841613t@aucs.AcadiaU.ca (Donald Tyzuk) (04/02/91)
So what is Ada9x?. (I am a know-nothing student, interested
in Ada). 8-)
--
Donald Tyzuk UUCP: {uunet|watmath|utai}!cs.dal.ca!aucs!841613t
P.O. Box 1406 BITNET: 841613t@Acadia
Wolfville, NS Internet: 841613t@aucs.AcadiaU.CA (131.162.1.51)
CANADA B0P 1X0 (902) 542-9960 mfeldman@seas.gwu.edu (Michael Feldman) (04/02/91)
In article <1991Apr1.203028.13158@aucs.AcadiaU.ca> 841613t@aucs.acadiau.ca (Donald Tyzuk) writes: >So what is Ada9x?. (I am a know-nothing student, interested >in Ada). 8-) >-- Undoubtedly there are other readers, students and others, to whom this alphabet soup is equally impenetrable, so I'm posting this publicly. Ada9x is the project sponsored by the Ada Joint Program Office (US Dept. of Defense) to produce a revised Ada language standard. Under ANSI rules, this project was started in 1988, 5 years after the original standard was adopted (Jan. 1983). In standards-speak, Ada9x means "revised Ada, to be adopted some time in the 1990's." The "x" means "we're not sure what this digit will be." The current project plan (see the transition plan I posted a couple of weeks ago) is showing x=3, that is we'll see a revised Ada adopted in 1993. If you'd like more info, write me by e-mail. Others more knowledgeable than me in the Ada9x process are cordially invited to expand on my brief note here. I have no formal connection to the Ada9x group; I'm just a professor who teaches a lot of Ada and likes it. Mike Feldman
jls@rutabaga.Rational.COM (Jim Showalter) (04/03/91)
>So what is Ada9x?. (I am a know-nothing student, interested >in Ada). 8-) The next generation of Ada. The Ada 9x process consists roughly of four phases: 1) Solicitation of Ada user feedback on desirable enhancements/changes 2) Mapping of that feedback into revised Ada syntax/semantics 3) Feedback on the proposed mapping, evaluation of implementation tradeoffs. 4) Standardization of the changes in a new LRM and ANSI/ISO standard There are also some other tasks involved, such as evaluating implementation tradeoffs when more than one possible mapping exists. Step #1 is completed. Step #2 is in process, and the first cut at the mapping documents (and accompanying rationale) are complete and in distribution, starting step #3. A summary of some of the goodies: 1) Inheritance and polymorphism. 2) Subprogram types and variables. 3) Lighter weight concurrency constructs called protected records. 4) User-definable default initialization, finalization, copy, etc. 5) User-definable attributes. 6) Hierarchical program libraries (subsystems). 7) Improved separate compilation features via child comp units. 8) Distributed programming support. 9) Hierarchical grouping of exceptions. 10) More flexible generic parameterization. 11) Assertions. 12) New constructs such as "when..raise" and named statements. I like what I see. They've managed to graft support for some new and powerful features onto the language in a remarkably upward-compatible manner, preserve the spirit of Ada '83, and yet come up with a language that is more powerful, flexible, dynamic, and extensible. And all with strong typing and tight binding. Check it out. -- ***** DISCLAIMER: The opinions expressed herein are my own, except in the realm of software engineering, in which case I've borrowed them from incredibly smart people.