Eric Weisstein (The Keeper of the Polyhedra for Wolfram) has pointed out to me that independently of my article he had recently explored the Chamfered Octahedron in preparation for its entry into Mathematica's PolyhedronData. It turns out that this is the same polyhedron as described above. So the same polyhedron can be arrived at from two different starting 'parent' polyhedra, simply by choosing whether to make the transition by (symmetrical or congruent) vertex slicing (truncation), or by chamfering (that is applied equally to each specified edge). Neat!