Re: [Debconf-team] Fixing the Debconf Delagation

[Tassia Camoes Araujo <tassia@acaia.ca>]

Hi Bernelle and team,

On Wed, Sep 30, 2015 at 12:21:24PM +0200, Bernelle Verster wrote:
> Hi
> 
> TLDR:
> * "You can set direction in Debian not by telling people what to do
> but by changing what is easy in the project"
> * "If there is one thing, it's getting people to talk to each other
> and find out which parts of *their* work they can think about
> differently in order to have it be easier. We can do this
> peer-to-peer. How can we make it more cooperation and less
> collaboration." We all need to think about making our tasks work
> better, not just the chairs. We need to be devil's advocate.
> * We need to build trust, I give some ideas on how.
> 
> B
> ___________________________________________________________________________________
> 
> Preamble:
> * I signed the 'on the local team' email because I wanted to let it be
> known that I consider the way things are currently working to be
> broken. The details are less of a concern to me. I think it is
> important to see how many people felt this way; it was not a single
> person being rebellious.
> * Theorising that it could work if people only give it a try is not
> good enough. Let's agree that things are not working as they are, and
> find ways to improve them.
> * I am taking my cue from Asheesh's talk [1], "You can set direction
> in Debian not by telling people what to do but by changing what is
> easy in the project" (20:35)
> 
> What is happening here has happened in previous events I was involved
> with too (One person made unrealistic promises to sponsors. We made it
> happen and it lead to an overall better event, but it damn near killed
> us). This is not unique to DebConf, and it is not unique to one person
> being cantankerous or wayward. So don't bash yourself up too much. I
> am writing this long email with the DebConf16 hat firmly on, but I
> hope it can contribute to all DebConf functioning.
> 
> From Tincho "The power to override decisions ... budgetary decisions,
> sponsors relationships, volunteers motivation, and even some basic
> tenets of the conference were at stake. Every time this happened, it
> was because somebody from the
> local team acted on their own, overstepping other people's
> responsibilities, and without prior consultation."
> I would argue that it may have been because the understanding of what
> was expected was lacking. If there is no structure that makes it
> clear, I would also tend to act in what I believe is the best way. It
> is hugely damaging to then have my best intentions questioned.
>
I don't agree that the problem was a lack of structure or understanding
of the issues and expectations. We are talking about a repetitive
behaviour of people failing to admit their errors and blaming the
structure when there are other underlying drivers of conflict.
 
> So going forward, we need to make the values more explicit for each of
> these things, and I use the examples Tincho listed as places where we
> can start - the 'easy' changes. (thereby my answer to Tincho's "I
> would like to ask fil or any of the people who think the chairs should
> not exist what would they do in situations like these.") I've
> rephrased these to be more constructive.
> 
> * The tension between short-term benefit for sponsors (e.g. being
> acknowledged on the website, authorising in-kind donations) vs
> long-term relationships with existing sponsors:
> 
> The issue: There is not enough time for the whole team to find
> consensus on a service that was not considered and therefore not
> discussed before the conference. This happens. There was no way to
> find input quickly, so he made the call.
>
Time cannot be an excuse over and over again for people making arbitrary
decisions. Teams should backup decisions, not individuals. It is not person
x or Y conference, it is the Debian Conference, and it needs to reflect the
project values, not X or Y's. In that particular case, the most problematic
in my opinion was not the decision per se, but the lack of accountability
and lack of a sense of respect for other people's work, and finally, there
is no recognition of error by the ofender, which makes us think that this
could happen again.

> How to fix this: To mitigate the tension, I would suggest that there
> is a clear understanding of what constitutes a sponsor, well in
> advance, with a bigger focus on the very small sponsors. I think a
> bigger issue that needs to be articulated, is how do we acknowledge
> the many, many material things that make a conference happen? I see
> this as a critical issue for the DebConf in Cape Town, because we may
> have many small companies who want to contribute and be part of
> something, but who fall far short of the official perks. We need to
> bring them in somehow - this is our community and directly part of the
> DebConf Goals.
>
I agree that local small companies should have a way to participate. If the
current arrangement do not fit them, maybe this should be an agenda point
for the meeting on Saturday?

> I can remember the sponsor being quite sad at how
> things went - we need to think about how to prevent that. (I can't
> find that thread now though).
> Then, I would think once there is a decision to be made in crunch
> time, to revert to a/the Chair to make the call. For that to work
> though, I would need to trust that the Chair would make a good
> decision - one that if there was time to bring in the whole team, they
> would agree with and support. I cover trust as an overarching thing
> right at the bottom of this email.
> 
For that particular case that was exactly what we did. We didn't override
any decision as Chairs (it was actually too late), but we tried to clarify
things to avoid similar errors in the future. The fundraising team had the
final words and I think there was a common agreement that the situation
was problematic.
 
> * The tension on what to spend money on, and the unclear Debconf
> goals. Taking sponsored food and accommodation vs 'more fancy perks'
> as example:
> 
> For a start (and wow, so many seemingly simple questions asked of me
> at DebConf15 make sense now!), looking at the budget and seeming
> conflict of interest between the conference dinner and sponsoring more
> people to come, the problem here is that it is unclear what the goal
> of DebConf is [2]. At the orga sprint this unclearness was
> articulated, and half of the people interpreted that the goal is to
> bring more people to Debian (aka sponsor more people to come) and the
> other half interpreted it to mean that the conference should be a
> place for Debian Developers (and those aspiring to become DD's) to
> work and be socially pampered (as the name of DebConf suggests: the
> Debian Developers conference). The former gives emphasis to the
> "sponsored food and accommodation" and the latter on "more fancy
> perks".
> 
> How to fix it: These goals need to be clearer. Is the primary goal to
> enable face-to-face interactions between Debian Developers? Or who?
> Who exactly are the proposed attendees? Let's decide on that and then
> work towards that.
> My personal feel for Debconf16: If DebConf is primarily for DD's, then
> DebConf should always be in Europe, or where most DD's are and that
> sponsoring people should be more vigorously controlled to (only) allow
> DD's and people applying for DD's to attend. But the next DebConf is
> in Africa, unless you want it back, so I think for us it would be
> reasonable that the budget would reflect bringing more awareness to
> Debian. From my side the conference dinner, or more fancy perks for
> example, is not part of that. This may change for other DebConfs and I
> suggest there is a set of questions making this explicit, that bid
> teams can answer and think about. We were asked these questions in
> vague terms only at DebConf15, which is too late.
> 
Yes we can talk more about goals, values and priorities, but I'say that
establishing a single main priority is hard, simply because our values
are not the same. Which is fine, better if we acknowledge it than try to
avoid the subject. We are all human beings with different and diverse
backgrounds, and DebConf should be able to accomodate that. So instead
of looking for a primary goal, let's envision a plurality of views?
If there is mutual respect, we should be able to work together.

> * Volunteers participation and motivation was at risk, example used:
> coordinating childcare
> 
> This seemed to be a communication issue, and personpower issue (I read
> all those childcare emails as they happened). The way I am trying to
> fix this is by making people more cooperative (see Asheesh's talk;
> collaborative needs a lot of communication, cooperative can work more
> independently) and clear on communications. These are currently coming
> across as exhausting and laborious, when they should be energising, so
> they are not perfect and I need help here. The way I am doing this is
> by the weekly meetings as a place to bond, the meeting template that
> should make things easy to read and hence easy to stay informed, as
> example.
> 
> As a meta-point, again from Asheesh's talk (Questions time): "If there
> is one thing, it's getting people to talk to each other and find out
> which parts of *their* work they can think about differently in order
> to have it be easier. We can do this peer-to-peer. How can we make it
> more cooperation and less collaboration." We all need to think about
> making our tasks work better, not just the chairs. We need to be
> devil's advocate. Madduck, looking at the points Tincho raised, how
> could you have done it better? What structures could have been in
> place that would have prevented you from taking action without
> consultation, that you would have agreed with (and not try to
> side-step)?
> Maybe this was the point of the orga sprint at DebConf, but it failed.
> I suggest we try one-on-one with people who are either not involved
> with DebConf at all, or who are not involved with our tasks.
> 
> * Setting up events external to DebConf, that is still associated with
> DebConf by design or otherwise, with tension of affecting the DebConf
> 'image'. Example: madduck's workship, and for DC16: ShowMeBox.
> 
> How to fix this: Tell the greater team about it, well in advance, and
> get input and then go-ahead, well in advance. Find a  constructive way
> to answer "Is lack of response a yes or no?" We need to understand as
> a team that there are other things happening, linking to, but
> independent of DebConf. This is unavoidable and a good thing. We need
> to find a way to make this work better (a 'wider-initiative' wiki page
> with tips and e.g. Disclaimer requirements?)
> The issue: Doing other things, e.g. outreach activities takes a lot of
> work, but they add to the bigger picture and should be encouraged.
> These things may be intimately linked to the personal 'agendas' of the
> people leading them - my main reason for being involved in Debconf is
> to learn about open hardware, for example. These things start very
> long in advance of DebConf, and the team members don't see it as
> urgent, so don't respond (and then later said they were never
> consulted at all). We need to work harder to think ahead. For
> DebConf16, a sub-group is planning to do a ShowMeBox competition. We
> need to start now to do it well, and by the time the greater team
> decides no (e.g. two weeks before DebConf) it's too late.
> I drafted this in response to some questions when I sent the ShowMeBox
> notification:
> 1. Why am I sending this if it's not really orga related and there's
> more urgent stuff to do?
> Because I want you to be aware of it and on board, so if you hear
> about it later you don't feel this was done behind your back.
> 2. Why am I sending it so early and asking for commitment by mid October?
> There's a lot of work required and I don't want people to get
> emotionally and otherwise invested in the project just for it to get
> trashed. We need to engage with schools before the end of the year, so
> mid October is the latest we can pick this up.
> 
> * Who decides how to spend the budget? (After budget approval)
> 
> Another thing that was not mentioned by Tincho, but may underlie the
> 'more fancy perks' discussion is who gets to decide on how budget gets
> spent, when (additional) money is sourced by people in the (e.g.)
> local organisation? If the budget is approved with a certain amount
> committed, and more than that amount gets raised, you need a clear set
> of guidelines (not a policy, but a set of questions to guide
> conversation between the team and chairs and DPL) to get everyone on
> the same page. Said another way, are the people who source extra funds
> in charge of spending it? What about the people who are not involved
> in fundraising but had ideas or feelings about it? Instead of saying
> this group or that group can decide where this money goes, what are
> the priorities that would decide where the money goes? What set of
> questions can make the thinking behind the decisions explicit?
> 
> In closing, people like madduck are enthusiastic and energised, they
> make things happen and we should encourage that. That they make
> decisions 'too quickly' or 'not consultative' is a symptom of the
> organisation needing to give better guidance, and the Chairs are not a
> catch all answer to this (perhaps rather addressing the symptom, not
> the cause). Their behaviour is not the core problem.
> 
> The trust does not currently exist, and I can be the first to admit
> that I'd rather get something through under the radar than submit it
> to bikeshedding and flamewars (following the 'it's better to apologize
> than ask permission' tenet). This is not healthy. I try to be devil's
> advocate to myself here by including it in the meeting agendas and
> emails *long* before it is required...
> 
> So how does one build trust?
> 
> It takes a long time, so I would argue that the chairs need to be
> involved a lot, yes. But one does not build trust by (just) doing
> stuff, but by capturing the gist of things. So I agree with Lucas that
> "they shouldn't become team leads, or be in a position to make
> important decisions" (and I'll add "when there is time for the team to
> make those decisions on their own" Emergencies may be different) The
> chairs should be able to articulate the values of DebConf in their
> every thought and action. As Gunnar said "During my time as a Chair,
> my *main* Debian task was to be as close as possible to DebConf
> organization, to be able to answer (or to point to right answers) to
> questions about DebConf, be they originated in outsiders, in Debian
> people, or in the orga-team(s?) themselves. "
> 
> In short, I see the chairs as the grandmothers of DebConf. They should
> know all about DebConf so that in a moment of crisis they can act in
> full confidence that they are doing what the rest of the team would
> agree with. The whole thing about grandmothers is that they've been
> around for at least two generations without getting expelled from the
> family.
>
> I find the power to override decisions damaging to trust. If you need
> this power, you are not doing your job as chair well. Your power to
> override should be the social pressure you can exert because of the
> respect afforded you. I can see how this power needs to be written
> into some agreement for real emergencies, but I am really worried that
> it is a sword hanging over our heads because the respect is not there.
> And respect is earned.
>
We don't claim the power of overriding decisions to the Chairs. But history
shows that without *any sort* of regulation, we risk having individuals (not
only local team) taking decisions regardless of the interest of the group.

If decisions are made in consonance with the group, no much intervention is
needed. The dicotomy Chairs versus the team is not honest, since we are usually
voicing other peoples concerns in our interventions. Our main goal overall is
to promote common values/interests instead of individuals'. What we saw in the
recent past was a repetitive damaging behaviour that compromises the group
coesion, and we are strugling to find a solution for it to not happen again.

The other point I'd like to raise is that as a Chair I try to voice the
non-dominant views, sometimes representing those who are not in conditions to
engage in a discussion, but can enrich our DebConf experience. For instance,
there are people who cannot be present in every meeting but still want to
contribute, or those who first language is not English and have trouble to
follow "super efficient" meetings where whoever has the strongest voice
dominates, or those who where hurt or got tired of damaging relations and
decided to focus on specific tasks to avoid more damage, but they still have
a lot to contribute.

> I suggested over a beer to someone during DebConf15 that we do
> expressive change exercises with the team, I don't know how people
> would feel about this as a way to get peer-to-peer input and gently
> rebuild the team? [3]
>
I liked what I read about expressive change and I'd be happy to contribute to
such a process, if people are open to it, but *right now I need a break*.

This whole discussion pushed things to far for me and I also have my doubts if
I'm in the right place. A DebConf Chair is also a human being and subject to the
same damage that we've been talking about. So maybe you'll need to find someone
else to "play the grandmother"...

Tassia.