On 9/18/2011 9:37, Warp wrote:
>    On earth we receive these two secondary beams, and here we make a delayed
> choice of whether to merge them or not before measurement. If we merge them,
> then the interference pattern appears at Alpha Centauri, and if we don't
> merge them (but measure which slit the photons went through) the interference
> pattern disappears.

Except you can't tell, in the case of this specific experiment, which 
electrons are part of the interference pattern and which electrons are part 
of the non-interference pattern until after you've detected the "B" idler 
photons. You're going to get all the photons, and what you have to do to get 
the interference pattern is to ignore the photons where I erased the 
information about my half of the experiment. Remember that you don't get an 
interference pattern from just one particle. You get an interference pattern 
statistically, when you let lots of particles build up. There's no way to 
look at one individual particle and say "is that part of an interference 
pattern?"

The way the particular experiment on the wiki page is described, 1/4th the 
particles randomly create an interference pattern, 1/4th randomly create an 
interference pattern offset 1/2 a fringe from the first one, and 1/2 of the 
particles create no interference pattern. But you don't know which group 
each particle is in until you look at the idler "B" emitted photon to see 
which way it went.  If you *influence* the emitted particle, you break the 
entanglement, and now you have results uncorrelated with the emitted 
particle, which (I think) means you'll get an interference fringe from the 
"A" particles because you erased what would let you distinguish them.



It's the same as a simpler question: Why can't I send any information at all 
over a quantum channel? Any experiment you concoct to measure a quantum 
property one way or the other to communicate is going to run up against the 
fact that the property you're measuring is influenced by the property you 
aren't measuring.

Let's say I decide to measure polarization, diagonal for 1, 
horizontal/vertical for 0. I send you half an entangled stream, then measure 
diagonal polarization or H/V polarization for each bit. The problem is that 
if you measure the H/V polarization when I measure the diagonal 
polarization, you just get random nonsense. You already have to know which 
polarization I'm going to measure in order to measure the same polarization. 
And if we both measure the same polarization, we'll get the same answers, 
but the answers will be a random stream of bits.




Or, for a super-simple analogy, let's say I have a glass table. You can see 
the underside of the glass table instantly, no matter where you are. So you 
go far away, and I stand next to the glass table. However, the only thing 
you can see through the glass of the table is a coin I have flipped.

So if I flip a coin and it lands heads up, I know instantly that you 
instantly saw the tails side of the coin. If it lands tails up, I know 
instantly that you know instantly that it landed heads up. That's FTL.

Now, what can we do with that data? Not much. It's a random stream of bits 
over which I have no control. If I set the coin down in a specific way, you 
can't see it. I know you saw the compliment of what I saw, but I have no way 
to influence that. If I try to influence it, the whole process collapses and 
you can no longer see the results from far away.




That's why, if you look at quantum cryptography, the quantum part of the 
process is used to generate a OTP, then to ensure the OTP hasn't been 
intercepted by someone else. But the actual communication, including the 
part where you ensure the OTP hasn't been intercepted, happens over normal 
non-quantum channels.

-- 
Darren New, San Diego CA, USA (PST)
   How come I never get only one kudo?

Post a reply to this message