Re: musings on "the release thing"
In article <19980529103626.B7957@molec3.dfis.ull.es> you wrote:
: If something doesn't work it is removed from the "frozen" distribution
: and moved to the "unstable" one.
I suppose this is the key difference in our approaches. The way we've just
copied all of unstable and declared it "frozen" is fundamentally broken to
my mind. We should implement a subset of the lintian checks and pkg-order as
part of ensuring that anything in a 'frozen' or 'stable' tree is correct by
construction for this class of policy issues.
If we do this, then the priorities really do mean something useful, and what
you're asking for falls out pretty easily. If we go one step further, and
restructure the tree so that packages are split into directories by priority
instead of by the ever-confusing concept of subsection, we'd get the best of
both worlds, I suspect...
I'm willing to help write code to do a better job of handling package release
logistics for slink. It is now way too late to worry about it for hamm, which
we need to wrap up as quickly as possible, so we can get back to having fun.
To UNSUBSCRIBE, email to firstname.lastname@example.org
with a subject of "unsubscribe". Trouble? Contact email@example.com