Re: Massive bug expiration spree

[Ian Jackson <ian@xxxxxxxxxxxxxxxxxxxxxxxx>]

Christian Robottom Reis writes ("Re: Massive bug expiration spree"):
> There will be warning. The Incomplete status has that semantic: it is a
> status that is intended to be driven to zero. If you know enough about
> the bug to acknowledge it is a bug, mark it Confirmed.

I'm terribly sorry, but this view is not compatible with the real
world.  Just because some bug status represents a state that we would
prefer bugs not to languish in, doesn't mean that they don't actually
languish.

We might as well say "It is policy that we aim to deal with all
confirmed bugs in Ubuntu within 30 days" and then because we are
"driving" the older Confirmed bugs "to zero", we should automatically
close them all.

A bug tracker is supposed to record the _real_ state of the world.

The real state of the world is that we are drowning in bug reports:
when we try to do bug triage we are drowning in inadequate reports
which aren't ripe for any kind of developer investigation; when we try
to work on fixing code we are drowning in reports of things needing
fixing.  This is the natural state of the world in software,
particularly in projects which strongly encourage users to file bugs.
This is unlikely to change any time soon.

It is the job of the bug tracker and its search facilities to help us
manage this situation.  It is not the job of the bug tracker to
"incentivise" people by throwing stuff away if they don't work hard
enough!

> On Mon, Sep 24, 2007 at 11:52:53AM +1000, Martin Pool wrote:
> > If Launchpad is now going to have a policy that bugs are not allowed
> > to be in state X for more  than Y days you have to tell people ahead
> > of time.
> 
> We did tell them ahead of time; truth is that people usually react to
> changes rather than to announcements. At any rate we will update the
> bugs closed incorrectly, so the inconvience to you should be minor, and
> we've all learnt from the experience.

I have to admit that this came as a huge surprise to me.  A change of
this nature ought to be highlighted more clearly.  (JOOI, where was it
announced?)

I agree with Martin that _all_ of the changes should be reversed.

Ian.