[Sugar-devel] Sugar upstreaming strategy and Activity Central upstreaming work

[Chris Leonard]

On Fri, Dec 9, 2011 at 2:13 PM, Pablo Flores
<pflores at activitycentral.com> wrote:

> 1. Many community members I talked to agreed that it would be better having
> only one single maintainer, if possible working full time on it. What do you
> think about that?

First thanks to sascha and silbe for the work they do.   From the L10n
perspective, I can say that the more "lang-admins" (committers) that
we have, the better.  I have seen no evidence of sascha and silbe
working at cross-purposes, it seems that they work together quite
well.  Both are diligent and respectful of community feedback and both
are entitled to take vacations (leaving the other to cover the job).
The industry standard scenario for the loss of key personnel is "What
do we do if X gets hit by a bus".  It is good to have "bus" redunancy
and I would simply ask them to avoid crossing the street together :-)

> 2. In AC we're a bit behind in having our developments upstreamed (I asked
> silbe to summarize Dextrose status, as he recently did). So, in addition to
> needing to put an important amount of work on it, we're trying to have
> better working processes for upstreaming:
> ** Sending patches to sugar-devel as early as possible (if incomplete, as
> RFC for a first round of reviews).
> ** Having AC developers dedicating some hours for reviewing and testing
> patches submitted by community developers, participating on the lists and on the Development Team meetings.
> ** Having AC developers trained for collaborating with and contributing to
> upstream themselves as much as possible. I asked silbe to help our
> developers on that.

Close integration and collaboration when working on core components
like Sugar itself (as AC does)  is critical.  Sugar does not follow
the kernel's "benevolent dictatorship by Linus" model and so there is
going to be an inescapable overhead of communication.  On the whole, I
believe this is good for Sugar.  The steps you outline are important
and will help keep everyone "pulling their oars" in the same
direction.

> 3. All this requires additional resources from AC, particularly silbe's
> time, which is not easy to find with all the work he already has. However,
> if we unilaterally take a decision on that (for instance, asking him to
> dedicate xx hours of his time for dextrose upstreaming, or on training our
> developers) we could be harming Sugar maintenance work. So, in my opinion,
> the current scheme for maintaining Sugar may not be sustainable in the mid
> term.

Training replacements or teaching newer contributors how to be
"better" contributors is an unavoidable drain on the best trained
resources in any situation.  In the long run however, if the training
effort is spent on people that continue contributing (as opposed to
taking that training and going elsewhere), there is a net return on
that investment.

Sadly the only alternative is to heap more and more work on single
individuals, leaving you with a serious "bus" exposure problem.  That
is a short term strategy and I believe that all of us here realize
that changing the world through the application of ICT to education is
a long-term gig.

Pablo, I want to thank you for openly communicating about the
challenges you face in you new role as CEO of Activity Central
(congrats again).  From my own personal perspective, I see the
emergence of separate entities like AC that seek to develop
sustainable business models from supporting Sugar as a positive
development.  Sugar Labs is producing a core toolset, but reaching for
a worldwide audience, the tools are designed for local appropriation
and mass customization (not one-size-fits-all), but Sugar Labs cannot
"do it all".  There may occasionally be "perceived"
resource-competition conflicts that come from a particularly talented
contributor being attracted by the opportunity to earn money for their
skills, but I believe that those "conflicts" are illusory and can be
dispelled by clear communication.  I do believe that in the long run,
simply seeing the possibility of turning Sugar development skills into
an income stream (typically after a long unpaid apprenticeship as a
volunteer) will ultimately serve to bring new resources into the
community, as well as preserve the availability of talents of
long-time contributors who might otherwise be forced to seek income in
other positions that would leave them with little time to contribute
to Sugar.

Just my 0.02 soles,

Warmest Regards,

cjl