[ale] Humorous (to me at least) note

[JD]

A one-line error is VERY DIFFERENT from a stack trace.

I've worked in teams where any memory leak was considered a sev-1 bug to be
corrected prior to shipping - no exceptions.

Hopefully programmers realize that when any tool/program spews more than a few
lines of information to the console (stderr or stdout), that is cause for
concern.  Recently switched to a newer version of perl here - lots of
deprecation warnings were shown - that means I have some work to fix them. ANY
warning is something to be addressed, if I'm the programmer.

On 09/03/2014 10:17 AM, Dylan Northrup wrote:
> This actually bugs this shit out of me, especially when I was dealing with
> some Java developers trying to stress test their software and track down
> the source of any failures.  Hearing "Oh, that stack trace is normal" was
> irritating as hell because it made determining the source of an issue very,
> very difficult.  If you have a known exception, FFS handle it!  You have
> the tools!  It's not like it's difficult, it just takes time and attention
> to detail. . . . and a modicum of care about the ultimate consumers of your
> software (which, sadly, seems to be low on many developers' priorities).
> 
> 
> On Tue, Sep 2, 2014 at 6:08 PM, JD <jdp at algoloma.com> wrote:
> 
>> On 09/02/2014 04:57 PM, Todor Fassl wrote:
>>> I was telling my (sort of) boss about how all these error messages on
>> one of our
>>> systems didn't mean anything and that she should just ignore them
>> because it
>>> would be too hard to make them go away. At the end of my little
>> soliloquy, I
>>> said, "Every thing is 100% under control -- probably."
>>>
>>> Which actually makes sense. I am not 100% sure everything is under
>> control but
>>> if it is under control, it's 100% under control.
>>>
>>
>> Some highly custom NASA manned spaceflight control systems had "standard
>> errors"
>> - NOT seeing the errors was an error to be researched - ASAP!