The SAT algorithm requires you to find the normal of each edge of each shape (essentially a vector perpendicular to the edge vector) to be used as the separating axes. This can be done very simply...
(x,y) => (-y,x)
OR
(x,y) => (y,-x)
Which should be used in the SAT algorithm? This is essentially a question of whether the left hand normal or the right hand normal should be used. Will it make a difference which is used? Should only the left or right hand normal be used? Should this change with different situations?