Open Stack

Hi,

With apologies to the specs repos we just created, the more I think
about this, the more I think that the right answer is that we should
stick with codenames for the spec repos.  The codenames are actually
more discoverable for potential contributors and collaborators.  If
you're looking for the place to submit a spec for swift-bench, you're
much more likely to find the 'swift-specs' repo than 'object-specs'.
And while some of our older programs have nice catchy names, the newer
ones can be a mouthful.  Here's a list of likely names based on the
program name:

Program Names
-------------
compute-specs
object-specs
image-specs
identity-specs
dashboard-specs
networking-specs
volume-specs
telemetry-specs
orchestration-specs
database-specs
baremetal-specs
common-libraries-specs
infra-specs
docs-specs
qa-specs
deployment-specs
devstack-specs
release-management-specs
queue-specs
data-processing-specs
key-management-specs

Note that "database-specs" is potentially quite confusing.

And here's a list based on the program's codename:

Codenames
---------
nova-specs
swift-specs
glance-specs
keystone-specs
horizon-specs
neutron-specs
cinder-specs
ceilometer-specs
heat-specs
trove-specs
ironic-specs
oslo-specs
infra-specs
docs-specs
qa-specs
tripleo-specs
devstack-specs
release-management-specs
marconi-specs
sahara-specs
barbican-specs

When I look at the two of those, I have to admit that it's the second
one I find more intuitive and I'm pretty sure I'll end up calling it
'sahara-specs' in common usage no matter the name.

-Jim