[clean-list] Re: (clean-list digest):Functions that look the same (smithll) (Richard A. O'Keefe)

Siegfried Gonzi siegfried.gonzi@kfunigraz.ac.at
Wed, 25 Jul 2001 15:59:13 +0200


clean-list-admin@cs.kun.nl wrote:

> From: "Richard A. O'Keefe" <ok@atlas.otago.ac.nz>
>

> The precision is thus one part in 9x10  .
> So an IEEE 754 double gives you about 16 decimal digits worth.
> To convert such a number to decimal so that it will be read back
> correctly takes a few more digits, but that's not the precision.
>

Thank you man. The literature is sometimes confusing or is using the term "precision" with
vagueness. I can remember (from programming courses) the phrase "do not trust the last digit".
In case of Mops (with 16 places) that phrase is appropiate but Clean takes care (in advance) of
this.

I think the Clean team should somewhere getting to start documenting "numerics". I wouldn't be
surprised if functional programming can contribute to some confusion when it comes to numerics.
One "bells and whistles" of functional programming is the saying: "We do not imperative code the
problem instead we describe the problem". I think "describing" can often lead to the believe
that with functional programming one hasn't to think about precision.

Even functional programming languages are based on v.Neumann architecture (at least in my small
world).


S. Gonzi
P.S. I find it conveniet that in Clean there are only double-precision and integer (compared to
C with many types of integer formats; complete insane in year 2001 with megabytes of memory).