Dan wrote:
> On 4/4/2007 9:41 AM, William Pociengel wrote:
>
>> well we did'nt actually lose the entire machine. we did have about 10
>> commands that worked hehehhehe
>> so I learned that testing means testing everything!! and geeksafe is one
>> of those things that you want to test really well. but not on live data ;-)
>
> Indeed. Always have good backups on hand. Ouch your backups didn't
> work? Was the backup media bad? Verify mode (if available) with
> backups may take longer but it is certainly worth it!
bad tape commands within that particular version of the OS. claimed it
was doing something when it wasn't =8O
IBM moved off that rev pretty fast as we had their guys from England on
the phone for the better part of 2+ days with short breaks to sleep ;-/
I also learned more about testing later in life when I built an app and
handed it over to our tester. the first thing he did after installing it
was to lay both hands on the keyboard. I found out about keyboard
buffers and controlling input, oppps!
so testing on anything that is considered mission critical needs to go
beyond the bounds of 'how do I expect this to be used' and into 'how can
an idiot (*&()&(*^ this up'.
And this would go for more than just geeksafe but into anything that
works at a low level.
william
-- -- special knowledge can be a terrible disadvantage if it leads you too far along a path that you cannot explain anymore. mentat admonition <http://jobs.collegerecruiter.com/PociengelWilliam> <http://www.linkedin.com/pub/0/934/483> -- This is the NewtonTalk list - http://www.newtontalk.net/ for all inquiries Official Newton FAQ: http://www.chuma.org/newton/faq/ WikiWikiNewt for all kinds of articles: http://tools.unna.org/wikiwikinewt/Received on Wed Apr 4 17:00:37 2007
This archive was generated by hypermail 2.1.8 : Wed Apr 04 2007 - 17:30:00 EDT