closing older bug reports without looking = bad practice ?

[Dave Jones]

On Tue, Feb 21, 2006 at 06:23:29AM -0500, gb spam wrote:
 > > Over the following few days, _dozens_ of bugs then get closed by the
 > > reporters (sadly there's a bunch that report "yup, is fixed",  but
 > > still leave me to go close the bug for them).
 > 
 > Closed for the reporter doesn't necessarily mean closed for everybody
 > who has been folowing the bug (withor without posting to it), or
 > closed to the satisfaction of the maintainer - who may have been made
 > aware of more gotchas than the reporter.

9 times out of 10, if its closed by the reporter, and others are
'still seeing it', it's a different bug that really deserves its
own bug report. Bugzilla makes this easy with its clone tools.

The usefulness of a bug actually goes down over time as more
and more people start adding unrelated issues to it thinking
they have the same problem.  Once it diverges from the original
problem enough, it gets to a state where you either close it
when the original reporter is happy, or you end up with a
bug with so much random noise, that it's next to impossible
to make enough sense out of the bug to make everyone happy.

It's why I try to get to the bottom of (or at least root
cause so I change the summary) bugs with summaries like
"the kernel oopsed", "my sound card doesnt work".
Making a bug summary as specific as possible stops other
people who are searching for bugs jumping in with
"Hey, I saw an oops once too!" and pasting a half dozen
comments on something completely unrelated.

Splitting off bugs after the reporter is happy to
'start fresh' on any other problems really is the
best way to keep things readable due to bugzillas
lack of comment threading.

		Dave