IMPLICIT PARALLELISM
   All hyperplanes competing simultaneously!
   But have a small finite sample in population.
   The sample is quickly not random (biased)
   note: selection is an exponential bias/heuristic

   participation in the population is "proportional" to the fitness
   across all hyperplanes!!

BUILDING BLOCK HYPOTHESIS (xover)
   small blocks of good values are joined to make larger blocks of good value
    (you wish)
   to have effective xover  (otherwise don't use xover)

EPISTASIS (property of the problem)
   interaction between genes.   you need to know the value of more than
   one gene at a time to determine fitness.   The "background" makes a difference on the value of a gene.

f(abcde) = a*4 + b*3 + c*(-4) + d*(-9) + e*3
f(abcde) = a*4 + b*3 + c*(-4) + d*(-9) + e*3 + (a & e ? 5 : 0) + (a & b & c ? -7 : 0)


HYPERPLANES

abcde
11001



10***
***11

10*11


3^n hyperplanes in an n dimensional cube

*101*

0***0
0***1
1***0
1***1

*0**0
*0**1
*1**0
*1**1

00
01
10
11

0*
*0
1*
*1