Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!rutgers!lll-crg!seismo!umcp-cs!chris
From: chris@umcp-cs.UUCP (Chris Torek)
Newsgroups: net.cse
Subject: Re: Re: quality software
Message-ID: <4115@umcp-cs.UUCP>
Date: Sat, 1-Nov-86 17:19:37 EST
Article-I.D.: umcp-cs.4115
Posted: Sat Nov  1 17:19:37 1986
Date-Received: Tue, 4-Nov-86 06:47:45 EST
References: <10410@cca.UUCP> <10528@cca.UUCP> <3877@umcp-cs.UUCP> <1874@uw-beaver>
Reply-To: chris@umcp-cs.UUCP (Chris Torek)
Organization: University of Maryland, Dept. of Computer Sci.
Lines: 27

[Richard Harter]
>>>(d)	The software should be rigorously tested.  It is not enough...

[Me]
>>This is the whole point of beta testing. ...

In article <1874@uw-beaver> bnfb@uw-beaver.UUCP (Bjorn Freeman-Benson) writes:
>No, no.  The software should be rigorously tested BEFORE beta testing.

True.  But note that no tests of large systems are ever thorough
enough.  The software should be `rigorously tested' before shipping,
but companies should expect, not ignore, bug reports from beta
sites!

>Beta test is a crucial part of the release cycle, but it must not be the
>only, or even only substantitive, testing that is done.

I had not meant to imply that.  I guess it was a bug that slipped
past the pre-posting tests. :-)  [The confusion clearly resulted
from differing definitions of `rigorous'.  In my opinion, for
example, no in-house tests of TCP/IP are ever rigorous, for there
are far too many variables for any one group to characterise and
test all possible effects.]
-- 
In-Real-Life: Chris Torek, Univ of MD Comp Sci Dept (+1 301 454 7690)
UUCP:	seismo!umcp-cs!chris
CSNet:	chris@umcp-cs		ARPA:	chris@mimsy.umd.edu