Category : Miscellaneous Language Source Code
Archive   : ADA-RATL.ZIP
Filename : RATL-C1

1. Introduction and Preface








1.1 Goals


This document, the Rationale for the design of the Ada programming
language, has evolved over a number of years.

A first version appeared in 1978 as the Rationale for the Green
programming language, and this was revised in 1979 at the time that
Green was finally selected as Ada. The purpose of these documents was
to explain the motivation for the language design, and to justify and
defend its position against the other competitive languages and the
Ironman (later Steelman) requirements. No corresponding document was
written in 1980 when Ada was proposed as a standard, nor in January
1983 when Ada finally became both an ANSI and Military Standard. The
present version completes and revises a draft issued in January 1984
and which was the subject of public review.

The goal of these documents has also evolved. The original goal was
both motivational and defensive; a major concern was implementability
(in potentially controversial areas such as overload resolution)
especially since there were no compilers then in existence. This last
concern is now substantially reduced since implementability is well
demonstrated by a number of production quality compilers.

Equally the goal is not to document and justify every language design
decision: it would be difficult to replace the archive containing the
hundreds of study notes and thousands of comments produced between
1977 and 1983.

The present goal is thus now more inspirational: to give the reader a
feel for the spirit of the language, the motives behind the key
features and to create the basis for understanding how they fit
together both globally as viewed from the outside and in detail as
viewed from the inside; above all to impart an appreciation of the
main architectural lines of the language and its overall philosophy.
It is only by knowing this philosophy that a real understanding for
the detail will be obtained.

It is now 1986, and it may well be asked why this ultimate Rationale
has taken so long to emerge. We must admit that the main reason is
simply that we have been deeply involved over the past years in a
number of activities aimed at creating an infrastructure that would
ensure the success of the Ada language, - in particular, the
development of training material and compilers. The success of Ada has
now relieved that pressure and allowed us to complete the work started
in 1978, and to make available what we hope will be an important
addition to the Ada literature.



1.2 Structure


This document is divided into chapters covering different aspects of
the language. Most chapters correspond to chapters of the Reference
Manual. Expressions and statements are here regrouped in a single
chapter, since the subject is fairly classical. Conversely, in view of
the importance of the subjects, special chapters are devoted to
numeric types, access types, and derived types, in addition to the
chapter on types. The chapters of the Rationale are fairly independent
(at the cost of some repetition) and can be read in any order.

Most chapters of the Rationale have a common structure. They start
with an introduction to the topic discussed. An informal description
of the language features follows. This description is made in terms of
examples chosen to reflect the major classes of uses of the features
considered.

We believe that the reader will get the spirit of the language reading
these examples. They should help the development of an intuition for
programming style in the Ada language.

A discussion of the technical issues follows, or in some cases is
interspersed with the informal description. Such discussions cover the
major design decisions, their justification, and the interactions with
other aspects of the language.



1.3 Acknowledgements


We would like to take this opportunity to acknowledge the
contributions of several others to the Ada Rationale: Jean-Raymond
Abrial, Bernd Krieg-Brueckner, Jean-Claude Heliard, Henry Ledgard, Ian
Pyle, Olivier Roubine, Steve Schuman, Stan Vestal, and Brian Wichmann
contributed to the 1978 and 1979 initial versions. Brian Wichmann also
contributed to the 1984 draft.

Other significant contributions were derived from the Language Study
Notes by Jean-Loup Gailly and Paul Hilfinger and by comments from the
Distinguished Reviewers.

We are also indebted to the comments on the 1984 draft from the Ada
Europe Review Group organized by Kit Lester, the comments by Henry
Dancy and Vincent Amiot, and the dedicated technical support of Marion
Myers.

The Ada Rationale was developed by Alsys and Honeywell under a
contract from the United States Government (Ada Joint Program Office).



Jean D. Ichbiah
John G.P. Barnes
Robert J. Firth
Mike Woodger