The origins of this decision lie back in 2008, when, as a college "senior," I
became terminally frustrated with the GNU/Linux community for some of the same
reasons.  The people- not the platform- had started to change.  The flame wars
and the asociality had always been there, but recently they had changed.  The
subject matter of flaming shifted (far more often than it previously had) from
technical challenges to people's parentage, upbringing, and choice of
wardrobe.  The asociality become a deliberate fa?ade of counter-culture, with
all of the genuine nerd-dom of a typical suited and well-pomaded Microsoft
sales rep.  In short, GNU/Linux had finally become popular.  It was the dog
that finally caught the proverbial car.  And in so doing, it seemed well on
the way to losing its soul.

That was in 2008, when I edged myself out of the GNU/Linux community, and
began exploring other FOSS platforms.  That was when I fell in love with
FreeBSD, and it was a good year to do so.  The ULE scheduler became the
default, drastically improving multiprocessor performance.  ZFS had just
recently been ported, and was surging in popularity.  Block-based journalling
had just been added to UFS.  SCTP (really exciting) was just coming into its
own, and FreeBSD was the reference implementation.  It short, it was a very
exciting time to be a geek in the FreeBSD community.  Innovation and
improvement was happening left and right, and yet the FreeBSD betas were often
more stable than the GNU/Linux releases (well- distro releases, at least).

Today, I finally came to accept the fact that the FreeBSD community has
changed- for the worse, in my estimation- as well.  When I read the forums and
bug reports, I see much less talk of deep-dive technical improvements and new
features, and much more diatribe about marketing, corporate appeal, and
standardization.  The FreeBSD community seems to have lost all passion for
innovation, unless there's a blue-chip company and a six-figure bounty behind
it.  We won't even go into what's happening to the "tier 2" architectures
(hint: it ain't pretty, and good luck if you try them).  While it is not a
surprise that Mellanox has no interest in ppc64, it is dishearting that the
FreeBSD community doesn't seem to give a damn about breaking it.  While it
is not a surprise that Shawn Webb's attempts to integrate industry-standard
low-level exploit mitigations based on the time-tested PaX tools met
resistance, it is disheartening that he basically had to fork FreeBSD
in order to make any progress.

The final "nail in the coffin," however, was none of those things.  The nail
came this morning, when I realized and admitted that I no longer cared.  As
many of the recent FreeBSD improvements either didn't impact me (at best) or
actually made my use cases more difficult (at worst), over the last two years
I've moved away from FreeBSD almost without realizing it.  Bug fixes for the
issues I encountered on Tier 2 platforms became death by a thousand paper
cuts.  I sat on a fix to get ipfw working on ppc64- which apparently, as of
about two years ago, nobody else had even noticed was long broken- because the
time to submit a bug report w/ the patch, deal with the inevitable questions
and delays, etc. wasn't worth it when clearly nobody upstream gives a damn.

FreeBSD, my disagreements with you aren't the problem.  The problem is that I
no longer even care that we disagree.  I'm leaving you for OpenBSD and
Illumos.  I wish you the best in life.  If you ever look back and regret
selling yourself out to the Fortune 500s, and think you might like to try
again, give me a call.