[IAEP] Release cycle - Goals

[David Farning]

Over the last weeks I have been looking at how we can improve our
release cycle. Today will be about defining and implementing goals.

Setting goals for any software project is hard, much less an open source
project.

Marco started the discussion last week at
http://lists.laptop.org/pipermail/sugar/2008-August/007838.html .  He
has also started a wiki page at
http://sugarlabs.org/go/ReleaseTeam/Roadmap/0.84 .

There seem to be as many methods for setting goals as there are
projects.  There is not a best practice.

Some of the variables are:
1.  Size of the project.  Larger project such as eclipse have developed
very formal methods.  Smaller project tend to be more informal.

2. Collaboration vs. Competition.  Projects differ in the attitudes with
which stakeholders regard each other.

3. Commercial vs. Commons.  Project differ with regards to their
interest in commercializing their product. 

4.  Dictator vs. Membership.  Some project such as the kernel do well
with the BD others such as Debian are entirely membership driven.

5. Planned vs. Evolutionary.

1.  Sugar Labs in in the category of small to medium sized.  We have
less than 10 committers responsible for 90% on the commits and less then
50 doing the majority of the majority of the commits.  This size implies
a less formal method of setting goals.  It is still quite easy for all
developers to coordinate on a single ML and use wiki pages to specify
new features.

2.  All of the current stake holds all share the same basic vision for
Sugar.  We should be able to decide on goals and priorities of goals
based on rough consensus.  If an individual stakeholder has an interest
in add a feature, it is up to them to implement and prove the value of a
given feature. 

3.  We are a hybrid of commercial and commons.  Our base product, the
Sugar desktop, is focused on the common good.  Yet, it will be to our
advantage to encourage organizations to advance Sugar as a platform on
which to base their own businesses.

4.  Base on some of the recent hits Sugar has taken in the media,
public trust for sugar is quite low.  Returning to a dictator,
regardless of his competence or benevolence is questionable.

5.  For a totally cop-out answer I will fall back on planned evolution.

Proposal:

1. Continue the discussion that Marco started on sugar at laptop.org .
2. Submit suggestions for future development via
http://sugarlabs.org/go/ReleaseTeam/Roadmap/0.84 .
3. Appoint Marco and Greg Smith to assign priorities to the list of
suggestions.
4.  Individuals volunteer to work on features.

thanks
dfarning