- #1

VantagePoint72

- 821

- 34

Now, at one point there's an unnecessary, but largely inconsequential, confusing of the uncertainty and superposition principles—but that's not the issue. The offending piece is this:

It turns out that classical computers are not very good at factoring large numbers, a weakness that has long been exploited by cryptographers to safeguard data on the Internet. It is easy to multiply two prime numbers in order to produce a much larger number, but it turns out to be horrendously difficult to engineer the same process in reverse, to find the two prime divisors of a large number, a process called factoring.

The only way classical computers can address the challenge is by systematic trial and error — trying out two numbers to see if they work, discarding them, trying out two different numbers, and so on. There’s no shortcut.

...

By contrast, a quantum computer could crack such privacy barriers in an instant, by the dazzling expedient of testing every possible combination of divisors, not one by one, but all at once, something no conventional computer could do. The right answer would reveal itself almost immediately.

This is not an oversimplification, it's quite simply wrong. Shor's algorithm, and every other quantum algorithm, does not run in a single step. Yes, technically every possible input is operated on simultaneously—but at the cost of getting every possible output simultaneously! Extracting useful results out of a quantum computer is considerably less straightforward and uses the fact that the quantum amplitude is what obeys superposition but its norm-squared is what gives the probabilities. Maybe this is too difficult to explain non-technically, but that's no excuse for saying something objectively false in its place. And yet, every article I've ever seen about quantum computing contains this same misinformation. The result is that to the extent quantum computers are a part of the public imagination, their potential computing power is

*grossly*over stated in people's minds.

I'm curious about the origin of this problem. Notice that in this article, the offending piece doesn't actually quote the researcher, Ray Laflamme. Still, I wonder if the journalist who wrote this got his awful description directly one of the other researchers he spoke to. I wouldn't be too surprised since I have, on some occasions, heard physicists—ones who should know better—saying things similar to what this article says. So, for those of you who work in quantum information: where do you think the blame falls? Do you think your colleagues generally give reasonably accurate explanation of quantum computing to lay audiences, but the message gets lost in translation? Or do they give into the temptation to give lazy, wrong answers because it's a difficult subject to explain? If it's the former, do you think there are steps the scientific community can take to remedy the problem? If the latter, is there any way to get the message to other physicists that this sort of thing is

*not*OK?

Right now, there exists a serious barrier of understanding between those who study quantum information and everyone else. I think we need to try to understand the origin of this barrier; hopefully, doing so will help us take it down. Appropriately enough, this article gets one thing right when it says that "no matter how difficult they might be to fabricate, quantum computers are even more difficult to explain." But that's no excuse for not trying.