On Fri, Oct 26, 2012 at 7:46 AM, Arno Töll wrote: > *) we have consensus that we are in need of such a rule set - which ever > it may be > > *) we have three orthogonally different ideas: > a) Bart's approach which was reformulated and proposed by Lucas in > this thread [1] > b) Mine - which was based on timeout arithmetics [2] > c) Michael Gilbert's approach to merge the concept of NMUs with > orphaning packages [3]
My proposal has nothing to say about orphaning itself, and explicitly leaves it untouched. Mine can be considered a more flexible co-maintainer process that importantly can be started far before orphaning becomes a package's last resort. It also seeks to handle the "hard" cases whereas a) and b) explicitly state that they're avoiding that problem altogether. Anyway, I and seemingly many others don't like the bureaucracy of a) and b), especially since there is already a common-law 4*7*24*3600 rule in existence that would probably get applied more often if it were actual devref-law. Finally, if a) and b) aren't meant to do something about the original issue, the "hard" maintainership questions, then what's the point? Why do we need more bureaucracy when we already have a common-law solution? Best wishes, Mike -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/CANTw=MNx1O0d0uQ9pLNgA=4z7ljslbssoqgcsolxjsg_4sr...@mail.gmail.com
