On Tue, 30 Oct 2012 17:47:53 +0100 Jerome BENOIT <g62993...@rezozer.net> wrote:
> On 30/10/12 17:36, Jon Dowland wrote: > > On Tue, Oct 30, 2012 at 04:53:24PM +0100, Jerome BENOIT wrote: > >> does it make sense to establish a list of candidates for reintroduction to > >> testing ? > > > > Is this not something best managed on a case-by-case basis? > > my experience as potential sponsoree for such a package answers me no because > it is hard to get a sponsor. As a former sponsor, the most common problem I saw was that packages prepared for sponsoring during a release freeze took no account of the freeze policy and introduced lame or undesirable changes which were not suitable as RC bug fixes. If there's little chance of getting an unblock, there is no point sponsoring. Sponsoring can be more of a burden for the sponsor than just fixing the bug, especially during a release freeze. I have, on occasion, moved on to investigate a different RC bug in a more interesting package rather than go through the process of sponsoring a stale "fix" merely because it exists. This is as much a reflection of the kind of packages involved - most are leaf packages which are of little interest to most potential sponsors. Just because there is a potential fix (which still needs testing by the sponsor), doesn't always mean that the package is worth keeping in Debian. -- Neil Williams ============= http://www.linux.codehelp.co.uk/
pgpfSzUwtD0pP.pgp
Description: PGP signature
