[nycbug-talk] OSViews review of the BSD family

Okan Demirmen okan
Tue Oct 19 17:45:13 EDT 2004

to top post:

grrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrr. i wish i had the time
to say more and make that voice heard - maybe later this fall. these
articles are so frustrating to read, it maybe worth the time/money
to drop a few projects to counter this stuff...

On Tue 2004.10.19 at 17:30 -0400, G. Rosamond wrote:
> For those who haven't see it. . .
> Not a very useful review, IMO, as it seems to rest on very dated 
> information on many points and utter confusion on others.
> From reading Bugtraq, I would hardly think that NetBSD "is not secure." 
>  And I hardly think that FBSD is in a "precarious" position due to the 
> simultaneous 4.x and 5.x branch existences. . .He probably thinks they 
> are competing "distros." <g>
> And actually, if I didn't know better, I wouldn't be very excited by 
> the future of the BSDs from this article. . .
> Really just posting for the records, as it's been Slashdot'd.
> And I do think that it would be useful to post a detailed response to 
> this in our library. . .
> g
> It's an exciting era in the Berkeley Software Distribution world; 
> indeed, things started off with a litigious bang over a decade ago, but 
> now BSD solutions are more varied than ever before and offer the user 
> heretofore unprecedented choice and power. So many are the options 
> today that it's time for a roll call from the various distributions. 
> Paul Webb submitted the following editorial to osOpinion/osViews which 
> takes a look at what each BSD has to offer and also looks at where each 
> is going.
> --
> Each of the four major BSD projects are pushing forward with 
> development and experiencing growth, diversifying the Open Source 
> playing field's offerings Let's take a look at what each project is up 
> to these days.
> FreeBSD
> FreeBSD is in a precarious state. While it has almost hit critical mass 
> in the corporate world, their latest growing pains have left potential 
> adopters confused. The new FreeBSD 5 branch offers some exciting 
> technology, generally regarded as comparable with or superior to what 
> is offered in Linux. The FreeBSD foundation is still upgrading its 
> FreeBSD 4.x line and suggesting its use for production environments 
> over FreeBSD 5. The reasons for this are very simple -- FreeBSD 5 won't 
> be ready for prime time until FreeBSD 5.4 or 5.5 -- but users are left 
> confused and timid.
> FreeBSD's last major release, which now sits highly optimized at 
> version 4.10, works just as well as always. For systems already running 
> with FreeBSD 4.x that see no need to adopt the new technology in 
> FreeBSD 5 or jump to Linux, this operating system is a godsend in 
> stability and continued support. FreeBSD 4.11 is scheduled for a 
> February '05 release, while plans for FreeBSD 4.12 are on the 
> backburner should FreeBSD 5 not achieve -STABLE status by the fourth 
> quarter of 2005. But what if you need the technology available in 
> FreeBSD 5 and don't want to jump to Linux?
> FreeBSD 5, currently available at FreeBSD 5.2.1 with FreeBSD 5.3 in 
> late beta, tantalizes the BSD world with the culmination of several 
> year's hard work and narrow escapes. Back in the late Nineties, when 
> WindRiver bought BSD/OS (a closed-source BSD operating system owned by 
> the now-defunct BSDI), FreeBSD users were promised a next-generation 
> BSD made possible by crossing the ultra-robust corporate OS with its 
> Open Source counterpart. While WindRiver let go of its plans leaving 
> the future of FreeBSD in peril, the realization of its goal is almost 
> here thanks to the FreeBSD community and Apple Computer, Inc.'s 
> contribution of FreeBSD code.
> That almost is a killer, though, in that it now causes potential users 
> to look elsewhere for modern operating system features elsewhere until 
> FreeBSD 5 is blessed as stable. Given FreeBSD's track record and the 
> corporate sponsors now behind its operating system, however, it has a 
> bright future ahead of it despite these stumbling blocks. Sadly, the 
> same can't be said for its two little brothers, NetBSD and OpenBSD.
> NetBSD
> NetBSD's claims to fame aren't its optimization or secure code -- it's 
> instead known for running on a wider variety of platforms than any 
> other operating system out there, including Linux. NetBSD's binary 
> releases include support for an amazing 40 platforms and an additional 
> 12 platforms in the source code. In other words, it runs on everything 
> but the kitchen sink. NetBSD forked from the 386BSD/4.4 BSD merger in 
> 1993 and continued on its own in parallel to FreeBSD since then, albeit 
> at a slower pace. It's currently at version 2.6.1, with aggressive 
> testing on the new NetBSD 2.0 promising fruition by the first half of 
> 2005.
> Those familiar with NetBSD swear by it, though its use in serious 
> environments is limited. It is not secure and device driver support is 
> paltry at best. NetBSD's true usefulness comes in providing developers 
> of other operating systems -- such as FreeBSD, OpenBSD, and Linux -- 
> with hardware support to base their own new ports off of. For instance, 
> much of the code for the PowerPC FreeBSD port comes from NetBSD. 
> OpenBSD implemented support for AMD64 by means of hefty imports from 
> the NetBSD source tree, and Linux runs on Motorola's ColdFire processor 
> family thanks to the work previously for NetBSD's port.
> Though it's the unsung hero of the BSD family and Linux, you can safely 
> ignore NetBSD unless you have old or obscure hardware or are looking to 
> port your operating system project to new hardware. Its desktop and 
> production applications are so limited as to be nonexistent and this 
> isn't likely to change even after NetBSD 2.0 is released.
> OpenBSD
> Forking from NetBSD in 1995 after a very heated -- and embarrassing -- 
> personal argument, OpenBSD's one and only focus is to offer security. 
> Every line of code is hand-audited and, as the site claims, there 
> hasn't been a hole in the default install in over seven years. Striking 
> a balance in hardware support somewhere between FreeBSD and NetBSD, 
> OpenBSD runs on very few platforms and even then only in 
> single-processor mode. Sticking with Intel and compatible chips is a 
> safe bet as its Alpha and PowerPC ports are still in their infancy.
> OpenBSD is updated every three or four months and doesn't experience 
> the major upheavals that FreeBSD is confronting now: When OpenBSD is 
> updated, there is no question as to whether or not it's secure or ready 
> for production. Oftentimes it stands in on a general computer to 
> emulate a specific network device, though in a highly secure fashion. 
> If you're in the market for a firewall, OpenBSD can make an aging 
> Pentium system do the job better than pricier hardware. OpenBSD isn't 
> acceptable as a desktop system or 3D workstation, however.
> One factor that mars OpenBSD's fair weather is its primary developer, 
> Theo de Raadt. This individual is known to be highly unstable and even 
> destructive at times. OpenBSD's very birth, as noted above, is owed to 
> one of his infamous tantrums and many users have been flamed off the 
> Internet due to his bad moods and compulsive control issues. Though 
> excellent for network equipment, developers may wish to remain wary of 
> this platform and its creator.
> Darwin
> Apple Computer, Inc.'s Darwin operating system is now the most 
> widely-shipped UNIX in the world, with a user-base of over 10 million 
> strong and growing. The current platform has been out for over a year 
> with Darwin 7.5 corresponding to Mac OS X v10.3.5. Darwin 7.6 will be 
> released before 2005 with another one or two follow-ons before Darwin 8 
> goes live, which has been in development since last January.
> Darwin 8.0b1, the first beta for Apple's next Mac OS X release, shows 
> many improvements over Darwin 7. First and foremost, it includes 64-bit 
> memory addressing and optimizations for Apple processors going back all 
> the way to the PowerPC G3. Many of its libraries and userland will be 
> synced with FreeBSD 5.2, while also enhancing Linux API compatibility 
> and support for AMD64.
> Other points of improvement are symmetric multi-threading (SMT), NetBSD 
> and OpenBSD binary support, next-generation on-the-fly file 
> de-fragmentation, integration of TrustedBSD security hooks, support for 
> Java 1.5, XHTML 2.0 and CSS 3.0, and a myriad of minor improvements 
> sure to make thousands of developers and end-users happy. Clearly, 
> Darwin is the most inclusive and feature-complete BSD -- and, indeed, 
> UNIX -- out there.
> With so much going on with Darwin, it might be hard to realize that 
> it's not right for everyone. There are certain groups who might not be 
> happy with it. Developers, for instance, have expressed frustration 
> over how fast Apple's evolves its operating system, which can sometimes 
> make it hard to create applications that run on more than one version 
> of Mac OS X.
> Another point of contention is hardware support. While Darwin supports 
> the PowerPC G3, G4 and G5 processors and all of Apple's mainboards and 
> other devices, it only runs on Intel's Pentium Pro, Pentium II, Pentium 
> III and Pentium 4 families. Darwin 8 will fix this, with support for 
> AMD chips, but it could be as long as eight more months off. The future 
> burns brightly for Apple's Darwin BSD.
> Final Thoughts
> If you're looking for a software solution in the Berkeley Software 
> Distribution family, you won't be disappointed. All four major projects 
> are continually updated and developed whether you need a general 
> workstation solution, network security, hardware development, or a 
> desktop operating system. The BSD world has never looked brighter than 
> now and each project is geared for major upgrades in the near future, 
> guaranteeing a continuity of utility in the years to come. ::
> _______________________________________________
> talk talk mailing list
> http://lists.nycbug.org/mailman/listinfo/talk
> % Be sure to check out our Jobs and NYCBUG-announce lists
> % We meet the first Wednesday of the month

Okan Demirmen <okan at demirmen.com>
PGP-Key: http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xB3670934
PGP-Fingerprint: 226D B4AE 78A9 7F4E CD2B 1B44 C281 AF18 B367 0934

More information about the talk mailing list