On 7/18/06, Oliver Zeigermann <[EMAIL PROTECTED]> wrote:
Hi, folks!
I currently plan an article on Open Source Software development. What
I would like from you are voices about my central question:
Why does the loose development process of OS work?
Or more in detail: Why are almost all commercial projects organized as
a strict hierarchy while most OS projects work with a loose
organisation or even none at all. Additionally, roles are always
switched on demand. This is something that would never be done in a
classically organized project.
What do you think? Opinions? Thoughts? Am I even on the wrong track?
(warning: generalizations and possibly under cooked ideas ahead)
I don't think OS development works any more than traditional closed
source development. There are as many more failed (stalled) projects
on SourceForge than successful ones. In both worlds failures rarely
make the news (because they are so common) and proponents tend to have
a selective memory about failures that don't further their motives.
I think project success is based on some fuzzy formula consisting of
passion, persistence, work and market demand. The difference between
open and closed source development is how those elements exist. In the
close source projects usually have to create the passion, persistence,
and work elements by using momentary incentives. In open source
development those elements occur naturally somehow.
As for the hierarchical nature of closed source vs organic nature of
open source, I think that open source is more hierarchical than most
would like to admit. It's still less hierarchical than a commercial
environment but it exists. Apache clearly has a structure. With in a
project there is usually one person or a small number of people who
are most passionate about it and lead the project forward. Since open
source usually doesn't have the momentary incentives available to
create passion, persistence, and work out of less involved people the
people leading a project tend to downplay the hierarchical differences
so they don't run the volunteers off.
Where open source is different from closed source is the controlling
entities of closed source derive their control directly or indirectly
from ownership. In the open source, controlling entities got their
either by longevity with the project or by nomination from others on
the project.
All that said, development doesn't have to be either open or closed.
There are many degrees in between of open-ness. JBoss and MySQL are
the first two examples I can think of where the demand for the project
has been enough that they were able to convert that demand into money
like a closed source project and then reapply those funds to maintain
control over the projects.
--
Sandy McArthur
"He who dares not offend cannot be honest."
- Thomas Paine
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
