Jere Lull wrote in
news:2008013121245350073-jerelull@maccom:

On 2008-01-31 14:27:03 -0500, Geoff Schultz
said:

Expecting the developers to add in code to check for rare events like
this just doesn't happen.

Sorry, but I'm a developer and I program the odd-event handlers first
because I *need* that info when things go "bang".

Okay, I'm an exception because my programs must work 100% of the time
or say what was wrong so the problem can be fixed fast. For some
reason, when you manage 10-20 billion dollars' investments, they want
things to balance to the penny every day.

Still, I can attest it really takes no real effort to output error
messages whenever some action gives an unexpected result.

An "odd event" is different from a program going "bang." I would be
willing to bet that you don't do range checking on all of the data that you
input from disk or a database. Most of the time you assume that it's
right. If a program goes "bang", it's easy to have specific or generic
error handlers to output that an error occurred and what was going on at
the time. The software for the C series has exactly such an error log. We
never got to that stage of debugging.

To me it's obvious that cognisense didn't want to fix the problem. He only
wanted his money back. I asked many questions and none of them were
answered, or if they were, only in the most generic sense. I, and others,
offered many suggestions and from what I could tell, none of them were
acted upon.

I really dislike that fact that he appeared in this group under the guise
of disclosing how awful the C120 is, when all that he wanted to do was to
use it as a bargaining chip with RayMarine. It amazes me that as a
"programmer and system architect" that he had never explored the System
Integration menus which can provide a tremendous amount of information on
what's going on within the system.

I used to run an Internet company and many times we had customers publicly
blast us over issues which they thought were ours. Sometimes they were our
problems, but MANY times they were on the behalf of the customer. I always
appreciated it when knowledgeable customers stepped up on our behalf. I
guess that I was trying to do that and at the same time, fix his problems.

-- Geoff
www.GeoffSchultz.org

On 2008-02-01 10:08:04 -0500, Geoff Schultz said:

An "odd event" is different from a program going "bang." I would be
willing to bet that you don't do range checking on all of the data that
you input from disk or a database.

You'd lose that bet, but I admit that I'm unusual.

I'm proud of the number of systems that I've given away to lesser (or
no) programmers as no further maintenance was required. A couple of
times, that meant that I had to find a new job, but most of the time
I've been grabbed up by a different division.

Two companies have paid me in 20 years (excepting a serious misfit I
stumbled upon between the two). Anyone who can claim similar will
almost assuredly have similar requirements (and accomplishments) to
mine.

--
Jere Lull
Tanzer 28 #4 out of Tolchester, MD
Xan's pages: http://web.mac.com/jerelull/iWeb/Xan/
Our BVI trips & tips: http://homepage.mac.com/jerelull/BVI/