Re:The envelope paradox (2006)
by dllthomas on Monday November 30, 2015 @02:06AM
(#10648190)
I still don't see why your analogy does anything but introduce additional complication. To me, the result appears no more nonsensical than in the original formulation - where it appears correct. So long as X and Y are chosen in advance - and independently - of my guessing, if I draw my reference R from any continuous distribution, there is some nonzero chance that it will fall between X and Y, letting me distinguish them in that case.
[ Parent
| on HN ]
Re:The envelope paradox (2006)
by logicallee on Monday November 30, 2015 @06:06AM
(#10648808)
you have no idea what possible values X and Y can have. It's like an alternative universe, or guessing a number first and then finding a function to numerically evaluate the universe whose range you don't know in advance. I find it nonsensical that this can yield any information.