[IAEP] [Sugar-devel] community influence on development

[David Farning]

On Tue, Jul 28, 2009 at 4:39 AM, Daniel Drake<dsd at laptop.org> wrote:
> quoting Tomeu from another thread (with no bad feelings at all):
>
> "Sugar Labs has currently no resources to focus on anything, it
> depends on volunteers doing whatever they want. I chose to spend my
> time to make easier for more people to bring their knowledge and
> experience to Sugar and the community has no say on this."
>
> Perfectly reasonable answer and this kind of development model works
> well for open source projects, including this one. However, I feel
> like it could be better if the community (who I might even stretch to
> call "customers") could have more influence.
>
> so..to create an open thread:
>
> What are the options for the community having more of an influence here?

It depends on what you mean by influence.  If you mean a producer ->
consumer relationship where Sugar Labs produces a produces a product
and Deployments tell developers what they need?  That is probably
_not_ going to work.  That model has never worked for open source
projects.  Open source does not mean free lunch.  It is a development
process which is particular effective when multiply parties are
willing and able to work together to collaboratively create a product.

What does work _very_ well is for consumers to shift their mindset
from consumer to community participant.  The two most clear cut
examples of this are the kernel and Eclipse.

A really funny example of this happened when Oracle wanted to get into
the Linux business a few years ago.  Oracle sent one of their lead
developer to a major conference (I can't remember which) to present a
laundry list of stuff the the kernel community should do for them.
Their approach was 'We are smarted than you and richer than you, and
more powerful than you.  You should be overjoyed that we sent someone
to tell you how much you suck.'  That turned out not to bet the best
approach.

Over the past couple of years, Oracle has become one of the leading
contributors to the kernel.  Oracle on Linux is now the preferred
platform.

Influence is directly correlated to contribution.

> One would be to somehow get sugarlabs to hire people, and somehow
> process "customer" feedback and assign technical tasks to payroll
> developers. Are there others?

I have attempted to contact several people at OLPC for information
regarding contacts at deployments to set up something like that.  The
responses were either 'Our deployments are none of your business' or
silence.  When organizations like Red Hat, Fedora and Solution Grove
have bent over backward to help Sugar Labs, does it come as a surprise
that more progress has been made on SoaS than projects which are more
interesting to OLPC.

> Having now visited 3 large deployments I feel frustrated that most of
> the features and changes entering sugar are not increasing
> deployability or increasing the educational impact of the platform.
> General technical and usability improvements are always needed (and
> are always of value) but I feel that the balance is wrong and I feel
> that I have not been very successful in getting community members to
> understand the needs of deployments.

If you are interested in 'Community Members' focusing on XO
deployments, I would suggest identifying and engaging participants who
have direct interest in solving those problems to participate directly
in the Sugar development process... Thus becoming 'Community Members.'

FWIW, over the next six months I would like to expand Sugar Lab's to
focus to supporting and working with deployments. But, I will continue
to encourage the two principles of:
1. Implementation over theory.
2. Contribution over consumption.

david

> Daniel
> _______________________________________________
> Sugar-devel mailing list
> Sugar-devel at lists.sugarlabs.org
> http://lists.sugarlabs.org/listinfo/sugar-devel
>