Re: An Ada Program Does What It Says?

[emery@mitre-bedford.arpa (David Emery)]

The point of Mendal & Bryan (or Bryan & Mendal, depending on volume)
is to illustrate the dark corners of the language.  As has been
pointed out, *every* language has dark corners.  (Didn't Goedel prove
this...) 

One of the big differences between the Ada "culture" and other
"cultures" is the tendency we (Ada-ites) have to make things specific.
This was clear in the POSIX business, where the Ada binding goes out
of its way to completely specify all behaviors, including
implementation-defined or undefined behaviors.  The C binding is
(deliberately) silent in many places, with the general caveat of
"anything not specified is undefined."  

Therefore it is much easier to find the implementation-defined things
in either the Ada language or POSIX/Ada, because they are explicitly
so stated.  It's a heluva lot harder to find such things when they are
not explicitly specified, such as in POSIX/C.

				dave