[Sugar-devel] Have we achieved consensus among activite Sugar developers?

[David Farning]

On Wed, Oct 30, 2013 at 4:59 PM, James Cameron <quozl at laptop.org> wrote:
> On Wed, Oct 30, 2013 at 10:28:35AM -0500, David Farning wrote:
>> Have we achieved general consensus that the three phase approach I
>> proposed earlier this week has the potential for establishing a
>> mutually beneficial relationship while progressively rebuilding trust
>> on both sides?
>
> I got lost in the discussion again; I couldn't see how your three
> phase approach answered Walter's question about your perception that
> Sugar Labs is not acting transparently.  ;-)

The big idea is that open sources projects thrive when they create
conditions and cultures where people with overlapping yet
non-identical goals can come together collaborate around a common
goal.

Sugar Labs has found itself in a position where there is a high degree
of conformity. This tends to create an echo chamber where similar
opinions are respected and encouraged. That can be effective at
building passion and energy, but it tends to crowd out dissenting
opinions and marginalize the people who hold those opinions. These
people can be the most productive members of the community in their
particular areas of interest.

The transparency challenge is that many potentially valuable members
leave in frustration when their voices are not heard. Conversations
escalate from civil to uncivil. This reduces the rate of development,
quality of support, and potentially the future viability Sugar.

Attempting to prove that via examples would create personal feuds
which are unproductive at all levels. Instead, I would ask you to talk
to people in the ecosystem, outside of the current core sugar
developers, and gather feedback about what they think works and
doesn't work.

Instead, I would like the opportunity to prove the premise by showing
the theory in action. My assumption is that if that we can work
together on a series of tasks which require increasing amounts of
acceptance for divergent opinions, we can identify and reduce the
sources of the underlying tension.

1. Phase one requires that we work together on a relatively straight
forward project. HTML5+JS is the current focus of Sugar Labs. While it
is not AC's primary focus, we consider it a key strategic project.

2. Phase two will be a bit more complicated as we ask various
developer to publicly agree on various core priorities for the next
release. This related directly to manq's post about being focused on
individual priorities. Without an understanding of everyone's
priorities and the value they bring to the project, it can be easy to
feel ignored, or even attacked, when one's own priorities are ignored.

3. Phase three -- Dig into the balance between stable and leading
edge. Historically, this has been a touchy subject because of the high
degree of interest in innovation by key Sugar Labs members. However,
large deployments consider stability and LTS very important.

My assumption is that if Sugar Labs and Activity Central can set an
example for working together, other marginalized parties with rejoin
the project.

David


> Regarding your need to rebuild trust on both sides; perhaps a
> quantitative approach; you could list the areas and extents in which
> Sugar Labs trusts Activity Central and Activity Central trusts Sugar
> Labs now.  e.g. feature discussion, design review, patch review, go
> no-go release decisions, support for released code.  Gain general
> agreement.  Then do a diff against past and future.  But this begins
> to sound like a developers' social contract, and not specific to
> Activity Central.
>
> My gut feel is that Sugar Labs treats all technical contributions
> fairly, regardless of funding source, and that promising funding gains
> no advantage except better phrasing of the responses; 'cause the
> funding bias is better understood to be present.
>
> However, looking carefully at your three phase approach on 29th
> October:
>
> 1.  you are funding work;
>
> fine by me, thanks, expect some responses to these developers to be
> coloured by the awareness of funding,
>
> 2.  you want more discussion about features and whether features as
> built are ready for release;
>
> fine by me, this is no material change to current process,
>
> 3.  you speculate that there is a conflict between supporting existing
> deployments and developing the next releases;
>
> this doesn't fit with me, the two workloads are very different vectors
> in the phase space of possible work, and Sugar Labs primarily operates
> on only one of the vectors, solving support problems in the next
> release.
>
> --
>
> Disclosure statement: the author provides consulting to OLPCA, and
> OLPCA does benefit from Sugar Labs releases.  The author receives no
> direct funding from Sugar Labs or any deployment.
>
> --
> James Cameron
> http://quozl.linux.org.au/



-- 
David Farning
Activity Central: http://www.activitycentral.com