Re: real numbers in ADA

[eachus@spectre.mitre.org (Robert I. Eachus)]

In article <WOODRUFF.95Mar13125852@tanana.llnl.gov> woodruff@tanana.llnl.gov writes:

  > I'm feeling better already! 

  > I have been having this same reaction to the Ada95 LRM, and was
  > beginning to wonder if a) my memory has become selective since I learned
  > Ada83 ("life sure was a lot easier then!") or b) a *lot* of cells have
  > taken early retirement.

  > I think the world is in greater need of quality Ada textbooks now than
  > was the case a decade ago.

    "The fault dear Brutus is not in ourselves, but in our stars..."

    I don't think it is the fault of the new RM, but that in the
process of extending the existing language, in many places we chose
truth, beauty, upward compatibility, and functionality over ease of
understanding.  There are some VERY powerful tools in Ada 95, and I
don't think anyone knows how best to use some of them.  In fact I have
convinced myself that many of the paradigms that will be commonly used
in Ada 95 applications are significantly different from the "standard"
object-oriented paradigms.  (We just don't know what the new paradigms
are yet. :-(

    To some extent the same phenomenon occured with Ada 83, but the
parts of Ada 83 which were new to most programmers were isolated in
chapters 9 (Tasking) and 12 (Generics).  It was possible, in fact
common, to teach Ada in three chunks: the Pascal subset (chapters 1
through 6 plus some I/O), the Pascal superset (add 7,8, and 11), and
full Ada.  In Ada 95, the same approach may be justified, but some of
the new pieces are found in the heart of the manual, in chapter 3.  Of
course one way to do this is to teach the Ada 83 subset--Ada 83 with a few
cosmetic extensions, then add tagged, controlled, protected and
abstract types.


--

					Robert I. Eachus

with Standard_Disclaimer;
use  Standard_Disclaimer;
function Message (Text: in Clever_Ideas) return Better_Ideas is...