Patrick Santucci wrote:
> Richard Gambrell added:
>
> > Just a quick comment: Before worrying over if a program will misbehave, I
> > think one needs to test if it will load and run with unchallenging data.
>
> Absolutely! But what I'm talking about is robustness and error handling,
> not basic functionality. After all, a program shouldn't even make it to
> the independent testing stage if the programmer can't make it do what
> it's supposed to! Your experience with Oracle (and mine with Microsoft)
> notwithstanding, most companies don't release products that flat out
> don't work; at least, not intentionally.

I have to clarify: my compliant isn't with Oracle.  Oracle is complex and
messy, but does seem to work.  My reference was to a particular company
making applications that run on Oracle DB which don't work 'out of the box'
and show little or no testing/QA work. While most companies probably don't do
this intentionally, I'm not sure what to make of one that seems to do it
repeatedly.  Calling it unintentional somehow seems to suggest worse than not
testing - ie, a totally out of control development/release process.  Am I off
base here?

>
> My point was simply that programmers often need someone else to test
> their programs who will use a little more imagination (or idiocy, take
> your pick) in selecting and using test data. Sorry if I offended anyone.

BTW, I agree wholeheartedly with your main point. As much as I hate doing it,
working to break software, while in testing, is necessary to produce reliable
products and many programmers just can't seem to do it.

--
Richard Gambrell
Database Administrator and Consultant to Computing Services
University of Tennessee at Chattanooga, Dept. 4454
113 Hunter Hall, 615 McCallie Ave. Chattanooga, TN 37403-2598
UTC e-mail: [log in to unmask]   phone: 423-755-4551
Home e-mail: [log in to unmask]