> At 11:13 AM 12/11/97 -0800, you wrote:
> >If I want to send a message to a receiver 2 things must happen:
> >1) I must change something in the receiver.
> >2) The receiver must be able to compare that change with something else,
> such
> > as the voltage on a wire now compared to what it was a nanosecond ago.
> >
> >Quantum Mechanics can do #1 but not #2 , so I have just changed one "random"
> >state in the receiver to another. The results are not really random but it
> >would sure look that way to the receiver. It's only when the receiver
> >compared his results with the records of what I did, and that can only be
> >done at the speed of light or less, would it become obvious that what I did
> >instantly changed what happened in the distant receiver.
>
> This doesn't necessarily seem true to me... You don't have to compare your
> photons against their previous states, but only against the states to which
> we had previously agreed. Suppose I tell you that every second I'm going
> to send you a photon, and that if the photon has agreed-upon property X you
> should interpret it as 0, and if it has property Y you should interpret it
> as a 1. This agreement would have to happen subluminally, but what if I
> then travelled 4E8 meters away and started up the QT machine? Wouldn't
> that be FTL communication?
>
If i read correctly, the problem is that the system doesn't always work.
It works 25 % of the time now and they think they can get it up to 75 %
(theoretical limit?) so you would never know which one of the photons was
correctly transmitted or not. Now, would CRC and parity
bit/photon checking help for such a low reliability rate?
chau
Alejandro Dubrovsky