• Google
    This Blog Web

October 2011

Sun Mon Tue Wed Thu Fri Sat
            1
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30 31          

RSS Feed

Bookmark and Share

Email Feed



  • Powered by FeedBlitz

« World Wide Web, circa 2015 | Main | A Technological Stampede! »

August 30, 2005

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

Brett Bellmore

You know, I'm not entirely clear on how quantum computers working is supposed to prove the many worlds interpretation correct.

Mike Treder, CRN

Have you read Deutsch's landmark paper?
http://www.qubit.org/oldsite/resource/deutsch85.pdf

Janessa Ravenwood

Me, too. Can you give us the short version? (The PDF link is 19 pages long.)

Mike Treder, CRN

Deutsch shows, in the paper, that there are some problems which a universal quantum computer should be expected to solve -- but which it could *not* solve without delegating subtasks to copies of itself in other universes.

That's where the speed comes from in a quantum computer: it is massively parallel with *itself*, doing work simultaneously on the same computer in different universes. Obviously, this could not occur unless Everett's interpretation holds.

Mike Deering

Mike, your answer, while correct, is still a little vague. Let me try to be a bit more specific.

Say you have a section of code that performs a logical test on a variable, and returns that variable if it passes the test, but returns nothing if it fails. What the quantum computer does is let each possible value for the variable be run simultaneously on parallel computers in parallel universes and returns the correct answer only, discarding all the wrong answers. The time it takes to perform this calculation is one loop of the algorithm, or the time it would take to test one number.

Mike Deering

I realize, after pressing "post" that you may be asking not "What does it do?" but rather "How does it do it?" which is a little more involved, but I'll be glad to take a stab at it.

Quantum computing is based on registers of qubits. Each qubit can either be a one or a zero, so a register of qubits can specify a number of states equal to two to the power of the number of qubits you have. You program your logical test and tie it to a register of qubits where each possible answer you are testing for is specified by one of the states of the register.

Once you have programmed your test and tied it to the register of qubits the correct answer appears in the register as if by magic. No, let me restate that. The correct answer appears in the register as if by a very complicated quantum mechanical process that, unless you are very up on your differential equations, is indistinguishable from magic.

Does this magic actually work? It has been shown to work in the laboratory with small registers of 2, 3, 4 qubits. Where the incredible time savings come in is when you have a large register of qubits.

One of the nice things about the quantum computer is that you don't have to understand quantum mechanics to program one, only to explain how it works.

Mike Deering

A quantum computer with a massive memory of qubit registers would be an oracle that could provide any answer, if only we can figure out how to ask the right question.

Sorry about the multiple posts. I really like quantum computers.

Chris H

One of the first big impacts of quantum computers will be the end of all existing computer encryption plus authentication systems like digital signing, that rely on encryption. All these systems rely on the fact that current computers can't quickly find the prime factors of large enough numbers. Quantum computers will be able to factor these large numbers quickly by trying to divide the large number by all possible smaller numbers at once.

Mike Treder, CRN

Thanks, Mike D.!

Martin Baldan

This is exciting news, but I'm with Brett in doubting that the "many worlds" interpretation will be widely recognised as the correct one as soon as practical quantum computing is demonstrated. The reason is that other interpretations (namely the standard, Copenhagen interpretation), unsatisfactory as they are, make almost exactly the same practical predictions, at least according to:

http://www.hedweb.com/everett/everett.htm#unique

where only three unique predictions are mentioned: exact linearity of the wavefunction, quantum gravity and a third one that requires *intelligent*, reversible quantum computer-enabled AI.

Indeed, if quantum computing could prove the other interpretations wrong, then the simplest laboratory demonstration should be enough, and they would already have been abandoned.

I am aware that David Deutsch thinks quantum computing can only be explained through Everett's interpretation, and I've found his arguments convincing, but others disagree and won't change their minds even if quantum computers are demonstrated. They will simply count it as yet another strange phenomenon of the wild quantum world that no one can hope to make sense out of.

michael vassar

No-one is "the authority on quantum computing". Peter Shore and Robert Grover have had the most success in deriving algorhythms, and I think the idea comes from Feynman, but the field has a LOT of heavy hitters. My impression is that Transhumanist thought has been severely stunted by failure to account for quantum computing in the past, and will probably continue to be. I honestly cannot, in good conscience, explain my reasoning on a public forum however. Interested people can contact me if they wish to discuss it, but no promises are implied.

I think that much of the resistance to Many Worlds comes from the fact that what Everett meant by it is very far from and does not imply the sf interpretation most of us know. Quantum computing would demonstrate that wavefunctions are ontologically primative relative to "observable" particles. If one makes certain other assumptions about qualia "binding" directly to ontological primatives, you get something sort of like the sf version of Many Worlds, but more like what Julian Barbour discussed in "the end of time".

Chris Phoenix, CRN

Chris H, I've been told that while public-key encryption may be cracked by quantum computers, other forms of cryptography may not be as easy to crack. I don't know the details, but I do know that most cryptographic methods do not rely on the difficulty of factoring primes.

Chris

Phillip Huggan

Yes, there is quantum cryptography on the horizon. Near term, with a length of fibre-optic, a tiny diamond, and a laser, it will soon be viable to create a secure communication which would instantly expose any hacker attempt.

Jay

I thought quantum cryptography was already commercially available.

Also, a few years back, I read an article that stated that researchers had proven that building a quantum computer was feasible. They didn't actually do it, because the machine's performance would equal that of a 1Mhz machine. So that was not worth the effort/investments.

The comments to this entry are closed.