Moore's Law may soon reach end of computing line

If you just assume that computers will keep on getting better, faster, and cheaper, most experts will tell you not to get your hopes up too high. The much-touted Moore’s Law is running out of steam.

Back in 1965, Intel’s Gordon Moore predicted that the number of transistors per integrated circuit would double every 18 months. He was pretty well right on the money. Since more transistors generally equal more horsepower, we’ve seen greater bang for the computer buck every year. This also explains why your kids are always screaming for a hot new machine and why I was just able to buy an aging Pentium for $15 because I needed a keyboard.

Computer gurus agree that Moore’s Law is actually a tribute to good engineering at places such as Intel and in no way an actual Law of the Universe. In fact, there are some real Laws of Nature that work directly against it. Things such as the speed of light and size of subatomic particles will eventually impose speed limits on computers that use silicon chips and logic gates. At the most optimistic end of the spectrum, prognosticators such as Sheldon Ronan (who admits it’s a “long bet”) gives Moore’s Law about 50 years to run, until 2052.

Others think that physical reality will catch up with computers within a decade. They point to the fact that processor-speed improvement is already dropping, apparently only doubling every three years. Rejoice, you can now hang on to that old laptop a little longer and not feel like a fool.

But wait a minute – the answer to “Will computers keep getting better?” has just changed from no to a strong maybe.

Barry Sanders, iCore professor of quantum information science at the University of Calgary, working with colleagues in Australia recently announced a breakthrough in the teleportation of quantum information.

Bear with me please. Quantum information is based on quantum mechanics, which was the part of high-school physics where you doodled as the teacher talked about cats that are simultaneously dead and alive and pebbles that turn to dust if light strikes them. Relax. After all, quantum information is a little easier to grasp than full-blown Bose-Einstein condensates.

“Standard computer science uses bits and standard Boolean logic,” explains Sanders. “Instead, we use quantum bits (qubits) which allow the simultaneous co-existence of all states of information in some superposition. We also use a different form of processing that preserves that simultaneous co-existence. This means that the foundations of computer science and communication theory no longer apply, because we’re working in a different space.”

In other words, problems that would baffle regular computers, or take a long, long, long time to finish, might be child’s play for a quantum computer.

While this all sounds like the stuff of science fiction, Sanders says quantum teleportation is a really like a common office tool.

“Faxing is a form of teleportation,” he says. “I have a piece of paper in a particular state (with a message on it) and you want your piece of paper to adopt the same state. The problem with quantum information is that if you try to learn what it says, you destroy it, and if you try to copy it, you can’t look at the copy because you’d destroy the information on it.”

So, is Sanders working on a quantum fax machine? Not exactly. His current project involves exchanging information between quantum computers, which, by the way, don’t even exist yet as far as we know.

So researchers like Sanders do experiments with lasers and think real hard about what quantum computers might be like – if we ever got one.

“About three years ago, my colleagues in Australia demonstrated the quantum teleportation of information using laser beams,” said Sanders. “When I saw what they had done, I suggested that quantum information could be used to share secrets by entwining information in the laser beams,” he says.

The experiment worked, yielding a four-page paper in Physical Review Letters, itself densely packed with (thankfully non-quantum) information.

According to Sanders, this research has huge implications for information security. If a quantum computer could be built, it might shatter all existing cryptographic algorithms that depend on mathematically gnarly problems such as factoring large numbers. An evil genius with such a machine could wreak havoc on the world’s banking and other security-dependent systems.

Sanders’ findings give us hope that, in his words, “quantum networks can be protected from component failures and malice.” In particular, it points the way to a secure method for sharing secrets among users of quantum computers. This uses Adi Shamir’s secret-sharing protocol, which divides data so a message can only be decoded by a majority of the recipients.

According to Australian National University researcher Andrew Lance, “if an encrypted message was sent to a spy network containing 15 individuals, a minimum of eight agents would be needed to access the message,” thus limiting the risk of a small number of double agents. An added bonus is that it’s physically impossible to eavesdrop on a quantum information transmission without being noticed.

The U of C’s Sanders has started to see business interest in his work, including the defence industry. “General Dynamics Canada recently gave us money because they regard this as a disruptive technology,” he says. “It may not be significant right now, but they know this is a technology that may be important in the future and they need to know about it.”

Asked if there are any true quantum computers in the world, Sanders laughs and says that people who could build one now probably wouldn’t tell us if they did. He predicts that it will be at least 10 to 20 years before we see a quantum computer that can solve problems that can’t be handled by regular computers.

So relax, your kids won’t be asking you for one soon. But, in a stroke of natural justice, their kids just might be clamoring for that hot new Quantum PlayStation . . .

Web watch:

www.qubit.org/
www.qis.ucalgary.ca
http://qis.ucalgary.ca/~bsanders

(Tom Keenan is a professor at the University of Calgary and an expert on technology and its social implications. He can be reached at keenan@businessedge.ca)

©2004. Business Edge. All rights reserved.
Powered By Knack Marketing Communications.
Hosted By Cyberwave Technologies.