Having worked both as an application programmer and a systems manager, Jeff's article on accepting that bugs are most likely in your code rings all too true.
One memorable situation, suggests that some programmers perhaps selected the wrong career.
Posted at March 24, 2008 7:20 PMSometimes simply describing the bug or creating a concise reproducer can help you find the bug.
Sometimes a concise reproducer identifies the bug, and sometimes it serves to identify where the bug is not likely lurking.
Should you need to communicate the bug to whomever you might be blaming or whomever might be assisting you with looking at the bug (a "second set of eyes" can sometimes be helpful), here is how one vendor documents the bug reporting process:
http://developer.apple.com/bugreporter/bugbestpractices.html
While ill-constructed bug reports are seldom ignored, such reports do have a higher likelihood of being marked "unreproducible". This if the engineer can't determine what is being reported or can't determine a way to reproduce it (if a concise reproducer was not provided), resolving the bug is anywhere from far more difficult to potentially unfeasible within the engineer's allotted time.
If you want the bug identified (even if it's your bug) and resolved, you don't want your bug delayed nor (worse) marked as "unreproducible."
Posted by: Stephen Hoffman at March 25, 2008 2:24 AM
My favourite bug squashing technique is a specific version of "describe the bug". Grab anyone passing in the aisle. It doesn't matter if they are computer literate or not. It's probably beneficial if they are not on the programming team and have no idea of what you are talking about when you start pointing at code and describing the bug.
Then describe what blocks of code do. Most likely, the person will look blank-faced at you, forcing you to break your description down into simpler terms. Hopefully, doing this will cause you to say something like "And this line of code does foo, see? Uh, wait..."
And another bug bites the dust.
Posted by: Jim Duff at March 25, 2008 12:54 PM
Comments are closed