Re: bug db cleanup time

Front page | perl.perl5.porters | Postings from December 2000

Thread Previous | Thread Next

From:

Jarkko Hietaniemi

Date:

December 11, 2000 05:49

Subject:

Re: bug db cleanup time

Message ID:

20001211074941.D26502@chaos.wustl.edu

On Mon, Dec 11, 2000 at 01:42:45PM +0000, Michael Stevens wrote:
> On Mon, Dec 11, 2000 at 07:38:00AM -0600, Jarkko Hietaniemi wrote:
> > Note that in the 2.32 month about one thousand emails will be sent out,
> > one thousand people will be reminded of their bugs...doesn't sound like
> > a good idea to open the gates so abruptly, at the worst case p5p might
> > get badly overwhelmed  (Okay, *I* would get badly overwhelmed :-)
> > 
> > I would suggest some ways to alleviate this:
> > 
> > (1) Do not send these messages for anything older than 5.005_03
> >     (these should be marked as 'abandoned' anyway, but just in case)
> >     (yes, one could argue for "anything older than 5.6.0")
> 
> At least the wishlist stuff should probably still be in there, if it
> hasn't been done in newer releases. I'm only happy with abandoning bugs
> where things have changed sufficently drastically to make trying to fix
> it silly - eg the install stuff in older releases.

Yes, isn't the 'wishlist' stuff supposed be perpetually in release n+1? :-)

> > (2) Some ideas on wording:
> > 
> > 	This is a reminder that a Perl bug you reported is still open.
> > 
> > 	... if you are not using the latest version, Perl 5.6.0, you
> > 	please try it first to see whether your bug has been fixed in it.
> > 	You can get the source code ... read INSTALL ... or you can see
> > 	whether your platform has ready-made binary distributions at
> > 	http://www.cpan.org/ports/ ...
> > 
> > 	... if you have new information to report on the bug, such as
> > 	new or improved test cases (simpler test cases are always better),
> > 	code or documentation patches, other kinds of workarounds ...
> 
> Idea for simplicity:
> 
> Only send messages out for bugs reported *after* we start the process.

Hmmm.  I would still like to get any remaining open 5.6.0 bugs to get
a nudge.  And as mjd demonstrated, there are still fatal bugs reported
in 5.005 that are still among us :-(

> Admittedly, it wouldn't help with the backlog, but it should help us
> keep future reports sane...
> 
> Michael

-- 
$jhi++; # http://www.iki.fi/jhi/
        # There is this special biologist word we use for 'stable'.
        # It is 'dead'. -- Jack Cohen

Thread Previous | Thread Next