Open Stack

On 09/08/2016 06:18 AM, Chris Dent wrote:
> 
> There's a governance proposal in progress at
> https://review.openstack.org/#/c/357260/ that I think is worth a
> visit by anyone interested in the definition and evolution of
> OpenStack's identity and the processes and guidelines used in OpenStack.
> 
> I'm assuming that not everyone regularly cruises the governance
> project so this thing, which is pretty important, has probably not
> been seen yet by many community members. It is full of many
> assertions, some probably controversial, about what OpenStack is and
> what we get up to.
> 
> At the moment a lot of the reviews are obsessing over the details and
> interpretations of various phrases and paragraphs. This is in
> preparation for a later presentation to the community that ought to
> engender a long email thread where we will discuss it and try to ratify.
> I fear that discussion will also obsess over the details.
> 
> The ordering here is backwards from a process that could be happening if
> what we want is effective engagement and a useful outcome (one where we
> agree). We should first have a conversation about the general principles
> that are desired, then capture those into a document and only then
> obsess over the details. The current process will inevitably privilege
> the existing text and thus the bias of the authors[1].
>
> I presume that the process that is happening was chosen to avoid too
> much bikeshedding. The issue with that is that the work we need to
> do is stepping back a bit and concerning ourselves not with the color of
> the shed, but with whether it is for bikes, or even a shed. Last we
> talked about it, it was a tent, but there's no consensus that that is
> going well.
> 
> [1] I don't wish to indicate that there's anything wrong (or right!)
> about the current text, simply that it is a presentation of a few
> authors, including some written in the past, not a summary of an open
> discussion in the present day.

tl;dr - We need to accurately capture the existing tribal knowledge as a
prerequisite for being able to have open and productive discussions
about changing it.

I totally hear what you're saying, and I agree with what I think you'd
like to see - which is a vibrant and substantive discussion about what
our principles _should_ be.

However, we're currently in a position (and have been for quite some
time) where there exists some shared understandings between people who
are in positions of technical leadership for OpenStack. Last time a the
majority of the TC was together, we realized that there were a set of
unspoken assumptions that we carried and used to judge things - and that
having those be unspoken is pretty terrible.

This is an attempt to capture the current state of the world and remedy
that problem.

Once that exists, it should empower everyone to actually challenge and
discuss them. It's pretty hard to suggest a change to the fundamental
fabric of something if there is no thing to change, and currently the
only way to really do so would be to corner a bunch of people one on one
at a Summit or something.

A conversation about what we _want_ the principles to be is an EXCELLENT
idea. I just don't think it can happen if people are not all aware of
the assumptions we've been operating under. Any of these being
controversial is an indication, to me, of the problem and why it's
essential that we take the first step of writing them down rather than
continuing to merely hold them as tribal knowledge.

Also - although I know the current review comments may seem like picky
wordsmithing, I think they are pretty essential. These are concepts that
have been around in OpenStack for a long time - but they haven't been
expressed clearly or at all. Making sure that what is written down
represents accurate communication is essential. Possibly the only thing
worse than having fundamental tenets that only exist as tribal knowledge
is having them written down inaccurately.

Monty