Reporting bugs: there is nothing to gain in frustrating reporters

[Roberto Bagnara <bagnara at cs dot unipr dot it>]

Stimulated by the discussion about fixing bugs and frustrated
potential developers, I thought it could be useful to briefly
share my recent experience of frustrated bug reporter with the
subscribers of this list.

Until the recent past, I have never given up investigating any
suspicious behavior of GCC and I have always tried hard to
come up with a testcase, check the standard documents and write
decent bug reports.  But this activity has recently been so
frustrating that I now see it as a loss of time.  In the past,
we had the list of open bug reports that was growing beyond
control;  to improve the situation now we have people sitting
on the bug database on a 24/7 basis, with an apparent anxiety
to close bug reports as quickly as possible.

The pattern I went through lately goes more or less as follows:

1) I submit a bug report;
2) someone looks at it superficially, too superficially, and
   posts a comment that tends to deny there is a problem;
3) I and/or someone else explain that the problem is indeed
   there, possibly citing the points of the standards that
   are being violated;
4) the person who said the bug was not (exactly) a bug does
   not even care to reply, but the superficial comments
   remain there, probably killing the bug report.

I wonder what is the rationale here.  Discouraging bug
reporters may be an effective way to avoid bug reports pile up,
but this is certainly not good for GCC.

Examples of the pattern I described above are in:

  http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21067
  http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21032
  http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19092
  http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12963

My advice to people filtering bug reports is: if you only had
time to look at the issue superficially, either do not post
any comment or be prepared to continue the discussion on more
serious grounds if the reporter or someone else comes back
by offering more insight and/or precise clauses of the
relevant standards.

All the best,

    Roberto

--
Prof. Roberto Bagnara
Computer Science Group
Department of Mathematics, University of Parma, Italy
http://www.cs.unipr.it/~bagnara/
mailto:bagnara@cs.unipr.it