Open Stack

On Mon, 2013-07-01 at 18:41 +0200, Thierry Carrez wrote:
> Monty Taylor wrote:
> > So I think that tying a definition directly to whether or not there is a
> > release artifact or when that gets produced is missing the point and
> > wrong. The point is that there are _efforts_ that we want to address,
> > and those efforts may not have a clear cut simple "main" source code
> > repository.
> 
> I can certainly agree that all those things are efforts which may or may
> not produce stuff that may or may not be part of the "integrated" subset
> of the things we end up publishing.
> 
> The two reason why I was trying to draw a line in the sand between
> "Projects" and "Programs" was that:
> 
> (1) "Projects" are mentioned in the TC charter (and the foundation
> bylaws), so they are not going away easily.. hence me trying to preserve
> them while introducing Programs

Ah, interesting - if I had thought about it, I would have presumed that
pretty much all the verbiage in the TC charter would apply equally to
programs.

> (2) "Projects" go through Incubation, while "Programs" don't need any
> incubation period. (They go through incubation because they are part of
> the "integrated release", which is not everything we release, but the
> subset of stuff we publish on a 6-month release cadence. They go through
> incubation so that we can make sure their addition will not screw up
> existing "integrated" projects and they are actually "integrated" with
> the other projects. They go through incubation so that we can make sure
> they can fit in the stable maintenance, vulnerability management and
> release management processes.)

Oslo libraries have plenty of opportunity to screw everything else too
and needs as much release management as everything else. So, that
distinction doesn't really make much sense.

The "exposes a user-facing REST API" distinction is enough to explain
the need for Incubation ... for now, at least.

Cheers,
Mark.