Comment Re:It's worse than that, it's physics Jim (Score 1) 345
Ok, just wanted to point out some incorrect statements here.
A. You said 'The trick to QC is that there's an algorithm which can calculate exactly which bits were sniffed, so that a key can be composed of the remaining safe bits.' Not true. This sort of processing is called privacy amplification and it is used to compensate for inefficient quantum channels. Basically if you are not sending a single photon for each bit of your quantum key, there is a possibility that an eavesdropper could use a beam splitter to observe some of your key bits. To account for this you use privacy amplification to throw out a certain percentage of the bits in order to stimy an eavesdropper. With a 100% efficient system this would not be necessary.
B. You say 'QC can be done with photons, molecules in NMR, electrons, etc. Anything that can be reduced to an EPR pair (or alternately, a Hadamard gate) is a basis for QC.' This sentence makes no sense to me. You produce EPR pairs and you perform Hadamard gates on objects, I don't understand what you mean by 'reduced to'. Also an EPR pair and a Hadamard gate are not at all the same thing. Hadamard gates are used in quantum computing to initialize qubits to initial states that are perfectly juxtaposed between 0 and 1 states, but I do not see how this could be used for quantum cryptography. Also, using QC to describe this is ambiquous as you could mean either quantum computing or quantum cryptography. Better terms are Quantum Information Science (QIS) and Quantum Key Distribution (QKD).
C. Grover's algorithm can do a search in 0(log(n)) not 0(sqrt(n))
A. You said 'The trick to QC is that there's an algorithm which can calculate exactly which bits were sniffed, so that a key can be composed of the remaining safe bits.' Not true. This sort of processing is called privacy amplification and it is used to compensate for inefficient quantum channels. Basically if you are not sending a single photon for each bit of your quantum key, there is a possibility that an eavesdropper could use a beam splitter to observe some of your key bits. To account for this you use privacy amplification to throw out a certain percentage of the bits in order to stimy an eavesdropper. With a 100% efficient system this would not be necessary.
B. You say 'QC can be done with photons, molecules in NMR, electrons, etc. Anything that can be reduced to an EPR pair (or alternately, a Hadamard gate) is a basis for QC.' This sentence makes no sense to me. You produce EPR pairs and you perform Hadamard gates on objects, I don't understand what you mean by 'reduced to'. Also an EPR pair and a Hadamard gate are not at all the same thing. Hadamard gates are used in quantum computing to initialize qubits to initial states that are perfectly juxtaposed between 0 and 1 states, but I do not see how this could be used for quantum cryptography. Also, using QC to describe this is ambiquous as you could mean either quantum computing or quantum cryptography. Better terms are Quantum Information Science (QIS) and Quantum Key Distribution (QKD).
C. Grover's algorithm can do a search in 0(log(n)) not 0(sqrt(n))
