[comp.lang.ada] 10 Commandments for Ada9X

mathis@CTC.CONTEL.COM (Bob Mathis x4232) (02/02/90)

This message contains some ideas for a "Ten Commandments" for the
requirements and revision of Ada. I have brought together some
ideas that I think are related and given them titles. If this
list doesn't provoke some discussion, then some net nodes must be
inoperative.

1. PUBLIC -- involve public in all phases of revision and
standardization; be responsive to various special interest
communities; continue focus on primary customers; coordinate with
Ada-related professional groups

2. CONSERVATIVE -- minimum change with maximum positive impact;
maximum compatibility with current programs and compilers;
generally conservative; revision not redesign

3. ISO -- revision of text to incorporate approved AIs; use of
large character set in strings and literals; adapt for better fit
with other ISO standards and guidelines

4. OOP -- increased support for object-oriented program
development paradigm while remaining consistent with Ada's
support for other paradigms and Ada's existing features
addressing similar goals

5. REAL-TIME -- improved suitability for and predictability of
real-time performance (dynamic priorities, distributed time-outs,
aborts, asynchronous communication, etc.); require more
documentation of run-time performance in Appendix F

6. RELIABLE -- features for improving reliability of applications
programs; maximize the possibilities for static analysis of
program correctness

7. INTERFACES -- improved facilities for interfacing to external
systems (GKS, CCIS, X, SQL, etc.); secondary and supplemental
standards for use with commercial data processing; adaptation to
multi-lingual environments

8. SIMPLIFY/UNIFY -- improve overall language consistency; remove
unnecessary limits and "silly" requirements in the current
language; reduce implementation freedom/inconsistency (clarify
and specify in RM); increase language orthogonality; increase
language integration; improve the similarity between user defined
and predefined constructs (types, exceptions, etc.); improve Ada
from the perspective of "least surprise"

9. STRICT CONFORMANCE -- continued strict enforcement of no
subsets/no supersets; but encourage compiler writers and tool
developers to experiment with new ideas for supporting program
development

10. KEEP IT ADA -- keep strong typing, encapsulation, separate
compilation, information hiding, static binding, packages, tasks,
rendezvous, generics, and other aspects of current Ada approach

billwolf%hazel.cs.clemson.edu@hubcap.clemson.edu (William Thomas Wolfe, 2847 ) (02/04/90)

From mathis@CTC.CONTEL.COM (Bob Mathis  x4232):
> 4. OOP -- increased support for object-oriented program
> development paradigm while remaining consistent with Ada's
> support for other paradigms and Ada's existing features
> addressing similar goals
> 
% 8. SIMPLIFY/UNIFY -- [...] increase language integration; [...]
> 
> 10. KEEP IT ADA -- keep strong typing, encapsulation, separate
> compilation, information hiding, static binding, packages, tasks,
> rendezvous, generics, and other aspects of current Ada approach

   My only comment is that in the interests of Sections 4 and 8,
   the keeping of packages and tasks (Section 10) should be done
   indirectly.  These two concepts should be integrated into the
   simple and intuitively appealing concept of an "object".


   Bill Wolfe, wtwolfe@hubcap.clemson.edu
 

mitchell@community-chest.uucp (George Mitchell) (02/07/90)

In article <9002021545.AA03784@ctc.contel.com> mathis@CTC.CONTEL.COM
(Bob Mathis  x4232) wrote:
`1. PUBLIC ....
`2. CONSERVATIVE ....
`9. STRICT CONFORMANCE ....
`10. KEEP IT ADA ....
The Ada9X effort has been underway for some time.  Do any of the above
"command" some kind of change to the way they have been working?

`3. ISO ....
`4. OOP ....
`5. REAL-TIME ....
`6. RELIABLE ....
`7. INTERFACES ....
`8. SIMPLIFY/UNIFY ....
Were any of the above supported by submissions to Ada9X?  If not, why are you
raising the issues now?  If so, do you have reason to believe that they
require greater consideration than you expect them to receive?
--
/s/ George   vmail: 703/883-6029
email:  mitchell@community-chest.mitre.org    [alt: gmitchel@mitre.arpa]
snail:  GB Mitchell, MITRE, MS Z676, 7525 Colshire Dr, McLean, VA  22102