On 3/11/19 10:44 AM, Thierry Carrez wrote:
Mohammed Naser wrote:
Hi everyone.
For quite sometime, there has been work on an effort of moving forward with building 'constellations' however this effort has stalled out for a few months since. It looks like we've tried a few times to get this document done
https://review.openstack.org/#/c/180112/ [3y9m ago, creating compute starter kit, by sdague] https://review.openstack.org/#/c/565466/ [7m ago, draft for scientific computing but in `governance`] https://review.openstack.org/#/c/586212/3 [6m ago, draft for compute starter constillation, by dims]
Do we want to continue this effort (do we see value in it right now)? Does the community find it a useful document to publish and does anyone feel like they can pick it up?
There doesn't seem to be much momentum around that. Here is a bit more context to kickstart the discussion:
Here is the part of the "2019 TC vision" that describes constellations:
https://governance.openstack.org/tc/resolutions/20170404-vision-2019.html#na...
As I see it, there were two key goals in there. First is to get the TC (and contributors in general) to wear more of a product manager hat and be more involved in defining common groups of OpenStack components matching a use case. The other is to expose those constructs more widely than just on the marketing website -- make them first-level objects that users interact with (in particular in documentation).
From what I vaguely remember about these conversations in the past, wasn't testing and validation of each constellation one of the harder parts to figure out?