[OSGeo-Discuss] scale of FOSS projects

[Frank Warmerdam]

Jo,

I'm having trouble responding to your email, I think since it touches on
a number of points, and perhaps just because I mostly agree with what you
have said.  So instead, I will just assert a few loosely related points that
come to mind after reading it.

1) I still fundamentally believe a bunch of enthusiastic and reasonably
skilled people can build a project with impact without the explicit backing
of one promoting enterprise.

2) For projects coming out of a "single backer" situation into OSGeo we
offer a level playing field to help turn the project into a fair community
where all contributors have some assurance of having an influence.

3) For projects coming out of a more chaotic origin - many contributors, or
at least no major enterprise associated with backing the project - we offer
some degree of "organizational legitimacy" that can be helpful in selling their
project to risk averse enterprise type users.

4) While this one of the things I like about geospatial open source software
is the participation of some folks doing it more for fun than profit, we
are still *mostly* an industrial software sector.  We make software used for
all sorts of gritty business / commerce / government / science as a sort of
"industrial IT input" to other things.  For this reason, I feel it is
inevitable that a substantial part of what we do will be about serving
various industrial needs.  This implies our primary users will be commercial,
government and academic/research - fields dominated by organizations of
various sizes that can be considered enterprises.

5) I absolutely do *not* think entrance into incubation for a project should
be based on having a substantial enterprise backing the project.  However,
to avoid being swamped in small immature projects, I think it is reasonable
to hold out for projects that are already reasonable mature, have a substantial
supporting community and are of a quality and utility that we think will
reflect well on OSGeo when we promote it. I would *prefer* a project coming
into incubation with six developers from six different organizations to one
with six developers all from one organization.

6) As Cameron mentions, consolidation is to some extent to be expected in this
and all software sectors.  I think that's ok and natural.  We have quite a few
desktop GIS software packages now for instance, and one imagines that while
some will grow stronger and grow, others will wither.

7)  On the other hand, I think there are other sectors where a small
projects can still fill a particular need without being big, heavily backed,
etc.  Utility programs, web mashups, mobile location aware applets, etc.
It behooves OSGeo to understand that these things play a role even if they
don't need our process-heavy project steering committees, incubation, etc.
Lets not hesitate to celebrate, and promote them as appropriate.

Ultimately, I'm left feeling that there is no explicit action item here.
The universe will continue to unfold, projects will bloom and die,
consolidation and ferment will both happen.  We don't need to predict it
all, or guide it.  We just help where we can, provide services where it
makes sense, and watch it unfold.  But then, I'm not really a very good
"big picture" kind of guy. A little too laid back in some ways. :-)

Best regards,
-- 
---------------------------------------+--------------------------------------
I set the clouds in motion - turn up   | Frank Warmerdam, warmerdam at pobox.com
light and sound - activate the windows | http://pobox.com/~warmerdam
and watch the world go round - Rush    | President OSGeo, http://osgeo.org