Re: Legislative Mandate for Ada

[bwb@sei.cmu.edu (Bruce Benson)]

In article <EMERY.90Dec17153931@aries.linus.mitre.org> emery@linus.mitre.org (David Emery) writes:

>Actually, I think there's something to be said for this idea.  Even
>more, I think that any DARPA or DoD-funded project should submit a
>waiver request to do their work in something besides Ada.  Such
>waivers should be reasonably easy to get, but there are a lot of
>research projects that could be used on gov't systems, except for the
>fact that they're implemented in languages that create a significant
>maintenance/adaptability problem (e.g. C, lisp).  

These kind of fuzzy statements, that imply all the problems of software
engineering will go away (or be *significantly* reduced) by using the
current king of silver bullets (i.e. Ada), are at the heart of why we
in government sometimes make such dumb decisions.  We don't know that
any of this is true, but we do know that the problems of software 
engineering are more centered on people and mangement problems than on
technology problems.  We focus on technology because it won't talk back.

>However, unless and until Ada compilers are as affordable as C
>compilers (e.g. Gnu C), such a requirement will be financially
>untenable, since most schools are unwilling to fork out the $$ for an
>Ada compiler.

What about the rest of the supporting environment: editors, debuggers,
libraries, code generators, readable books, jobs, etc.?  (I had a boss who
insisted that WordStar should be a fine programming editor since it can
export ascii files!) Affordability of the compiler must be one of the least
important issues when selecting a development environment (or a teaching
environment).  COBOL, FORTRAN, and now C, are still the most widely used
and *supported* languages in the commerical world, and this dwarfs the 
government sectors need for Ada programmers.  Universities like to not only 
educate their students, but possibly give them a skill that makes them readily
employable.

When Borland creates TurboAda or Microsoft does QuickAda, then the language
is probably mature enough to hold its own.  Until then, it remains an
interesting experimental language by the government.

[All the above is, of course, IMHO and does not necessarily reflect anyones
thinking but my own :-)]

* Bruce Benson                   + Internet  - bwb@sei.cmu.edu +       +
* Software Engineering Institute + Compuserv - 76226,3407      +    >--|>
* Carnegie Mellon University     + Voice     - 412 268 8469    +       +
* Pittsburgh PA 15213-3890       +                             +  US Air Force