Book Conflates Affine Types and Regions

[anasazi]

In the Boxes subsection of the Pointers section, the book claims "boxes are a specific kind of affine type, known as a region". This is not correct. Affine types are not regions. Affine types are a kind of substructural type system. Regions are a kind of type and effect system. They're two very distinct ideas that can be used for similar purposes.

I would suggest moving the reference to the Cyclone paper to the References subsection at the end of the discussion of region points and lifetimes. The paper Experience With Safe Manual Memory-Management in Cyclone could be good to cite in the Boxes section. It discusses adding unique pointers to Cyclone.