Open Stack

On Thu, Dec 19, 2013 at 4:14 AM, Sean Dague <sean at dague.net> wrote:

> On 12/19/2013 12:10 AM, Mike Perez wrote:
> > On Tue, Dec 17, 2013 at 1:59 PM, Mike Perez <thingee at gmail.com
> > <mailto:thingee at gmail.com>> wrote:
> <snip>
> > I reviewed the TC meeting notes, and my question still stands.
> >
> > It seems the committee is touching on the point of there being a worry
> > because if
> > it's a single company running the show, they can pull resources away and
> > the
> > project collapses. My worry is just having one company attempting to
> > design solutions
> > to use cases that work for them, will later not work for those potential
> > companies that would
> > provide contributors.
> >
> > -Mike Perez
>
> Which is our fundamental chicken and egg problem. The Barbican team has
> said they've reached out to other parties, who have expressed interest
> in joining, but no one else has.
>
> The Heat experience shows that a lot of the time companies won't kick in
> resources until there is some kind of stamp of general approval.
>
> If you showed up early, with a commitment to work openly, the fact that
> the project maps to your own use cases really well isn't a bug, it's a
> feature. I don't want to hold up a team from incubating because other
> people stayed on the sidelines. That was actually exactly what was going
> on with Heat, where lots of entities thought they would keep that side
> of the equation proprietary, or outside of OpenStack. By bringing Heat
> in, we changed the equation, I think massively for the better.
>
>         -Sean
>


To make my message more clear, I would like to see the TC thinking of this
problem as well. In Cinder for example, there was a push for a shared
service.
One of the problems that the core saw in this feature was it was a
one-sided
project because only one vendor was really contributing. The API they
provided
may work great for them, but may not work for other potential contributors
that
come from another company where their storage system works differently. I
see
this as causing potential serious rewrites that really just sets a project
back.

I'm not at all saying this stops incubation, but just something else to
consider besides
a company pulling out the main resource from a project.

-Mike Perez
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack-dev/attachments/20131219/ce6996e7/attachment.html>