As this discussion demonstrates, there is a lot of subtlety around what constitutes a "byte" in Rust's abstract machine. By contrast, it is sufficient to define the bit validity of most types in reference to u8s without actually specifying much about the bit validity of u8 itself. Since most of our conversions operate in terms of existing u8 types (usually [u8]), this isn't a problem for us. We should replace most uses of "byte sequence" with a similar phrase involving u8s.
As this discussion demonstrates, there is a lot of subtlety around what constitutes a "byte" in Rust's abstract machine. By contrast, it is sufficient to define the bit validity of most types in reference to
u8s without actually specifying much about the bit validity ofu8itself. Since most of our conversions operate in terms of existingu8types (usually[u8]), this isn't a problem for us. We should replace most uses of "byte sequence" with a similar phrase involvingu8s.