El 2014-09-26 10:18, Matthew Miller escribió:
Wow, it got kind of quiet in here. Sorry for the delay — between F21
Alpha
and the bash vulnerability, it's been a busy week. But let's get back
to the
big issue!
I can see a large open map of possible ways to go forward, but here's
my
suggestion for what I think is best. I've tried to take into account
the
concerns noted in the previous board ticket¹. Let me know if there are
big
areas that seem to be missing and we'll work those out. I know we need
an
official document on consensus process and how exactly that will work;
that's a next step. Like everything we do, it's okay if we don't get
all of
the details right at first; we can always refine.
As we've discussed, it's really hard to represent the entire sprawl of
the
project on a board with a limited number of seats. In addition to the
representative seats we've talked about, I want to have two to four
members
each actively connected to an 18-month goal; for example:
- Fedora.next products (Marketing + WGs + QA + Metrics)
- Fedora.next rings (Base WG + Env & Stacks + RelEng)
- Automate all the things (QA + RelEng + Infrastructure)
- Improved project communication (CWG + Design + Hyperkitty)
These are example goals (although not completely arbitrary). Here, the
person selected would be actively driving and coordinating each thing.
The
parenthetical groups are non-exclusive examples of areas of the project
each
"objective lead" would work with.
These would be relatively high-load roles — this is where we really get
into
reducing stress on the FPL. In most cases, it shouldn't be hard to find
someone, as these should generally represent big ongoing effort in the
project anyway. Sometimes, though, there may be something which we all
agree
is a good idea but no one has made their focus — as Eric was saying, if
we
only draw people from things we're already doing, it's hard to do new
things.
In any case, these would be focused on specific goals, and once each
goal is
met, changed for a new position reflecting a new community goal. They
would
be more than just advisory members, but their votes would only be
binding in
areas related to their objectives.
Then, we want to have representative seats, as we've been discussing. I
propose four:
- Engineering (representing FESCo, QA, RelEng, Infrastructure, WGs,
FPC, etc.)
- Outreach (representing FAmSCo, regions, Marketing, etc.)
- Elected (at large)
- Elected (at large)
The engineering and outreach members would be selected by the people
active
in those areas (probably FESCo and FAmSCo, as the existing steering
committees for each). One seat covers a very broad area, because that's
the
only way to keep the numbers small. The responsibility would be to
represent
those groups collectively, _not_ to be an individual voice that happens
to
be voted-in by some subset of Fedora. Representatives of more specific
groups would be encouraged to participate in discussion, but only these
positions would have a final binding vote. (We could formalize these as
"auxiliary members" as Christoph suggests; my preference is to keep it
informal.)
The elected positions cover Fedora's subprojects not under the
engineering
or outreach banners (docs, l10n, etc.), and the community at large. I
don't
see a better way of representing the groups that don't fall neatly
under
"building the distro" and "building the userbase" while keeping the
number
of seats to a level which feels functional.
Finally, we would keep a few formalized appointed positions. Overall,
this can be thought of as just reframing the currently-appointed board
positions as connected to specific named roles:
- Fedora Program Manager (FPgM)
- Community Action and Impact
- Diversity Advisor
- Fedora Project Leader
The FPgM is something that we actually have already, and which we
actually
depend on a lot for getting releases out the door, so I agree with
Jaroslav
(the current FPgM) that it would be nice to formalize that role within
Fedora itself.
Community Action & Impact is a new one — Red Hat's Open Source and
Standards
group is interested in funding a new full-time position to lead
initiatives
to grow the Fedora user and developer communities, and to make Red Hat
/
Fedora interactions even more transparent and positive. The Fedora
community
budget comes to us through OSAS, and this position will facilitate
decision-making on how to best focus that to meet our collective
objectives.
The diversity advisor role also doesn't currently exist, but it's an
area
where we have a lot of work to do, so I would like to see it exist.
(And I
think it's much bigger than an 18-month objective.)
And the FPL... that's me. I'd still like to stay involved. :)
Overall, we wouldn't be meeting-driven, although we would probably have
semi-regular meetings in order to discuss what's going on and to clear
through anything outstanding which can be quickly resolved. We might
make
the regular Monday meeting into "office hours", instead. All members
would
be expected to regularly communicate what's going on in their area,
through
blog posts or other public updates. We'd use the lazy consensus model
extensively, with strong expectations for communicating about decisions
or
plans that might affect others.
I feel strange that I have agreed with all the above proposal. So I
force myself to read carefully two more times.
Until the third time I spot a concern, which I think under the category
of try and refine. Something to keep and eye to see hot it eveolves. I
see an overlap in some area that may incline the decision in one way.
- Automate all the things (QA + RelEng + Infrastructure)
- Engineering (representing FESCo, QA, RelEng, Infrastructure, WGs, FPC,
etc.)
- Fedora Program Manager (FPgM)
This overlap may be seem as a duplicity, but probably the work load may
be too much for converting those three into two seats. These are 25% of
the proposed board, but this is a technical driven project. Probably
instead of being too much, it is an under-represented side.
Having said so, there is two wild cards, people elected. That may change
proportions in any direction. As I said above, this is more something to
watch how evolve, rather than something to change in the proposal.
I also think that the outreach will be a real burden time consuming
role. This seat need to be keeping tabs on many groups. Either live at
several regular meetings or catching up with long reading logs. For sure
will have to be subscribed to many many mailing list. This looks like a
full time job and probably will be a quick burnt out role. Again,
something to keep an eye and refine if needed.
Neville
_______________________________________________
board-discuss mailing list
board-discuss@xxxxxxxxxxxxxxxxxxxxxxx
https://admin.fedoraproject.org/mailman/listinfo/board-discuss
