Original URL: http://www.theregister.co.uk/2006/05/04/quanutm_analysis/
QIT: quantum hope or quantum hype?
Certain uncertainty surrounds quantum information technology
Posted in Physics, 4th May 2006 08:37 GMT
Free whitepaper – Application Performance Management:
Quantum information technology (QIT) is here already. Judging by the impressive turn out at the Cambridge-MIT Institute's recent Industry in the Quantum Age chinwag session at the Royal Society, it's here to stay.
Also clear, though, is that despite repeated forecasts that we're on the precipice of a revolutionary base jump, even the basics of QIT are poorly understood outside the research community. The research community aren't necessarily always the best people to turn science into real-world applications, either, so the questions remain: will it change the world, and if so, when and how?
FundaMental
A quantum is the smallest unit energy can be divided into. So for light, that's a photon, and thanks to Einstein, we know mass and energy are interchangeable, so a quantum can also be a particle like an atom or an electron carrying energy.
QIT is the party where physics meets computer science. Applications take advantage of the Addams family of freaky effects that come in to play on the quantum scale.
As it continues to pack more and more transistors onto silicon, traditional digital technology will have to deal with quantum effects soon (see the interconnect bottleneck here [1]), but the game for the old guard will be how to dodge them, not embrace their potential.
The science of secrecy
Quantum cryptography exploits Heisenberg's uncertainty principle to make for unbreakable security. It states properties of quanta are mutually inaccessible to an observer. Most commonly so far in quantum cryptography, the polarisation direction of a photon has been used.
The principle is at the heart of the conundrum posed by quantum properties and means that just by breaking into a quantum encrypted key exchange and determining the direction, a hacker is instantly detectable, and anything they get from the communication is worthless. The polarisation direction the photon is emitted in is random; a hacker would have to guess which direction to measure it in, is bound to make mistakes, and because of uncertainty, affects all subsequent quanta.
Quantum cryptography is already here [2], of course. It's high-end stuff, and MagiQ, which makes and sells a box to do it, says most of its customers are exploring the possibilites, like US telco Verizon, or too hush-hush to talk about. Former MI6 chief Sir Richard Dearlove was at the Industry in the Quantum Age workshop, so its pretty clear the sort of government agencies that have an interest.
Current quantum cryptography systems suffer from some limitations. The distance the emitters can send light through optic fibres is an issue, currently at a maximum of around 100km. There's a need for true single photon detectors and emitters to be deployed to supercede current hardware compromises too. Such road humps will be summited in pretty short order if the industry buzz is anything to go by. Along with start-ups like MagiQ, big boys Toshiba, IBM and HP have quantum cryptography specialists beavering away at making the technology more stable and accessible over the next five years. A system that uses quantum cryptography to secure cashpoint transactions is in development.
Spooks aren't just interested in the quantum encoding of information. The flipside is that a quantum computer's theoretical processing grunt could smash their top secret communications channels in no time. Quantum cryptography pioneer Gilles Brassard once said: “If a quantum computer is ever built, much of conventional cryptography will fall apart.”
The party planner's trilemma
A working quantum computer with such practical applications is further away, but the experts reckon the implications are profound enough that the IT industry should start educating itself today.
Bob's got a problem. His cellar has three lights in it, and the three switches which operate them are upstairs. Bob needs to know which switch works which light so he can set the perfect mood for an upcoming cheese, wine and gimp evening he'll be having with a few close friends.
“Simple”, he thinks at first; flick one switch, nip downstairs, note which bulb lights up, repeat once and by a process of elimination all three circuit paths are illuminated.
Bob's an idle fetishist however; just the thought of walking up and down the stairs of his suburban faux-Gothic pile twice gives him a bout of angina. After several days of feverish Da Vinci-esque kitchen brainstorming, a whole pack of jumbo chalks and 62 cups of weapons-grade Java, the answer reveals itself.
Bob flicks one of the switches, leaves it on for ten minutes while he slips into something more uncomfortable, and then turns it off. Then he quickly flicks another and squeaks downstairs. Obviously, he knows which bulb is powered from the switch he just flicked by looking.
Bob's genius is to feel the other two bulbs. One is warm, so he knows that was the one that the first switch turned on, and therefore, by elimination which bulb the final switch operates.
That, in a somewhat abstract nutshell, is the power behind quantum computing. Part of the reason quantum IT is so poorly understood is that things get pretty brain-melting when the experts talk in real-world specifics. Thought experiments are key to understanding the field.
Bob's first thinking is traditional and binary; the bulbs are on or off, one or zero. His solution is the essence of quantum computing; there is more than two states the bulbs can take.
“No Noah, qubits”
Bob's quantum solution is simple. The state of each bulb is a qubit – a quantum bit. The thing that makes qubits more powerful than regular ol' bits is another of those weird quantum effects: entanglement. Entanglement means that added together qubits can express more than one state simultaneously, just like Bob's warm bulb. For a heavier reading of how quantum changes things in terms of logic gates, see here [3].
The physical form qubits will finally take in quantum computers isn't yet clear. There's a gaggle of contenders, and some analysts think the reality may be that different quanta will be set on different problems. Photons, ions, electrons, regular atoms, carbon nanotechnology and silicon dots are all working as qubits, and in varying stages of development. They use quantum properties like polarization, magnetism, spin, and phase transitions to hold the information. What they have in common is being a way off hitting the streets. Estimates vary fairly wildly as to when we'll see the first quantum powerhouse proper with upwards of 40 qubits crunching Very Hard Sums, but it's difficult to find anyone in the know who'll say fewer than 10 years. Most punt between 15 and 25.
Money is finding its way into the field, but too slowly for some. HP's quantum guru Tim Spiller said quantum R&D investment needs to increase by an order of magnitude to bring about the “Quantum Age”.
Pipeline hypetime
When there's an Age on its way, or even the dreaded paradigm shift, the hype factor inevitably enters the fray. Analyst firm Gartner's hype curve is already vigilant for signs of industry reaching “the plateau of disillusionment” with quantum computing (see the 3G blunder). Overselling quantum could be to book the undertaker before the baby's even walking.
Science has been burnt of past controversies too; veteran Princeton quantum scientist Herschel Rabitz recalls a debacle in the early 80s over a misreported finding in his field of controlling chemical reactions through quantum manipulation. He thinks the fallout set it back 15 years. GM crops and Prince Charles' infamously Luddite “grey goo” comments on nanotech mean scientists can get suspicious when outsiders start taking an interest.
Media reports when scientists announced they had “teleported” the quantum state of a photon precipitated inevitable Star Trek screen grabs in the press.
Insiders appreciate quantum's killer application may not come from research. The people who invented the laser used it as little more than a lab toy. Tim Spiller said: “We should at least be prepared for the fact that the inventors of major QIT applications may not have PhDs in quantum physics!”
In the meantime, interested parties are concentrating on the more mundane aspects of building an industry. Another Cambridge-MIT meeting later this month will set about trying to establish standards. Quantum cryptography buyers have a problem that they cannot observe their system working, for example. A standards-based stamp of approval might lay minds at ease.
Phew. The last word goes to Nobel Prize winning quantum physics genius Richard P Feynman [4], one of the high priests of the quantum cult. He said: “Nobody understands quantum theory.” He was including himself. ®