Quantum Computerat, galaxy

[sn0m]

Quantum computers

Dream code


Programming languages for quantum computers are now being written

THERE are times when putting the cart before the horse is actually sensible. Quantum computers, which rely on the arcana of quantum mechanics to do many computations in parallel, are a long way from actually being useful. But researchers are already trying to work out how to write programs for these almost non-existent devices, in the belief that learning how to do so might help engineers to design the computers in useful ways. A paper by Stefano Bettelli of Paul Sabatier University in Toulouse, France, and his colleagues, which has just been accepted by the European Physical Journal, describes the latest effort to come up with a quantum programming language.

The basis for existing “classical” computers is the binary digit, or “bit”, which can have a value of either 0 or 1. In a quantum computer, bits are replaced by “qubits” which are in a “superposition” of states—partially 0 and partially 1. It is this superposition that allows calculations to be performed in parallel. Measuring the value of a qubit causes it to collapse into one of the two classical bits, 0 or 1. In a well-organised quantum computation, that should not happen until it becomes necessary to find out what one of the values actually is.



Quantum heads and tails
Jan 2nd 2003
Why quantum computing is hard
Mar 8th 2001


Computer technology


See “Toward an architecture for quantum programming”, by Stefano Bettelli et al.





That, at least, is the principle. Converting it into practice will be tricky. Dr Bettelli and his colleagues have, nevertheless, tried. The key elements of their language are things called quantum registers and quantum operators. The quantum registers are ways for a program to interact with specific qubits. They act as “pointers” to the locations of qubits within a machine. Those qubits can then be manipulated by the program.

The manipulation itself is done by the quantum operators. These are the equivalent of the logical operators, such as “and”, “not” and “or”, that are the basis of classical programming (in which an instruction might say “when A or B and not C are true, do D”). Quantum operators rely on what are known as unitary transformations (the origin of the name is buried in the algebra of matrices). The trick is to find a way to describe, in a manner useful to computer scientists, the unitary transformations that underlie a program. Dr Bettelli has managed to do it using object-oriented programming—long a buzzword among software developers.

Object-oriented programming works by combining both commands and data into individual bundles known as objects. These objects can be used to bridge the gap between the classical and quantum worlds. Because a working quantum computer is likely to be a specialised portion of a larger classical computer, any successful language will have to be able to handle registers and operators in ways that allow them to be integrated with old-fashioned classical computations. Representing a unitary transformation as an object makes it fairly simple to translate programming directives at the classical level into physical control instructions at the quantum level. Dr Bettelli's language should do this.

It is still early days, clearly. But it is worth recalling that much of the past half-century's progress in classical computing was both predicted and made possible by such prescient theorists as Alan Turing and John von Neumann. Although programming languages may change a great deal before usable quantum computers can be built, it is worth constructing the theoretical underpinnings now. Quantum programmers would do well to heed the words of Dwight Eisenhower: “I have always found that plans are useless, but planning is indispensable

[SH_H]

Keta kompjutera do jene shume te fuqishem.