Re: [Gnu-arch-users] Re: practical questions of archive ownership

[Joshua Haberman]

On Tue, 2003-09-30 at 23:14, Andrew Suffield wrote:
> On Wed, Oct 01, 2003 at 02:48:09PM +0900, Miles Bader wrote:
> > Still, I'm curious what will happen in the future with emacs and other
> > projects where there _isn't_ a strong central maintainer, and where the
> > notion of a gate-keeper for the canonical sources is unlikely to be
> > acceptable.  It may be that, as now, people do initial development/testing
> > in their own archive/branch, but merge into a shared central
> > archive/branch.
> 
> Here you want a patch queue manager. Your central archive then belongs
> to a program which does all the merging (conflicts are "handled" by
> bouncing the group of changes - no conflict resolution occurs in this
> archive). Developers just pitch bundles of changesets at it; it can
> also be responsible for running test suites and/or something like
> tinderbox. Arbitrary criteria can be implemented for what changes are
> allowed (like "must not cause it to fail to build").

This seems equivalent to giving all the developers write access to the
central archive, modulo the addition of hooks in the commit process.  Is
this patch queue manager intended to offer benefits beyond enforcing
arbitrary criteria for commits, such as not breaking the build or
failing the test suite?  If not, this concept of enforcing criteria for
changesets appears to be orthogonal to the question of how the role of
approving updates to the canonical tree can be distributed to more than
one person.

Josh