Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site watdaisy.UUCP
Path: utzoo!watmath!watdaisy!gvcormack
From: gvcormack@watdaisy.UUCP (Gordon V. Cormack)
Newsgroups: net.lang
Subject: Re: IF KICK-BACK GREATER THAN 10 AND LESS THAN 100 THEN ...
Message-ID: <6761@watdaisy.UUCP>
Date: Thu, 22-Nov-84 15:34:30 EST
Article-I.D.: watdaisy.6761
Posted: Thu Nov 22 15:34:30 1984
Date-Received: Fri, 23-Nov-84 02:43:36 EST
References: <440@ima.UUCP> <6172@mcvax.UUCP>, <1241@utah-gr.UUCP>, <1345@ihuxq.UUCP>
Organization: U of Waterloo, Ontario
Lines: 27

I am tired of people saying that this is the programmer's mistake.
COBOL is wrong.  Using the "rules of logic" gives the same result
as everyday common sense:  the opposite of what COBOL does.

The rules of logic say the following:  when you distribute a 
positive comparison like "less than" across a logical operator 
there is no problem.
 e.g.  A less than b and c   MEANS  A less than b and A less than c

However, if a negation is distributed across the logical operator,
DeMorgan's theorem applies; AND is transformed to OR and vice versa.
 e.g.  a not less than b and c
   
     MEANS (except in COBOL) a not less than b or a not less than c

but COBOL interprets it as

     a not less than b and a not less than c

An everyday example:

   Compare the meanings of:

     "I am not male and female"
     "I am not male or female"

-- Gordon Cormack, University of Waterloo