[alexnking]

  "Such systems make it impossible to intercept 
  and read messages sent over a network, because 
  the very act of measuring a quantum object 
  changes it, leaving behind telltale signs of 
  the spying.

  'It’s unlikely everyone would need this type of 
  technology...'"

I wonder if they're saying it would be impractical at a large scale, or that it's not necessary because it's not like everybody is being spied on or anything :)

[beloch]

Google vacuum chambers and optical tweezers if you want to see why these switches might be impractical for home users. The ease of scaling they refer to is likely just the number of switches they can fit in one chamber.

On the other hand, you don't need a classical switch in your house to get on the Internet. If quantum repeater networks ever roll out to people's homes (a very long time from now), the switch will only need to be within a hundred kilometers or so of the user. By then, this sort of construction won't be out of the question for telco substations. Still, an entirely solid state solution would obviously be preferred.

[qwerty_asdf]

I'm assuming that the "switches" they're currently experimenting on also probably aren't designed to function as telecommunications switching equipment might. I'd imagine they have all the frills and features of a circuit breaker or a knife switch.

So, in that sense they really are "classic" switches in terms of representing the raw fundamentals of functionality for the purposes of experimentation, but not in the sense of "classical physics" where the off state is caused by introducing a physical air gap to the circuit.

[jerf]

http://en.wikipedia.org/wiki/Quantum_cryptography

In theory, it's actually impossible to secretly snoop on such a connection, not merely "very hard".