[petsc-dev] PETSc Bug Database, Issue Search

[Sean Farley]

> Barry doesn't believe in having a place to keep track of this stuff because
> he thinks it causes bug reports to be closed prematurely or allows bugs to
> go too long without being fixed. So we discuss this stuff on petsc-dev or
> petsc-maint and it sits in our inboxes until someone fixes it.
>
> Every once in a while, Barry reiterates his contempt for bug trackers and I
> reiterate my perspective that it is actually useful to users to be able to
> watch bugs or feature requests without the high volume of following
> petsc-dev or reading the Hg commit log to get news of action on it. ;-)

I will, again, add +1 to having a searchable bug-tracker. Even if
something is closed prematurely / ignored, then at least there is a
place for this to be found. As opposed to having users sending
duplicate emails about ignored issues.

Case in point (even though Barry used the exact opposite reasoning),
how would we have known that mpich2 is ignoring the issue that Jed
found without it being publicly online? At least it saved us the time
of finding out the hard way.