That isn't necessarily true yet, though. They solve an NP problem by exploiting physical phenomena. Quantum computers aren't programmable -- yet. You create one to solve a problem.
However, a problem is considered NP-Complete if it is equivalent (ie, reducable) to every other NP-complete problem. Thus, An NP-Complete problem must be reducable to any single other proven NP-Complete problem (which is the same property). Thus, if a computer can solve a single one, it can solve any type of NP-Complete problem. So we just need to find a way to make it take proper inputs for the Travelling Salesman problem, and bam! we can solve any NP-Complete problem.
NP-Hard problems are also problems that cannot be solved in polynomial time, but NP-Hard problems have not been proven reducable to an NP-Complete problem.
I knew I took that horrible algorithms course for a reason...
I have heard of proposed methods to create programmable quantum computers, but I am not sure that it was a practical method of doing so.
I highly suspect that programmable quantum computing will be seen in a desktop environment sometime in the future.
it really depends on the manufacturing cost, which I have no idea what it is.