Once a large number of bugs exist, a strategy can be used (or not) to
resolve them unless resources exist to fix them all within a reasonable
amount of time. It could be FIFO, ascending level of effort, squeaky
wheel/vote, product centric (e.g.,fix all reported Duet issues) or
focused on core functionality/features. I have listed these approaches
in the reverse order of how I believe they impact customer perception of
the software. A weighted approach could be taken such as 80% core
functionality/ 20% low level of effort.

If a bug sits unresolved long enough, it can be rationalized as
unneccessary to fix because users have lived so long with it. When
debugging a new release, goals can be set such as 70% of new bugs
resolved prior to release or 100% of critical bugs resolved prior to
release. If the 70% resolved are easy bugs or the critcal bugs are
mis-characterised, well...

Regression testing helps discover when fixing a bug has side effects.
When a fix to a bug gives rise to another, the bug count does not
diminish.


-- 
DotSystem
------------------------------------------------------------------------
DotSystem's Profile: http://forums.slimdevices.com/member.php?userid=15213
View this thread: http://forums.slimdevices.com/showthread.php?t=68947

_______________________________________________
discuss mailing list
[email protected]
http://lists.slimdevices.com/mailman/listinfo/discuss

Reply via email to