Signature Events Strategy for 2017 (Part I: Flock)

[Brian Exelbierd]

Warning: This email is long.  I don't know how to avoid that.  I debated
making this a wiki, but I think we need more discussion that just
editing.  There is no tl;dr.  If you want a tl;dr perhaps you should
wait a while and participate near the end of the conversation.

Warning: The opening text is the same for both Part I and Part II.

During 2016 I had the privilege of helping to plan and execute Flock and
attended both FUDCon LATAM (Puno) and FUDCon APAC (Phnom Penh).  Talking
to people has led to me to believe that these events may not be meeting
all of our goals.  Specifically we are spending a lot of time and money
on them and may not be getting our full value.  I also heard a lot of
comments about how the events are not equal from people who thought they
should be and not well distributed across the world from people who
thought they should be.  This led me to believe that there is a lot of
confusion about why we do these events and what we want out of them.

Since it is easier to edit than to create, I am going to propose some
ideas.  My goal is less to present this as a polished proposal ready for
a vote and more to allow us to have a discussion around the finer points
of the events strategy.  It would be nice to see this discussion come to
an end by the end of March so that it can have impact this year.  The
planning for Flock will begin soon (see my another of my emails today)
but the FUDCon processes are not heavily started, as far as I can tell.

Additionally, while history is important, I think it is equally
important that we consider what we want to accomplish today and in the
next few years, not just what we tried to accomplish in the past. 
Therefore I've written this without too much reference to what is being
changed and instead as an idea of what we should be doing.  I look
forward to folks bringing forward suggestions for continuing activities
from the past or better incorporating lessons learned that I may have
missed.

I look forward to your feedback and input.

regards,

bex

# Flock

Flock is described by most people to me as the "Fedora Contributor
Conference."  Therefore, I'd like to see us put together a structure to
continue to make that statement more and more accurate.  If we are going
to have a contributor conference, I think we need to understand why
having one is important.  I have been told that we need Flock for these
reason (I am sure I am missing some):

## Goals

* Increased Contributor Bonding - Contributors should meet each other in
person and strengthen their relationships.  We are a far-flung remote
community and at times we forget about the person behind the FAS ID. 
Fedora works best when we work together.

* Increased Project Cohesion - The Fedora Project is a huge complex set
of moving pieces and fantastic people.  At times it can be hard to know
or remember what is going on and what we are working on.  Flock should
provide a touchstone to keep all parts of the project interconnected.

* Presentation and Discussion of Large Ideas - Some ideas just aren't
well served by a wiki page and an email thread.  When we make major
changes (think rings or editions) we need to talk them over.  We need
presentations and forums for face to face discussions.

* Work - Yes, this is what FADs are for, but we can save money and time
if we schedule work sessions in a place where we are already going to
have the right people present (see more later).  Flock has been
described by several people as needing to be more of a "do conference"
instead of a "talk conference."  I agree that if we are having a
*contributor* conference we should focus on *contribution.*

* Cross-pollination - Having contributors from all over the project in
one place creates the opportunity for a fantastic hallway track
(informal, unplanned conversations and working groups).  Problems being
solved in one area may be the answer for problems being discovered in
another.  We may find out that we have multiple groups working on the
same problem in different ways and that working together may help (or
conversely allowing them to have a compare/contrast session may help).

## Programming

Moving Flock more toward a "do" conference means that we should change
the way we develop the program.  Specifically, I believe that we should
ask people why their talk and workshop proposals should be included in
greater detail.

### Theme

I believe the council should, with community input, set some themes for
each Flock.  These themes are not absolutes, but instead should be one
of the many factors considered when looking at programming submissions. 
This way we can also help keep Flock focused on doing.

### Talks

Talks will always be part of Flock, however, we can ask a few more
questions to help the committee choose them.

1. How is this talk focused on contributors?  Why should contributors
attend it? Which contributors should attend? How will this talk foster
greater contribution to Fedora?
2. If this talk is about a specific piece of software or technology, why
is presenting it at Flock better than presenting it as a Fedora-inspired
talk at another conference?
3. What actions will you be taking as a result of the feedback you
receive on this talk?  What actions are you looking for your audience to
take?

### Workshops/HackFests/etc.

We need more focused work and planning sessions.  However, for them to
be successful, we need to make sure they are planned for success from
the beginning.  Therefore, I believe we should ask questions like this:

1. What specific goals do you have for this session.  What is your
specific agenda?  What problem(s) are you trying to solve?
2. Who specifically needs to be present in order for the goals of this
session to be met?
3. What activities will be done prior to Flock to ensure success and
that all attendees are ready for this activity?

## Funded Attendance and Costs

I was told by several people that they felt like most non-EMEA and
non-NA contributors were not encouraged to attend Flock.  Several people
related stories about how they had been told that because their region
had a FUDCon they should attend that instead.  This sentiment was
usually relayed to me during a conversation about funded attendance.  I
want to address this through our structure.

I believe that Flock should be explicitly open to anyone and built for
Fedora Contributors.  If someone is not a Fedora Contributor they should
recognize they may not get a lot out of attending Flock.  That is OK. 
We will have other opportunities for them.

The flip side of this is that we had a significant number of registered
attendees for Flock 2016 who did not attend.  This cost money in terms
of wasted catering and prepaid expenses.  I am told that this year was
exceptionally heavy in terms of no-shows, but that we routinely have a
significant number of no-shows and unregistered attendees.  We need to
find a way to ensure that when we budget for someone's attendance that
they really come.

I think that we should consider funded attendance from several angles:

### Tickets

Flock has expenses that are not necessarily part of the "funding" of a
funded attendance.  This could be meeting room costs or the costs or a
group lunch or evening activity.  Some of these costs are fixed for the
event (e.g. Meeting room rental) and some are per attendee (e.g. Evening
activities).  Therefore I think we should set a ticket price equal to
the cost of all per attendee costs.  This means that a ticket will cover
the cost of these items, and not make a profit.

However, not everyone should have to pay for their ticket.  I believe we
should adopt a strategy similar to OpenStack.  OpenStack provides free
admissions for all active contributors.  We should define contributor
activity and provide a similar benefit.  Defining an active contributor
is hard, but we can probably start with something like "generated X
fedmsg events in the last year" or "acknowledged as a contributor by a
subproject."  I think it needs to be more than "cla+1" but flexible
enough to account for work that may not be commit oriented.

If only active contributors attend Flock then no tickets will actually
be sold.  This is OK.  However, non-contributors who attend will pay the
costs of their attendance.  This may help those folks understand that
there are better Fedora events for them to attend.  This should also
reduce the issue with no-show non-contributors.  It won't help with
no-show contributors[0].  I also believe it is reasonable to have a
refund policy in place for those with a documented reason for not
showing up (illness, etc.).  I think we can solve this by adopting a
policy like many other events of offering refund schedules, etc.  This
should not be a heavy process and it should actually not be a huge
burden.

### Funded Travel

We cannot afford to cover all of the costs of all of our contributors
for Flock.  There is currently no budget probability of this changing. 
Therefore we need to figure out a way of choosing who gets funding and
how much more deliberately.  I believe we should do this by creating a
more complete funding application.

We can assign points or use some other method to rank applications.  I
think that points of consideration should include:

* Is this person an active contributor?  (see above)
* Is this person on "people needing to be in the room" list (see above)?
* Is this person a contributor in the general areas of focus of Flock
(we will know this once the program is set)?
* How will the Fedora project benefit through your attendance at Flock?
* How will you benefit (specifically) from your attendance at Flock?

As far as funding levels, I believe our default funding for active
contributors should be the cost of the ticket mentioned above and 0%
travel.  After that travel funding can range from partial to full.

I believe we also need to strongly encourage attendees to seek outside
funding.  I believe we can do this by explicitly asking some questions
about employment.  Note: Employment is not a factor in funding
consideration for anyone.  All community members are equal regardless of
employment or lack thereof.

* How does your work in Fedora relate to your work, if you have a
$dayjob?
* Have you spoken with your employer and specifically asked for some
funding to help cover your costs of attending Flock?
* If you are an employee of Red Hat, please provide your Red Hat email
address and your manager's name and email address.  This question is
specific to an employer because of the nature of our funding.  Almost
100% of Flock is paid for by funding provided by Red Hat.  Being able to
accurately demonstrate the costs incurred for Red Hat employees at the
event may allow us to have a stronger case for increased funding for
Flock in future years.

## Geography

Today Flock rotates between NA and EMEA.  For now, I think that works,
however I believe that we should look to allow Flock to float "around
the world."  I think that we should give consideration to the
distribution of our contributors and the location of likely attendees
when thinking about where too hold Flock.  I am not ready to suggest we
commit to having Flock outside of NA and EMEA, but I believe we should
be prepared to think about it when a compelling case is made.

[0]: Having consulted with Fedora Legal recently, it appears that we can
legally maintain a list of people who are not allowed to obtain funding
in advance.  This is going to be explored as I work on the reimbursement
and advance purchase policy.  However, I suggest we use a similar
mechanism here.  If a funded ticket contributor no-shows, they are asked
to pay for their ticket to the next Flock they attend.