In researching this week's story on quantum cryptography, I had the opportunity to speak with IBM research scientist Charles Bennett, who teamed with University of Montreal's Giles Brassard two decades ago to invent the underlying technology. Remarkably, their BB-84 quantum-crypto protocol has withstood the test of time and remains the foundation for the first commercial products for encrypting communications in this unconventional way: with single photons.

Comprehending quantum crypto is not for the weak. It's said not to be quantum computing but is part of quantum mechanics. But what the heck is a quantum anyway? In physics, according to one dictionary, it's "the smallest amount of a physical quantity that can exist independently, especially a discrete quantity of electromagnetic radiation."

Quantum crypto relies on manipulating a photon-"the quantum of electromagnetic energy regarded as a discrete particle having zero mass, no electric charge and an indefinitely long lifetime" - to route it in a string of bits using state-of-the art optoelectronics, such as light-emitting diodes. The end result desired is a quantum transmission of information that's encrypted in a manner wholly unlike traditional cryptosystems, except to say it's based on private key.

And indeed, it turned out that speaking with IBM's Charles Bennett about quantum crypto - analogous, maybe, to discussing the incandescent light bulb with Thomas Edison - it does mean talking about very small particles. It's about taming them. It's also about how weird things get with really tiny stuff.

"Small particles behave less predictably and less controllably than you would want," offered Bennett in an explanation that at times rang with mystical overtones. "A newspaper can be read and you don't hurt it by reading it. For really small things, the process of observing them disturbs them."

People several decades ago who first began to notice this "thought this was mostly a nuisance," says Bennett. But of course it was intriguing as well, and eventually led to the idea of the "uncertainty principle," that if you measure one quantum accurately, it disturbs another.

But how did Bennett and his colleague Brassard ever stumble onto the idea of making photons dance the dance of crypto? Bennett, a physics student at Brandeis and later Harvard, gives credit to a friend named Steve Wiesner for that eureka moment.

Wiesner, said to be scientist living in Israel now, back in these university days kept coming up with novel but impractical ideas, like "quantum money" and "unforgeable bus tokens," says Bennett. In the ferment of the imagination, wild ideas like these inspired Bennett to work with Brassard to nail down the BB-84 protocol, which wasn't tested out extensively until the 1990s.

The first quantum crypto products based on BB-84 are finally becoming available for use with fiber-optics lines. Much more research is still needed, especially for wireless "free space" quantum crypto that might reach up to secure satellites.

All this discussion about quantum information is a bit dizzying for the uninitiated. But perhaps in acknowledgement of that, Bennett concluded by saying, "Quantum information is like information in a dream. Pretty soon you only remember what you said about the dream, not the dream itself. It intrinsically exists only in one copy."

He said he didn't particularly care for that saying attributed to scientist Niels Bohr, that "if quantum mechanics hasn't profoundly shocked you, you haven't understood it yet." Bennett said that sort of remark probably only discourages people from learning more about quantum crypto.