[ExI] Cramer going backwards

[scerir]

> "If the experiment works, then I will be on to some very interesting
> roads to success," Cramer said, "but I'll probably end up writing the
> novel rather than making the discovery. In a sense, doing the
> experiment is background for the novel."

He is trying a difficult experiment. As far as I know
he is trying to transform a two-photon interference
experiment (the correlation between two position/momentum 
entangled photons) in a one-photon interference "at a distance"
experiment. There are papers showing that this is impossible
(for essential reasons). There are experiments showing that
this is -at least- difficult. 


Another conceptual experiment (to entangle past and future,
and to modify one of the two) would be the following (according 
to Afriat).

The observable used to violate Bell's inequality is a function 
of four quantities (2 dichotomic parameters), two for one side, 
two for the other. These are usually angles or directions, but 
could also be times or quantities concerning the Hamiltonians. 
Provided realism is granted, a violation of Bell's inequality 
indicates the 'non-separability' of entangled systems.
If the quantities are physical directions, a value possessed 
by an entangled system can be modified by the physical rotation 
of a distant apparatus. If the parameters are times, one could 
conclude that the value possessed by an entangled system at time t 
depends on whether a measurement on a distant system is performed 
at time t' or at time t''.