IN DEFENSE OF NETBSD
From David Maxwell
<[email protected]|[email protected]>

Dear Rik,

You followed the 386BSD->FreeBSD history, but not the NetBSD (<ftp://ftp.netbsd.org/pub/NetBSD/misc/release/NetBSD/NetBSD-0.8>)one: see <http://www.netbsd.org/Misc/history.html>. NetBSD 0.8, also a derivative of 386BSD, was announced/released April 19, 1993.

The NetBSD and FreeBSD developers (to be) had discussed things at length before that April, and in essence, agreed to disagree. The FreeBSD folk wanted to focus on the i386 platform and put their full energy into it; the NetBSD folk wanted to continue in the UNIX heritage of portability, and take it further.

In 1994—95, Theo de Raadt, who had contributed to NetBSD, had an argument with some of the NetBSD core team. Not all the details are public. I became aware of this when Theo posted a message to one of the NetBSD lists claiming that his messages were being censored, and so he had put up a Web page to publish his complaints. I followed the link to the page, and I was convinced quickly (in large part by Theo's tone) that Theo was in the wrong.

Theo took the current release, NetBSD 1.1, and relabeled it OpenBSD. Theo has since then worked to raise the visibility of OpenBSD, focusing on security. (Part of his argument with the NetBSD core team was over the lack of speed with which his changes were integrated into the code.)

Theo calls NetBSD an "academic research platform" — that is, not worth considering for production systems. My usual response is to ask, "If NetBSD is just an academic research platform, why was OpenBSD completely based on it?"

I mention this because in your article you said you were told by your anonymous informant that "The NetBSD group . . . is more interested in experimentation than in having a rock-stable version of BSD." That sounds like a small twist on the "academic research" line that OpenBSD tries to sell.

So here's the sales pitch: Linux and FreeBSD claim to support multiple platforms. I looked today at <http://www.freebsd.org> and <http://www.linux.org>, and I couldn't find any mention of a non-Intel installation. I know that Red Hat and some of the other Linux distributions do provide support for other platforms; I'm just indicating that it's still not an important thing for them.

OpenBSD has support for Alpha; Amiga; HP300; I386; Mac 68k; MVME 88k; PowerPc; PMax; and Sparc.

NetBSD has support for Alpha; Amiga; Arm32; Atari; Bebox; HP300; HPCMIPS I386; Mac 68k; Mac PowerPC; MVME 68k; NEWSMIPS; Next 68k; OFPPC; PC532; PMax' SH3; Sparc; Sparc 64; Sun 3; VAX; and x68k.

NetBSD has not been (and probably won't be in the future) about being popular, it has been about doing it right. The same source tree compiles on all the above platforms. Yes, there are some platform-specific driver files, but they are kept separate in the system.

Linux and FreeBSD may claim they support multiple platforms, but the code is not integrated into their source at this time. As changes are made to Intel Linux, the Alpha Linux and Sparc Linux and PowerPC Linux
( . . . ) development teams must rework that code into their out-of-date copy of the source tree. This may change at some date, but only if it becomes important enough to the Intel developers.

NetBSD was the first free UNIX with USB support, the first to do binary emulation, and the first in other arenas, but I've babbled on long enough.

NetBSD is not about experimentation or research, but it's not about doing the "popular" thing, either, or having a pretty graphical install utility, or having version 0.001 drivers for some new card. NetBSDers (and core in particular) aren't into wild self-promotion, but what they have accomplished deserves a tremendous amount of respect.

Incidentally, part of the rationale for the research tag comes from having a very clean source-code base. Things are very well organized, such that someone who wants to write something new has a good idea of the structures they're diving into. Many advanced things such as RAIDframe were developed and integrated into the base code set. IPV6 support is standard as of November 21. IPSec is already available as patches and will be in the base very soon. UVM is amazing too.

I hope I've demonstrated that NetBSD deserved more than 4 lines of mention in your article.


DIVERSITY
From Max Southall, MIS Director, Kelme USA Inc.
<[email protected]>

Hi Rik,

Interesting that your October Musings on StarOffice have been somewhat fulfilled and beyond, now that it's being given away by Sun and the subsequent proliferation of the Windows-based port. I think at this point it must be over a million downloads, as well as shipping-charges-only CDs ordered.

From experience with most of the office systems that have pretensions of being upwardly mobile (UNIX, Windows, Mac), I have to draw that painfully obvious (to us sysadmins, anyhow) conclusion that the proliferation of Windows has brought with it uncontrollable administration costs.

Pretty well the only practical solution I've actually seen implemented by those who've tried to stick by Windows is the hiring of additional and progressively cheaper bodies to try to keep it all going somehow. And concomitantly, the laying off of the fewer more expensive bodies, because, as they say, it doesn't matter how smart you are under the MS scenario, because it takes just as long to reboot yet again. Could this be the real meaning of "Windows for Dummies"? In any case, costs keep rising and the level of service sinking.

You know something's got to give when it takes more time to resuscitate a user's scrambled PC than it does to restore a well-managed Sun server that serves dozens or hundreds of such users.

Now in my humble opinion, what's prevented management of the desktop has been essentially the abandonment of the desktop-application market to Microsoft by the major UNIX vendors. UNIX finally matured as the OS platform best suited to the thoroughly networked environment we all find ourselves in, but at the same time, ironically, with none of those PC-styled "killer apps" left that are needed to woo away the disenchanted PC shops.

Except, maybe, StarOffice. A UNIX clone of the lumbering toad MS Office, transformed into a charming thin-client prince? Hey, that's the ticket!

So I think that Scott McNealy's thin-client application-services vision for StarOffice is a mite convenient and maybe disingenuous. Not yet. As you noted, this is pretty compatible to MSOffice, right down to being a fair imitation of bloatware installation heft. Also, StarOffice wasn't even Sun's idea, although as my friend Gerry Singleton points out, the synergy with ex-Sunner Andy Bechtolsheim can't hurt. Trying it out last year, my opinion was, after finding it just didn't quite cut it, that Sun Microsystems ought to buy it and make sure all the rough spots were shined, so that there would be user-level office software available that wouldn't end up telling everyone in the enterprise where Microsoft wanted them to go today. That and a cup of Java could eventually get us all off the MS dime.

The lesson from the emergence of Linux is that the only strategy with any chance of competing with Microsoft, regardless of merit, is one that gives software away to gain significant market share. Because Microsoft with its enormous accumulated wealth can afford to dump its products until its competitors go out of business, we have seen over and over companies who have pioneered successfully in the Microsoft arena be absorbed or disappear soon after Microsoft decided to enter their markets.

In the case of Linux, whose distribution is modeled in a way even Microsoft can't compete with economically, Microsoft is worried enough to speculate internally in hysterical fashion.

McNealy's free distribution of StarOffice punishes Microsoft in the only way it understands — becoming subjected to the same strategy it aimed at everyone else, namely, amputation of cash flow from key product sales.

OK. So that's the fun strategy for the folks in Mountain View. What about the strategy for MIS?

We need to have manageable systems that encompass the desktop. We can't have systems becoming ever more unmanageable under an unworkable PC paradigm, or, in the case of what Microsoft has disingenuously offered as enterprise management solutions, with all the important decisions outsourced to Redmond and made with full attention to Microsoft's cashflow needs rather than MIS's.

We're currently running StarOffice on all three platforms to which it's ported. Because of Sun's stability and also its focus on its core products, we are deploying Sun servers. And we chose Sun earlier this year after some soul-searching, before the StarOffice acquisition.

There is a window of opportunity right now for vendors like Sun. People are dissatisfied with the Microsoft enterprise path for very serious manageability reasons and are willing to entertain a shift to a more viable approach at this moment. I think that the acquisition of StarOffice and its release in this way are showing that this is the way to go. Personally, I hope that they will build it better, and the customers will come. I just can't stand the thought of system administration being reduced to carrying a CD fanny pack from user machine to machine, forever. And that's what's happened to some of my formerly UNIX colleagues.

cfengine Alert!

<www.cfengine.org>

<www.cfengine.com>

<www.cfengine.net>

He warns people to be wary of them because they might be used to disseminate Trojan-horse versions of cfengine. The official Web site for cfengine remains at Oslo College:

<http://www.iu.hioslo.no/cfengine>

complete with checksums for your safety and assurance.