Subject: Re: Lisp Programmers in X Months
From: Erik Naggum <erik@naggum.net>
Date: Mon, 15 Apr 2002 00:36:46 GMT
Newsgroups: comp.lang.lisp
Message-ID: <3227819806254921@naggum.net>

* Bruce Hoult <bruce@hoult.org>
| You don't.  Read my lips: "Integer arithmetic using IEEE FP is exact"[1].
:
| [1] add the proviso "for integers less than 2^53" if you wish to be
| pedantic.

  Pedantic?  Integer arithmetic is _not_ exact with IEEE FP _unless_ you
  confine yourself to integers in a fairly small range, and there is not
  even any indication that you have lost precision when it happens.  This
  is not dealing with _integers_, but with a severely restricted subset of
  integers under optimistic conditions.  _Integers_ is what we have in
  Common Lisp, defined so as not to truncate their precision or work only
  modulo some "word length".  A guarantee that you work within the range
  that your "integer" supports is hard to come by and failure to get it is
  the source of many programming errors.  Just increasing the fixed number
  of bits in the representation constitutes no such guarantee.

///
-- 
  In a fight against something, the fight has value, victory has none.
  In a fight for something, the fight is a loss, victory merely relief.

  Post with compassion: http://home.chello.no/~xyzzy/kitten.jpg