From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me
X-Spam-Level: 
X-Spam-Status: No, score=-1.3 required=5.0 tests=BAYES_00,INVALID_MSGID
	autolearn=no autolearn_force=no version=3.4.4
X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 103376,de5dfd6df880dd3
X-Google-Attributes: gid103376,public
From: Michael Stark <mstark@cs.umd.edu>
Subject: Re: Bugs or Mistakes?
Date: 1999/11/20
Message-ID: <38371226.758B@cs.umd.edu>#1/1
X-Deja-AN: 551049847
Content-Transfer-Encoding: 7bit
References: <19991112202656.2368.rocketmail@web216.mail.yahoo.com>
 <80i9la$i9e$1@nntp9.atl.mindspring.net> <3832E75B.2B6E7E1E@pwfl.com>
 <812p3s$2ei$1@nnrp1.deja.com>
Content-Type: text/plain; charset=us-ascii
X-Complaints-To: abuse@cs.umd.edu
X-Trace: cronkite.cs.umd.edu 943133223 24577 128.8.126.43 (20 Nov 1999
 21:27:03 GMT)
Organization: U of Maryland, Dept. of Computer Science, Coll. Pk., MD 20742
Mime-Version: 1.0
NNTP-Posting-Date: 20 Nov 1999 21:27:03 GMT
Newsgroups: comp.lang.ada
Date: 1999-11-20T21:27:03+00:00
List-Id: <comp.lang.ada>

Robert Dewar wrote:
> 
> In article <3832E75B.2B6E7E1E@pwfl.com>,
>   condicma@pwflcom wrote:
> > Perhaps "design defects" is a more accurate description of
> > what we commonly call "bugs"
> 
> I don't like lumping all bugs into design flaws. To me a design
> flaw is quite a different beast from a typo introduced during
> implementation, and it is useful in preventing, finding,
> tracking, ande removing bugs to make the distinction.
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.

Folks --
1.  Robert's objection can be taken care of by calling any problem a
"defect" without the qualifier "design"

2. IEEE already has some useful definitions relating to defects -- I am
paraphrasing, since I don't have the text in front of me

failure -- the program doesn't run as expected and a user sees a problem
fault -- the defect in the product that needs to be corrected
error -- the conceptual problem that explains why a fault is in the
system.

Note -- a failure can occur when there is no fault in the system (e.g.
user inputs values in English units rather than SI), and a fault can
exist in code that isn't executed and thus doesn't cause a failure. 
Faults can be found either by running tests and noting failures, or
inspecting one product against another (e.g. design vs. requirements,
unit test plan vs. code,...)

Thus ends my bried description of the IEEE view of defects ;)

Mike

-- 
Michael Stark                          Goddard Research & Study Fellow
University of Maryland, College Park   e-mail: mstark@cs.umd.edu
phone (UM) 301-405-2721                phone (gsfc) 301-286-5048
"Half the world's population has never even made, or received, 
 a telephone call" -- Kofi Annan