Fwd: The success of Grants

FYI from the Squeak list.
Sorry if you've already received this.
I believe this can be applied to the formation of the Consortium...
particularly Brian Behlendorf's point about development 'market' forces.


---------- Forwarded message ----------
From: Ron Teitelbaum <[hidden email]>
Date: Aug 30, 2006 4:02 PM
Subject: The success of Grants
To: The general-purpose Squeak developers list
<[hidden email]>


Hi All,

I received another interesting reply from Brian Behlendorf.  He said I could
share this with you.  In our conversation I mentioned that our non-paid
contributors are far from casual.  He agreed but didn't really know what to
call not paid contributors.  He said hobby, or amateur didn't sounds right.


I really like his points though, and agree that there is an interesting
dynamic between those that Squeak for a living, and those that are
contributing in their spare time.

Over all I like the idea of trying to work out what appears to be a switch
from the professional SqC group to a more Community based Squeak by trying
to encourage more people like Avi to actively support development.  The loss
of the professional paid group needs to be countered with more private
investment.  In return we get a more stable platform from the mile stone
makers and we encourage more innovative contributions from us causal
contributors.

Did anyone take up Avi on his suggestion?  It doesn't look like a difficult
project.

Ron Teitelbaum

From: Brian Behlendorf
Sent: Tuesday, August 29, 2006 11:45 AM

On Mon, 28 Aug 2006, Ton Roosendaal wrote:
> My current conclusion is that - when money gets involved - it's important
to
> participate in an existing professional environment, or to create one
> yourself (like we did for studio Orange), or to help people to setup a
> business to become 'professional'.
> We didn't give out large grants yet, but if we will do I would look at
> sponsoring companies/organizations to hire Blender developers/artists for
> projects.

That's very similar to the perspective at Apache - if a company wanted to
make a "donation" to us so as to further the work in some area, we'd say
no, and that they should instead hire their own person with that same
money and have them work on ASF projects as an individual contributor.
I was personally very interested in making sure that there was a healthy
ecosystem of companies providing services and products around httpd,
because I felt that would help ensure we weren't dependent on charity, and
that mutual market dependence would balance interests better than any
rule-setting we could do.  Early on we were very generous with press
releases and other public speaking to highlight companies building
products/services around Apache.

The salaried full-time contributors do have issues, of course - they've
got to justify their participation to their employers on the basis of some
measurable return, and if you can't promise release dates or features,
that's challenging when your employer is only used to those kind of
metrics.  If instead your employer is more tangential - a web site design
company that happens to use Struts, or something - then participation is
easier to defend, but tends to be less than full time.  But I would say
that a lot of the tension in Apache comes from conflict between the
fulltimers who want to hit ship dates and fulfill promises they've made
versus the more casual contributors who care more about quality and
creativity.

I'd actually like to believe that the creative tension between the two
results in something better than when either side dominates.  Classic
problems in software engineering come from projects that are either too
button-down and can't take risk, or the other extreme, big-vision projects
that never reach practical usability.  When Open Source projects have
both, you can have the casual/creatives taking big risks with ideas that
might fail and only waste their own time, and the fulltimer-production
types who make sure the successful new ideas work well and get bugfixed
and matured.  That kind of applied innovation model is really cool.

        Brian