Re: Benchmark (Phoronix): FreeBSD 9.0-RC2 vs. Oracle Linux 6.1 Server

On Sat, Dec 24, 2011 at 04:04:31PM +0100, O. Hartmann wrote:
> There maybe serious reasons having the Linuxulator, i do not know. But
> if not, why spending rare developer resources on that?

This is a classical misunderstanding of the FreeBSD development model.

There is no "staff" standing around waiting for assignments, as with
a commercial company.  When committers join the project, they usually
(almost always) already have a long list of things that they want to
work on.  And then they go work on them.

Neither the core team, nor the FreeBSD Foundation, "direct" the project
and its course of development.  Some of the members of each do post
emails, or stand up in front of conferences, and say "you know, I think
it would be really neat if someone did xyz."  Sometimes this leads to
results, sometimes not.

As for the companies that have their own FreeBSD-derived products,
often their goals are tightly focused, e.g. "improve the number of
packets we can pass" or "support our specialized hardware".  Some,
but not all, of the resultant work makes it back into FreeBSD.  We
get to say "it would be really neat if ..."; and, in addition, point
to possible future minimization of merging and duplication of effort
as a way to save costs long-term.

But with these exceptions, development is primarily driven from the
bottom-up (individual committers find something they are interested in
working on, and then go work on it), and not the top-down as in "real"
companies.  This is the way the overwhelming majority (90+%?) of the
work on FreeBSD gets done.

So, there's no one "assigned" the tasks of closing PRs, nor working on
coordinating code with the other BSDs, nor working on the Linuxolator,
nor even supporting high-performance computing.

It's a cooperative anarchy, not a hierarchy.

mcl