On Sun, Sep 01, 2002 at 07:13:33PM -0500, John Goerzen wrote: > On Sun, Sep 01, 2002 at 07:35:25PM -0400, Joey Hess wrote: > > Standard procedure for NMU's was considerably relaxed (or rather, > > clarified) this winter by the release manager. The developers reference > > is just that: a reference of typical best practice. Sometimes it's > > expedient to not follow its every letter, and that's ok. > > I think that doing things for "expediency" can come back to haunt us. If > NMUs are not properly logged in the BTS, maintainers can miss them and end > up uploading newer versions later that lack the NMU code.
Yeah. I did several perl 5.8 NMUs this weekend (all to the delayed queue), but all of them are filed in the BTS. I don't think aj intended that routine procedure to be changed. > I'm concerned that a single person has the power to dictate such dramatic > changes in our procedures. There was quite a bit of discussion about it, mostly centred around the fact that there are a quite unreasonable number of bugs mired with inactive maintainers. The DELAYED queue is a fantastic tool to encourage people to fix bugs while still giving maintainers warning of the proposed changes well before they're actually made. -- Colin Watson [EMAIL PROTECTED]