SQL::Statement unusable in major Linux distributions

Paul Beardsell paul at beardsell.com
Tue Nov 3 22:49:18 UTC 2009


H.Merijn Brand,

Thanks for running, against recent versions of SQL::Statement, the test code
I supplied.  You do not mention actual version numbers but I think you have
confirmed the bug exists too in 1.22, the latest release version.
Accordingly I have added 1.22 to the list of versions affected by this bug.
https://rt.cpan.org/Public/Bug/Display.html?id=50788

I think you are wrong about how bugs should be managed in Perl packages.  It
seems to me that the way you and Jens think it ought to work is not how it
works best for other Perl packages.

I assert:

* It is beneficial to the Perl community (developers and users) that bugs
are held centrally. I am sure this is also the position of the Perl
community. The discoverer of a bug cannot be responsible for contacting each
distro.  Distro maintainers cannot independently triage *all* bugs in all
the packages they include.  They need (we need!) "upstream", a central
repository.  rt.cpan.org has the facility to record bugs against older but
still current versions.  It should be used as I think Jesse Vincent also
intended, for the benefit of the wider community, as this central
repository.

* It is detrimental to the community of users and to the package itself to
only accept bugs against what the maintainer arbitrarily decides is the good
version.  Especially when others typically do not have that version.
Package maintainers need to accept that they at least share responsibility
for the overall health of the package as it is actually in use in the
community.  If a package maintainer knows a version, even an old version,
which is in widespread use is broken and prevents a central log of this
being kept then this is not good.

* The best package maintainers take the community of users along on the
journey to better software.  They care that users of former versions ought
to upgrade to more current versions, and they make it attractive for them to
do so.

* The worst package maintainers make it impossible to upgrade by
discontinuities in functionality or by introducing bugs in the new versions
of the code.  Essentially the maintainers fork the package, leaving the user
base effectively orphaned.  There is seemingly no vacancy - there is a
package maintainer - but nobody is maintaining the code actually in use, and
upgrade is impossible.

* Even if someone does nothing more than report a bug this is of benefit.
If one reports a bug and supplies test code that is even more useful.  If
the bug reporter does both those things and supplies a patch that is great!
But if all you do is report a bug, that should be welcomed by a package
maintainer.  Picking holes in the way it is reported, refusing to consider
that the bug might exist in the latest development version, making
unrealistic and non-specific demands on how a bug test script is packaged,
none of these is helpful.

Paul Beardsell
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.alioth.debian.org/pipermail/pkg-perl-maintainers/attachments/20091103/1826e1d3/attachment-0001.htm>


More information about the pkg-perl-maintainers mailing list