I don't think it would be a good idea to include BYTE-VECTOR-FILL!,
BYTE-VECTOR-COPY, & BYTE-VECTOR-SUBSEQUENCE (which might be better
named BYTE-SUBVECTOR) in the initial proposal; the first has been
deemed not very useful, and the latter two are both trivially built
atop BYTE-VECTOR-COPY!.  My suggestion was only that the bare, minimal
interface be augmented by an important primitive; adding library
functionality to the base interface is probably not a good idea.