Path: utzoo!utgpu!jarvis.csri.toronto.edu!cs.utexas.edu!samsung!usc!snorkelwacker!mit-eddie!uw-beaver!Teknowledge.COM!unix!hplabs!hpfcso!mjs
From: mjs@hpfcso.HP.COM (Marc Sabatella)
Newsgroups: comp.lang.misc
Subject: Re: Modulus (Re: hashing function for strings)
Message-ID: <8960012@hpfcso.HP.COM>
Date: 22 Feb 90 16:44:14 GMT
References: <E1wt*4@cs.psu.edu>
Organization: Hewlett-Packard, Fort Collins, CO, USA
Lines: 9

>How did we get into this mess? Once upon a time (c. 1950) some machines
>(mostly in Europe) did integer division "right" and some (mostly in America)
>did it "wrong". The difference was essentially between machines built by
>mathematicians and machines built by engineers (note that, if you carefully
>follow a textbook on Algebra, you will see how mathematicians treat it).

Gee, my text books imply that a/b should give a DOMAIN ERROR if "a" or "b"
is negative.  Is that "right"?  [ Note: later, they define an optional possible
interpretation if "a" is negative, but never for "b" ].