LTV only makes sense if every data type you represent requires a length field, and that's usually not the case. T(V/LV) allows you to represent types with impl

LTV only makes sense if every data type you represent requires a length field, and that's usually not the case. T(V/LV) allows you to represent types with implied lengths (and thus save space).