Xref: utzoo comp.sys.mac.misc:10604 comp.windows.ms:11275
Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usc!cs.utexas.edu!hellgate.utah.edu!dog.ee.lbl.gov!nosc!crash!orbit!pnet51!pj
From: pj@pnet51.orb.mn.org (Paul Jacoby)
Newsgroups: comp.sys.mac.misc,comp.windows.ms
Subject: Re: software testing
Message-ID: <4521@orbit.cts.com>
Date: 7 Apr 91 00:25:01 GMT
Sender: news@orbit.cts.com
Organization: People-Net [pnet51], Minneapolis, MN.
Lines: 23

John Yen (jyen@apple.com  7.0 kernel test team) writes:
 > '- nor is it difficult to test'?! Have you ever tested non-trivial
 > commercial software?

After working Tech Support for a 'major mainframe/Unix company' for 2 1/2
years, I can tell anyone who wants to ask that software testing is (1) a pain,
(2) extremely difficult to implement properly without LOTS of people and
resources, and (3) even more difficult to do when management could give a
rat's-ass about quality.

A proprietary database product of said 'company' never ceased to amaze us, as
it would crash, destroy data, and generally just not work properly in some
spectacularly simple configurations.

Of course, said management team had some odd ideas of quality--(1) The release
date is the driving goal; (2) Quality lies with Continuation Engineering (the
after-it-is-released bug fixers), NOT with the Development team; and (3) you
really don't need many people to properly support a product.

Said company has gone down the toilet in a big way over the last number of
months....

INET: pj@pnet51.orb.mn.org   -OR-    pejacoby@3m.com