The quantum computer

Printable View

07-27-2012
rogster001

The quantum computer

Hi all,

I am interested to know your thoughts on the *possible* advent of quantum computers and what would that mean, to us as programmers for one thing but also from a user point of view - it is interesting that in this article they suggest that it may well have limited usefulness - fields like code breaking would benefit, which i have read before, and can understand, but in terms of factoring out huge numbers then surely that kind of number crunching power leads to further application? There is also an interesting idea that it would be more like a new graphics card - a hardware add-on to enable quantum computation, rather than replacing the traditional computer wholesale.

Quote:

"For many questions it's not going to be superior at all. There is simply no point to use a quantum computer to do your word processing."
08-12-2012
C_ntua

It seems that a second processor makes more sense as indeed it would take much time to completely replace what is existing. Then consider that with parallel programming advancing it makes sense to use your heavely threaded GPU as well. This starts then to get complex as you would have three different processors each one with multiple cores. But since it seems that a single core processor is reaching its limits going with multiple might be the future in which case having specialized processors might be also part of the solution. Then you either need a very smart framework to choose for you or the programmer will have an additional challenge and a lot more room for optimization, which will be interesting.
08-12-2012
Elysia

To be more specific, we need better frameworks, compilers and hardware. They all have their parts in the equation.
I can already see us going down this path with more specialized circuits in the future.
08-12-2012
manasij7479

Quote:

Originally Posted by Elysia

I can already see us going down this path with more specialized circuits in the future.

Bringing a whole lot of new incompatibilities !
08-12-2012
Elysia

Yes, but like we have said, the free lunch is over. Adapt or die.
08-12-2012
whiteflags

Or not. I could probably avoid quantum programming for the rest of my working life, especially if we don't know what to do with it if/when it does show up in computer hardware.
08-14-2012
C_ntua

Quote:

Originally Posted by whiteflags

Or not. I could probably avoid quantum programming for the rest of my working life, especially if we don't know what to do with it if/when it does show up in computer hardware.

Or you will have to worry! Think that great computational power means that you might be able to break too quickly known cryptographic algorithms. Then the only solution is to use a quantum computer so you can compete. Since security is a must you might end up with a quantum computer just for encrypting and signing your data. But since it is there why not use it for something more?

But if you don't have that serious a reason to use it I don't see the market going towards more power for desktops. Mobile devices, touch screens, better graphic cards and in general more convinience and functionality is more wanted right now.
08-15-2012
grumpy

My view is that quantum computers may eventually find use in some important, but niche, areas. I can't predict what might replace conventional computers, but I'm pretty confident it won't be a quantum computer.

In rough terms, quantum computers only run probabilistic algorithms (i.e. algorithms that include randomness as part of their inherent logic, and it is not possible to confidently guess what the solution might be). So they're suited to problems like encryption, decryption, factorisation of large integers, some types of search, and a few similar things.

Most people - or businesses - don't need to do those things on a regular basis. Those that do will eventually adopt them, but only if they can receive certain assurances of reliability (eg cryptographers will want algorithms that reliably do encryption that is tough to break, even with a quantum computer, unless one knows the key). But for most other people or businesses, conventional computers are enough, and quantum computers will be a curiosity. At most, they may be a dedicated and probably expensive add-on, simply because there won't be major economies of scale in their manufacture.

The other concern with quantum computers is that programming them is significantly more difficult than programming a conventional computers (which is beyond a lot of people, as it is). So programming quantum computers will always be a niche skill, probably based - eventually - on a small number of algorithms.
08-15-2012
Elkvis

but the compilers will abstract all the hard stuff away from the programmer, so that all you have to learn is a few new libraries related to quantum processing.
08-15-2012
grumpy

It takes effort to design libraries, languages, compilers, etc to abstract hard things away from the great unwashed.

One of the joys of niche disciplines with "hard stuff" .... those in the know are less motivated to make it easier for others to join in.
08-15-2012
Matticus

I suppose it's only natural on this forum for a discussion on quantum computing to lead to the high level programming aspect of said technology.

There is an IEEE paper about quantum programming ("Semantics of Higher-Order Quantum Computation via Geometry of Interaction") that seems like it might be interesting - unfortunately, the paper must be purchased. But even the abstract is worth a peruse.

-----

Abstract

While much of the current study on quantum computation employs low-level formalisms such as quantum circuits, several high-level languages/calculi have been recently proposed aiming at structured quantum programming. The current work contributes to the semantical study of such languages, by providing interaction-based semantics of a functional quantum programming language, the latter is based on linear lambda calculus and is equipped with features like the! modality and recursion. The proposed denotational model is the first one that supports the full features of a quantum functional programming language, we also prove adequacy of our semantics. The construction of our model is by a series of existing techniques taken from the semantics of classical computation as well as from process theory. The most notable among them is Girard's Geometry of Interaction (GoI), categorically formulated by Abramsky, Haghverdi and Scott. The mathematical genericity of these techniques - largely dueto their categorical formulation - is exploited for our move from classical to quantum.
08-15-2012
Subsonics

Personally I think we will probably see memristors being used in general purpose computing a lot sooner than there is a quantum computer for sale. From what I have gathered they seems to have potential for significant change as well, and not only for storage but also for CPUs and the Von Neumann architecture.
08-16-2012
Elysia

Quote:

Originally Posted by whiteflags

Or not. I could probably avoid quantum programming for the rest of my working life, especially if we don't know what to do with it if/when it does show up in computer hardware.

It is our job to embrace new technology and tame hardware for users of our programs.
It is also our job to ensure that our programs do not compromise the security, or reliability of users' computers.
I would be very upset if I find that a program that runs on a crappy computer runs just as slow on my new shiny super computer (possibly with quantum computing).
I'd be equally upset if I find that someone reverse engineered my private key and imitated me, stealing, say, my business (we'd have to make sure our security can't be compromised by quantum, or whatever).