Commentary: Is BSD the tortoise?

ZDNet columnist Stephan Somogyi wonders whether all the leaps that Linux has made in recent history will wind up being compared against the slow, steady progress of the BSDs.

This week reminded me of nothing more than the fable of the Tortoise and the Hare. For months--nay, years--we've been inundated by stories of the remarkably fast growth of Linux, its wholehearted embrace by IBM, and it's general disruptive goodness. All of which is great stuff, no doubt.

But I have to wonder whether all the leaps that Linux has made in recent history will wind up being compared against the slow, steady progress of the BSDs. The BSD-based OSes all look to be doing better and better at the moment, even without Linux's marketing fury behind them.

Validation from Microsoft?
FreeBSD got some serious mindshare via Microsoft this week when it was picked as the unix target for Microsoft's Common Language Infrastructure. Microsoft's views on the GPL are by now well-known; this is the first time the OS giant has done anything public with an open-source OS other than disparage it.

While the open source community and its proponents will bristle at the notion that Microsoft's choice of FreeBSD is a form of validation in the eyes of the mainstream, one shouldn't examine gift horses' dentition too closely. This is an excellent opportunity to generate visibility for the BSDs with those who are familiar with Windows OSes, know of Linux, but may never have heard of the BSDs due to their relative obscurity.

KAME and ALTQ, to name a few
Not only are the BSDs serious workhorse OSes, but rather a lot of research and development happens on BSD OSes even today, a fact that the pro-BSD faction(s) may consider entirely self-evident, but which the majority of even relatively techno-savvy people are unlikely to know about.

A shining example of BSD-based research is KAME, an IPv6 protocol stack and IPsec (for v4 and v6) implementation, which is explicitly built to run on top of various BSDs. KAME is developed by researchers and engineers from a consortium of seven Japanese companies, and is being used as a testbed for all sorts of network research.

ALTQ is one of these projects. It's integrated into KAME, but has also recently become known to a somewhat broader audience as a result of being imported into OpenBSD-current.

Not to be left out of the inexorable march of progress in the world of BSD, NetBSD recently announced a port to IBM's PowerPC 405GP, with the apparent support of Linux-proponent IBM.

FreeBSD and Apple
This past March 24th brought us the release of Mac OS X, which, if Steve Jobs's prognostications are to be believed, will be the largest volume unix by the end of this calendar year. OS X's lineage is undisguisedly BSD.

Apple has even gone so far as to say that it's not just any ol' BSD that it used, but that OS X's BSD layer is based on FreeBSD 3.2, which by now is a good two years behind the state of the art.

One of my ongoing gripes with OS X is that much of its unixy side appears rather stale, with old versions of useful tools being installed by default. Imagine my pleasant surprise when Apple hired FreeBSD's Jordan Hubbard this week, and he actually mentioned this very issue--albeit in a somewhat oracular way--in his message to freebsd-announce.

It's obvious even to the untrained observer, ie me, that it's in Apple's best interest to use as much unmodified source code from the outside world as possible. For both Apple and us on the outside, it would be so nice to be able to build, eg, OpenSSL without any modifications required instead of being stuck with Apple's increasingly aged custom version.

If Apple didn't have to allocate engineers to maintaining Apple-only variants, it could spend more time improving and innovating its OS. Hiring Jordan Hubbard will hopefully add considerable momentum to the improvement of this situation. Such improvement would benefit Apple and its customers, but would also provide ample benefit to the FreeBSD community in the long run as well.

Better to be the turtle
BSD and its growing tree of descendents are no longer a fringe, and their importance in the tech industries is growing steadily. Considering that unix was declared dead, often plausibly, on numerous occasions in the past, Linux's rise was remarkable enough in itself. The vitality of BSD's various offspring is a clear sign that the ecosystem of OSes still has plenty of spacious niches.

ZDNet columnist Stephan Somogyi thinks that if a big commercial studio like Fox can underwrite a film like Moulin Rouge, there's hope for Hollywood yet.