On Aug 6, 2014 2:49 AM, "Carl Baldwin" <c...@ecbaldwin.net> wrote:
>
> I have a spec proposal in play that crosses the Nova/Neutron boundary.
> I split it in to two specs:  a nova spec [1] and a Neutron spec [2].
> There is a little duplication between the two at a high level but not
> in the details.  Each of the specs references the other at various
> spots in the text and in the references section.
>
> This isn't the optimal way to write a cross-project spec.  There is
> difficulty involved in keeping the two consistent.  Also, reviewers
> from one program often don't bother to read the spec from the other.
> This is unfortunate.
>
> However, given the constraints of the current process, I believe that
> it was necessary to split the spec in to two so that the cores
> responsible for each program review and accept the design for the
> proposed changes in their realm.
>
> Would it make more sense to submit the exact same monolithic
> specification to both?  At the time, I chose against it because I
> thought it would make it more difficult to read in the context of a
> single program.
>
> I'm open and looking forward to hearing others' thoughts on this.
>

While I think we need to flesh put how to do cross project specs, and
better cross project communication in general, as this will become more
important in the future, this email thread doesn't seem like the
Appropriate place to do it. This sounds like a new topic, that deserves a
new thread.

> Carl
>
> [1] https://review.openstack.org/#/c/90150/
> [2] https://review.openstack.org/#/c/88623/
>
> On Mon, Aug 4, 2014 at 8:33 PM, Jeremy Stanley <fu...@yuggoth.org> wrote:
> > On 2014-08-05 01:26:49 +0000 (+0000), joehuang wrote:
> >> I would like to know how to submit cross project spec? Is there a
> >> repository for cross project cross project spec.
> >
> > Specs repositories are about formalizing/streamlining the design
> > process within a program, and generally the core reviewers of those
> > programs decide when a spec is in a suitable condition for approval.
> > In the case of a cross-program spec (which I assume is what you mean
> > by cross-project), who would decide what needs to be in the spec
> > proposal and who would approve it? What sort of design proposal do
> > you have in mind which you think would need to be a single spec
> > applying to projects in more than one program?
> > --
> > Jeremy Stanley
> >
> > _______________________________________________
> > OpenStack-dev mailing list
> > OpenStack-dev@lists.openstack.org
> > http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
>
> _______________________________________________
> OpenStack-dev mailing list
> OpenStack-dev@lists.openstack.org
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
_______________________________________________
OpenStack-dev mailing list
OpenStack-dev@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Reply via email to