Notes from San Diego mtg

[Bart Decrem <bart eazel com>]

Hello everyone,

Following up on Havoc's posting, I wanted to send out another copy of my
notes from the San Diego people, so the new people on the list will be
up to speed.  I have edited these notes to preserve the anonimity of a
corporate partner who hasn't announced their support of Gnome yet.

At this point, we have decided to retain Collab.net to facilitate this
process.  So expect to see quite a few posts from rob@collab.net.  Brian
Behlendorff (apache) and Frank Hecker (mozilla) both work at Collab.net
and will also be assisting us.

Bart
----


Hi everyone,

Thanks for joining us at the meeting in San Diego.  Again, thanks in
particular to the folks who flew to San Diego just for this meeting
(Brian, Jonathan, Maciej, Bud and John)!

Here are some draft notes.  Please send me suggested edits.  The next
steps are:
1.  Edit these notes
2.  Inform Gnome Steering Committee
3.  Decide whether to retain Collab.net

Let's try to get these steps done this week.

Bart

1.  Background
----------
Until March, the governance structure of Gnome was very informal.
Miguel provided overall direction; the release coordinators and
maintainers of individual packages worked together to resolve conflicts
and ship releases.  In response to the growth of Gnome, the emergence of
a Gnome industry, and feedback from third-party software developers, a
group of Gnome hackers sat down at the Guadec conference and created a
Gnome Steering Committee, with representatives of the main components of
Gnome.  The Steering Committee was created to oversee the next 3
releases of Gnome (April, August and November Gnome - aka Gnome 1.2, 1.4
and 2.0).

The Steering Committee asked Jonathan, Elliott and myself (Bart) to
create a Gnome foundation.  Meanwhile, a group of people from a major
industry player had
approached Miguel about embracing Gnome.  These people
started talking to Miguel about creating a Gnome foundation or
industry consortium.  Meanwhile, a group of people from Eazel had a
meeting with Brian Behlendorff from Collab.net, where he shared some of
his experience about running major open source projects.

In response to all these developments, we brought together a group of
people during Usenix to talk about Gnome governance.  We invited the
Gnome steering committee and other people who were either major
stakeholders in Gnome or had an interesting perspective to contribute.
Here's who was at the meeting:
- Miguel De Icaza, Gnome, Helix Code
- Jonathan Blandford, Gnome, Red Hat
- Ettore Perazzoli, Gnome, Helix Code
- George Lebl, Gnome, Eazel
- Maciej Stachowiak, Gnome, Eazel
- Bud Tribble, Eazel
- Bart Decrem, Eazel
- Brian Behlendorff, Apache, Collab.net
- Jim Gettys, X Consortium, W3C, Compaq (not representing Compaq)
- John, representing a corporate partner
- John Gilmore, Cygnus, Electronic Frontier Foundation

.2.  The meeting
------------
The purpose of the meeting was to have a discussion about Governance of
the Gnome project.  We talked about how to structure Gnome so it could
continue to be successful, embrace a growing number of corporate
partners, and maintain the energy and passion that the hackers bring to
the project.  We also talked about the process of setting up a Gnome
foundation: who should be involved and timeline.

3.  What we came up with
-------------------
Everyone in the room agreed that it is important to create a more formal
structure to support the Gnome project.

We came up with the idea of creating a Gnome Foundation that would have
4 main goals:
1.  Providing a forum to determine the overall technical direction of
Gnome
2.  Promoting Gnome
3.  Foster collaboration and communication among Gnome developers
4.  Manage funds donated to the Gnome project

The most important function of the Gnome Foundation is that it would be
a forum for discussions about the future of Gnome and a legal space
where companies and hackers could talk to each other without the risk of
violating anti-trust laws.  The foundation would also be an entity that
could speak for the Gnome project, and it would provide stability to
Gnome.

The Gnome Foundation would have a Board of Directors that runs the
foundation.  The Board of Directors would replace the current Gnome
Steering Committee.  The Board of Directors would oversee the release
schedule and roadmap of Gnome, but would have no control over individual
developers or the packages they maintain.

We also talked about the planning process for creating the Gnome
Foundation and we felt:
1.  That the Gnome Steering Committee needs to be actively involved in
this process
2.  That the broader Gnome community (ie. all the Gnome hackers) should
have an opportunity to participate in this process
3.  That it would be good if Collab.net faciliated the project

We propose the following timeline:
1.  In July, online discussions about the structure of the Foundation
2.  Before LWE in August, post a specific proposal for the Gnome
Foundation on Gnome-Hackers for discussion
3.  At LWE (August 14-17) have a meeting with a larger group of people
to create the Gnome Foundation and put together its Board of Directors


4.  Collab.net's proposed involvement
---------------------------
Brian suggested that Collab.net be retained to facilitate this planning
process, and Collab.net put together a proposal for $25,000 to do this
work. The advantages of using Collab.net, instead of doing it ourselves
are:
1- Collab.net has a ton of experience in organizing major open source
projects that include both open source hackers and major industry
partners (including Apache and Mozilla)
2- They will be able to dedicate more resources to this than a group of
volunteers can
3- Having a neutral third party facilitate this avoids conflicts of
interests and will make it easier to get feedback from other groups such
as ISVs

The idea would be that Eazel, Helix Code and Red Hat would split
the tab for the consulting engagement, but Collab.net's 'customer' would
be the entire Gnome community.

5.  Things we propose NOT to do
------------------------
- The Gnome Foundation would be the main forum for discussions about
Gnome.  There would be no need for an industry consortium in the short
term.
- We talked about the pittfalls of having a foundation that would be
dependent on a small number of funders and would require significant
ongoing funding.  It would make the foundation beholden to these
coporate interests.  So it's not a good idea to have a foundation with a
bunch of staff, and funded by just 4 companies.

5.  Things that need to be resolved
-------------------------
There's a ton of things to figure out before we create the foundation,
including:
- composition of the board of directors: who's on it, terms, elections
or appointments, regional representation, balancing hackers
participation and people representation corporate interests...
- detailed charter of the foundation
- budget and/or staffing
- relationship with regional organizations (Gnome Foundation Europe...)
- incorporating the foundation, tax-exempt status
- involve major corporate partners in Gnome.

6.  Functions of Gnome
------------------
Here is a list of all the functions that need to be handled by the Gnome
community generally.  Many of those are best handled by individual
contributors or by corporate partners, but some of these are best
managed by a Gnome Foundation:

1- Development, including documentation
2- Technical direction
   - boundaries of Gnome (what is Gnome?)
   - APIs/standards/protocols
   - Roadmap
   - Release coordination
   - Reference distribution/QA
3- Marketing
   - Web site
   - Press relations
   - Evangelism
   - Trade shows
   - Policy/legislation
4- Collaboration & communication
   - Developer site, IRC, mailing lists
   - Conflict resolution
   - Developer conferences

The foundation would play a key role in 2 (Technical Direction) and
would also oversee 3 and 4, although most or all of the actual
implementation of the marketing and communication may be done by Gnome
contributors or corporate partners.

7.  Next steps
-----------
1.  Edit these notes
2.  Inform Gnome Steering Committee.  We need to make sure that the
Steering Committee will maintain the confidentiality around the Sun
announcement.
3.  Decide whether to retain Collab.net