The Case for FreeBSD

essdodson writes "Scott Long of FreeBSD release engineering team describes some of the finer points where FreeBSD continues to innovate and display its mature development environment. Items such as netgraph, geom and incredible desktop support by way of Gnome and KDE." From the post: "While I strongly applaud the accomplishments of the NetBSD team and happily agree that NetBSD 2.0 is a strong step forward for them, I take a bit of exception to many of their claims and much of their criticisms of FreeBSD."

Comments

hmmm

[Quasar1999]

I just installed FreeBSD this morning... I must say, straight off the iso, a quick install had me up and running pretty darn fast... much quicker than any linux distro I've tried in the recent past... Now if only I could figure out how to get visual studio to run under it, I could ditch windows... stupid work... stupid requiring development on Windows...

One serious thing about FreeBSD over linux distro's... It feels like it has more of a structure, especially when installing utilities and apps... I find with linux distros, the stuff included feels like it's all over the place, hard to find where things end up installing... but I'm really a vxworks fan... so take what I say with a grain of salt... ;)

Re:hmmm

[drinkypoo]

Some linux distributions are more fragmentary than others. Gentoo linux in particular tends to put things in the same place every time; /etc/conf.d for commandline and environment options, and /etc/ for that package's config files. On the other hand I've been mulling over the possibility of putting QNX on my laptop, which has only 128MB ram :)

Re:hmmm

[bcmm]

As you say, Visual Studio is a real killer...
I am forced to program in VB for an AS level course, and Visual Studio is what keeps me dual-booting (Linux and Windows).

Re:hmmm

[dadragon]

FreeBSD 5.3 and 4.10 didn't like my computer, either. 5.2 and 4.9 worked fine. The problem is IEEE1394. There is a bug in its disk code, and it pauses looking for hard drives on the 1394 bus. Disable it, and it worked fine for me.
YMMV

Re:hmmm

[agraupe]

Conceivably, you could run an emulator such as bochs or qemu. I agree this isn't the best situation, but when you're dealing with VB in any capacity, you're pretty much doomed to disappointment.

Re:hmmm

[Sentry21]

I find with linux distros, the stuff included feels like it's all over the place, hard to find where things end up installing... but I'm really a vxworks fan... so take what I say with a grain of salt... ;)

Generally this is true - this is the reason I restrict myself to Debian. With Debian, everything is packaged in the same manner, to the same standards, and it all makes sense. The structure makes it the only Linux distribution I'm willing to spend any time on.

When I tried FreeBSD, I felt that it had much more of a UNIX feel to it - I felt like I was dealing with something classic and powerful. I wasn't (there's only so much a P133 can do), and I had no use for FreeBSD whatsoever, but even just at the console, it felt more responsive and powerful. All subjective, but interesting.

Re:hmmm

[forkazoo]

Oddly enough, I actually did my first FreeBSD install early this morning. I just got an AlphaServer yesterday, so I figured I'd try something new. While the basic OS setup went fine, it just rebooted itself when I tried to get into X. Dunno why. :( And, aparently, lynx isn't part of the base install, so I just gave up, and tried to install debian. For some reason the debian installer is fine, but I get a kernel panic when I reboot after the initial install. I guess BSD wins on ease of installation on my Alpha, but not by much.

Of course, I've owned the alpha for less than 24 hours, so I assume it is something of my inexperience that makes it not work right...

Re:hmmm

[cdcarter]

I highly agree. I have been mucking around in new OS's and distro's and FreeBSD was really easy for me to install. No mem-hog graphical installer, but it wasn't extremely cryptic either. But most importantly (to me at least) it got my non-standard, onboard video card properly configured in *one* try! FreeBSD was a godsend.

Re:hmmm

[MPHellwig]

I guess you didn't followed the excellent FreeBSD handbook?

Re:hmmm

[ckaminski]

Spend the $180 for VMware Workstation 4.5. No only can you run Windows, but every version of Windows out there, DOS, etc...

Re:hmmm

[bcmm]

Yes. I've tried to run win 98 SE under qemu. I think the CD may be messed up though. Any idea if windows 98 SE should install OK in qemu?

Re:hmmm

[Aeiri]

That's the way people feel about Linux the first time they try it, it's all just because you are used to Linux, and the something new is tricking your brain into thinking it's something absolutely amazing.

I'm sure that if it were the other way around, you being used to FreeBSD and now trying Linux, you would feel the same way.

(Granted, this is all just my thoughts on this matter... My only experience of FreeBSD was it freezing after the first display of the install ("Welcome to FreeBSD, now starting installation" or something like that))

Re:hmmm

[blindbat]

I used to use FreeBSD and loved it. I had to switch when I purchased a new laptop and couldn't get the pccard services to work. I tried many linux distos but never felt at home until Slackware. It was recommended to me from someone that appreciated Slackware's similarity in layout to FreeBSD.

Re:hmmm

[dkh]

For years I used a 133 box as a workstation running X with fvwm and later windowmaker without problem. The box also served a few mailing lists and hosted a small mud and web services.

I still have that box working deligently and capably as a firewall machine. 133Mhz provides plenty of cycles to get real work done when the operating system actually works well.

That being said - buildworld on such a box is a day long afair, (thankfully its simple to build on another machine and install over NFS). I wouldn't even try to start KDE, Gnome, or Firefox on it. I think that says more about "big software" then it does about the os.

Thanks to the *BSD's old hardware can still be put to good use.

Re:hmmm

[DrSkwid]

OpenBSD installs quickest, and that includes X & Apache

Re:hmmm

[DrSkwid]

getting lynx installed is a real pain

# cd /usr/ports/www/lynx
# make install

or perhaps you'd prefer a pre-compiled binary

# pkg_add ftp://ftp.freebsd.org/pub/FreeBSD/ports/i386/packa ges-stable/All/lynx-2.8.5.tgz

n.b. you'll have to remove your own spaces though =)

Re:hmmm

[forkazoo]

I didn't have another system handy at that moment to get any docs, and I wasn't familiar with BSD. I'm sure if I had a handy web browser, it would have been trivial to use google and figure out how to install lynx... But, I'm sure you see my dilemma. I'm sure I'll install some form of BSD on it again before I am through with the box, I just won't start at 1:00 am, without any docs handy the next time I do it... :)

Re:hmmm

[evilviper]

Some linux distributions are more fragmentary than others. Gentoo linux in particular tends to put things in the same place every time; /etc/conf.d for commandline and environment options, and /etc/ for that package's config files.

This is besides the point, IMHO. Even if one particular Linux distro is (mostly) consistent with where it puts files, where does that leave you when you go from one distro to another?

Where is KDE installed again? /opt? /usr/local? /usr/X11R6/share/X11/libs? I think you get the point. I use Linux, but I must admit my annoyance with the, almost random placement of file from distro to distro. To say nothing of the searching game you have to play even with just a single distro.

Re:hmmm

[DrSkwid]

If you won't read the instructions, what's the point of complaining ?

Re:hmmm

[Glen Ponda]

Where is KDE installed again? /opt? /usr/local? /usr/X11R6/share/X11/libs?

qpkg -l kde-base

Re:hmmm

[gothfox]

Have you heard of package managers?

With the ability to easily tell what file belongs to what package and to see all files of specific package there is nothing hard "to find where things end up".

Re:hmmm

[jo42]

> qpkg -l kde-base

-bash: qpkg: command not found

Re:hmmm

[forkazoo]

I didn't mean to sound like I was complaining. If you will note, BSD actually got the easy-install win in my 1:00 am install fest. :) Debian wouldn't boot on the thing.

Eventually, I will read the docs, and get everything set up real purty, but I didn't care to take the time to do things right at that particular moment. Hell, I jury rigged the chassis switch so I could use a hard drive without the proper drive bracket, with the skins off! :)

It was a new toy, and I intended o play with it directly, after all.

Re:hmmm

[kernelistic]

I disagree.

I've used various os' and distros inc win3.11, win95, Win98 (rtm&se), winMe, win2k, winXP, bsdi, solaris, redhat, debian, suse, gentoo. And always end up coming back to FreeBSD. It just seems to make sense, don't need to guess (I started on SysV), tried others still end up coming back to FreeBSD.

It's all a matter of personal preference. Just use whatever tool you are most comfortable with!

Re:hmmm

[knapper_tech]

I have an nForce 3 chipset and no way to update the BIOS (HP's fault). FreeBSD's netinst cd won't even boot on my machine.

Who cares about this battle?

[drinkypoo]

I don't see why people are so worried about advocacy. If you're not making money, what is the difference? Continue to refine the thing and get what you want out of it, and if other people don't get it, who loses? Personally I have a use for only a couple of operating systems now, and they are Linux and netbsd. netbsd because it runs on just about everything, and Linux because it's most supported. It's nothing against FreeBSD, which I simply don't need. The point is, I use whatever fits the job and if that was FreeBSD then I'd use that. The best fit is determined partially by functionality and partially by familiarity...

Re:Who cares about this battle?

[thepoch]

You see, people are worried about advocacy because these create mindshare. Without advocacy, people won't understand what the advantages are with using/supporting whatever it is you are advocating.

Without advocacy, your product/whatever will seem inadequate, small, meaningless. This will make your whatever simply useless in the eyes of those who have not decided for themselves at the moment.

People who are not making money out of this have all to lose if they don't get the advocacy they need. They don't have marketing might, and advocacy is all they have. The moment they lose advocacy, they lose mindshare, they lose users. They will them either wither and cease to exist, or become mediocre and simply unimportant, a relic of the past, with the people unwilling to just move on.

You have already decided what you need/want. This makes advocacy useless for you. For the rest of those who have not finalized that decision, they need this stuff to understand the advantages as viewed by those who use the stuff.

Of course, you are also advocating Linux and NetBSD by stating you use those. You didn't give hard facts, but it's still advocacy in a simpler form.

Re:Who cares about this battle?

[jjohnson]

Advocacy is to free software what marketing is to commercial software: the driver to bring in currency. In free software, that currency is warm bodies, both users and people who actively contribute to the product. Free software projects benefit mightily from community involvement, to the point where a lack of involvement can kill an otherwise promising project.

Re:Who cares about this battle?

[debilo]

I don't see why people are so worried about advocacy. If you're not making money, what is the difference?

Donations. Many (maybe most) FOSS developers don't get paid, this is especially true of FreeBSD (or any of the BSDs) since there's less corporate backing than with Linux. A more vocal advocacy will surely change that by drawing more companies' attention to FreeBSD (look what IBM does for Linux) and get them to support the development, and a larger userbase will surely increase much needed donations, be that money or hardware.

Continue to refine the thing and get what you want out of it, and if other people don't get it, who loses?

The FreeBSD community loses, for the reasons laid out above. The more attention FreeBSD draws to itself, the more donations will flow, the more corporate backing they will get, the quicker native drivers will be written, etc. etc. Advocacy is important.

Re:Who cares about this battle?

[cperciva]

[If other people don't know about FreeBSD], who loses? Personally I have a use for only a couple of operating systems now, and they are Linux and netbsd.

To answer your question: You lose.

Linus Torvalds has said that the idea behind Linux is "do it yourself". The idea behind BSD -- coming, as it does, from an academic background -- is "there's lots of trash out there. Let's give people something better".

As far as providing people with a better alternative is concerned, writing FreeBSD doesn't accomplish much if everyone keeps on running the Linux distribution of the day.

Re:Who cares about this battle?

[NutscrapeSucks]

Advocacy is to free software what marketing is to commercial software

Actually there's a key difference. Most marketing is carefully directed at potential new customers. Most "advocacy" takes place in forums specifically designed for advocacy (comp.*.advocacy, slashdot, ars technica battlefront, etc), where a tiny number of relatively knowlegable users quibble amongst themselves for kicks.

Let's take this very article as an example. Both FreeBSD and NetBSD have relatively small userbases which primarily consists of Unix and BSD-saavy users. Neither project has very much to gain by converting the other's users. (Unless there really is some threat of one or the other dying.) Either project would have much more to gain trying to convert the HUGE market of fleeing commercial UNIX users instead of arguing amongst themselves. You'll notice that's what RedHat is doing rather than trying to pick off Debian customers.

Re:Who cares about this battle?

[spiritraveller]

Advocacy in open source is important because it indirectly affects the speed of development and ultimately the survival of the project.

If hardly anyone were using FreeBSD because they thought it "sucked", then there would be far fewer people willing to develop FreeBSD.

Those who are developing it would find their efforts less fruitful, because fewer people would exist to fix bugs and improve on their work... eventually FreeBSD would go the way of the Amiga... a few diehard users stll existing, but essentially dead.

Not to imply that that isn't already the case. (I'm kidding... I'm kidding!)

Re:Who cares about this battle?

[CDMA_Demo]

I've been using FreeBSD and NetBSD for over one year now for two good reasons: 1) They both run faster that any linux distro on my 150 Mhz 32 MB ram machines. Even slackware and gentoo fail me here, regardless of how much optimizations I include in the 2.6.* kernel 2) The security is outstanding compared to linux distros. My two machines serve as routers and firewall plus I needed something light and fast. I also use these systems to play around with socket programming, and they are great. With both NetBSD and FreeBSD, you get a core system, and then you can use ports to install whatever it is that you need. Its more clean and organized compared to all Linux distros out there. You will be surprised how much junk mainstream linux'es install on your system. Someone will ask, "what do you mean by faster?". Try compiling a small c++ program, for example. I say people should use Linux as a bootcamp, sort of a training field, so you are better prepared to handle classic unix OSes like BSD.

Re:Who cares about this battle?

[CDMA_Demo]

Anonymous Coward says:

"Maybe if you got an after-school job at McDonalds, you could save up for a new Dell."

here you invoke two obnoxious brand names, McDonalds and Dell (a man is known by the company he keeps). Obviously you are a "supersize me" windows guy, so why don't you chill out in the M$ area?

"Also, try asking your mom to increase you allowance."

What works for you will not work for others. Come back in a couple of years when you can share some new experiences.

Re:Who cares about this battle?

[mollymoo]

Mindshare? You are a PHB and I claim my $5.

You advocate your fave OS so more people will use it. More users attracts more developers: some users are also developers, developers often want their software to be widely used (kudos, cash, karma...). With more developers your fave OS gets better quicker.

Not to mention...

[elid]

...FreeBSD is getting a new logo (well, 0 submissions to date, but still !

Re:Not to mention...

[Anonymous Coward]

Heh, give it time. The contest just started five days ago.

Still, from reading the mailing lists, creating a new logo is pretty unpopular in the FreeBSD community. That unpopularity stems from people thinking that this logo will *replace* Beastie, which isn't the case.

Re:Not to mention...

[laffer1]

I submitted a logo this weekend so they have at least 1. :)

More people need to try and use FreeBSD

[Kip Winger]

Repeat a lie enough, and it becomes true. That lie, mostly being, that FreeBSD is dying, or is some arcane system only to hack around on, similar to Plan9.

In fact, for those who haven't tried it, it's quite an excellent full-featured Unix, with everything you'd find under Linux. In fact, it's fully binary compatible with Linux.

The only difference is that it does things the old way -- vi is vi, not vim, and you get sh, csh or tcsh instead of bloated bash. It doesn't have anyone pushing for "ease of use," though it's about at the level of slackware, except with ports, the greatest package management system known to man. Gentoo's portage doesn't even come close to the flexibility and reliability of ports.

Internally, it runs great, because it's not doing things the kernel shouldn't do to boost benchmarks. It's not deeply involved in corporate America, but remains strong due to good management.

Plus it's far more secure. With how much Linux websites are hacked these days -- see http://zone-h.org/ and check out the statistics section, at least 70-80% of website hacks are Linux based -- I wouldn't run it on Linux. FreeBSD is the obvious choice, as it runs its services flawlessly.

Re:More people need to try and use FreeBSD

[drinkypoo]

Amusing that you should mention gentoo, in which vi is vi, not vim. It's not installed with the system by default, though. I also don't know why you say that ports has more flexibility, but it probably does have more reliability.

As for linux websites being hacked, that's because they're not updated. If you fall behind on your FreeBSD updates, you'll get rooted too. Usually it's not a kernel hack, it's an application hack that would probably happen to FreeBSD too.

Re:More people need to try and use FreeBSD

[oxygene2k2]

I wasn't aware that the original vi (by sun) was released as open source. are you sure that gentoo's vi isn't just nvi or elvis?

Re:More people need to try and use FreeBSD

[drinkypoo]

"AUTHOR William Joy. Mark Horton added macros to visual mode and was maintaining version 3. This version incorporates changes by Gunnar Ritter."

The ebuild seems to have fled, though; I got it by emerging vi, but emerge -p vi says emerge: there are no ebuilds to satisfy "vi". so I don't know where the hell it went :/

Re:More people need to try and use FreeBSD

[ArbitraryConstant]

"Gentoo's portage doesn't even come close to the flexibility and reliability of ports."

Portage's flexibility is considerably better.

However I agree on the other point. Portage's reliability is much, much... much worse.

Re:More people need to try and use FreeBSD

[Cthefuture]

Until I can run the latest VMware under FreeBSD, I'll be sticking with Linux. The same holds for IvTV (hardware PVR) support under FreeBSD.

I generally like BSD but it suffers even worse than Linux when it comes to available software (Linux being worse than Windows).

Plus it's far more secure. With how much Linux websites are hacked these days -- see http://zone-h.org/ and check out the statistics section, at least 70-80% of website hacks are Linux based -- I wouldn't run it on Linux.

Yeah, but isn't this related to the fact that more sites run Linux? I mean, most BSD servers are running the exact same hackable software (Apache, ssh, etc.) Rarely do I see kernel level hacks. Especially not remotely exploitable kernel hacks.

Re:More people need to try and use FreeBSD

[jefp]

>it's fully binary compatible with Linux

It's not just compatible, it actually runs Linux programs faster than Linux does.

Re:More people need to try and use FreeBSD

[LurkerXXX]

There is a difference between what is the default and what is available. You have to realize that other folks may have think different applications are 'better' than what you do. Better is a relative term, depending on what you want to do and how you like to do it.

In short, if you want vim, you can install it. If you want bash, you can install it.

Re:More people need to try and use FreeBSD

[Pedrito]

With how much Linux websites are hacked these days -- see http://zone-h.org/ [zone-h.org] and check out the statistics section, at least 70-80% of website hacks are Linux based

Without being able to state what % of that market Linux is, your percentages mean nothing. For example, if Linux is 70-80% of that market, then it would say that they're all pretty equally hackable. I'm not saying they are equally hackable, I'm just saying your statistics don't provide any reference by which to measure.

Re:More people need to try and use FreeBSD

[sunwukong]

I was benchmarking *native* binaries on my old rig (dual PIII Xeon 700MHz 2MB w/ 640MB PC100 DRAM) on both Gentoo 2004 (gcc 3.3.2) and FreeBSD 5.2.1 (gcc 3.3.3) using:

nbench
- FreeBSD slightly faster on memory & integer (3% and 1% respectively)
- Linux much faster on floating point (17%)

scimark
- FreeBSD faster on FFT (11%), SOR (3%) & LU (5%)
- Linux faster on Monte Carlo (2%) & Sparse Matmult (16%)

stream
- FreeBSD faster on copy (1%) & scale (2%)
- Linux faster on add (7%) & triad (10%)

ubench
- Linux faster on both cpu (5%) & mem (60%)

I haven't pulled apart these programs in minute detail to account for any discrepencies, but it seems that most operations will operate at the same speed (within a margin of error) at least on this old machine.

On my newest rig (dual Opteron 242s w/ 2GB PC3200 DDRAM each) FreeBSD wasn't stable enough at the time of setup for me to test it properly.

Re:More people need to try and use FreeBSD

[jefp]

You're benchmarking CPU and physical memory operations there. There's not going to be much difference with any OS on those. Where FreeBSD wins is in things like network & disk I/O, virtual memory, context switching, etc.

Re:More people need to try and use FreeBSD

[ArbitraryConstant]

It's better at handling more diverse dependencies and upgrading existing software.

It's also more up to date, but that's a very mixed blessing, because they don't do proper regression testing and stuff breaks a lot as a result.

Re:More people need to try and use FreeBSD

[Brandybuck]

I'll bet you a lot that it's really nvi, which just about everyone calls plain old "vi". This is the vi that the BSDs use, since it was the replacement for AT&T's vi in 4.4BSD. From the README for nvi: "This is the README for nex/nvi, a freely redistributable implementation of the ex/vi text editors originally distributed as part of the Fourth Berkeley Software Distribution (4BSD), by the University of California, Berkeley."

Re:More people need to try and use FreeBSD

[sunwukong]

Lesse, lmbench 2.0 ...

"Simple" stat, fstat, read, write, open/close
- Linux at least 100% faster than FreeBSD

"Select" fd's (incl. tcp)
- FreeBSD at least 100% faster than Linux

"Process" fork +exit,+execve,+/bin/sh -c
- Linux up to 100% faster than FreeBSD

"Bandwidth"
- tmp write: Linux 30% faster
- localhost socket: Linux 13% faster
- sock stream: FreeBSD 16% faster
- pipe: FreeBSD 33% faster

Re:More people need to try and use FreeBSD

[LurkerXXX]

You mean the version that's easier for 90% of Linux users switching over? I think the BSD folks would rather cater to the current BSD users than add junk to a default install just for switchers.

Re:More people need to try and use FreeBSD

[thejuggler]

"vi is vi, not vim, and you get sh, csh or tcsh instead of bloated bash"

I have bash on my FreeBSD installs. vim is in the ports collection. So it doesn't install everything by default. That's a good thing isn't it?

-----
Happy user of FreeBSD, Slackware, Solaris 10 and OSX. I still have Windows 2000 Server to remind me why I use the others.

Re:More people need to try and use FreeBSD

[setagllib]

Define 'better'. I love vim, in its ease of use and power, but nvi has a MUCH smaller footprint in every way, and for some of the lower-end systems you may run on (applies more to NetBSD which fits on these systems, than FreeBSD) it's more than enough.

FreeBSD also boldly cater for newbies with ee (easy editor - don't confuse with electriceyes) which is very painful to use (at least compared to pico/nano) but also has a small-enough footprint to be worth keeping just for the broader audience.

At least you get tcsh, which has 90% of bash's features and some bash doesn't, and itself is very graceful and efficient. The other BSDs stick with csh, but anyone in their right mind installs their favorite shell from pkgsrc/ports within their first boot of a new system.

Personally what I think FreeBSD severely LACKS is a CVSup functional equivalent (preferably in C, so as not to need ezm3 in the base too), which is necessary to fetch sources and ports trees in any successful way. NetBSD uses CVS AND has the client in the base package, so you don't have a chicken-egg problem getting pkgsrc.

Base systems get a lot of arguments surrounding them, but of all the BSDs I like FreeBSD's the most. It keeps up-to-date and relevant, and besides the CVSup dilemma it's enough to get a good system up with minimal fishing in ports.

DragonFly BSD's will be better once gcc 3.4 is the default and works for everything, which makes it more consistent. The rest of their work is good, like OpenBSD's new BSD-licensed replacements for GNU junk entering the base system.

Re:More people need to try and use FreeBSD

[jwr]

Would someone care to explain how Gentoos portage doesn't come close to FreeBSD ports?

This is a serious question -- I'd really like to know what's better about the ports system.

Re:More people need to try and use FreeBSD

[Neil]

The original vi wasn't by Sun. It was written by Bill Joy at Berkeley. The command-line version of the editor, ex, was in the very first "Berkeley Software Distribution". The first vi for display terminals was in 2BSD. (source: Twenty Years of Berkeley Unix, Marshall Kirk McKusick's chapter in the O'Reilly Open Sources book).

The vi that FreeBSD uses these days is nvi, a "bug-for-bug compatible" rewrite of the original, which was produced for 4.4BSD (presumably the original vi/ex was "encumbered", derived in some way from Bell Labs Seventh Edition Unix sources?).

Re:More people need to try and use FreeBSD

[yamla]

It is NOT fully binary compatible with Linux. I cannot get VMWare 4.5 to run in FreeBSD, for example. Crazy counterexample? Perhaps, but VMWare was a critical application for me and as a result, I had to go back to Linux.

I had some other problems as well. valgrind had problems but my understanding is that these have been resolved now and may not have been anything to do with FreeBSD in the first place.

Re:More people need to try and use FreeBSD

[lachlan76]

Check out the stats on OpenBSD vulnerabilities sometime...I doubt you will find that Linux is more secure anytime soon.

I agree

[bogaboga]

Yes, because the BSDs continue to shine where Linux and Windows seem to fall short IMHO. This is software pakgage management. I am using Debian now and was shocked to find that even for Debian, with its much acclaimed apt tool, Debian got confused and made my system unstable when I decided to upgrade it.

I also heard that Windows used or at least used some BSD work in it's internet capability push years ago. One question will always dog me: Why aren't the BSD's as popular with their very good license at least in the eyes of the IBMs and HPs?

Re:I agree

[Mr Bill]

I use both FreeBSD and Debian regularly, and have to say that I've had my problems with the ports collection as well. Has anyone tried to upgrade perl lately? Here is an excerpt from /usr/ports/UPDATING on how to upgrade perl (just a minor version upgrade)

* run some magic incantations to upgrade all ports depending on perl,
that is run something like :
portupgrade -f `(pkg_info -R perl-5\* |tail +4; \
find /usr/local/lib/perl5/site_perl/5.[68].[1245] -type f -print0 \
| xargs -0 pkg_which -fv | sed -e '/: ?/d' -e 's/.*: //')|sort -u`
This is likely to fail for a few ports, you'll have to upgrade them
afterwards by hand.

Not the cleanest way to do a package upgrade. When I upgrade perl on debian, it just had to download a few packages, and everything was handled properly.

I like both systems for what they provide me, but neither system is perfect...

Re:I agree

[che.kai-jei]

because its not as popular. [? probably ?in open source cheerleader terms]
no one cares about the 'good'
apple do by releasing code back. even though they arent
really 'GPld' into doing it!
its not seen as a 'good' and they still keep all their nice toys. [as well they should if they want to/ esp in it benefits them to do so!]

you see?

how many open source developers are shaping the future of the next OS X release? [ i really don't know- it could be thousands of netBSD and freeBSD and openBSD devs -- but the point is i dont know that and i doubt its more than small handful. not that apple needs or cares! ]

but SUSE, redhat, ubuntu, whatvernux 2.0- or any open source package / project *LIBERATED* by ibm/*whateverOpenSourceDarlingCorpoftheMonth??
pe rsonally i know people who are actively helping
the 'community' eg i personally am one of those people . and end users? - they can help too - any wya they can.

i am not trying to malign BSDs or apple. i am reporting on the perceptions of others as related to me and my own somewhat idiotic perceptions.

ps these are also the perceptions which inform me and keep me merely as a user of the wonderful freebsd, openbsd and pf, rather than contributing [as opposed to linux.and tbh.. i'm not THAT thrilled with linux distro's. they just leave me content.]

Re:I agree

[dermusikman]

the BSDs aren't as popular with heavy hitters because if they contribute code, there is absolutely NO promise of return with that code. the GPL puts everyone on equal footing - IBM contributes code, and I can contribute code, and both parties may take advantage of it.
in BSD, IBM contributes code, and I can improve upon that code and close off my source. IBM gains nothing.

Re:I agree

[destiney]

Yup.. My Perl has been semi-broken ever since the 5.6.8_2 release several weeks ago. I had to rebuild all Perl modules myself ( which should have updated themselves as dependancies if they are dependant on a specific version of Perl, right? ), and I still have 4 Perl modules that refuse to rebuild at all. I thought RELEASE was the most stable version of FreeBSD, that's why I upgraded from 4.10. I gotta say 5.3 isn't living up to the name as far as I can tell.

Re:I agree

[poopdeville]

People say BSDs are better than Linux and Windows because of the It's-a-less-popular-OS-so-it-must-be-good-and-I-mu st-be-cool-for-using-it syndrome. The same thing goes for a lot (most) of Mac users. Face it. x86 and x86-based OSes are better than Mac and OS X. Linux is better than *BSD. And in some areas Windows is better than Linux.

Good troll! Thankfully, I think people say BSDs are better than linux because they fit their needs better. Same with OS X and windows. Nobody really cares which OS you're using anyways, outside of 12 year olds on usenet.

Re:I agree

[Homology]

Why aren't the BSD's as popular with their very good license at least in the eyes of the IBMs and HPs?

One view from Why Researchers Should use a BSD-style License Instead of the GPL :

Linux is most attractive commercially to (1) small companies selling CDs primarily to end-users, not developers, in an environment where ``buy-low, sell-high'' may still give the end-user a very cheap product (this does not mean that Linux is not attractive to programmers); (2) hardware companies that intend to undermine software companies in the OS business; and (3) companies that expect to survive by providing various forms of technical support (including documentation) for the GPLed intellectual property world.

Re:I agree

[cpghost]

I thought RELEASE was the most stable version of FreeBSD

You are not running RELEASE, if you updated your ports tree :). At least, not *exactly*.

Good luck with the remaining Perl modules.

Re:I agree

[Hope Thelps]

Why aren't the BSD's as popular with their very good license at least in the eyes of the IBMs and HPs?

They are. If IBM or HP want to take freely available code and relicense it on other terms then BSD fits that need perfectly and is very popular for that purpose.

If, on the other hand, IBM or HP want to work on a collaborative open source project without competitors being able to relicense code under more restrictive terms then BSD is less suitable.

It's horses for courses.

Re:I agree

[Mr Bill]

that magic worked for me, but it is simply a shortcut! that's why it looks like magic.

I know what the command was trying to do, but would suggest that it should be unnecesary in a system as 'amazing' as the ports collection.

Upgrading perl on FreeBSD is a non-trivial task that has the tendency to break the system. All I am saying is that this is a deficiency in the ports system!

you can't compare a perl upgrade with packages vs. source

I'm not talking about having to compile verses not compile. I am talking about a very basic concept in package/port management, and that is depedencies. Appearantly the ports collection doesn't handle dependencies as well as some people would make you think. If it did, it would not allow you to upgrade (without using force) your perl port without also upgrading all ports that depend on it. You have to do a bit of hoop jumping to get things done, and even then you are told that it might not work.

Re:I agree

[Badanov]

There is a text instruction at the end of every perl install/upgrade which reminds you to export some kind of indication your newest install of perl is now the default version the OS should use.

I forget the command, but like you I was kinda upset about older perl modules not working with the newly install perl until I read that.

I run a web server that uses CGI scripts and several database modules.

FreeBSD has been very good to me.

Re:I agree

[geminidomino]

Cute. I ask an honest question (I don't know if 5.x has a -STABLE branch or not) and some asshat mods me "Overrated". Idiots...

(Posted so that someone who knows the answer but reads at +1 can actually see the question: Does FreeBSD 5.x have a -STABLE branch yet?)

Re:I agree

[Brandybuck]

If you contribute code under the BSD license, everyone still gets to use it. You can't both contribute and close it off. It doesn't work that way. Both parties still get FULL advantage of all contributed code.

Of course, you're not required by the license to contribute anything, but so what? I've always preferred voluntary contributions over sour grapes myself.

Re:I agree

[Brandybuck]

Appearantly the ports collection doesn't handle dependencies as well as some people would make you think.

Actually it does. It just doesn't cover all the oddball corner cases.

If it did, it would not allow you to upgrade (without using force) your perl port without also upgrading all ports that depend on it.

When the upgrade doesn't break the dependents, you don't need to upgrade the dependents. If the upgrade does break one or two, then you mark those for a mandatory upgrade by bumping the revision level. But then you get to this particular Perl upgrade. For whatever reason it broke too many dependents, and there were simply too many to mark them all. At least not in a timely manner.

This is a corner case because usually you don't need to upgrade packages that haven't changed. After all, I migrated from XFree86 to X.org without having to upgrade ANYTHING. I've used some Linux systems where such a migration would have required an upgrade to everything on the system.

Re:I agree

[JamesTRexx]

It does. I'm running 5-STABLE on my laptop (Dell C810). (as far as I know :-P) I run a cron job daily to update ports and source for RELENG_5.
As an aside, I've heard 6-CURRENT runs pretty decently already and is set for "release" in june or july.

Re:I agree

[PurpleWizard]

Because with the BSD license if IBM added code (functionality) and released it its competitors would be able to take it and use it without needing to release it.

The GPL means they can add to Linux and it can't end up being used by Windows or other competition.

As for the apt destabilising of your system I sort of accept that eventually my system will need polishing back up. I just celebrate it doesn't happen to my with the frequency Windows used to. (Not having experienced more recent WinXP mind you, but all the service pack woes of that doesn't really make it sound any better).

The upgrade problem, does it really have an answer? If you want stability don't make changes (except security I suppose). Doing anything generally carries more risk than doing nothing in the upgrade sense. The only answer is perfection.

Re:I agree

[bedessen]

Microsoft used the BSD networking code in the kernel for its first version of Windows NT. Starting with NT v3.5 in 1994, the TCP/IP stack was completely rewritten from scratch by Microsoft using no BSD code. There is still some BSD code in use in the userspace utilities, like ping, telnet, et cetera.

Can we PLEASE put to rest this notion that any version of Windows made in the last 10 years has any kind of BSD TCP/IP stack code in it?

SOURCE

Requiem for the FUD

[Anonymous Coward]

// Please *don't* mod this up. It has already been done! Thx

... facts are facts. ;)

FreeBSD:
FreeBSD, Stealth-Growth Open Source Project (Jun 2004)
"FreeBSD has dramatically increased its market penetration over the last year."
Nearly 2.5 Million Active Sites running FreeBSD (Jun 2004)
"[FreeBSD] has secured a strong foothold with the hosting community and continues to grow, gaining over a million hostnames and half a million active sites since July 2003."
What's New in the FreeBSD Network Stack (Sep 2004)
"FreeBSD can now route 1Mpps on a 2.8GHz Xeon whilst Linux can't do much more than 100kpps."

NetBSD:
NetBSD sets Internet2 Land Speed World Record (May 2004)
NetBSD again sets Internet2 Land Speed World Record (30 Sep 2004)

OpenBSD:
OpenBSD Widens Its Scope (Nov 2004)
Review: OpenBSD 3.6 shows steady improvement (Nov 2004)

*BSD in general:
Deep study: The world's safest computing environment (Nov 2004)
"The world's safest and most secure 24/7 online computing environment - operating system plus applications - is proving to be the Open Source platform of BSD (Berkeley Software Distribution) and the Mac OS X based on Darwin."
..and last but not least, we have the cutest mascot as well - undisputedly. ;)

--
Being able to read *other people's* source code is a nice thing, not a 'fundamental freedom'.

Re:First post "bsd is dying"

[Bigos]

I've been using linux for couple of years, several times i thought about trying one of bsd systems, but didn't have enough courage to reccucitate it. If i see something that i like about them, then i might give it a try though.

Is it just me?

[ndogg]

Is it just me or do BSD people dole out more insults to each other than the Linux community does to them?

Not only that, but most of the jokes I hear from Linux people are often in jest, and not serious in any manner.

Re:Is it just me?

[endx7]

The trolls probably give out the most insults, and who knows whose side they are on. Probably their own.

Re:Is it just me?

[mickyflynn]

No, that is in fact the case. Theo is usually the butt of many jokes and jabs, at least in my experience. While Linux users tend to focus on crap like 'GNOME V KDE,' we often make it very personal very quickly. 'Your software sucks, and so do you 'cause its your fault'

Re:Is it just me?

[Brandybuck]

It's just you.

Now pardon me while I mop up the buckets of blood here. That's the last time I ever invite both Redhat and Debian advocates to my annual Friendly Package Manager Discussion party...

Re:Is it just me?

[Nelson]

At the end of the day, that's the difference between Linux and the BSDs. Alan, Andrew and numerous others are content to just publish a patch set for their stuff if it's not accepted in to the mainline, Linus is fairly consistent, you want a debugger, for example, then make it work on all of the supported platforms because that's what he wants, it's nothing personal and it's usually a fairly technical decision when you get to the end of it... (It chaps me that so many people point out areas where patches aren't being accepted without understanding that.) So then you maintain it outside the main kernel tree or you finish the job.

In BSD land, it seems like it quickly becomes a matter of personal pride and other things so you take the whole tree and start your own distribution..

I like the BSDs, I think they are doing great things and bringing good things to the world, same for Linux. I think that there may be a certain psychology at work though. I'm just throwing an idea out there. As Linux has become larger and more accepted and successful, certain types of geeks felt the need to switch to something else, like it was too mainstream for them. Some of them chalk it up to a "corporateness" of Linux, whatever that means, some chalk it up to whatever the masses are using must suck, some say that with the bigger companies and larger user base they don't like the directions that Linux is moving in (I'm not entirely sure I know that that means but I've heard it) and I think some people just like to be different, it makes them feel better for whatever reasons. Linux went through this phase a few years back when there was seemingly a different distribution every week, there are still a huge number of them but for a while you could go to different LUG meetings and such and there would be some new fangled linux dist every week, most have died LSB has come to be accepted and nobody really cares that much any more, Linux is Linux in many ways. BSD has been stuck in this mode a while; to be honest, I expect it to become much worse before it starts to get better, if it ever does. The difference was that the linux distribution frenzy was largely made up of single people or smaller groups trying to brand something, trying to come up with a name for themselves and some small piece of vision that wasn't being satisfied by Redhat or the other big boys. In BSD it's the actual kernel and distribution maintainers that have become angry with each other and taken their toys to a new snadbox to play in. It's an advantage (not that there aren't costs for it) for splitting the kernel from the distribution, now if there were a set of rifts within the linux development community and then a set of rifts within the binutils team and then a set of rifts within the glibc team and then a set of rifts within the compiler team then you could see some real fractures to the Linux process.

Re:Is it just me?

[evilviper]

Is it just me or do BSD people dole out more insults to each other than the Linux community does to them?

I certainly understand how it might look that way, but it's really not the case.

The BSD community at large are painfully honest. When somebody complains about some missing feature, you usually hear "Yeah, it's too bad we don't have that, you should use something else if it's important to you." Meanwhile, in the Linux world, even with practically the same complaint, you'd hear "You shouldn't be using that, and we don't want it in our system."

Now, that could be because a program is missing certain functionality, because the kernel is missing some feature, etc.

Unfortunately, what I've seen of Linux users bashing BSD, is always uninformed nonsense. I think the most popular one is complaints about the lack of a GUI installer from people who have probably never even used it. As if the BSD installers somehow aren't usable just because they don't have RedHat/SuSE logos for you to look at while your partitions are being formatted.

So, there is a huge difference between the criticisms you hear about BSD from BSDers, and the criticisms you hear about BSD from Linuxers.

To be fair, 5.x has been botched

[Ars-Fartsica]

It has taken the FreeBSD team literally years to get 5.x to an acceptable stage, which is reminiscent of the 3.x issues. Contrary to popular myth, FreeBSD goes through sustained periods in which the latest release is a very weak product.

Also, the development is getting very political, this also scares off people.

Re:To be fair, 5.x has been botched

[cperciva]

the development is getting very political

It is?

Ok, I can't say that I'm the most politically savvy of people, so maybe there's a lot of politics which has whooshed over my head, but... jeez, I had no idea.

It's a sad day when a FreeBSD committer learns something about the internals of the FreeBSD project from slashdot.

Re:To be fair, 5.x has been botched

[bcrowell]

It has taken the FreeBSD team literally years to get 5.x to an acceptable stage, which is reminiscent of the 3.x issues. Contrary to popular myth, FreeBSD goes through sustained periods in which the latest release is a very weak product.
Sure, there have apparently been a lot of very difficult problems with SMP in 5.x. But why is that an issue that we should be concerned about as users? Personally, I don't use SMP, and 5.3 has worked great for me as a desktop system. If 5.x doesn't work for you, keep running 4.x, which is very stable, and is going to be supported for a long time to come.

Re:To be fair, 5.x has been botched

[puzzled]

You still sound a bit like you're posting by the light coming through the planks in the bridge, but this one is a little more fair.

I've used all twelve versions of the FreeBSD 4.x OS. I've tolerated ... briefly ... the first three releases of the 5.x OS. I now have a 5.3 box in production, which I'd slay this afternoon with a 4.11 install disk if I were just a tad less lazy.

4.x good. 5.x becoming good, but not yet ready for prime time.

Re:To be fair, 5.x has been botched

[gstoddart]

It's a sad day when a FreeBSD committer learns something about the internals of the FreeBSD project from slashdot.

See, if you weren't so busy making sure I have a reliable OS you'd have had time to read the memos.

Re:To be fair, 5.x has been botched

[Anonymous Coward]

A specious argument. 4-RELEASE is the "latest release". 5-* is the next version up to bat, as it were.

Re:To be fair, 5.x has been botched

[Brandybuck]

Don't worry about it. The people who where trying to make it political went off and created their own fork with a bug for a mascot :-)

That's the only political brouhaha I can think of recently, and to be fair, it's largely been confined to advocates and not the developers.

Innovative death cycle

[Anonymous Coward]

What really sets FreeBSD apart is its robust death cycle. No other BSD at any price dies so reliably and consistently, with painless migration between deaths. It's clear that the FreeBSD development team has death as its highest priority and the result is easy to see in the product.

Re:Incredible desktop support?

[Nine Tenths of The W]

Why FreeBSD? Why go half-way? Why not switch to Mac OS X? It has everything that FreeBSD has, plus the best GUI there is or ever will be!

Because you're forced to use Apple's overpriced hardware.

Don't focus on microbenchmarks.

[HEMI426]

Scott has several good points. FreeBSD still has the same level of polish, the same amount of "professional" feel as it always has and it's just as consistent as before. The documentation is fabulous, Netgraph can do a lot of neat tricks, GEOM handles storage pretty well, vendor support is improving, etc. However, I think the most important one is discovered if you read between the lines: "don't focus on microbenchmarks."

Getting defensive?

[idiotnot]

NetBSD 2.0 is a higher-quality release than FreeBSD 5.3 on the IA32 platform. There's just no other way to put it.

My experience with FreeBSD is that the 4.x branch is rock-solid stable, fast, and everything works as it's supposed to.

NetBSD has basically reached that level of quality, with better performance.

FreeBSD 5.x has been unstable for me at best. While the userland programs are pretty much the same, the kernel-level changes have killed reliability. Furthermore, some of the much-touted new features simply do not work yet. I'm sure the SMP performance is much better, but I don't have many SMP machines. I've had problems with hard lockups, just doing things like trying to combine vlan and pf. The bridge interface, afaik, also, still doesn't work with pf.

As far as packages go, ports has more packages, true. Still, rarely has there been something not in pkgsrc that I absolutely needed. Pkgsrc is also much easier to work with, and far more friendly when it comes time to upgrade things. Portupgrade is an abortion, especially compared to even *gack* portage from ricerloonix.

There are reasons there's a buzz around NetBSD these days -- and reasons FreeBSD isn't getting the love it used to. I don't know whether the FreeBSD developers bit off more than they can chew, or if they just are rushing things out the door. But until they get their act together and put out a 5.x-RELEASE that truly is release-quality (by which I mean, all the features *work*, and the drivers are supported the same way), I'm going to be using NetBSD and advising my friends to do the same.

Re:Getting defensive?

[krreagan]

All I know is that I use FreeBSD 5.3 on three different machines, A laptop, a server and a workstation and on all three they are _very_ rock solid.

I use the portupgrade facilities all the time and have not found anything else as easy to use. On several occations since 5.3 was released I'd set off my workstation to upgrade all userland ports (portupgrade -a) on Friday as I leave and have come to work on Monday and have a complete updated system. This is with both KDE and GNOME being updated along with many other ports. I also build my laptop stuff (kernel, world & ports) on my server and only install on my laptop. All of this with less then 5 minutes at the command line.
I have never tried NetBSD or OpenBSD but have a lot of respect for both of them. I find FreeBSD brain-dead simple to maintain and is as rock solid as ever FMP. I have not found 5.3 to be any less "solid" then 4.10, Which is the last 4.x that I used (also on all three of my machines).

We are BSD! lets not let the Linux factor creep in!

Re:Getting defensive?

[Mr.Ned]

As far as packages go, pkgsrc just doesn't cut it. The quarterly branches simply do not get updates in a timely manner if at all; there's no other way to put it. Upgrading packages was a pain - there are several utilities to help with the process, but none of them do a particularly good job, certainly not like portupgrade does. Getting the newest versions of sotware is also a mess - some people test things in pkgsrc-wip on sourceforge, others just put it in the tree.

FreeBSD-5.3 wasn't the most successful release they've had. Most of the complaints I've seen on the mailing lists were fixed in 5-STABLE shortly after release, and the fast-approaching 5.4 will incorporate those changes.

Re:Getting defensive?

[hugo_pt]

pkgsrc is easier to work than ports? I don't understand what could be easier than cd /usr/ports/category/portname ; make install clean But that's just me..

Re:Getting defensive?

[molnarcs]

Yeah, they may have been a lots of problems with 5.3-release - I don't know any for I didn't experience them, but still. On the other hand, most of those problems have been solved in STABLE, and 5.4 will be a kick-ass release. I described one improvement here

As to FreeBSD not getting the love it used to - well, the massive amount of flames developers/users get by anonymous cowards might have given you that impression, but I don't think this is the case. The community/user-base is growing at a fast pace (if bsdforums membership growth can be taken as a sign of that - even though I realize bsdforums covers all bsd's, still the largest number of questions come from users trying out FreeBSD) Also, how much of that buzz is created by NetBSD folks?

Re:Getting defensive?

[Ragica]

NetBSD 2.0 looks to be awesome. We still have one development machine running it; though really that server doesn't get much action. Even so, it was a fairly painful process at times getting to the 2.0 level...

We moved most of our servers from NetBSD to FreeBSD a while back. The main reason was ports vs pkgsrc. As an ISP we use them a lot. How many times were we messed up by someone hastily typing "make update"... NetBSD deleting all dependencies and then failing to re-build? Ouch.

Not sure what your problem with portupgrade is. It is a bit of a resource hog, but it works beautifully with an excellent feature set: support for defining fine grained control over the build processes, including hooks for running scripts before and after builds, setting up the environment on a per-port basis.

And while the NetBSD pkgsrc team work very hard and I have a lot of respect for them and the help the provided me in several instances in the past, there's just a lot larger community working on the BSD ports. Linux has a (much) larger community again, but Linux isn't BSD...

The other reason we moved to FreeBSD was for stability, as NetBSD was having a lot of problems with our machines at the time (around 1.6, with SCSI in particular). I believe these have been resolved by now, but they were extremely painful at the time.

We moved to FreeBSD 4.x, which we are still on. I regret though that I didn't have the nerve to go with the relatively new 5.x branch at the time. I've run 5.x on my desktop for a long time now, and it certainly has had some problems along the way... but nothing insurmountable (for me at least)... and the track that 5.x is on, though it may be inching towards it, is really exciting in a way that NetBSD never was for me (but this is merely a matter of personal perspective).

Anyhow, have fun on your NetBSD. It's a great O/S! And 2.0 is extremely impressive at what it does, as you point out.

As for me, I'm looking forward to the latest SMP developments, which already are good, but are looking better and better all the time. And GEOM is another FreeBSD technology, as the linked article describes, that I'm really hoping to utilise for our infrastructure...

But all the BSD's are simply great.

Re:Getting defensive?

[laffer1]

Some users report problems with NetBSD as well. I think all of the BSD projects are going through a transition period. There is a need. FreeBSD is getting ready for dual core processors. The SMP support is good. I have a dual xeon 2.0ghz box here and its very nice. On a single cpu system, it is much slower than 4.x or presumably netbsd. (i've only used sparc ports of netbsd) Fine grained locking will slow down a uni.. they knew that going into it. The primary market for freebsd is still servers and many servers have 2 processors or more. Some have HTT enabled. It makes sense to focus on SMP at this point. In 2 years everyone will wish netbsd had done it too! Personally, I think all the bsd systems need work on the desktop front. Linux is still by far ahead in this area. Since XORG and others have stopped work on DRI for modern video cards we are stuck with no hardware acceleration in freebsd at the moment. (ati at least) That really hurts with KDE or gaming. My gentoo install runs circles around freebsd 5-stable (post 5.3) and i built a custom 2.6 kernel in gentoo. :) With time, FreeBSD will mature. I think by 6.0 release, we will have a good chance to compete with Linux. Linux development seems stalled as its so commercial. Look for a Lamer friendly distro of linux thats FREE!!!!! I tried to get my mother to switch to linux, but she saw it costs money and didn't want to switch. I could not convince her otherwise. Thats a real issue. In contrast, hit any bsd site and look for a "fee" attached. (apple excluded of course)

VPS Services?

[KidSock]

Anyone know where I can get root on FreeBSD for ~$20/month? Right now I'm using a Linux hoster and I'm happy with it but I'd be happier with FreeBSD for something on the Internet.

Re:VPS Services?

[vena]

i use jvds.com and have had no problems in the year and a half i've had a freebsd vps there.

Re:VPS Services?

[Offwhite98]

That looks like a great price on hosting. I am very much interested. I would like to know more about your experiences with that service so far. I have been hunting for a new hosting service on FreeBSD. Once they can run FreeBSD 5.4 with Mono and Java 1.4 I am going to switch.

Currently I run my own server and co-locate it. It is 4.10 now and I really, really want to run Java and ASP.NET apps.

Re:VPS Services?

[vena]

well i've never actually noticed any downtime, to be quite honest. i initially used the service to learn FreeBSD, and while the service isn't so managed that they'll hold your hand, they are extremely helpful with support by email or AOL IM. i'm currently on an unmetred freebsd vps for $30/m, i get transfer rates of about 150kb/s. they're true to their word about no more than 16 VPSs per CPU, which gives me pleanty of processing power for my needs

Re:Reliability of ports?

[cperciva]

A decent number of them are marked BROKEN.

If by "a decent number of them", you mean "1.5% of them" (192 / 12396 at last count), sure.

Gentoo has superior coverage in portage.

Gentoo may have fewer ports which are marked as BROKEN at any given time; but does it actually have fewer broken ports?

Their new logo says it all

[Ober]

The one here pretty much says it all.

Re:Their new logo says it all

[mickyflynn]

This says it better.

Guys, please!

[Epsillon]

This is BSD. Sod avocacy if it means in-fighting, mud-slinging, politics and such. We're not Linux or Microsoft so just STFU, code and enjoy. Don't make me come over there... ;-)

Re:Guys, please!

[Brandybuck]

Here's the new rundown of the BSDs:

FreeBSD: Most performance
NetBSD: Most portable
OpenBSD: Most stability
DragonflyBSD: Most trolls

Re:Guys, please!

[setagllib]

FreeBSD: Most engineering towards technology for corporations (hardcore SMP, GEOM, etc.)
NetBSD: Cleanest code of any system in existence, very portable (though not yet port/ed/ to all archs), very good performer on UP.
OpenBSD: Security you can bet your manhood on, and possibly the biggest advocate of openness and freedom (won't accept anything non-free for a minute longer than necessary).
DragonFlyBSD: Most innovative and outspoken, least real-world results to date ;)

Re:Guys, please!

[Epsillon]

Indeed, I recall a "showdown" between NetBSD and FreeBSD where the [stupid] reviewer left WITNESS and all the debugging symbols in the kernel of FreeBSD, then proclaimed loudly that the performance sucked. Well, forgive me for being realistic here, but the reviewer was in need of percussive maitenance with the size 10 clue stick.

Now, some of these folks may have a point in certain circumstances such as tightly controlled benchmarks but, please, let's retain a little objectivity. So there's arguing on the lists. So what? It was ever thus. I give you COLA as an example of what "advocacy" does for an OS and rest my case.

IMHO, Scott could do with a knock from the size 5 clue-stick, too. Rebutting these claims lends credence to them and tends to encourage folks to engage in mindless discussions (like this one) instead of doing what the BSDs do best: Just work. Let's just concentrate on getting things working and leave this style of "advocacy" for the Linux distros, eh?

Re:Guys, please!

[Epsillon]

Smorgreff

Nice. If you're going to try to look smart at DES' expense, at least get his name right. It's Smørgrav. I suppose you're going to start on PHK next? How about the completely mindless sport of picking on Søren?

These three are amongst the most productive developers on the project. Arguments are inevitable, especially on group projects like the BSDs. The real trick, which these have mastered, is not to let them get in the way of high quality code. Your comments are typical of the way this style of "advocacy" damages reputations in fairly short order, usually the reputations of those engaging in it.

Re:Guys, please!

[Epsillon]

How about you just don't worry your pretty head about Linux at all, eh? For fuck's sake why does every BSD zealot have to bring up Linux all the time. It is really damn annoying. Enough with Linux. This is a FreeBSD thread, got it? Unless you are going to make an objective technical comparison of the two, just shut the hell up and leave the uninformed, biased, anecdotal crap that spews out of your ass in the toilet where it belongs.

Good point, well made. I apologise unreservedly for doing exactly what I proclaim to be against.

Re:Guys, please!

[setagllib]

Did you even read my post? What the hell did I say that was incorrect and trolling? You need a really, really big chill pill. Or a lethal injection.

just to be clear

[mqx]

The NetBSD team were not criticising FreeBSD: basically, NetBSD stepped up their advocacy as part of NetBSD 2.0 release, including some whitepapers on performance comparision between NetBSD and FreeBSD. If anything, the BSD camps all have decent respect for each other, and honestly, Scott suggested that there was more animosity from the NetBSD camp that I think is the case in reality. All of the BSD camps could do with better advocacy, and Scott's post is more an indication that none of them are doing very good marketing, and as soon as NetBSD stepped up the marketing, the other camps (i.e. FreeBSD) felt they weren't getting a good rap: but really, the issue is, that FreeBSD guys just haven't been out there pushing their case as hard as they should really be.

I'm Trying OpenBSD...

[eno2001]

...due to the heavy trolling I got last week regarding my comment that OS security and usability are 50% admin skill and 50% OS distributor integrity.

I'm learning more and more that OpenBSD definitely needs an admin that is more highly skilled admin than most Windows or Linux admins. I've definitely made progress in my implementing of OpenBSD, but I still say that my axiom holds true (see my SIG): With most OSes, if you have a competent admin, then you can have a secure system. OpenBSD might up the ante with oddball features to ensure security, but until those are implemented in other mainstream systems, they don't apply. Additionally, you really need to have very strong Unix skill to use OpenBSD, so it flies right in the face of my theory. Where most OSes would have the admin skills required at 50% competency, OpenBSD requires something more on the order of 80% competency in order to get a usable box.

Re:I'm Trying OpenBSD...

[ulib]

>With most OSes, if you have a competent admin, then you can have a secure system.

This is totally false. If there's a bug that's exploitable as a remote hole and a malicious fellow discovers it before you, even if you're the most competent sysadmin on earth, you're still screwed.

For a sysadmin that's paranoid about security there's one, easy answer: OpenBSD. "Only one remote hole in the default install, in more than 8 years": it's a slogan, but it's true.
For somebody who's not paranoid and who's normally concerned about security, any BSD would do. This study suggests that when security matters the *BSDs (& the Mac) are the best choice.
The world's safest computing environment (Nov 2004)
"The world's safest and most secure 24/7 online computing environment - operating system plus applications - is proving to be the Open Source platform of BSD (Berkeley Software Distribution) and the Mac OS X based on Darwin."
--
Requiem for the FUD

Re:I'm Trying OpenBSD...

[eno2001]

This is totally false. If there's a bug that's exploitable as a remote hole and a malicious fellow discovers it before you, even if you're the most competent sysadmin on earth, you're still screwed.

This is where I got with the troll. However, what you said is also false. If I set up a Windows box and do *everything possible* to secure it, then it can be 100% secure. The primary things I would do with a Windows box would be:

1. Patch it up to the latest SP/hotfix level
2. Turn off all unneeded services
3. Put it behind a decent firewall (dedicated hardware or a *nix firewall)
4. Make sure to set up auditing both within the OS and on the firewall that traffic passes through
5. External IDS, etc...

With a very few exceptions, there is nothing that can get past this kind of setup. Even if an exploit of, say, IIS happens but I only allow port 80 traffic in, then there is no way that an attacker can do much with that box. 100% secure by most non-paranoid definitions. By a paranoid definition, NO OS is secure. I don't trust OpenBSD 100% just because it has a secure track record. In fact I would say that I wouldn't trust an OpenBSD installation unless I completely recompiled everything from source myself. I don't trust software that is behind the latest version because older stuff tends to have holes in it no matter how much it's audited.

Re:I'm Trying OpenBSD...

[timmarhy]

only one remote hole in 8 years is total FUD sure maybe in the base system, but hell i can do the same with freebsd by just blocking everything, but people install shit like bind and apache and think the above applies to it as well. oh an i noticed a poster above complaining about ego? fuck me have a look at the openbsd mailing lists. you want ego try theo on for size.

Re:I'm Trying OpenBSD...

[alpha_foobar]

So you just said that any OS is secure, as long as you set it up sensibly... and then place it behind a *Nix based Firewall?

I disagree... it really does depend what your talking about though. If you are running a Microsoft Web Server... and you are Microsoft, then sure, on occassion you are going to have to protect your services with some good linux or *bsd based firewalls... But for most of the time and for most Windows Desktop users, just setting up their machine sensibly and locking down unused and unrequired ports will be satisfactory.

I am also unsure what the big difference for an end user is between the main BSD's. They seemed pretty similar to me, as far as setup and configuration and compatibility with other *Nix open source projects. But I would be interested in clarification of real differences.

Re:I'm Trying OpenBSD...

[0racle]

Bind and Apache are included in the base system. They are audited and patched just like the rest of the base install, so having them in OpenBSD does not change the exploitability of the system. Secondly, the phrase is 'Only one remote hole in the default install, in more than 8 years!' Which means that while you could install FreeBSD and bring up a firewall that blocks everything, it is not the default distribution method. The benefit to the OpenBSD default everything off is that the admin must explicitly enable services and so its much easier to keep a secure system, whereas a system where you have to start off with shutting things off there is a higher chance that you could miss something.

Re:I'm Trying OpenBSD...

[eno2001]

That's pretty much my sentiment. I can secure any OS as long as I take the proper precautions, so OS doesn't really matter. It's 50% admin abilities and 50% target OS integrity.

The original thread last week was specifically addressing web servers. I argued that I could take a Windows server (even though I'm not an MS fan and don't trust their security) and secure it properly so that it would be just as secure as any other well implemented web server. The [ahem] "conversation" devolved into issues regarding desktops, etc... My main point still stands though. If you know what you're doing, you can put any OS online as a web server and have it be secure even if it is Windows/IIS.

Regarding any OS for desktop use, I would never put them directly online. They should be behind a hardware firewall of some kind as well as a local software firewall. With servers, I trust most *nixes to be able to handle directly on the internet connections more securely. For example, I've been using a Linux based firewall since 2001 with no problem. But it's highly customized and has NO external services running. No telnet, web, ssh, nothing. The only way to access it is via a serial console cable. That cable is plugged into another box that is on my network and that box is only accessible via a secure encrypted tunnel to specific IP addresses. I feel confident that this setup is just as secure as OpenBSD because there is little that anyone can do to get into the box. And in the even that they did, there is little they can do once they are on the box.

i found one

[Anonymous Coward]

this is a really nice case for BSD!

Embedded Features?

[sixoseven]

As one who is not particularly up to date on these matters, I wonder why so much development must occur? It sounds like Bill Gates argument for bundling IE with Windows: If we don't add more features to the core operating system, we'll die.

But hey, I still use ksh and vi, so what do I know?

Re:Does FreeBSD really need to prove itself?

[ArbitraryConstant]

"so of what consequence are NetBSD's criticisms?"

Just because NetBSD has fewer users doesn't mean its criticisms are without consequence. After all, by that logic FreeBSD's criticisms of Linux would also be without consequence.

indeed

[FidelCatsro]

if BSD is dying /dead , then its one hell of a zombie.
I use three OSs, debian GNU/linux , freeBSD and Mac OS X.. and i think all three are as healthy as ever
im not sure on the whole of apples market share I think about 5% , but considering that OS X has its roots firmly in BSD from its NeXT heritage not to mention the programs it has from the FreeBSD project, then its safe to say that BSD is more alive than ever .

Re:indeed

[nightski]

Oh, yeah, FidelCastro is using it, so it must not be dying! :-) Seriously though, I always think it is funny when someone argues that something is not dying just because they have it on their machine.

Re:indeed

[FidelCatsro]

yes me and the other few million apple users as i said above.
you do know its commen on here to say things like that under the guise of a anon.
even if i were the only person still using it , that would still mean it was alive .

Re:indeed

[ShieldW0lf]

Perhaps they judge it based on the quality of it and on the committment of the manufacturer. It doesn't need to have a growing userbase to remain a healthy project, you know.

This isn't Wall Street.

A one time try is all that's needed for success

[Zedrick]

I've been using Linux since around '96 something, first Redhat, then Slackware and recently Gentoo when I got my AMD64. I tried FreeBSD for the first time a few months ago when I had an old 200mhz machine that I just wanted to use for something, and since that seemed to work ok (a very basic install, no X or anything like that) I decided to give FreeBSD/AMD64 a try when I had to do a reinstallation due to hardware changes.

I downloaded a minimal boot CD, burned in, booted installed the base system over FTP and then X, KDE etc via ports...

After only a few hours I was totally confused. Everything just worked!! Well, almost everything. I had some problems with the soundcard, that was solved thanks to great documentation pointing me to a very logical solution.

I'm still a bit lightheaded. An operating system just can't be this good, I'm probably going to wake up soon.

Re:A one time try is all that's needed for success

[molnarcs]

Zedrick, you are not alone. I'm not a comp sci student (what I do is on my user page), but a few years ago I got curious about linux. So I installed rh 7.3 (which just came out fresh at that time). Later, I went through two mandrake releases (9.0 and 9.1). By that time, this whole nix thingy picked my interest, so I decided to learn the unix way - with the intent of setting up my first server. So I installed debian. Everytime I switched, I was presented with different sets of problems (different places for different setup files). I was dedicated to learning to use the command line and not rely on various frontends. One of the first concerns when one is intent to set up a server is the firewall. So I went to netfilter homepage, and bang! I tried and tried, and finally I went the frontend way: shorewall.

Downloaded documentation of shorewall, began reading it, noticed that it needs a newer kernel than the one installed (yeah, it was 2.4.18) with Debian. So I went on to pull that from the STABLE branch - which resulted in the disappearance of my /etc/networks. By that time I was so frustrated (by my attempts to get my usb mouse working) that I was beginning to look for alternatives. One sysadmin (knowing that my english is pretty good - not my native tongue mind you) recommended FreeBSD. I was pretty tentative at first, decided to try it out on a spare partition (luckily I had one primary) on my home computer, before putting it on my server. A week later, it was running my server (and instead of editing 3 separate files in shorewall, I could set up NAT + a pretty tight firewall in a few days). Two weeks later my linux partition was gone. A few months later I became active in the freebsd community at bsdforums. That was my second big surprise: it was the friendliest community I ever participated in (and I'm saying that coming from a mandrake background). And I've been happy ever since.

The reason for this longish post is that I think there is another potential user-base for FreeBSD: noobs who want to learn the unix way. It is one of the user-friendliest unix-like operating system. The entire layout, the configuration is so clean, elegant and easy (because it is logical) that learning it was a joyful activity, while I remember that I was quickly bored when trying to follow this or that howto or tutorial to learn linux. It is not for those who are looking for a quick replacement for windows (like I was at first) - but for anyone interest in the internals of a unix-like OS, FreeBSD is among the easier to understand and learn ones.

Have a nice stay in FreeBSD land :))))

Re:A one time try is all that's needed for success

[releppes]

Yes, you're not alone. That's the exact same sucess story I've had with NetBSD. I had an old laptop I wanted to make use of. I've tried many minimal Linux distros (Luit Linux being the best), but I was disatisfied with all of them. Under the hood they were little more than a horrible hack. I then knew I had to go with Debian to get any sort of control that I wanted. Even the much overhyped Debian I felt was just a hacked up unix. The package system was granular, but stupid. The dependancies were so intermixed that even the most trivial package install had dependancies going right back to installing the whole Gnome system. That and it was just unstable. I then downloaded two NetBSD boot floppies. I couldn't believe my eyes when jsut after two boot floppies, all my hardware was detected and working! I then proceeded to install the whole OS via internet. Smooth and fast and never a crash. Adding X and a few utilities was so damn easy. I was surprized that all apps were fairly recent. Not like the far outdated Debian stable releases. I'm getting a new laptop soon, and I'm concidering FreeBSD so I can see if I like that more, but from my general experience, the BSD world is a real treat from running a hacked up Linux box.

Re:A one time try is all that's needed for success

[RedWizzard]

I just saw your journal entry mentioning TV-capture:

I've captured the more interesting matches, using mencoder, and my very very cheap capture card (PlayTV MPEG II). Image quality is excellent. On my amd xp 2400+ I can capture into 528:396/1500Kb/s mpeg4(divx)avi, if I don't do anything else.

Just thought I'd mention that you should be able get better resolution than that. I'm running MythTV under Linux on a XP 2400+, and I can record two 576x576 MPEG4 streams simultaneously (2 capture cards) at about 95% cpu. I've got the "bitrate" set to 4400. Not sure whether that's Kb/s or what. I record the audio unencoded and then convert to mp3 later. Recording a single source I can do 768x576 at the same bitrate with mp3 audio and it still only hits about 50% cpu usage.

Re:Reliability of ports?

[bcrowell]

• A decent number of them are marked BROKEN.
  If by "a decent number of them", you mean "1.5% of them" (192 / 12396 at last count), sure.

It only takes one to make someone's life miserable :-) If I have 100 ports I want installed on my system, and one doesn't work, well, that's a big problem if that's a port I really need. This was what prompted me to switch my dekstop systems from FreeBSD to Debian recently (while leaving BSD on my server) -- I needed to run a recent version of Inkscape, and it was broken.

I don't really think there's one system that has the best availability of all software. Some stuff's broken on Debian. Some stuff's broken on FreeBSD. I'm sure some stuff's broken on Gentoo as well. (Dunno first-hand, because my attempt at installing Gentoo failed.) It's just a question of whether you can get the stuff running that you personally really really need.

One thing I am not happy with about the FreeBSD ports system is that updating one port tends to break lots of other ports in complicated ways that are hard to fix. Portupgrade is the solution that everyone seems to push for solving this type of problem, but in my experience, portupgrade is a good way to hopelessly mung your system. To be fair, I don't think this is entirely the fault of the FreeBSD project; the stuff that gets really difficult to deal with is the Gnome libraries, and I think it's because Gnome just isn't careful enough about QA and maintaining backwards-compatibility (or they value rapid development more than stability).

Re:Incredible desktop support?

[ArbitraryConstant]

"Why FreeBSD? Why go half-way? Why not switch to Mac OS X? It has everything that FreeBSD has"

Except PF, jails, ports, etc.

MacOS is an excellent desktop OS, but it can't touch FreeBSD as a server. Even the server edition is behind FreeBSD.

Re:Incredible desktop support?

[bcrowell]

Why FreeBSD? Why go half-way? Why not switch to Mac OS X? It has everything that FreeBSD has, plus the best GUI there is or ever will be!
Because if you're interested in running open-source desktop apps, OS X is a lame platform, and if you're interested in running a server, you don't need the eye candy of OS X.

Coincidentily

[defile]

I just posted an article that's been sitting around on my hard disk for awhile now (I'm testing out nanoblogger). It's about how I'd improve LAMP, but it ended up becoming an advertisement for FreeBSD.

Have a look if you can stand an honest critique of Linux (I love and run Linux on everything, so don't accuse me of FreeBSD shilling).

Re:Reliability of ports?

[epine]

To say that A is better than B does not demand that anything B can do, A can to better. One might in that case say that A "owns" or "dominates" B. As you observe there is no single "dominant" package system. There are still grounds to ask whether or not A is better than B.

Your argument becomes more interesting when one considers that "B" doesn't even exist. Does B refer to Woody, Woody plus random backports, or Debian testing? What is Debian testing if not a crap shoot in motion. Debian package system, on a good day, if ever released again, could be compared on rough equal terms to the FreeBSD ports collection. But please, let's not pretend that three year intervals between stable releases a package system makes.

Why?

[JustNiz]

Linux appears to be way more popular than BSD these days, and there appear to be more apps and hardware drivers available for Linux than BSD.

So why would anyone consider BSD over Linux?

Re:Why?

[ulib]

Already gave my answer to this - right here. :)
--
Being able to read *other people's* source code is a nice thing, not a 'fundamental freedom'.

Re:Why?

[hugo_pt]

6) no root exploits every month 7) decent codebase 8) organized filesystem layout 9) commits to the OS are closely monitored and quality-assured, unlike linux 10) an OS as a whole, not just a kernel.

Re:Why?

[hicsuget]

Windows appears to be way more popular than Linux these days, and there appear to be more apps and hardware drivers available for Windows than Linux.

So why would anyone consider Linux over Windows?

Furthermore, 27 million AOL users can't be wrong.

Jeez..

[ulib]

.. it's a *logo contest*. Kinda difficult to submit a new logo 2 days after the official announcement, I think.. :)
Anyway, for NetBSD's logo contest over 400 logos were submitted. Given FreeBSD's much wider user base, the numbers will probably be higher - and making a choice will be even harder.
--
Requiem for the FUD

Re:Jeez..

[ulib]

I disagree. To me, NetBSD logo looks very clean, tasteful and appropriate - come on, it's a *logo* we're talkin about. It should be sthg like that, IMHO.

Where's the Java

[kwerle]

I've been nothing but disappointed with FreeBSD's inability to get a binary ditribution of Java for the platform. I have used, and continue to use, FreeBSD for my main server, but next round will be OSX.

There are plenty of reasons why it hasn't happened, and plenty of workarounds - but I don't care (welcome to the customer).

Re:Where's the Java

[ririarte]

Yes, but there is a very dedicated team that manages to produce a very high quality J2SDK implementation, even if the installation involves compiling Java from scratch. 1.4.2-p7 runs a production Tomcat site like a dream here. I am very happy with OS X as as desktop OS, but i would not rate Apple's SDK as highly as FreeBSD's one, YMMV, of course. BTW, nobody knows if Apple intends to ship a 64bit JVM with Tiger, au contraire, FreeBSD's Java team has an already working, if early one, for AMD64.

--------Quick recipe to get up and running with Java under FreeBSD ------

a) Make sure to be running a modular kernel OR a kernel with linux compatibility enabled (Compile phase only)
b) Read http://lists.freebsd.org/pipermail/freebsd-java/
c) Make sure to have a relatively recent ports tree
d) cd /usr/ports/java/jdk14
e) make install
f) follow instructions.

Re:Where's the Java

[ririarte]

Time is never free, neither mine nor anyone i know.
I prefer to go the extra mile when setting up a server to know that it will work exactly as i want it to, and for that, FreeBSD is a magnificent platform; you seem to advocate for a prepackaged solution. Who is right ?. If it gets the job done, both of us.
Don't get me wrong, i do appreciate Linux, especcialy lightweight non RPM distros á la Slackware, i will, as time permits, experiment with Solaris 10 as well.
Still, if there is one thing lacking from FreeBSD, it's not Java but a native Oracle port (Not FreeBSD's fault of course)

Not Marked as Broken is Even Worse!!!

[Noksagt]

A decent number of them are marked BROKEN. The usefulness of ports is overrated. Gentoo has superior coverage in portage.

I use FreeBSD on my servers and gentoo on my desktop. I like both of them. But your argument is flamebait.

Gentoo uses more bleeding edge packages than FreeBSD. Even in using the stable branch, I've downloaded borked packaged more than once. While the ports in FreeBSD are order, they are tested MUCH more & the broken packages are actually labeled broken!

Portage does have some advantages over ports. Package stability is not one of them.

Re:My OS is better than your OS. Sheesh!

[Wild_dog!]

Here Here

And what about just having a choice or preference.
Doesn't it make sense to just use what you like.

I have never used BSD, but I know lots of folks who do and love it. Great!

I know lots of people who love Windows XP....great!

I know lots of people who love various flavors of Linux. Awesome.

Why is it bad for people to choose what they like?

I would hate to think that I had to choose something cause someone else decided it was best....for everyone.

Re:Reliability of ports?

[shutdown -p now]

Relative to the size of the ports/portage tree, yes, it does.

acpi

[phrasebook]

"A team of FreeBSD developers works closely with engineers at Intel to provide the best ACPI power management support available in an open source operating system."

Is this true? I would really like S3 suspend/resume to work. I can't make it happen cleanly with linux 2.6.10. Does FreeBSD do a better job? From reading section 11.16.3.2 Suspend/Resume in the FreeBSD manual, it doesn't sound like driver support is much better than Linux. Anyone have good ACPI experiences?

Re:acpi

[Ecks]

I have S3 suspend/resume working on both a Gateway 450ROG and a Toshiba 4600. It was as simple as adding a few lines to /etc/sysctl.conf. I don't have any experience with Linux but the ACPI support in FreeBSD 5-STABLE is what convinced me to upgrade from 4-STABLE.

--Ecks

Why perpetuate myths

[Wild_dog!]

Everyone says apples hardware is overpriced. Prove it.

Maybe their hardware is actually worth more for longer.

Both chrysler and BMW make cars, but BMW's cost more generally...why?

It would be nice if people would be more rational about hardware and quit parroting lame statements that don't make sense.

Re:Why perpetuate myths

[vox_gabrieli]

I don't need a BMW car, as my Chrysler meets my needs and wants. I don't want to spend twice as much for a BMW. I test drove a BMW once, and I liked the leather seats. I think it's a reasonable expectation that I can pay a little extra for a Chrysler with leather seats without having to buy a BMW.

I love OSX, but buying Apple computers is like buying a BMW just to get leather seats.

Re:Why perpetuate myths

[evilviper]

Both chrysler and BMW make cars, but BMW's cost more generally...why?

Because BMWs are a status symbol, and those with lots of money want to show off their money by buying an expensive car.

Between BMWs and Chryslers, the differences are most superficial... In one you'll get leather seats, but they both function equally well as a CAR, and one in much less expensive than the other.

Re:Why perpetuate myths

[mollymoo]

If you think all a BMW has to offer you is leather seats then get leather seats for your Chrysler and be happy.

I've got a Mac, an iBook (and home-built x86 boxen running Linux and Windows, I'm no Mac zealot). They aren't that expensive compared to comparable Windows laptops. A bit more, yes. You can certainly get an x86 laptop for much less, but comparable models aren't hugely cheaper. If I keep the iBook for two years the extra cost ammounts to about half a pint of beer a week.

I know where all the extra money went, it went on quality materials (plastics and mechanics - the computing hardware is pretty generic of course) and trillions of little details. That's where it goes in a BMW too. That's what separates great hotels from OK ones and sucessful spacecraft from lithobraking demonstrations. It's the details, and getting them all right is both hard and expensive. I think it's worth the cash. This may be my first Mac but it won't be my last.

Apple aren't the only maker of more expensive computers of course. You can pay more for a Sony x86 box than a Walmart special too. And I bet the Sony gets more of the details right.

Re:Why perpetuate myths

[merdark]

Although ironically, if you compare the price of a BMW to a car of 'equivalent performance and features', it's not over price at all. In fact, they're generally cheaper than cars such as, say, Mercedes.

If all you want is leather seats, you could just buy a Lada and put leather seats in it. Hell, maybe Ladas even have leather as an option these days.

It's all about what level of quality you want to settle for. If you want to criticise apple for not using shitty parts... uh... well... ok. I suppose they are guilty.

Re:Why perpetuate myths

[_damnit_]

Spoken like someone who has never owned a nice car. Chryslers are just not the same as BMWs, Mercedes or other high-end cars. I didn't buy mine to impress you or anyone. I tried one at my wife's insistance and fell in love with the handling and attention to details. In fact, the high end cars are nearly always the first to get the best new safety features.

As far as the high-end analogy applying to computer hardware, why not? How many inane box mod stories are posted so we can see someone who fit their PC in a fishbowl or added xenon headlights to their old AT case? People are willing to spend more to mod their box, just not pay more to start. I liken this to people who spend thousands to rice out their Dodge neon. Why?!? At least start out with something nice and improve on it.

Damn type-R Neons keep getting in my way when I try to get anywhere fast.

Re:Why perpetuate myths

[releppes]

So true, and it's why I'm still surfing the web on an intel/amd and not a G5. I just purchased a new laptop. I could have gotten anything I wanted as long as it's non-Windows. I seriously concidered making the jump to the Apple world. I so want to be there. Nice hardware, stable OS,....but then reality....I'm that low/middle income guy...I drive a $20k car...I wish I could drive a BMW....and I too wish I owned a Mac. Granted, I could buy a very cheap iBook for less than a grand, but compaired to a similar priced intel, that iBook looks like crap. When money becomes no object in my life, I'll be driving a BMW and owning a Mac. For now, I'm sadly living within my means. I'm not down playing Apple. I think they're the best, but it's like people who can afford a BMW saying to an average income guy like me, "you really need to own a car like this..." Why even say crap like that? When you can afford the best, you buy the best.

Fix the threading model and I am on board

[gatkinso]

'nuff said.

Re:Fix the threading model and I am on board

[Brandybuck]

What? You want the old Big Giant Lock back?

Re:Reliability of ports?

[puzzled]

What bridge do you live under when you're not posting to Slashdot?

http://www.freshports.org/

12,397 ports available
192 broken
600 others impaired in some sense

So that makes 11,600 of 12,400 or 93.5% ready to roll right out of the box. OK, maybe the stuff on the install CD isn't the most current - use the cvsup app provided on the CD, sync to one of the master sites, and *then* you've got 93.5%. Its a simple process that takes less than five minutes on a high speed connection.

Contrast this with the miserable mess that even the best Linux distros display and you'll see why FreeBSD is the choice for mature operators. Its all there, it all works, and you just don't have to mess with it.

Moving from Windows to Linux reduces perceived OS chaos by a certain amount. I got the same sort of relief when I converted from Redhat 6.2 to FreeBSD 4.0 some years ago.

GEOM IS BLACK MAGIC

[QuietRiot]

Where are all the geom HOWTOs?

The linked man page is "tasty" n'all, but details on implementing such magical wonders, until recently, have been rather scarce.

This man page is better than the one linked to in the original post. There's also some information from committer (read: major contributor to ggate ) Pawel Jakub Dawidek in Poland.

Not that the info isn't there now, right under man, but for a while it was all very vague.

When searching about all that is BSD, don't forget Google's special google.com/bsd section.

You can also search the freebsd-geom mail list archives to learn more.

geom-gate sure looks nifty! It's akin to block-level NFS (though that's most likely an extremely oversimplified view). All the fun things you can do with geom you can do over your network. Need worldwide distributed, encrypted, multi-level RAID? Go right ahead!

Pretty slick. We'll be hearing more about this.....

Zesiger License = GPL + BSD

[qwasty]

It's a shame no one uses a BSD-style license that's time-limited. That way, companies could do their proprietary thing, make their money, and only release their source after they've already milked their products for what they're worth.

The Zesiger License is the only license I know about that seems to achieve the goals of both the GPL and BSD licenses.

zerg

[Lord Omlette]

I'm happy that the FreeBSD people like their OS. Call me when they fix SMP.

Re:zerg

[Lord Omlette]

I think that's the problem. 5.0 cries for mommy under even moderate load, unless SMP is disabled... which sort of defeats the purpose. When version #s go up, things are supposed to get better.

Maybe NetBSD is sparse on features compared to FreeBSD, but NetBSD 2.0 was an improvement over previous versions of NetBSD, at least!

Re:zerg

[hawk]

I think that's the problem. 5.0 cries for mommy under even moderate load, unless SMP is disabled...

???

I've had this machien running SMP 5.0 for over three years, running 24/7.

I have no clue what you're talking about. Sure, mozilla can be slow to respond when the load is at about 30 or 50, but X remains crisp.

And yes, I have kept it at double-digit loads for days on end while still working on other things.

hawk

Re:zerg

[Lord Omlette]

eh, I have no experience running FreeBSD as a user machine. Try setting up apache and your choice of db-driven web application on it... Then have a few dozen people try to use it at the same time.

Re:zerg

[hawk]

Like phpbb for a busy class?

Yes, it does that at the same time.

hawk

Re:Incredible desktop support?

[jurv!s]

You're a little hasty calling OS X lame on the desktop and server. Browse here and here and here. OS X covers a much wider range of uses than any other *nix out there. On the desktop it can run most of the X11 sw out there + the big name sw from Adobe, MS, Macromedia, etc. On the server, you can tweak it just like any other OSS *nix since you have a CLI and free access to the source. OTOH, the GUI is an option for those lacking your elite hacking skills who would have used MS Server but heard this was better...

Re:Incredible desktop support?

[man_of_mr_e]

Off topic, but for your information:

Multiple desktops are trivial in any version of Windows. There are lots of free or very inexpensive 3rd party tools to give them to you.

There's of course the official MS powertoy but this is largely a piece of crap.

There's DeskSelect $9.95

There's Cool Desk $24.95

There's a whole section of them at tucows.

There's Multidesktop

There's Virtual Desk

There's Enable virtual desktop

Open Source, there's Virtual Desktop, Virtual Dimensions, VirtuaWin, etc... etc... etc..

Unless your company won't allow you to install any software on your local computer, there's no excuse to be whining about lack of virtual desktops.

Re:My OS is better than your OS. Sheesh!

[adavidm]

You're new here aren't you?

Wild_dog! (98536)

Oh, wait....

About politics rather than functionality

[David's Boy Toy]

One thing that always turned me off about the BSD projects was the amount of ego and politics. The only thing the BSD crowds hated more than Linux was the other versions of BSD. Meanwhile Linux supporters where much more concerned with the real threat, Microsoft.

Re:About politics rather than functionality

[geminidomino]

Meanwhile Linux supporters where much more concerned with the real threat, Microsoft.

Man, the old truism still rings true as ever...

"Linux is for people who hate Microsoft.
BSD is for people who love Unix."

Re:About politics rather than functionality

[setagllib]

Microsoft is not a real threat to BSD, since they don't really care about the end-luser market (as noted by some BSD advocates, the end-most of end users don't contribute code and most don't even donate funds, and so there's nothing tangible to gain from appealing to them). The server market IS where Linux, Solaris and the BSDs compete (I'm ignoring OSX at the moment, it's nowhere near free enough to count as a competitor), and the highest-end servers only run Linux and Solaris (right?), so BSD is only on the low->medium market... but the security and development model give it useful competition against Linux (which seems to be the highest performer).

The BSDs compete in a productive way. They share code and designs where practical (unless something conflicts with philosophy or goals), and their constant benchmarking against each other forces each to adress their problems. Some harsh words get said and some personal arguments occur, and this has even caused forks, but that's typically a fact of life. It shows the developers have human personalities, not just senses of humor and coding abilities.

See? Bright side to everything :)

FreeBSD / NetBSD

[Anonymous Coward]

Brian: Excuse me. Are you the Judean People's Front?
Reg: Fuck off! We're the People's Front of Judea

I'm always struck by the similarity.

Geom howtos

[dougnaka]

Root software raid via geom. http://people.freebsd.org/~rse/mirror/

And the short version of the same thing, but using a recovery CD instead of a live system http://dannyman.toldme.com/2005/01/24/freebsd-howt o-gmirror-system/

Kind of a coincidence that this gets posted today on /., as I've spent most of the morning setting up geom on a new 5.3 box, had used Vinum in the past on 4.x, and have loved FreeBSD for servers since 2.2.5

Re:Geom howtos

[QuietRiot]

Spectacular. Parent deserves a "+1 : Something". Many thanks for the link.

Have never used Vinum but have always been a fan of FreeBSD since my first intro (3.4??). Two production 4.x boxes in service at the moment with uptimes peaking above 200 days; downs due only to scheduled power outages and required hardware additions. Not overly impressive, but I certainly don't stay awake at night worrying about them.

[....Psst.....((random /. reader))..... If you haven't tried it, "the beast[ie]" makes a __really__ nice server...]

Re:Incredible desktop support?

[equilith]

Because you're forced to use Apples overpriced hardware.

I can't speak to the hardware issue, but you also have to pay Apple for major OS updates. In fairness, they do offer small, incremental updates for free, but I've gotten pretty used to having nigh-perpetual features and OS improvements for free.

-royce

Who knows?

[roystgnr]

When I was on an "advocacy kick", I spent a year (well, a few minutes every day for a year) answering Linux questions on Usenet. A year after that, I took a good part time job offer from someone who remembered seeing my name and college in those newsgroups. A year or two later, when I was hunting for finite element software to help with a class project, I downloaded the most appropriate program I could find and was surprised to find my name on the acknowledgements page, because apparantly I'd helped fix the author's first Linux installation.

Of course, this could be "random good luck" as much as "bread on the water", and it probably helped that my "advocacy" was helping others rather than just preaching to them, but I think the lesson was clear: free software users don't give you money, but some can give you respect and some can give you more software. That wouldn't be worth it if the respect and software were all you were interested in, of course; it's just a bit of added reward for doing something like rooting for a baseball team that some people find fun to begin with.

Re:Not Marked as Broken is Even Worse!!!

[archen]

I use Gentoo on the desktop, and BSD on servers as well. I'm not sure what you mean by "using stable". The base install has nothing to do with ports, differences between 4x and 5x aside..

Most of the popular stuff is about the same as far as how up to date they are. Apache, Postgresql, Samba - they're pretty much the same. Some of the fringe stuff is pretty far out of date though, but as you say, I would rather have something tested well before release then someone throwing the source in the ports tree just to keep it up to date.

My biggest problem with Gentoo is that many things for me are broken, but not because of something broken in portage, but because portage itself is part of the problem and causing screw-ups. Seems like I can't go more than 2 weeks without some mysterious breakage of an application that I will have to dig up a fix. Some things (like torsmo right now for me) will just sit broken until the next update. No one else has problems, just me. I see a lot of Gentoo users report the same sorts of problems, and I hope they'll work on that area a bit more in the future.

Re:QNX

[drinkypoo]

Your roadmap leads here... or perhaps here :)

Let's not forget that BSD is about cooperation

[ulib]

I think the other messages in the thread on FreeBSD-current are also worth a glance, especially this one by Robert Watson, which stresses the strong cooperation and code sharing that is actually connecting the BSD projects.
--
Being able to read *other people's* source code is a nice thing, not a 'fundamental freedom'.

Re:MODS?

[molnarcs]

Yeah a totally related copy and paste job of a few years old mailing list message posted over and over and over again here on slashdot as flamebait. Mode down please.

"FreeBSD 5.3 ships with x.org, which is **BROKEN**

[crmurphey]

A quote from the 2005/2/22 review FreeBSD vs. NetBSD: ready for primetime?:

...
* FreeBSD's installer has always been, and still is, better than NetBSD's.

* FreeBSD 5.3 ships with x.org, which is **BROKEN**. Sorry, guys, but you don't earn any rating but broken when you can't even produce a usable TWM display on a Radeon 9200.

* XOrg on FreeBSD 5.3-RELEASE (or -STABLE, take your pick) still dies on this machine. This indicates that insufficient testing was done before the switch to xorg. Obviously they can't test everything, but a radeon 9200 isn't THAT old!* FreeBSD 5.3 ships with x.org, which is **BROKEN**. Sorry, guys, but you don't earn any rating but broken when you can't even produce a usable TWM display on a Radeon 9200.
...

Re:Not Marked as Broken is Even Worse!!!

[molnarcs]

I agree with you - just some additional remarks about ports and portage.

In my experience, ports doesn't lag too far behind portage - it's somewhere between the portage stable branch and current. As the complexity of a package (and it's impact on other ports) grows, so does the time port maintainers spend testing them. Just to give a good idea of how much ports is up to date (or not):

If we take the GIMP for example, usually it is in ports the day it is announced. That speaks volumes of it's portability/cleanness of the code base. On the other hand, as important as it may be, it doesn't affect much other ports.

Then let's see KDE. KDE becomes part of ports (and the package repository) usually a few weeks after the announcment. If I remember correctly, one of the 3.2 releases was in ports some 3-4 days after release. On the other hand, 3.3.2 took 2 week to get there. That isn't much of a lag, now is it?

And finally: Xorg. Xorg affects many many ports, so there is usually a lot more time spent in testing then with the packages I just mentioned. We are still at 6.8.1, although 6.8.2 is coming as you can see from this mailing list post.

Generally, ports is quite up to date. There are weekly updates to OpenOffice.org 2.0 - probably because it is independent from other ports. Also, the most important package are updated pretty fast (I had PHP 5.0.3 running before the announcment of the security fix release hit slashdot). Others, however, lag behind somewhat. We don't have KDE 3.4beta in ports for example, while I guess it is already in portage.

As I said, I don't disagree with what you wrote, I just wanted to give a general impression about the freshness of ports for other readers (check out freshports to learn more.

Re:Incredible desktop support?

[ArbitraryConstant]

" Oh, really. I run OS X Server, with Darwin Ports, with Open Directory, etc., etc., etc.."

Darwin Ports and Fink are 3rd party addons, not supported by Apple.

"It blows FreeBSD away in ease of administration"

Ease of administration comes down to opinion.

"comes shipped on enterprise-class hardware"

FreeBSD can run on enterprise class hardware if you feel like buying enterprise class hardware.

"and runs the same software under the hood."

Except for the kernel, which is different, and the cool features like jails and PF.

Re:Not Marked as Broken is Even Worse!!!

[Noksagt]

I'm not sure what you mean by "using stable".

You're right. I wasn't clear at all. I meant that gentoo packages in say x86 are sometimes unstable enough that I would expect them to be in ~x86.

Most of the popular stuff is about the same as far as how up to date they are. Apache, Postgresql, Samba - they're pretty much the same.

The most popular apps are well tested for both distros. I still like FreeBSD's patches & obsessive maintenance--the packages that need to work together just seem to be more tested and mature.

My biggest problem with Gentoo is that many things for me are broken...Seems like I can't go more than 2 weeks without some mysterious breakage of an application that I will have to dig up a fix. Some things (like torsmo right now for me) will just sit broken until the next update...I see a lot of Gentoo users report the same sorts of problems, and I hope they'll work on that area a bit more in the future.

I agree completely.

Re:Developer Laments: What Killed FreeBSD

[speedbump]

I feel your pain Mike, but you haven't gone into specifics. It just looks right now like you've not gotten your way about certain directions the core is going, and you've taken your marbles home.

Good luck to ya, I hope you can take your expertise with BSD and make Apple's offering that much better. I just am saying that your post is lacking specifics.

Re:"FreeBSD 5.3 ships with x.org, which is **BROKE

[Brandybuck]

X.org on FreeBSD 5.4-RELEASE with a Radeon 9200 works just dandy for me. It's not broken, it works fine. Just because NetBSD is one of the lone holdouts for XFree86 doesn't mean that everything else is broken.

Re:"FreeBSD 5.3 ships with x.org, which is **BROKE

[Crawlin' King Snake]

I'm sorry, but that article is just crap. It's not even an article, it's something someone would scribble in a notepad while trying FreeBSD and NetBSD for the first time.

And he's completely wrong too. Just because he couldn't get X.org to work on his Radeon 9200 in FreeBSD does *NOT* mean that it's broken. I know for a fact that it *WORKS* because I installed it on a machine with with a Radeon 9200 and a GeForce FX5200 and it worked perfectly, right out of the box. Both the initial FreeBSD 5.3-RELEASE and the current patched release, not to mention -STABLE.

That "comparison" really is a poor choice to be used for making a point against FreeBSD

It's more a case of PEBCAK than anything else.

-CKS

P.S. PEBCAK == Problem Exists Between Chair And Keyboard, for those of you that don't know.

Poor Java support

[drodver]

Tried FreeBSD recently and gave up on it due to the difficulty getting Java working. On a machine of limited capabilities trying to compile it is too much. It claimed to need 1.7GB of space for the compile! I didn't believe it until the system slowed to a crawl and I found the build directory eating all free space on the partition.

Too bad, seemed good otherwise. I'm falling back to good old Gentoo for now.

Re:Poor Java support

[rainer_d]

> I didn't believe it

Oh-man. You didn't believe it and then you complain because the factual information you received from the system beforehand turned out to be correct ?
I thought that's a lesson a child learns, the first time they try to touch the hot cooker even though Mummy told them not to !
That being said, compiling Java keeps my Dual Xeon busy like no other compile-job, SCSI-drives and 1 GB of memory not withstanding...

Re:Please learn how to make links.

[DrSkwid]

fuck off, idiot

why would I link to a binary !

and, on top of it all, why bother linking for YOUR modus operandi ?

why not

wget -n `echo 'ftp://ftp.freebsd.org/pub/FreeBSD/ports/i386 /packages-stable/All/lynx-2.8.5.tgz' | tr -d ' '`

"Redundant" ?..

[ulib]

The guy asked: "So why would anyone consider BSD over Linux?"

Modding "redundant" my one line answer is just stupid - or dishonest.

Re:Reliability of ports?

[setagllib]

Gentoo plays dangerously with blockings and confusing dependencies, if you run 'tested' software only (and non-tested basically sets your machine on fire).

I have two examples. Firstly, when XFce 4.2.0 was just released, it was in Portage in a heartbeat. Some days later HALF of it was marked tested, and half was still untested (at least for x86). Since parts that were marked tested needed the untested ebuilds, they couldn't emerge; and since some other software (e.g. panel plugins) needed the older versions altogether, emerging them killed your newer versions (and hence your whole xfce4) and brought back older, incompatible ones.

While this is as much a failure of XFce for being such a disjointed piece of shit in the first place, Gentoo Portage falls way behind FreeBSD which Doesn't Care About Versions. Rightly so.

Second example, copy-n-paste:
[blocks B ] [ebuild U ] media-libs/freetype-2.1.9-r1 [2.1.5-r1] -bindist -debug -doc +zlib 969 kB
[ebuild U ] media-gfx/gimp-2.2.3 [2.0.4] -aalib (-altivec) -debug -doc +gimpprint +gtkhtml +jpeg +mmx -mng +png -python +sse -svg -tiff -wmf 13,486 kB

I can either have the new GIMP (2.2 was in the tree for one day buildable, then impossible to emerge without hacks for a month or so, now back as untested - even though every other package manager concedes that it is in fact very stable) and lose Firefox because a new freetype is needed, or keep the old GIMP and be able to use Firefox. Charming.

The worst example ever, is that even the tested ebuilds are often completely broken. libusb needs user intervention to work at all (try to emerge hpoj, for instance), and this is a known bug nobody has fixed in portage. It was marked tested in spite of this. The new openmotif (might have been fixed, but was still bad when first marked tested) couldn't be linked against at all - try emerging nedit. It was marked tested.

It just goes to show that, no matter how big a community you have testing your software, if they're complete f#%^ing morons it doesn't help your case. pkgsrc and FBSD Ports have recently shown a MUCH better stability record, even using bleeding-edge software, and without these mysterious breakages and blockings.

I like Gentoo Portage since it gracefully lists what it will do and has USE and all, but sometimes it makes me want to break down and cry.

Re:Reliability of ports?

[setagllib]

Grr @ HTML nature of Slashdot... the first line of that portage had "net-www/mozilla-firefox-1.0-r3 (from pkg media-libs/freetype-2.1.9-r1)" after a certain bracket I won't repeat.

Hint to Slashdot: Wake up, you can format things without needing HTML.

Sorry for parent

[ulib]

Parent is indeed redundant: my apologies to readers (especially BSD users) for posting it. I did it because I thought that the other equivalent message wouldn't get modded up, but it did.
I'll pay more attention in the future. :-)

Re:Incredible desktop support?

[ArbitraryConstant]

"FreeBSD runs none of them, so what enterprise class hardware *does* it run on?"

HP Proliant DL585
Sun Fire V40z

It doesn't run on the really big machines, but then neither does MacOS. So if I'm wrong about enterprise class hardware, then the person I was responding to is also wrong about enterprise class because the best XServe doesn't compare to the machines I've linked to.

"FreeBSD doesn't even understand NUMA, which is basically indispensable for Opteron, POWER, or any serious Intel IA64 or x86 platforms."

FreeBSD does understand NUMA with the ULE scheduler in 5.x.

I'd be shocked if MacOS X understood NUMA though, as Apple has never sold a NUMA box.

pros and cons

[discogravy]

Like all OSes -- including Microsoft Windows and Apple's OS X as well as the various Linux distributions and other BSDs-- FreeBSD has it's pros and cons. Choosing which to use boils down to prioritizing what you need the system to do and what's less hassle for you. If you're a Windows admin primarily, it's going to be immeasureably easier for you to set up LDAP on an AD box; if you're primarily a unix admin, you can just as easily do the same thing on a *nix.

Re:We need a +1 troll modifier for this stuff.

[Nelson]

Sorry I hurt your feelings; what was the troll? Any chance you were a BeOS user or an Amiga user?

Re:Please learn how to make links.

[WebCrapper]

I was reading your post and I figured some idiot would do that... I understood the reasoning behind it and I would have done the same.

Re:Please learn how to make links.

[DrSkwid]

oh *I* would, would *I* ?

Re:Please learn how to make links.

[DrSkwid]

I think they search through for unlinked URI's just to flame

misinformed

[Anonymous Coward]

Ha! That article is so misinformed. In fact, FreeBSD has the worst most buggy networking stack of all the BSDs due to the buggy "SACK" code which does not properly implement SACK, according to the authors of the SACK spec. FreeBSD 5.3 Release went out with a thoroughly buggy TCP stack. It's hilarious how FreeBSD is touting their SACK code when it's actually something to be ashamed of.

Ditto for the SMP locking they're touting as an achievement. They've been fumbling around for a year and it's still not right.

Qt

[Santana]

It is available on many platforms including BSD, Linux and Windows

*The* case for BSD?

[Compact Dick]

I recommend this.

Re:"FreeBSD 5.3 ships with x.org, which is **BROKE

[Neo-Rio-101]

You sure you don't have a motherboard with one of those onboard ATI IXP chipsets?