From: Reason (reason@exratio.com)
Date: Tue May 07 2002 - 16:44:38 MDT
---> Eliezer S. Yudkowsky
> Actually, this is something that I've been wondering about myself since
> roughly the age of 16. Whether or not a measurement is made is something
> that has experimental consequences, and measurements of entangled
> states are
> supposed to cause distant collapses. So why can't you use the fact of
> measurement, whether something is in an "entangled state", to transmit
> information? You can't use Bell inequalities to transmit information
> depending on *which of two* measurements you make, but it's
> always looked to
> me like you should be able to transmit information based on
> *whether or not*
> you make a measurement. I seem to recall posting a question about this
> once, but I can't find it in my mail archives.
In order to establish what "whether or not" actually means, you have to have
previously communicated by normal methods. To set up the apparatus, you have
to have previously communicated by normal methods. So the information (as
opposed to data/noise) still took STL methods to get there. Otherwise, you
just know that something happened, but that knowledge hasn't transmitted any
information.
Blah. That's not all that good. Try again: the actual Bell inequality part
of this is just a part of the process of transmitting information. That
process starts when you agree (via STL) how the Bell inequality measurements
are to be interpreted. Or when you organize (STL) the process of setting up
the apparatus. So even though portions of this communication may appear to
be FTL, the process of sending information is still coupled to STL
methods -- so you can't violate causality.
You have to include the process of setting up the apparatus in this to get
around the idea of using cryptographic analysis to obtain meaning from Bell
inequality measurements that were not previously agreed upon. (I think it's
fair enough to ignore the possibility of using random Bell inequality
measurements on random (apparently) preexisting systems and then subjecting
the data to cryptographic analysis to obtain messages that you didn't know
were necessarily going to be there. Though it would make a good plot point
for SF).
Reason
http://www.exratio.com
This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 09:13:53 MST