Ada 9X objectives

[wtwolfe@hubcap.clemson.edu (Bill Wolfe)]

   Ada is now entering the revision process, and much has transpired
   over the last ten years.  Many ideas have emerged, both from inside
   the Ada community (e.g., Hilfinger's ideas to the effect that literal 
   recognition should not be limited to the predefined types, but should
   in fact be definable for user-defined abstract data types as well),
   and from outside the Ada community (notably multiple inheritance).  

   When Ada was first designed, the objective was to provide the maximum
   possible support for good software engineering practices, given what
   was then known about programming language technology.  Unfortunately,
   compiler maturity is just now reaching levels which make the language
   attractive as a production tool, and in the meantime the continuing
   advancement in the theory and practice of programming language design
   have made the delay very costly to Ada's prospects for widespread
   acceptance.  

   Moreover, some persons who have been in a position to observe the
   Ada community seem to be of the opinion that for a significant
   number of members of the Ada community, Ada 83 has become an end
   rather than a means; in their estimation, Ada will fall victim to
   the resistance to change which seems to plague virtually all human
   endeavors.  They cite the recently expressed view that inheritance
   would result in too great a setback for Ada compiler maturity as
   an example, saying that inheritance has already been proven to be
   efficiently implementable and that this is merely a smoke screen
   for resistance to new concepts.  

   Therefore, I would like to pose this question to the Ada community:
   what are the basic objectives of Ada 9X?  Specifically, is Ada going
   to be something that is tied to the idea of maximum support for the
   practice of software engineering?  Is the DoD *really* committed to
   making Ada the primary vehicle of software engineering technology?
   If so, what is being done to close the gap between the dream of 
   widespread Ada use and the reality of the widespread use of COBOL,
   C, C++, and Eiffel instead?  

   I do not mean to imply that tremendous progress has not been made;
   as John Foreman (I believe) recently noted, Ada is much farther along
   than COBOL was at a similar point in its existence.  However, it is
   extremely frustrating to see people choosing C++ and Eiffel instead
   of Ada because of this ten-year revision cycle (and perhaps a bad bet
   that inheritance would not turn out to have a bright future).  It is
   extremely frustrating to see employment advertisements calling for
   knowledge of the likes of C or COBOL, with nary an Ada advertisement
   in sight.  But the worst of all is listening to people assert that
   Ada is obsolescent and will never be able to catch up. 

   As a software engineer, I strongly believe in the Ada dream.  But
   there have been big problems associated with realizing that dream.

   An explicit linkage of Ada 9X to maximum support for the software 
   engineering process, combined with a shortened five-year revision 
   cycle, would go a VERY long way toward making that dream a reality. 
    

   Bill Wolfe, wtwolfe@hubcap.clemson.edu