It is somewhat strange that all these articles/blogs claiming that Geometric Algebra is inherently superior to Quaternions spend so much time on how Quaternions are isomorphic to the even-subalgebra of GA3 (hence for all intents and purposes they're the same) and so little time on the odd-subalgebra (exterior algebra IIRC), which as far as I can tell is the main (only?) advantage of GA over Quaternions.
On the other hand, almost nobody mentions the nice geometric perspective that unit quaternions offer that are somehow "lost in translation" in GA: as a compact Lie group, unit quaternions come endowed with a bi-invariant Riemannian metric which means you can do interpolation, clustering, blending, statistics with them in a metric-consistent manner. And since the metric is compatible with the group structure, the geodesics are cheap to compute.
On the other hand, almost nobody mentions the nice geometric perspective that unit quaternions offer that are somehow "lost in translation" in GA: as a compact Lie group, unit quaternions come endowed with a bi-invariant Riemannian metric which means you can do interpolation, clustering, blending, statistics with them in a metric-consistent manner. And since the metric is compatible with the group structure, the geodesics are cheap to compute.