On 02/13/2012 01:54 PM, inode0 wrote:
Sorry, this is long. Really, sorry this is long.
And with this mail, John, I forgive you for being long, really long,
because I think I was just longer.
For the past couple of years Fedora has had a policy governing this
developed by Paul, Max, and others. At times the process has worked
very well and at other times the process seems rather dysfunctional
and arbitrary. Many of those involved in trying to implement this
policy have at times been very pleased and at other times
uncomfortable about its execution in practice. I suppose that isn't
all that surprising but as some of us look back at the policy after a
bit of a break-in period we are looking for ways to improve things and
I'd like to ask for any comments or suggestions that the Board might
have to that end.
For reference here is the full policy as developed back in 2010.
It is rather long and I'd like to include here the executive summary
of the point of the policy from its beginning.
This document explains the process for obtaining sponsorship for event
attendance. We use this model for FUDCons, FADs, and other events
where the Fedora Project provides partial or total travel subsidies.
Almost everything on this page can be summarized as follows:
* In every case where Fedora Project funds will pay for events,
there is a responsible party -- a person or group -- who will handle
any requests for travel subsidies.
* That party will consider requests based on a number of criteria,
including relevance to the event, how critical that request is, the
specific deliverables it will enable, the proximity of the traveler,
and the amount of the request.
* Decisions are made in an open, transparent process that
complements the rest of Fedora's processes.
While there have been various issues with the formality of requests
and the tracking of results both of those can be improved by the
concerted effort of the "responsible parties" who approve subsidies.
Most of the difficulty that I see and that has been expressed to me by
others revolves around those responsible parties. The section of the
policy that defines them is here
I've received feedback and have personal experience with feeling we
haven't quite lived up to the blurb at the end of this section.
Openness and transparency
Decisions on sponsorships, regardless of who makes them, are made in
an open and transparent manner. All Fedora contributors should feel
comfortable with the stewardship of event funding.
I believe these decisions are made in open and transparent ways, but I
also know that all Fedora contributors are not always comfortable with
the results and I'd like to improve that. More troubling to me is that
not all the people involved directly in the decision making process
are comfortable with the results.
I haven't been involved in FAD funding decisions so setting those
aside I can speak to the process used for FUDCons and other events.
FAD funding has generally been fairly ad-hoc; the budget for such things
comes out of the Premier Fedora Events budget, and not the regional
support budget, so it's typically been a "Go ask Max/Harish/Jared/Paul"
type of thing. The FAD page usually needs to have been made, with a
budget laid out, etc.
(That said, we have seen a dramatic drop in the number of FADs
sponsored; I don't think this is necessarily a sign of anything
concrete, except perhaps (a) people have forgotten that this resource
exists, or (b) we have fewer people working on new projects or solving
problems that they are excited about, and can produce results
face-to-face more rapidly. I suspect the latter may be the case, and is
certainly troubling, but is a topic not really for this email.)
For FUDCons there has been a concerted effort to follow this process
and that much has worked well. Where it seems to break down at times
is in the composition of the "responsible party." That has been a
moving target over time and probably over geographic region. CommArch
originally made these decisions as I understand it, then CommArch
working more in public with the FPL and the team organizing the
FUDCon, then more the FPL with the organizing team, then more the FPL
with the organizing team and whoever (usually very interested parties)
else shows up and starts voting.
For other events in some places funding requests have gone through
FAmSCo. In North America they have normally been approved by FAmNA
directly or delegated to an event's owner. Often this has worked well,
although delegating a travel budget to an event owner is usually
putting the cart in front of the horse when there isn't a clearly
defined need for travel subsidies. I don't think we've done as good of
a job as the FUDCon planners at implementing the other parts of the
process (requests, reports, etc.) as we could and often regional
groups of ambassadors may not be familiar enough with the overall
budget to know where things stand so there is at times the appearance
of a default approve policy.
So we will continue to do our best to improve our execution of this
policy over time at all levels. The one issue that keeps coming up in
all of these situations though is some contributors are not
comfortable with the composition of the "responsible parties." Are
there ways you can imagine where perceptions of fairness or at least
disinterestedness could be improved by some means? Possibly some more
formal notion of who the responsible parties are or how they are
constituted? Perhaps a Fedora Council?
Every time I begin thinking about this I keep banging my head on the
fact that the closer the decision is made to the event the more some
see the decisions as self-interested and sometimes showing favoritism
and the more removed it is made from the event the less the decision
maker understands about the people and needs of the particular event.
Perhaps there isn't any great solution and we will all just have to
continue to do our best. After two years of trying I thought this
might be a good time to look for ways to improve the process for the
As someone who has sat in countless FUDCon subsidy meetings, I
understand your frustration level here; the meetings are incredibly
tedious, tension-filled, and often the only participants are the ones
that want to attend the FUDCon. The same can be said of, at least in
NA, FAmNA meetings where event sponsorship is requested; those wanting
to go show up, and nobody else does.
WRT regional events, more specifically, I think we've been lax in the
past with follow-up simply because we had a BOATLOAD OF CASH, and we
didn't really have to worry about prioritizing; we simply made an
attempt to ensure that the event had some value, and an owner, and some
We are no longer in that situation. The regional budget for FY12 was at
least pushed to the absolute limit, and more likely exceeded the limit.
Additionally, whereas in the past the regional budget was largely
utilized by EMEA and NA, we now are starting to see rapid increases in
community activity and requests from APAC and LATAM, which is a good
thing, but will certainly require those in EMEA and NA to be more
cognizant of impending events and requests in other regions.
I do also want to say that in my personal experience I do believe that
everyone involved in making these decisions does his/her best to make
the right decision for Fedora. I would just also like to see fewer
people questioning that in the end.
As far as solutions go (or at least thinking towards them), I have a few
* Choosing who goes where
The Council is a reasonable idea for managing some of this, particularly
as we encourage Non-ambassadors to consider requesting/utilizing
Fedora's financial resources, as I know Christoph has mentioned having a
more diverse group of folks available who have more knowledge of
people's contributions in either projects or regions of the world. My
fear here is this - we have had an incredibly tough time just getting
enough people to run for positions as it is, will this be any different?
One solution is to simply require a non-biased quorum of $some number of
folks to be present to make the decisions. An easier way to do this, at
least WRT regional events, is to just do all event approvals once a
month, a quarter, etc. and make sure that we are singing from the
heavens that the meeting is happening, and that people should be present
to make their cases, and that we need people who are not the proposed
attendees to attend as well. Of course, we have FAmSCo to handle
requests as well, and many regions simply default to FAmSCo's
decision-making process rather than proposing in a regional meeting.
Same goes for FUDCons - simply having a quorum of people who are not
attending, and screaming from the sky that unless people show up, there
will be no fudcon. Period.
However, I have some additional input here. For the Tempe FUDCon, we
made the case that we were having additional people from each region
come to participate to learn how to run a FUDCon, and to bring that
knowledge back to their respective regions. And to that, I say, MISSION
ACCOMPLISHED, we now have plenty of people with the knowledge. And yet,
for Blacksburg, we had numerous people applying from out of the country,
with requests like, "I'm coming to teach about X," or "I'm coming to
learn about how to run a FUDCon," "I'm coming to engage with other
people from the teams I work on," etc., without any very specific,
concrete deliverables. I think these requests (and grants) need to be
cut down drastically, or we should reconsider the idea of just having
one or two large fudcons a year, bring in as many people as we can, and
push people to enable smaller one-day events for outreach in their regions.
* Budget Management
As I mentioned previously, the Regional budget does not go as far as it
used to, and I believe we need to be *far* more careful with it going
forward. This mostly involves planning ahead.
In EMEA and NA, we have a fairly good handle on what events we
traditionally go to. Less so in APAC and LATAM, though there are
certainly events like LCA and FISL where we traditionally send people.
What I would love to see is a proactive approach to spending each year -
where each region gets a handle on (a) What events they likely expect to
attend, (b) How much they expect to spend at each one, including any
sponsorship (and I don't believe we should shy away from .org or
lower-level sponsorship of some events), (c) Swag planning, (d) Buffer
for additional, non-listed events, or thoughts of allocating towards
additional events that we haven't attended in the past, (e) Other stuff
- shipping, media, etc.
At some point, we'll get a general idea of what we plan on spending.
There is also additional money that FAMSCO has laid out for things like
release parties, etc. and we've traditionally said "if it's under $X
amount, just do it," and it may be time to at least take a look and see
how much we have spent in that area.
I would like to see FAmSCo take the initiative, as we start launching
into the next fiscal year, to reach out to each region and ask them to
compile these numbers, and come back and say, this is what we're
expecting to spend. I agree that everyone is always acting in Fedora's
best interests, but I think it would add an additional level of at least
comfort with what we're all spending, or planning to spend.
Additionally: I would like to see a return to accountability here. That
means: Having complete event proposals, submitted more than 3 days
before the event. Ensuring that people understand that, unless there
are exceptions, having a Red Hat employee or community card holder pay
for your hotel room does not mean they will be picking up breakfast,
lunch, dinner, and 6 cocktails. Seeing a final tally on costs for events
IN THE EVENT PAGE ITSELF, not loosely hanging on another budget page.
People who don't follow up with event reports, blogs, etc. as requested,
or finalize the event page with that material post-event, shouldn't be
considered in the future, or should at least be put on the "don't screw
it up again" list.
And I don't know what the shape of Famsco reporting is right now;
perhaps we need to get back to a focus on it, and if famsco isn't
willing to do it, get a group together who is willing to put in the
effort. But that's been a major piece of the accountability, and
additionally, an important piece that Max, or Harish, or the FPL can
show to Red Hat and say, "Here's the list of where your money went, and
how it was used, and why continuing to invest here is important."
Again, being ahead of the game here is key: We've seen, at least in
FAmNA, far too many event requests come in at the last second, when
people aren't aware that they are on the agenda, don't know to show up
to contribute feedback, and it is such an emergency that we rubberstamp
it in the name of not blocking people from doing things in Fedora, and
while I'm all for not blocking people from doing things, I think that
there should be some responsibility on the part of the event owner to
get it submitted in a timely fashion. Period.
In short: Well, I suppose a specific, less wordy proposal for at least
working towards a solution might be in order. I'm happy to take a cut
at it, and while I don't think it's something the Board necessarily
needs to approve or disapprove, I'm happy to work towards it, assuming
other people also think there are problems in this area.
advisory-board mailing list
advisory-board mailing list