incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Dirk-Willem van Gulik <di...@webweaving.org>
Subject Commerce and open-soure (Was) Proposal to join Apache OpenOffice
Date Mon, 06 Jun 2011 17:06:24 GMT
Having finally caught up with most of the discussion so far - I am wondering if there is a
fundamental disconnect between how the various communities model commercial interests and
open source.

Perhaps it is fair to surmise that Apache rules of engagement matured during the start of
the dot-com boom. And where virtually each and every person involved was there for what I
would almost call very 'selfish' reasons[0]. Fierce competitors (on content) easily conceded
to collaboration (on technology and open standards). Driven by clear tradeoffs around time-to-market,
interoperability or cost. Driven by clear 'win's for the contributors (and not necessarily
considering a win-win at both contributor and nascent ASF) as the amplifying force. 

Now above is probably too stark of a caricature - and a lot of people where motivated by a
lot more (cool technology, the joy of collaborating and learning, access to smarter people,
the challenge) and generally well attuned to the 'lets make the world a slightly better' place
of internet engineering dominant in that period.

However I'd still argue that the fact that a lot of people then (and now) were driven by powerful
commercial forces became part of the ASF its fabric. And then since then - the  apache community
has learned how to work with that.

You may have noticed that above mentions 'people' far more often than companies. 

And that is part of that lesson - Apache tends to work with individuals - who get their 'commit
bit' based on merit, based on the opinion of their peers and their visible contributes. As
opposed to corporate access or dealing with companies[1].  We trust people.

And unless they specifically state otherwise (and this is rare!), when an Apache member or
committer posts on any mailing list - they do it as themselves. It is their personal point-of-view,
wearing their personal hat and not as a mouthpiece for whatever company happens to be signing
their paying their salary at that point. Likewise - VP's and ASF directors very rarely use
their 'hat' - and if they do so - will identify themselves clearly. . And we do see a lot
of ASF committers move from company to company - over periods of decades even - loyal to the
codebase and apache[2] . Some have even managed to make a full circle.

But none of this makes the ASF a counter balance or a shield for- or from- corporate interests.
It just makes it a place where individuals can safely contribute to code, release that code,
get the benefit of proven processes and know that they shielded form the usual liabilities.
And it makes it a place where anyone, individuals and companies alike - can pick up release
- and where they know that their exposure is as it says of the tin.

So this is somewhat in contrast with other possible community structures. Where the collaboration
structure _itself_ is there to protect, to shape; or where the contributors and interest sitting
at the coding table are companies, rather than people. And where the collaboration structure
needs to be strong enough to keep this in check.  Or where strong licenses, like the GPL,
are needed to keep certain undesired commercial land grabs at bay.

The ASF its structure, culture and bylaws are simply not conductive to the latter. All it
is, can do, is considering to accept a donation (software grant[3]) under very specific terms
and then allow a self managing[2] group of individuals who are peers, work on that code within
a fairly narrow set of processes[4] following a defined path[5].  And the ASF will only do
this when that group of individuals is there. People. Willing to do work.  Only during that
first bootstrapping phase is there some help[6] - but beyond that - projects are self manage,
self select their PMCs, self propose individuals for commit access and so on.

And I think that this difference in expectation is at the heart of some of the current debates.


I'd personally expect that the Open Office world - which its sizeable impact on a very commercial
enterprise world with expensive demands will need to garner a solid and balanced support ecosystem
which is far beyond the ASF - where the free and strong ideological chops of, say, LibreOffice
balance commercial product and support companies.

Hope this helps,

Dw.
-- 
Dirk-Willem van Gulik <dirkx(at)webweaving(punto)org>

0: I'll be the first to admit that - though arguably in my case it was research money and
getting satellite pictures distributed by other means than faxed request forms and large boxes
of tape.
1: In all fairness - we do have Company Contributor License Agreements - partly as to make
things easier on the process side for individuals which work in (large) companies. Companies
do not contribute code - people do.
2: http://www.apache.org/foundation/how-it-works.html
3: http://www.apache.org/licenses/software-grant.txt
4: http://incubator.apache.org/guides/graduation.html#releases
5: http://incubator.apache.org/incubation/Process_Description.html
6: http://incubator.apache.org/incubation/Roles_and_Responsibilities.html


Mime
View raw message