Re: Ada Readability, Book Reference

[seas.gwu.edu!mfeldman@uunet.uu.net (Michael Feldman)]

In article <9212281507.AA20416@ajpo.sei.cmu.edu> SAHARBAUGH@ROO.FIT.EDU writes:
>I received one request for a pointer to Bryan & Mendal's
>book so I may as well post it (same amount of keystrokes).
>
[stuff deleted]
>
>For many examples the book asks, "What will this
>program print?"  For example 7.3.12 the answer is
>"The answer to the question is indeterminate."
>
>So I ask "Is Ada readable if you can express an indeterminate
>program using it?".
>
The particular example depends on the evaluation order of the sides
of the expression, but more fundamentally, speaks to the approximations
inherent in floating point. Ada's readability has little to do with it -
it's characteristic of digital computers that floating-point arithmetic
has indeterminate results sometimes, so that (A*B)/C may yield a different 
result than A*(B/C). 

In the specific case, (6.0 * 1.0)/3.0 can yield a different result from 
6.0 * (1.0 / 3.0) because 1.0/3.0 cannot be exactly represented.
Therefore an expression like
  6.0 * (1.0 / 3.0) = (6.0 * 1.0) / 3.0
may not evaluate to True. Traditional floating-point gotcha.

You chose an example that doesn't make your point very well, because a
programming language can't fix problems inherent in trying to approximate
the real numbers in a finite number of bits.

Mike Feldman
------------------------------------------------------------------------
Michael B. Feldman
co-chair, SIGAda Education Committee

Professor, Dept. of Electrical Engineering and Computer Science
School of Engineering and Applied Science
The George Washington University
Washington, DC 20052 USA
(202) 994-5253 (voice)
(202) 994-5296 (fax)
mfeldman@seas.gwu.edu (Internet)

"Americans want the fruits of patience -- and they want them now."
------------------------------------------------------------------------