Since access via the subscript operator to multiple consecutive single bits is more expensive than accessing them directly in the underlying data structure, there should be an option to fill a range in the bit vector (using only the underlying data).