Double package:base-compat

Double-precision floating point numbers. It is desirable that this type be at least equal in range and precision to the IEEE double-precision type.