rules are correct. And the process needs to be automated, Hickey says, because there are hundreds of millions of lines of source code, and it is physically impossible to check it all.

The existence of a global network that is guaranteed to be secure would open up other possibilities for its use besides communication and information gathering. For one, a secure global network would behave more like a utility. When traveling, for example, you wouldn’t need to lug your laptop, files, and software. Instead, your files and software could remain on a network, accessed through a simple terminal at your destination. Hickey has a large ongoing project he calls “transparent mobile computations” that would make a person’s working environment completely mobile.

Another possibility, would be eliminating the perpetual replacement of computers and software. Instead, an individual could pay a single fee to a company that would periodically supply hardware and software upgrades.