This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Reporting bugs: there is nothing to gain in frustrating reporters
- From: Michael Veksler <VEKSLER at il dot ibm dot com>
- To: Steven Bosscher <stevenb at suse dot de>
- Cc: gcc at gcc dot gnu dot org, Gabriel Dos Reis <gdr at integrable-solutions dot net>, Vincent Lefevre<vincent+gcc at vinc17 dot org>
- Date: Mon, 20 Jun 2005 12:50:07 +0300
- Subject: Re: Reporting bugs: there is nothing to gain in frustrating reporters
Steven Bosscher <stevenb@suse.de> wrote on 20/06/2005 11:13:35:
> On Jun 20, 2005 09:51 AM, Michael Veksler <VEKSLER@il.ibm.com> wrote:
>
> > Despite being descriptive and friendly, bug masters
> > frustrate me and other users by being too eager
> > to close the PR. I would suggest a policy change,
> > a PR should be closed (as duplicate or as INVALID)
> > only after discussion was exhausted.
>
> And you going to provide the extra man-power required to track bugs this
> way, right? This means keeping an eye on many more bugs where a
discussion
> is still going on. Unless you have convincing proof that it happens often
> that valid bugs are closed as INVALID, I think we should change nothing,
You are probably right that it will take extra person-power (even
after automating part of the process).
>
> I have seen a discussion similar to your description of a bug discussion
> only once, and in this case the bug master was right. It is equally
> frustrating for gcc bugmasters that some user thinks it is OK to keep
> re-opening a bug report because his/her opinion is The One Opinion. This
> is something that happens a lot. "There is nothing to gain in frustrating
> bugmasters" ;-)
>
Look at PR 21951 for such a discussion that ended up NEW.
I also have users, and I am also get frustrated by repeated bogus
bug reports when the user is to blame. I have learnt that it is
simpler and more efficient to teach few coders to be tolerant
(and not easily frustrated), than to teach many users.
It is probably an axiom that at least one person is going to
be frustrated in a big enough community (GCC). The processes
of the community should minimize the amount, the cost
and the impact of such frustration. It is a fine balance to be
maintained.
> Maybe every once in a while a bugmaster closes a bug report too quickly,
but
> at least the bugmasters get a useful job done. If you compare the state
of
> our bug database now with the mess of a couple of years ago, we are much
> better off now.
>
It is much better now than what it used to be, can it be even better?
Is it cost efficient? I really can't tell.
Michael