[User-committee] Efficiency of WGs?

Michael Krotscheck krotscheck at gmail.com
Tue Aug 30 17:59:05 UTC 2016


Responses inline:

On Mon, Aug 29, 2016 at 5:36 PM David F Flanders <flanders at openstack.org>
wrote:

>
> One of the recent suggestions has been to converge some of the WGs to
> help ease the burden of these logistical tasks.
>

Focus is good.


> Other options include:
>
>   * having a shared IRC channel for all WG activity to help create
> more water-cooler conversation between chairs?
>

Having a common communication channel is how the other teams in openstack
have become so cohesive (I'm thinking of infra for example). Whether that's
IRC or something else I don't care, though using IRC will certainly build
familiarity with the tool so that there's less friction across the board.
Also, it's on a common ground with other teams, so asking a question to
nova, keystone, etc becomes super easy.


>  * sharing of logistical duties between WG chairs, etc
>

How about reducing the necessary logistics? Maintaining a WG requires a
minimum overhead / fixed cost, based on how many timezones need to be
served, how often the group meets, and how many documents and
communications need to be maintained. That should be adjusted up or down,
based on whether the market supports the additional cost: Are there enough
active contributors in other timezones to warrant adding the additional
cost of a timezone-sensitive meeting? Is there not enough activity to
justify meeting every week? Judge these things against what the WG
produces, and advice can become pretty clear; the trick is recognizing when
demand is shrinking, and to contract overhead to a manageable level. It's
easy to expand bureaucracy, it's hard to shrink it.

As an example, the Hackathon WG started in the App Eco WG, however very
quickly demand and visibility increased to warrant them creating their own
group. More demand makes the overhead worthwhile, and now they're one of
the most productive WG's we have.

In the meantime, the App Eco WG shifted to focus on supporting the SDK
efforts, which have turned into a heavy dev/documentation effort. Perhaps
it's time to create an SDK team under the TC, on par with Infra and
OpenStack itself? That would reduce the duties of the App Eco WG to a level
where it may be rolled back up into the Enterprise WG, until research
results from UX and the Hackathons warrant starting a new effort.

Michael
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/user-committee/attachments/20160830/86a69596/attachment.html>


More information about the User-committee mailing list