🚀 go-pugleaf

In article <5064@internal.Apple.COM> rmh@apple.com (Rick Holzgrafe) writes:
>In article <317@panix.UUCP> alexis@panix.UUCP (Alexis Rosen) writes:
>usually too many combinations to be tested in reasonable time, so you can 
>only test the more obvious ones. The Monkey is an *extra* test, one you 
>can run at night for example, that might through luck stumble across a bug 
>that your planned testing missed. *If* your planned testing is thorough, 
>*and* it finds no bugs, *and* the Monkey can't break your product, you can 
>ship with confidence!

*And* still have a buggy product!!

If the application is large and complex, I doubt that any amount of
conscientious testing can span the entire state space. That's why people
who claim that large-and-complicated-program-X is bugfree are invariably
talking through their hats. Why do you think they put those 'it's your
problem, pal' disclaimers in warranties. For example, reading the fine
print on my Word 4.0 disk package, I see a heap of 'thou shalt nots', plus
disclaimers which absolve the company from all liability in States whose
consumer laws are not that strong.

Note that I am not picking on Microsoft. Most other software products sold in
the USA contain almost exactly the same guff. Are people really that eager
to sue at the drop of a hat? Aww, I'm crapping on now....

I just want a program that does what it is supposed to do, and does it
properly. I can take the occasional crash, as long as it is not catastrophic,
i.e., wipes out my life's work, etc. etc.


Ben Lian

Dept of EE & CS, Uni of Tasmania, GPO Box 252C, Hobart TAS 7001, Australia
UUCP: ...!{uunet,ukc,mcvax,hplabs,nttlab}!munnari!tasis.utas.oz!ben
ACSnet: ben@tasis.utas.oz  Ph: 002-202380  Fax: 002-202713