OSNews

Well, it's in the nature of quantum computing to help in areas where traditional machines can't. For example, SETI or medical or math problems could be solved much faster. And cracking encryptions is now substantially easier.

Edited 2007-02-14 01:21

Except that now quantum encryption is possible, which is IMPOSSIBLE to decrypt, because the process of decryption would alter the payload, thus rendering it invalid.

I think you're right that quantum encryption/decryption is useless, but quantum decryption of digital encryption is possible, and this could possibly be executed so quickly as to render digital encryption useless. Therefore encryption of any kind becomes useless?

I don't know whether to be overjoyed or frightened by this...

Edited 2007-02-14 05:40

Quantum encryption and VALID decryption works fine. It's basically unbreakable (as far as we know.) In other words, using quantum encryption, I can send a message anywhere, and *know* it was only readable by the intended RX.

The stream cannot be "touched" between sender/reciever without the payload being invalidated, that's what I was trying to get at.

So, in a way, it's great - no government agencies spying on us. In another way - it's bad. No more spying on enemy agencies. Ever. Well, unless we come up with some new ideas that radically change the face of quantum mechanics. :p

The stream cannot be "touched" between sender/reciever without the payload being invalidated, that's what I was trying to get at.

Sounds like a pretty easy DoS opportunity. Just touch the stream every time and the intended recipient will never receive it.

How will you know it's being sent? Good try, though.

That applies to any form of communication. What the data contains is irrelevant if you can interfere with the transmission.

No, the greatest trick is to allow the communication to go on apparently untouched and safe, while actually eavesdropping. That is impossible with quantum technology, because it's not possible (at least for now) to reproduce the same data stream once you've touched it.

Sounds like a pretty easy DoS opportunity. Just touch the stream every time and the intended recipient will never receive it.

Not true. The message will arrive, but it will be rejected due to the fact that the message was 'altered'.

So, in a way, it's great - no government agencies spying on us. In another way - it's bad. No more spying on enemy agencies.

I think you're being over-optimistic. As with any new technology, the price will be prohibitive for regular consumers. Government agencies will afford it, while you will not. This means that the government will be able to (a) use unbreakable encryption; (b) make the "old" encryption that you're using completely useless.

And they might also pass laws saying that quantum technology qualifies as "weapons" and you may not own and use it even if you can afford it, so it won't fall into the hands of "terrorists", God forbid. Sounds familiar?

And they might also pass laws saying that quantum technology qualifies as "weapons" and you may not own and use it even if you can afford it, so it won't fall into the hands of "terrorists", God forbid. Sounds familiar?

Hmm, isn't it so that in a democratic society, *we* are able to pass laws restricting how authorities can use this technology?

Ahh, the naive. The US already has very strict regulations on what kind of encryption can be exported overseas (only extremely weak stuff now). It's only a couple of small steps away from limiting what American citizens can use as well. After all, if you have nothing to hide then you won't mind, will you? And if they let this out it's only a matter of time before the terrorists and child porn viewers start using it.

Supposedly someone has figured out how to entangle (or whatever the magic is) the photons in a stream and then send them happily on their way and so eavesdrop on communications. Not sure about altering them though. Sorry I can't remember who announced it.

the question is, is this really where computing as we know it is heading?

Obligatory: the question is, does it run Linux? ;-)

The question is rather: "will it run Vista?"

Yes, because Microsoft is known to make version of it's operating system compatible for almost every architecture out there... *head -> wall*

"The question is rather: "will it run Vista?""

no, i think it won't run vista, vista is too heavy also for a quantum computer

Well, have a quantum computer and run vista in it, is as stupid as buy a ferrari for run faster and atach to it a trailer

The question is: will it run Duke Nukem Forever?

"NP-complete" is a designation of something a computer cannot compute precisely in a fashionable time.

You made a couple mistakes there - first, quantum computers can't efficiently solve NP Complete problems, only BQP.

From Wikipedia:
BQP is suspected to be disjoint from NP-complete and a strict superset of P, but that is not known. Both integer factorization and discrete log are in BQP. Both of these problems are NP problems suspected to be outside BPP, and hence outside P. Both are suspected to not be NP-complete. There is a common misconception that quantum computers can solve NP-complete problems in polynomial time. That is not known to be true, and is generally suspected to be false.

Second, computers can compute a NP Complete problem, they're just too slow at it to be useful. It isn't impossible though if you made one fast enough.

NP-Completeness is about how the complexity of the program (run time) scales with the problem. You can have a NP-complete problem with a certain amount of input and have it solved quickly with a certain machine (doesn't matter how fast it actually is). Now increase the inputs, and the run time of your algorithm will be increased *a lot more*.

So computers can always start computing NP-complete problems. They can compute everything that isn't incalculable; the halting problem would be an example for something incalculable, for which no algorithm is possible (proven by mathematics).

But fast computers don't help with NP-complete problems. They can compute a certain problem, now you make the problem slightly more complex and your computer will take a lot more time.

Quantum computing is very difficult to understand; the point is that incredibly high numbers of ways can be tried simultaneously, but if you actually want to see "the" result, you run into a lot of issues.

Dwave is apparently having another demo at their HQ in a few days, that Ars Technica will be attending.

From what I read about it, it could be a sham, but you wouldn't really know until later models, which really should outperform regular chips.

Not sure if I really trust Ars in this situation to begin with. They have some bright folks there, but I doubt any of them are all to familiar with quantum computing. Reading a book on it certainly won't make you an expert. I'd like to see some real researches get some hands-on time, possibly after being given all the specifications/interfaces/whatever they needed to write controlling software - and give it a REAL workout.

D-Wade has one too. Charles Barkley is jealous.

That can be said about a lot of today's techology if you stop to think about it. It's just that we've become too used to it to notice. It stops looking like magic once you understand how it works, but how many of us can say that about everything?

Lol! Would be very cool!

I'd like to see how many cats in boxes they could fit on a PCI-E card. I kinda get the impression that all the noise reduction steps I have taken will be negated by all the meowling.

I know of one person who would not be happy with this at all! When I tried to explain Schrödinger's problem to my girlfriend, she told me as a hypothetical question, it is a pill of sh*t and no longer wanted to hear any more nonsense from me.

The things a man puts up with for luv...

Running an OS on this just seems like a waste. Wouldn't it make more sense to run the OS on a "classical" computer and just use this as a coprocessor for quantum problems?

Some problems need to be fully simulated using quantum logic from inception to execution. Problems such as observance and environment feedback / noise elimination would be better simulated using a quantum universe as the base. Perhaps this logic chip could be properly used with the brain simulators currently in development, which makes your point a cost effective solution. I still think however that if you have a solid operating foundation you can attack really complex problems.

Edited 2007-02-14 20:10

Aren't Quantum Computer programs probabilistic in nature? I may have misunderstood, but I thought that you arrived at a solution for a problem by repeatedly executing problem and measuring the result until you get a reasonably accurate answer.

I think also that using standard program style logic is impossible, since you have to perform each part of the algorithm in exactly the same order every time. Writing a Quantum OS that functions in any way like an OS how you are used to them is probably impossible.

Aren't Quantum Computer programs probabilistic in nature? I may have misunderstood, but I thought that you arrived at a solution for a problem by repeatedly executing problem and measuring the result until you get a reasonably accurate answer.

Yes, that's right. Although I assume the hardware itself automatically reruns the problem until it can say with certainty it has found the right one, but maybe this still depends on software.

Say goodbye to the digital age and welcome back the analog one.

Edited 2007-02-14 23:04

"Aren't Quantum Computer programs probabilistic in nature?"

Yes, but technically digital computing is as well. The difference is that we don't use, and try to minimize, that property when computing something digitally. In fact, a digital system can be seen as nothing more than an analog system with ECC. The ECC helps to prevent 'errors', but when one occurs, it's typically catastrophic.

Much like what happens when you observe a closed quantum system before it's ready to be seen.

On the other hand we try to add that property back in certain applications that use floating point and randomization. FP gets less accurate the greater the magnitude (it was never designed for 'precise' results and you can't use it to 'drive' logic), and the current generation of graphics HW uses a lot of chip real estate to properly randomize rendered results. Such applications could benefit greatly in terms of heat, power, and speed from quantum computing.