Open Stack

On Mon, 23 Apr 2018, Doug Hellmann wrote:

> What aspects of our policies or culture make contributing to OpenStack
> more difficult than contributing to other open source projects?

Size, isolation, and perfectionism.

Size in at least three dimensions:

* the entire community
* individual projects in terms of humans and scope
* individual projects in terms of lines of code (per repo and per
   file)

Isolation in at least two dimensions:

* For someone who is not "of OpenStack", OpenStack is kind of "over
   there, doing its own thing". Non-OpenStack colleagues wonder about
   the tempestuous teapot I'm in.

* Individual members of project teams sometimes self-identify as
   members of that that team, not of OpenStack.

Perfectionism:

* In at least some teams project teams (see, look at me
   identifying and isolating project teams) proposed specs and code
   can be nitpicked to death and forward progress delayed while every
   edge case is considered. We should strive to iterate more.

* At the same time there is too strong and attachment to master
   needing to be perfect. A bug on master is an invitation addressed
   to a potential new contributor.

> Which of those would you change, and how?

I think we've started making a more conscious effort on all three of
these areas. We talk more often about incomplete bug fixes being
adopted experienced contributors. Decomposing repositories to harden
contractual and social boundaries is increasingly common. Actively
working with other communities (notably Kubernetes) is on the rise.

But there is plenty more to do in each of these areas.

> Where else should we be looking for contributors?

I agree with Thierry that academia is a good place to look and that
we made a mistake when we highlighted and enforced an artificial
boundary between developers and operators. Ideally many features and
bug fixes would come from people who _use_ OpenStack as their day
job. The people who think of _developing_ OpenStack as their day job
should be most focused on enabling those other people and cleaning
up and refining what already exists.

I also think that we need to figure out, if possible, some way to
make OpenStack relevant and interesting to individuals who are
technically curious enough to want to try playing with their own
mini cloud at home. If we can make OpenStack accessible to amateurs
(not amateurish!) there's a big world of good input to come.
Something as one stop, integrated in the documentation and official
seeming as minikube is for Kubernetes.

-- 
Chris Dent                       ٩◔̯◔۶           https://anticdent.org/
freenode: cdent                                         tw: @anticdent