[openstack-dev] Split of the openstack-dev list (summary so far)

Morgan Fainberg m at metacloud.com
Mon Nov 18 06:41:56 UTC 2013


A couple of quick points.
1) I think that splitting the list is the wrong approach.
2) Perhaps we need to look at adding a mechanism that enforces the use
of tags in the subject line (send a nice "sorry, but you need to
indicate the topic(s) you are mailing about" error back if it doesn't
exist, keep an active list of these via infra?).
3) It might also make sense to have all stackforge projects include
[stackforge] in the topic.  That will help make filtering easier.

Finally, I notice the difference in a threaded client from a flat
client.  I don't think I could subscribe to this list without a
threaded client.

TL;DR Don't split the community, work to improve the tools for those
who are overwhelmed. (Email clients, enforcing use of subject tags,
etc)

On Sat, Nov 16, 2013 at 8:01 AM, Nick Chase <nchase at mirantis.com> wrote:
> I am one of those horizontal people (working on docs and basically one of
> the people responsible at my organization for keeping a handle on what's
> going on) and I'm totally against a split.
>
> Of COURSE we need to maintain the integrated/incubated/proposed spectrum.
> Saying that we need to keep all traffic on one list isn't suggesting we do
> away with that. But it IS a spectrum, and we should maintain that. Splitting
> the list is definitely splitting the community and I agree that it's a
> poison pill.
>
> Integrating new projects into the community is just as important as
> integrating them into the codebase.  Without one the other won't happen
> nearly as effectively, and we do lose one of the strengths of the community
> as a whole.
>
> Part of this is psychology. Many of us are familiar with broken windows
> theory[1] in terms of code.  For those of you who aren't, the idea is based
> on an experiment where they left an expensive car in a crime-ridden
> neighborhood and nothing happened to it -- until they broke a window.  In
> coding it means you're less likely to kludge a patch to pristine code, but
> once you do you are more likely to do it again.
>
> Projects work hard to do things "the OpenStack way" because they feel from
> the start that they are already part of OpenStack, even if they aren't
> integrated.
>
> It also leads to another side effect, which I'll leave to you to decide
> whether it's good or bad.  We do have a culture of "there can be only one".
> Once a project is proposed in a space, that's it (mostly).  We typically
> don't have multiple projects in that space.  That's bad because it reduces
> innovation through competition, but it's good because we get focused
> development from the finite number of developers we have available. As I
> said, YMMV.
>
> Look, Monty is right: a good threaded client solves a multitude of problems.
> Definitely try that for a week before you set your mind on a decision.
>
> TL; DR Splitting the list is splitting the community, and that will lead to
> a decline in overall quality.
>
> [1] http://en.wikipedia.org/wiki/Broken_windows_theory
>
>
> _______________________________________________
> OpenStack-dev mailing list
> OpenStack-dev at lists.openstack.org
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
>



More information about the OpenStack-dev mailing list