Slashdot Mirror
*BSD News
Woody writes "There is tons of stuff happening on the BSD front. The NetBSD Foundation is scheduled to release version 1.4 of their multi-platform OS tomorrow, May 12. The OpenBSD group is scheduled to release version 2.5 of their super-secure multi-platform OS on May 19. Finally, those boys over at FreeBSD are scheduled to release version 3.2 of their i386 optimized OS on or around May 15. Lots of rumblings in the BSD world! "
I've heard Berkley-San Diego and Berkley System Design before, but what is the right one?
Man I just bought FreeBSD 3.1 a few days ago this makes me angry. Ooh well I guess I will get 3.2 also.
:0)
As for BSD, saying that the existence of OpenBSD or NetBSD undermines FreeBSD makes about as much sense as saying that MkLinux undermines the mainline kernel. Russ, you're an angry, bitter guy who needs to get away from the computer a little more often. I've seen your postings in various free-software-centric forums for the last 5 years (at least), and you always seem to have an axe to grind- if not with MS, then with some other free software project or with some supposedly evil ideological foe. Take it easy for a while. If you think that things like software are what really make a difference in the world, and that you are really being some kind of idealistic freethinking kinda guy based on what software you run, then you suffer from the same brand of bizarro tunnel vision that seems to afflict everyone else of any public stature in the free software world. Turn off the computer for a few days, or a few years, and go read some books or hang out with non-computer-using people. It will do your tortured soul a world of good.
Call freebsd i386 optimized because it only runs on i386? ;)
Usenix has contributed funds to all three *BSD projects to help them get their new releases out. Free copies of all the CD's will be distributed to the Usenix attendees as I understand it.
On a 33.6 kbs connection? Wouldn't that take a while? How big is it? I have no idea since I've never done it before.
I'm a linux geek and a BSD newbie.
frankhale@yahoo.com
Usenix provided funding for the releases. See the news column in the May 1999 issue of Daemon News .
The BSDs didn't have any money to make CD ROMS so Usenix duplicated the CD ROMS and will give some of them away at Usenix. So yes, this is really a Usenix story.
I wonder if any of the Linux's or BSD use the pthreads library in dveloping the SMP kernel thread support.
I was told by someone that the Corollary Algorithms are the best commercial threading algorithms out there and that most [all] the Unix
flavors use them in the development of their threaded OS'.
Does anybody have any more details on the Corollary Algorithms?
NetBSD - lottsa experimental stuff here.
OpenBSD - The FreeBSD ports with NetBSD's selection of different hardware. Has auited the code for security.
FreeBSD - Optimized for x86 opcodes. Easy to use ports/packages.
Hehehehe...
On a related note: anyone know where I could get an el cheapo CD of OpenBSD (in the US)? I'm guessing that the wonderful crypto laws we have in the US scare off potential sellers (e. g. cheapbytes, cdrom.com).
/me imagines back-alley CD burners, with "bootleg" CDs of SSL, OpenBSD, etc.
Posting as AC because I'm the moderator in question
Because, to the best of my knowledge it's not true. See JKH's original announcement at the FreeBSD web site.
Note: /. has probably mangled that URL. If you follow it, open it in a new window, then scan the URL for any spaces inserted by the posting process, and remove them.
I was regrettably forced to conclude that the posting in question was probably some sour-grapes posturing from a hairy-knuckled Linux user who mistakenly links the exposure of his favourite OS distribution with his sexual prowess, and therefore feels the need to beat off about it at every opportunity.
Since this is the new caring, sharing /., I knocked a point off it.
Hope that answers your question.
So, Russ, how do you treat your enemies?
Shoot this guy for the BSD FUD, but make sure it's in the kneecaps for the Debian FUD. :)
Some might say that, but some would hope that people disregard the piece of crap that you linked to. Speaking as a BSD-license advocate, that thing is the best piece of GPL advocacy I've ever seen. I'd love to switch licenses just to disagree with this guy!
Other platforms (e.g. sparc) have other COMPAT_*
options.
That article is a piece of flame bait worthy of Brett Glass.
One of the major points it leaves out is that _both_ the traditional BSD license and the GPL are incompatible with other licenses (BSD because of the advertising clause and GPL because it can't be combined with more restrictive stuff).
Instead of complaining about the 'restrictiveness' of the GPL, the author should point out that there are other licenses out there designed to be compatible (XFree86 and LGPL). Moreover, many authors have dual-licensed their code under both the GPL and a BSD-like license just so that it can be shared between both groups.
It will never happen because the BSDs are distributed under BSD-licence.
Nice statement. What does it mean ?
> the need for kernel threads on FreeBSD is a lot lower than on Linux
Well actually it's the same, seeing as both FreeBSD and Linux do SMP, and user-mode threads cannot work over multiple processors (among other things).
From users point of view, they are ideantical. Most applications are the same.
From sysadmins point of view they have some differences in system administration. You will find that recompiling kernel is slightly different, package management is done though ports collection (as opposed to rpm on rpm based distros, or dpkg or debian or no package management at all on slackware), etc, those little things.
*BSD is the continuation of a failed idea -- that quality matters.
So does quality count in your use of qmail?
But why is BSD a failure? Why does it continue to lose marketshare? These questions are never answered. Only a complete idiot would think that BSD is winning. BSD is in an ongoing downslide. As the marketplace consolidates, BSD has become one of the first casualties.
BSD out of cash? Why is that? I thought BSD was doing very well--I guess BSD is not doing as well as its publicity. Oh well, another urban legend bites the dust.
I don't agree. Since 1993, my home system has always been mainly Linux, and independantly of marketshare. Remember that by the marketshare metric, Windows still crushes all the others OSes. *BSD, most notably FreeBSD, can easily get better performance than Linux, at least on some benchmarks, have some different features implemented, ... and this technical merit alone to make BSD a living success. And my guess is that the total number of FreeBSD users is been growing, not declining like you are implying.
I think that the NT designers had ambitions to learn from the 25 years of UNIX and other OS's and develop an updated OS based on what was learned. Whether they succeeded or not is a different story. So NT has many relatively new concepts which modern OSes are all aiming for.
NT (and many UNIXes) have these key technologies which Linux is sorely lacking:
1) Logical Volume Manager
2) Journaled File System
3) Access Control Lists
4) A updated security model beyond file permissions
So I'd say the ambitions lie in very different places. Most commercial OS developers think that these are very important things to have in an OS which is why most commercial UNIXes have most if not all of these technologies.
Now you can peg me a MS lover, but I'm just telling it like it is.
If you have access to a Mac you might be interested in Darwin. It's basically the FreeBSD core sitting on top of the Mach kernel. Non-Apple development for Darwin has only just started so there's no closed or hard to enter core development group yet. Good hackers can even get write access to the Apple CVS system. Start hacking!
The main difference is that Linux is a success and *BSD is a failure. Here is a link to Intel's take on the success of Linux. One of the major side effects for the BSD problem is that BSD has almost no commercial quality software available for it. But the most practical reason to avoid BSD, is that its long term outlook is even poorer. Most BSDs will be gone within a year or two due to fincancial problems. For example, the BSD groups have been so strapped for cash that they can not even afford to manufacture their own CD ROMS anymore. But for the generosity of Usenix, these foundering projects would be completely shut out of the CD ROM market. The good folks at Usenix have agreed to a one time charitable subsidy lest these wayward experiments be lost forever. But only a fool would switch to this BSD sinking ship, matey.
BSD a failure? Depends on your point of view. I guess in that case Linux is also a failure compared to Solaris and MS Windows, if we are only looking for marketshare.
Maybe there are more important things, like technical merit?
Sure *BSD has less users, but it has certain niches in which it excells, like multiplatform (one distrib that compiles on all supported platforms, the same source on all machines, try that for a change, Rathead) and security.
Maybe the *BSD community is not that much out to convert all sorts of people, instead of attracting a lot of newbies and clueless people, I found that the *BSD community consists of very knowledgeable and dedictated people.
As for applications, most stuff compiles or even Linux binaries run on my OpenBSD systems.
If you don't want to use *BSD, fine. Like somebody is going to miss you.... Stick to Linux then.
> I heard that BSD has run out of money and that
> they can't even make CD ROMS cause the cash has
> run out.
Developping a multiplatform secure OS costs money.
I don't think we need to argue about this point. Yes it's more expensive then throwing thogether a cheap (and reliable?) Linux distro.
> Usenix was forced to cover the expense at a
> loss.
Nobody forces USENIX to do anything, they are an independant organization. It's true to USENIX gave a bunch of money to the *BSD organizations, because they want to give away the CD's to people who are attending their conferences. That might have something to do with the fact that USENIX is really pleased with the quality and the support from the *BSD community.
USENIX has been supporting FREE software for a long time. Ever been around in the 1970's and 1980's, people?
Check the terminalroom (PC farm) next time you attend a USENIX conference, last year at LISA 12 they were all OpenBSD systems, at USENIX Technical they are believed to be FreeBSD systems. They worked *excellent*, people were very pleased with their performance and especially, maintaining them was fairly easy, at least one did not have to spend the night reinstalling them to get rid of all sorts of installed trojans. People could *trust* them too.
Seeing the big picture?
Maybe USENIX is so happy with what the *BSD community has done in the last decade for the FREE software that they decided to give something back.
There is more BSD stuff floating around doing very important stuff. This is not an issue of "it's not GPL so it must suck" I hope... Ever heard of Sendmail? Bind?
Maybe all those Linux biggots should check out their UNIX history a bit... there *was* life before 1991!
It's true that the marketshare of Linux machines is bigger then that of *BSD machines, but that does not mean there are *less* *BSD around, it just means that the growth of the *BSD market is not as fast as the Linux market. So what...
"4) A updated security model beyond file permissions"
Just out of curiousity, what is the updated security model NT uses?
But I was mostly referring to their commitment to component-based computing and object-oriented stuff, something that I'm personally very interested in on a theoretical level. This is all stuff that people in the industry have been talking about for years and years and years, and it has really been only recently that components have begun to be seen as "trendy" in the Unix world (and elsewhere in the non-MS world). But MS has been pushing actual implementations of this stuff since the early '90s. Again, you can always argue that the ideas are good but MS has somehow poisoned them with their evilness (or whatever), but there are some unquestionably sharp engineers there who have been attempting to do the Right Thing for years now.
It's not fair to say that MS is wholly evil or that their engineers are incompetent across the board. They're just way to much of a lumbering, enormous, schizoid company to apply any of those sorts of blanket statements to them. Likewise, the Unix world (or even just the free software world) isn't all black and white. There's no shortage of jerks with agendas pushing backward- thinking designs in the free software world, or at Sun, or wherever. No more or no less than there are at MS or elsewhere on "the dark side". There's no shortage of idiots in all segments of the industry.
may 19th is perfect! everyone in line, you can
just set a cron job to download it for you if
your also going to be in line.
-pixel fairy
Open source projects the size of the BSD's don't die. That's just a simple fact of life.
As far as their success goes, could it be that BSD isn't trying to beat windows/Linux/MacOS in the desktop market? Ever think they might not be trying to "succeed" the way you envision it? Maybe they're just trying to write the best server OS out there.
Just because they aren't jumping on the bandwagon doesn't mean they're a failure. It means they're staying true to their original goals.
As far as them being strapped for cash, let's see
the linux kernel development team start pressing CD's. They're a non-profit organization. They aren't trying to make money, they're trying to make an operating system.
The only thing Linux has over BSD is hype. If you're the kindof person who thinks hype is the equivilant of success, you shouldn't even be using Linux, let alone BSD.
What every Linux user needs to know about BSD
http://www.daemonnews.org/199904/editorial.html
1. Your argument that BSD has no quality software was not only irrelevant, but it was also incorrect. BSD runs linux software, in most situations just as fast as linux, in some it runs it faster, and in some it runs it slightly slower. However the strength of its commercial software really isn't all that important. For every commercial product for Linux, you can find a superior open source project, which is what you should do. If Linux became a set of proprietary programs on top of an open source kernel, then it would be a failure. Thankfully, no open source developer is going to roll over and die at the site of commercial competition, so we'll have great open source software whether you like it or not.
2. I would find your argument that BSD is going to go out of business quite humorous, except for the fact that someone might read your post and actually believe it. BSD is an open source project. The developers aren't in it to make money, they're in it because that's what they love to do. You can always get a CD elsewhere (like www.cheapbytes.com).
A bit inaccurate... It has a lot of the NetBSD netwoking code (since it really is the best out there), but most of the rest is strewn from various places, especially ol' Next.
Apple fucked themselves with version 1 of their APSL. Unitl they make the yellow box open source, fuck 'em. They're forever gonna be the bitch of the industry because they have no clout in business or server markets. And rightfully so; they've yet to make a good business platform or a good server.
If you want to sell your soul, go hack Solaris
Blowfish encryption integrated in all libraries. Integrated source available with anoncvs checkout or binary snapshot on all ftp mirrors for several platforms including sparc, i386, mac68k, m68k, alpha...don't trust me though, I just installed OpenBSD >8P on another harddisk. OpenBSD installation was the most accurate and verbose configuration I tried so far. Most important question for you: If not OpenBSD, try FreeBSD or NetBSD before deciding whether all *BSD operating systems in the BSD Lite category could be usefull or useless ;) Good luck Nat.
.. is the "enforced free redistribution" clause in the GPL.
This makes it possible for large companies like Yahoo to modify BSD code and then see a return on their mods. They can sell modified versions to other corporations without restriction. You can't do this with Linux under GPL.
Conversely, it makes code forking more likely, which is bad for more general development (c. Halloween docco).
So the licenses kind of balance. Advantage in corporate development versus more general development of code base.
MolochHorridus-- (license weenie)
When you say " a failed idea -- that quality matters. " are you telling I should be running
Windows?
Hmmm ;login: to members.
2 years ago USENIX had a "mindcraft effect" where
larry mcvoy, using mindcraft methods, showed
some very good numbers of linux against all other
systems around. well, now we see not only mindcraft but several other 'independent' benchmarks showing NT, Solaris and even the BSD
faster ("but it seems the new linux kernel will correct all problems").
May be USENIX (that was as proficient about opensource unices two years ago as bill gates) is
trying to give a chance to other OS's they killed
by complete ignorance then.
Unfortunately I can't got there to get all CD's to
try at home, I hope they'll send copies with
There are always differences between theory and practice, and though NT may be THEORETICALLY more advanced, in PRACTICE it does very poorly (damn BSOD). Though I do believe that there are some highly skilled designers at MS (I heard two people from the Mach project are there, is this true?), but their OSes still suck. They don't lend themselves to a networked computing environment, and the performance and stability have always been an issue. Sure, their DESIGNERS may be good, but what about the coders?
Anyways, such things you speak of are implemented rather well in BeOS, which is the OS I would use, but it isn't open source, and I like opensource because of frequent upgrades and fixes.
The correct answer is that the various BSD versions have diverged and do not share the exact same set of system calls. There is no binary portability between the various BSD platforms. In order to run an executable for one version of BSD under a different BSD variant, the executable must be passed through a BSD emulation layer similar to the Linux emulators used by the various BSDs. Emulating one BSD under another BSD involves trapping the incompatible system calls and remapping and adjusting them for the host BSD system. And also adjustments must be made for executable format (Zmagic, Qmagic, ELF, and so on). And executables which depend on internal private knowledge of VM and other kernel subsystems will fail even if they can be started.
In contrast, Linux executables are completely portable between Linux distributions. All Linux distributions share a common kernel. But all the BSD kernels are different and incompatible without an emulation layer.
> What can OpenBSD offer over what I already have?
A system that is difficult and cumbersome to use even by Unix standards, that achieves its security by omitting all but 5% of the programs on a typical Linux distribution, and a computing experience that is generally slow.
Seriously, OpenBSD will set up everything 'right' security wise to begin with, and a lot of encryption stuff is built into the kernel and libraries. however, lots of stuff (like ssh) doesn't come standard anyway and has to be built from source after you install the system. (it's in the ports system)
One nifty thing it does is kernel-level transparent firewalls, i.e. bridging with the ability to filter packets. That is cool.
One problem with OpenBSD is that, while they proudly proclaim they are not subject to US laws regarding exporting crypto, many useful and common crypto algorithims are PATENTED and they are still bound by Canadian law regarding patents. (for instance they can't include RSA so you will probably have to download and build a different version of SSLeay to do anything useful even after installing OpenBSD)
Actually the GNU license is better for a company that writes a software package and wants to make it open source. They can release it as a GNU program and still sell commercial products based on the same source (since they wrote it, they own the rights!) On the other hand, their competitors can't take the source and sell a proprietary version. Therefore, they can maintain a monopoly on commercial versions of the program.
If they release it under a BSD license, on the other hand, any other company could take their code, modify it, and sell it- something which they obviously would want to avoid.
Now, let's say that someone modifies the program and places the modifications under the GNU license. The company can't take the modifications and sell them in a proprietary product, but they can pay the guy who did the work to allow them in the proprietary version as well. In other words, they are forced to compensate their developers, which is a good thing.
If someone modifies the program and places the changes under the BSD license, on the other hand, the company can take the changes without having to compensate the person who did them.
The only advantage the BSD license has for a corporation is when the software is written by free developers in the first place-- a company is then free to take it and sell a proprietary version without having to pay a dime to the people who actually wrote it.
I couldn't agree more.
I'll be buying my *official* FreeBSD 3.2 CDs the day they're released.
These people dedicate a lot of time and effort to their OS projects. It's only fair to help them cover costs.
-jake
I run a linux box and a freebsd box at work, next to an NT box, the freebsd box gets the slowest cpu btw, but still worth having around.
Get of ur ass and give it a go, it slike in 1996, you got off your ass to try linux, try fbsd for gods sake, its not gona kill ya. In some ways its boot system kicks the crap out of the shiity lame lilo effort.
I will be upgrading to 1.4 on my 1.3.3 box
//d
can't be bothered logging in
The *BSD's can all share code with one another, providing it was compiled on the same architecture as your host (you're not going to be able to run VAX binaries on your pentium).
The real difference between the BSD flavours is their "goal" - NetBSD with its ports to every known piece of computer hardware, OpenBSD which concentrates on high levels of security (sometimes at the expense of performance) and FreeBSD which is fairly x86-centric (Alpha port seems to have stabilized pretty well too) but architecturally advanced and high-performance.
FreeBSD is the most popular because it's the most "mainstream" (almost everyone has an x86) - but if your interest lies in non-x86 (& non-Alpha) hardware then one of the other two is probably suitable for you.
Oh, and I forgot to mention that there's a lot of code sharing which goes on between the various BSD's, because they are so close together and therefore easy to port to.
Often software written for one platform will compile and run just fine on the other BSDs without modification. It's really in the kernel internals that they differ.
ROFL.
If Usenix said to you "here's $10000, go burn some CDs on us", would YOU say no? (I pulled the $10k figure out of a hat, don't know how much it actually was).
Duh.
> I will give the *BSD folks an ounce of
...
> credibility when they reconcile their
> differences.
*LOL!*
RedHat, Slackware, Debian, OpenLinux, MkLinux, Mandrake,
I will give the Linux folks an ounce of credibility when they reconcile their differences.
FreeBSD is not losing market share in absolute terms (i.e. people are not walking away from it). The number of installed FreeBSD systems has grown consistently at a good approximation of an exponential rate over the past 4 or 5 years.
Linux has been growing at better than an exponential rate (as the media gives it free publicity). Because of this, Linux has been 'stealing' percentage points from every other OS, because it's growing at a faster rate.
This doesn't mean that the other OSes are not still growing (they are), just that they're growing slower than Linux is.
USENIX supports *BSD because they are impressed by the technical side and the support they are getting from *BSD.
They think it's so good they even want to give the CD's away for free to attendants of conferences.
What's you problem? Not pleased there is no bunch of Linux CD's in the bag?
> Cheapbytes has it
Sure.. a cripled version, not the fully bootable double CD... it's not really exportable but the did disable some of the encryption stuff to be safe...
And where are the cool stickers? One of the reasons I keep buying OpenBSD CD's are the stickers... Kickass artwork!
> What can OpenBSD offer over what I already have?
What about a bunch of possible security bugs less?
Buffer overflows, various races and buggy libs.
http://www.openbsd.org/security.html
The new-kernel-every-day isn't an issue. Because the kernel are mostly interchangeable: whever you run 2.2.0 or 2.2.7 is doesn't change anything except for some minor bugs: I even happen to boot indifferently on 2.0.36 or 2.2.3 depending on what kernel-boot diskette is the nearest.
As far as the different free BSD's go, they all have different goals for what they want their operating system to do, but they also cooperate with each other. The Linux kernel hackers have been known to borrow from the BSD's from time to time also.
True, but the BSD split is still very annoying. I understand that the OS have different goals and that the system utilities should be different, but the remaining 97 % of the programs (including all the ports), why should they be made gratiously incompatible ?
[...] The moral of the story, kids, is use what you like, and don't worry about which is better.
That's true. BSDs are cool. But it won't stop people like me whining because sometimes some Linux executable won't work on other distributions directly, or that the *BSD are incompatible.
Last time I looked, either you used redhat unadultered, or you lost most of their system administration stuff each time you installed a new kernel.
I don't understand what are the "patches for the system management utilities". This is completly news to me ; maybe is it with RedHat 6.0. I have tried at least 2/3 of the RedHats that ever existed, and RedHat didn't do something like that. They have sometimes patched kernel in order to include support for not-officially-supported hardware in their boot disks, but loosing system administration stuff with a new kernel seems completly unbelievable.
It makes no sense either: why adding a user, adding a new partition, changing IP address, adding a printer, etc... would have anything to do with the kernel ? These are in distinct files from the kernel.
The only thing which might be remotly close, is if you had configured some modules for some drivers (tell IRQ, DMA, ...) ; then you'll have to read the RedHat documentation to see where module configuration is stored. But I never bothered to check because everything works out of the box on my systems (with kernel automatically loading the needed modules). The only problems were for sound and ISA PnP, but I just wrote a script to configure it ; this is explained in isapnp, and 2.2.x kernel sound Documentation.
For your information I administrate some Debian 2.0, Debian 2.1, RedHat 5.2, RedHat 5.1, RedHat 4.2 systems and of course I compile the kernel and the modules on only one machine (Debian 2.0) and copy and install them on the others, and this works perfectly (well there are some warnings on machines with older utilities for newer kernels, but since it works...). Besides, installing new kernel source usually completely destroys your rpm database... pretty hard problem to solve, I know.
This is incorrect: the Linux kernel is really independant of the distribution. Especially with newer kernels (2.2.x), you can extract your kernel source anywhere and just do "make xconfig ; make bzImage ; make modules ; make modules_install". It will install the modules in /lib/modules/<kernel_version>/... and you'll have a kernel in arch/i386/bzImage that you might install manually or automatically with "make install" if you have the proper lilo. I personnally prefer now to keep the kernel standalone on a separate floppy disk with "cat arch/i386/boot/bzImage > /dev/fd0".
You may overwrite modules listed in the rpm database but I don't see it as a problem, as long as you are not manually removing the old one.
Your problems have probably another cause than just changing the kernel. But you're acting as though this kind of stuff didn't exist.
But it doesn't exist ! Look, I've copied the same Linux kernel in more than 10 Linux machines with different distributions, and that works.
Because of ongoing financial difficulties with the BSDs, Usenix has been asked to fund the pressing of BSD cdroms. In addition to the struggling BSD market, Debian Linux is among the other troubled organizations which can not finance its own development. A collection for the "homeless" was taken up, with Usenix footing the bill.
*BSD is the continuation of a failed idea -- that quality matters.
A failed idea that I can accept.
The lesson of the last 20 years has been that marketing is more important.
I really pity you if you honestly think marketing is more important than quality. It is obvious that which matters more is purely dependant on your goals and if your goal is only to sell lots of something you wind up with Windows. Well I have news for you. Not everybody is only concerned with quantity. Some of us actually care about quality. I know that there are some out there that have a deep need to have their OS outnumber the rest but that is foolish. That is why lots of people run Windows and I have even heard it used as a reason to run Linux rather than the other Free *nix products. I personally choose the platform based on what I need it to do. If Linux's only goal were quantity it would turn into crap as well. I obviously know that isn't true but that is the implication of your premis.
Unix vendors (and this includes the *BSD people) have responded by creating Yet Another Unix which will have higher technical quality than the opposition.
Oh HORROR!! What Fools!! Don't they know they need to produce crap or it won't sell?? After all how much it sells is all that matters.
Me? I use Linux, because I want to make a difference in the world.
If it would make you happy to run something well marketed you must be running Windows right? No? Your running Linux? What for? Its not marketed like Windows and it is far more concerned with quality! After all just like *BSD, Linux is Yet Another Unix. But you say that *BSD is competing with Unix rather than Windows. Linux's goal was to be a free nix Just like *BSD.
I will give the *BSD folks an ounce of credibility when they reconcile their differences.
Why should there only be one BSD? The three projects have different goals. BTW Which Linux distribution do you use?
I personally use whichever OS fits the job at hand. I use FreeBSD on my servers and RedHat on my desktop. I also keep a Windows partition for games.
I'm not sure exactly what you're asking, but possibly the answer is this: both NetBSD and OpenBSD will run binaries compiled for each other, as well as FreeBSD, BSDI, most (all?) Linux distributions, SCO, and Solaris x86. This is done through emulation modules for each of the other OSes (and possibly loading the shared libs from the other OS into /emul//usr/lib - not needed if the application is statically linked). Likewise, the Sparc ports of Net & Open will run SunOS 4.x and Solaris 2.x binaries, the Alpha ports will run Digital Unix binaries, the m68k ports will run SunOS, etc.
I'm not sure what emulation capabilities FreeBSD has, but I imagine they are similar.
So quick to tell me that BSD sucks?
I thought linux was about CHOICES. I thought that you people screamed that we all should be able to CHOOSE what we should run. Now that there's another Unix (A REAL unix, not a flavor) out there, you are instantly telling me that it's WRONG to like choices? You people are so fucking hypocritical. Take a look in the mirror, kiddies, and wonder WHY the big boys like Yahoo, and Hotmail, and such are running FreeBSD. Maybe THEN you'll understand that Freedom of CHOICE is more important than pushing a fucking COMPUTER program.
Jarrod
The main difference is that Linux is a success and *BSD is a failure
This is why FreeBSD is used by people like Yahoo, Hotmail, Xoom, Geocities, Inktomi, Verio, US West, the IMDB, and just about every other major ISP and large content provider you can name. The only bright star on the Linux map is DejaNews, and you can tell it's a Linux system by the "Document contains no data" errors you get using it.
A recent, conservative, estimate suggested that FreeBSD systems might well be serving as much as 50% of the content currently available on the 'net. That's not "failure" by any stretch of the imagination.
And just to correct the other misapprehensions in your post; while OpenBSD (and possibly NetBSD) did indeed require funding to press their Usenix CDs, these aren't commercial organisations. They don't exist to make money - they exist to write software. You might as well suggest that Linus press "linux kernel" CDs to go with all of the commercial distributions.
The FreeBSD project has never had any trouble pressing massive numbers of our immensely popular CDs; we've just released a 6-disk companion toolkit, for example, to take the pressure off our regular 4-disk release.
But feel free to avoid our very buoyant and popular vessel - we can do quite happily without people like you.
How many people use your operating system AC? None? What do you mean you haven't written an operating sytem? Well... that makes you a failure, doesn't it?
What was the point of that line of questioning? Simple, don't impose your goals on other people. Just because you think world domination is the only acceptable goal for a software project doesn't mean that it has to be the opinion of the people who write *BSD. I don't know any of them personally, so I can't say this for sure, but I would assume that like any good open source project, their goal is to write good code simply for the love of hacking.
Why is *BSD loosing market share? Markets fluctuate, people are fickle, etc... But it sure as hell isn't going to go away. The BSD's weren't written for the average user. They were written for people who needed an operating system they could trust as being stable and secure enough to run on their servers. Those are the kind of people who stick with an OS, not the kind who always have to use "The Next Big Thing." Now I'm not saying people don't use Linux for the same reasons, however alot of the new Linux users are just using because of the hype. Who knows what OS they'll be using in 5 years. When the market consolidates as you put it, percentage wise, BSD will take the smallest hit.
Oh yeah, and BSD is about to inherit a few million users from an OS called MacOS, which just switched its base to BSD. That'll help their market share a bit I would assume.
Well, you can order the FreeBSD CD direct from Walnut Creek CDROM (the people who sponsor ftp.cdrom.com, maybe you've heard of them). Check out their homepage at http://www.cdrom.com
I read the internet for the articles.
But why is BSD a failure? Why does it continue to lose marketshare? These questions are never answered. Only a complete idiot would think that BSD is winning. BSD is in an ongoing downslide. As the marketplace consolidates, BSD has become one of the first casualties.
Your comments sound oddly familiar to the condition of Unix in general, circa 1995:
But why is Unix a failure? Why does it continue to lose marketshare? These questions are never answered. Only a complete idiot would think that Unix is winning. Unix is in an ongoing downslide. As the marketplace consolidates, Unix has become one of the first casualties.
If you're choosing Linux over BSD because of marketing, why aren't you running Windows instead?
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
I'll have to disagree with you.
Linux is, and aims to stay, Multi-platform (truly difficult to accomplish, and certainly highly ambitious)
On the user-interface site of things, projects like Enlightenment that allow full user-customization of the work environment aim to take "windows" to a new level...while maintaining the beauty and availability of a command line interface...another ambitious goal.
That Linux can run on a 386 is testament to the ambitious ideas of its designers as well. NT scoffs at the need to support old hardware...but to me, supporting it is ambitious and commendable.
So I guess...what it really comes down to, is what turns you on. I would say that UNIX developers are more ambitious than NT developers....just in different ways.
Werd.
So if FreeBSD's threads are implemented entirely in userland, how do they avoid putting all of a process's threads to sleep when one thread blocks on IO?
Actually.. NetBSD and OpenBSD (1.3.3 and 2.4) can be FTP'ed in 7-9 hours, depending on your connection (took me around 8hr each at 24kbps). It's not really likely there'd be a new version released while you were downloading, unless you're downloading the -current release. :) Dunno how long FreeBSD would take, though.
If you have the money (and/or patience), though, get a CD. If nothing else, you'd be helping them continue working on the OS.
As soon as they replace the BSD utilities and C library with the GNU equivalent. Which would be apoproximately as pointeless as Tom Christiansens project to create a BSD/Linux.
Pah! VMS had clustering back in the early 80s.
NetBSD: the cathedral vs the bizzare.
Basicly, the difference is this:
...
...
...
- FreeBSD is the fastest (you can argue about this, but it is that way 85% of the time), and has better Hard and Software support (running WordPerfect or Doom or Quake (even the beta 3) for Linux is no prob
- NetBSD runs on a lot of different platforms
- OpenBSD has more encryption/security options
I would recommend FreeBSD in your case
Scalability is a very ambitious technology, and something that only high-end Unix systems really have a claim to. I keep laughing when I hear about NT clustering. Can they really call two nodes a cluster whilst keeping a straight face? If it seems "sexy" on NT, it's probably been around with Unix for ten years already.
Do you have any evidence to back this up? The story cited in another posting says only that they're "providing grants to the OpenBSD, FreeBSD, NetBSD, and Debian Linux development projects, to support each of them in issuing new releases.", and says that this is "because CDs of their distributions will be given away for free at the Usenix Technical Conference in June".
If the continued existence of the BSDs (or of Debian GNU/Linux, for that matter) makes you unhappy, I wish you many years of continued unhappiness....
By wrapping a pile of API calls (e.g., read(), write(), etc.) so that descriptors run in non-blocking mode (with the non-blocking mode that the API calls return being maintained by the threads library, rather than being the kernel's value for that mode), using an unwrapped select() in the thread scheduling loop, etc.. See the stuff in the lib/libc_r/uthread directory in the FreeBSD source tree.
No, it doesn't keep the entire process from blocking if, say, a file read() reading from a file server blocks because the file server isn't responding, but it catches the common case of reads or writes from serial ports, network connections, and the like.
Besides, it is well known that BSD's NFS implementation is better than Linux's.
support gun control: take guns from cops
I'd be running it on a P-150, 64MB. I have a copy of FreeBSD 3.0 but I was wondering if NetBSD or OpenBSD is better/more interesting for any purpose ...
support gun control: take guns from cops
There is currently support in FreeBSD (3.1->) for a native port of the LinuxThreads kernel threads package, but the need for kernel threads on FreeBSD is a lot lower than on Linux, because the userland threads (libc_r - which should be POSIX compilant) has a very quick context switch. The performance between the two is pretty much the same, and should perform very similarly to libpthread on Linux.
There has been a lot of talk about a lightweight kernel threads implementation, especially to use SMP effectively accross threads, but there is no code.
-Jeremy
This reminds me, did that project ever get off the ground? I would be interested in helping if possible.
Anyway, why is it that every time there appears a *BSD related message on
If you want to run Linux, fine. But don't flame people who don't.
Intosi
Intosi
NetBSD supports a wide range of binaries.
:)
Here's some of the kernel config file:
# Compatibility options
options COMPAT_NOMID # compatibility with 386BSD, BSDI, NetBSD 0.8,
options COMPAT_09 # NetBSD 0.9,
options COMPAT_10 # NetBSD 1.0,
options COMPAT_11 # NetBSD 1.1,
options COMPAT_12 # NetBSD 1.2,
options COMPAT_43 # and 4.3BSD
options COMPAT_SVR4 # binary compatibility with SVR4
options COMPAT_IBCS2 # binary compatibility with SCO and ISC
options COMPAT_LINUX # binary compatibility with Linux
options COMPAT_FREEBSD # binary compatibility with FreeBSD
# Executable format options
options EXEC_ELF32 # 32-bit ELF executables (SVR4, Linux)
Also, there are various packages in the NetBSD package system to automatically download and install the appropriate libraries for compatibility with these systems, and an ld.so to handle the magic.
hmm... openBSD.... may 19.... Star Wars.... do I smell a conspiracy?
but I thought you were supposed to pray to the great lord Lucas while waiting for the holy movie...
hmm... I see your point... install the holy OS, while waiting for the holy movie and praying to the holy director.
ummm... the BSDs don't generally rely on the GNU tools as much as Linux. They have their own original BSD versions.
Of course you CAN install the GNU versions if you like.
I fully respect and honor people who find the BSD license preferable to the GPL, or who object to the GPL because they think its retrictions are unnecessary. I have no use for jerks who argue against it on the basis that it is "communist", especially from proponents of the BSD license.
After all, the BSD license is more communist in the sense of Marx than the GPL. "From each according to his ability, to each according to his need." If you tell me I should use the BSD, you're saying that I should contribute work to you to do whatever you want to do with, because I have the ability and you have the need.
The GPL is more market-oriented. The author who uses the GPL gives software to other developers to make derivative works from only if the other developers are willing to give back their changes. Many developers who use the GPL also sell proprietary software; in some cases (e.g. Aladdin Software, makers of Ghostscript) they use dual licensing: folks who don't like the GPL restrictions can pay $$$ to get other terms.
As Russ Nelson has said, the reason he uses the GPL is "When I write commercial software I want to get paid." Those who claim that developers must use BSD-like terms when they write free software are saying that they must give up their work without any compensation at all, to whover needs it, in accordance with Marx's dictum.
There's no question that RMS has a political agenda, but then so does the Daemon News, which has repeatedly demonstrated that its political agenda is to try to get people to release software on BSD-like rather than GPL-like terms.
Does any of the *BSDs have a proper pthreads implementation based on kernel threads? Last time I checked, only FreeBSD had a pthread.h in /usr/include.
In fact, threads on linux are kernel based, so the blocking IO problem doesn't exist there.
Assuming you've installed the source tree from your 3.1 distribution, a cvsup to 3.2 will not take long at all, since it only updates the things that were changed. I wouldn't expect it to take more than 30 mins on a 33.6 connection.
If someone posted anti-Linux propaganda replies to every Slashdot story with a penguin on it, it'd be war.
Doug White
Tech Support Volunteer
questions-at-freebsd-dot-org
I'm very interested in security stuff.
I currently use debian's non-US security stuff; primarily, ssh, ssl, apache-ssl, ssltelnet, etc.
What can OpenBSD offer over what I already have?
Go to the web sites and look around a little bit...
For instance, OpenBSD's site is pretty small and to the point: you can order CD-Roms from a variety of places, including Amazon. You can also get rather nice T-Shirts.
Oh yes, and for buying OpenBSD in Europe, you have a choice of Germany, Norway, Sweden, Finland, Britain, and even France now ! :)
Not only is this FUD, but this looks malicious as well.
Either you're hell-bent on destroying OpenBSD's image, or you're completely clueless and terminally brain-dead, take your pick.
Yes, it's true, OpenBSD does not cater to complete idiots, as you've found out.
- ssh ?
patent issues: can't be included on the CD if it's to be used BOTH in the US, and elsewhere. *however*, there is a port, and this is probably the most prominent package that gets built for all architectures... Installation is just a matter of getting either the intl or usa flavor, and doing a pkg_add. Shesh !
- sslay ?
why don't you try `man ssl' some time ? maybe you'll find out that there *is* a patent issue as well, and that the true, complete sslay libraries with RSA code are *also* available for download... same problem as with ssh, and that *everything*'s ready for use (secure web server, and all that).
I should know, my OpenBSD box uses only ssh, and I'm setting up a secure-webserver next month... tried all the experiments on OpenBSD with no problem at all.
Wow, does this mean redhat has stopped patching kernel and pcmcia sources to add support for their system management utilities ?
Last time I looked, either you used redhat unadultered, or you lost most of their system administration stuff each time you installed a new kernel.
Besides, installing new kernel source usually completely destroys your rpm database... pretty hard problem to solve, I know. But you're acting as though this kind of stuff didn't exist.
There's a large difference between the Linux and the BSD model of development.
:) It's just a bit less accessible for newbies, which just means you can't feel like a hacker without *becoming* one.
In Linux-land there is about one kernel (with slight tweaks) and loads of distributions with lots of minute to large changes everywhere).
In BSD-land, you've got three distributions, which noticeable changes everywhere: kernel, user-land, ports.
If you look at BSD logs, you'll often see the same names, or stuff taken from one BSD to the next. I think there might even be a higher level of cooperation between BSD distros than Linux. I know for sure that we don't hesitate to pick useful stuff from Net/FreeBSD when it comes along... in some cases, we wait until stuff is stable enough to include.
Sure, you might miss some of the `bleeding edge' that's soooo keewl when running linux with a beta kernel, a beta glib, and beta stuff everywhere.
Not to say that this is impossible in BSD land, my laptop running full-tilt egcs-990502 and soon binutils-990427 is about as bleeding edge as it goes
Binaries are usually portable if you have to.
:)
:)
One point of the matter is to try to run from source when you can, which is why the ports systems of all three BSD are highly functional.
Just cd to the right directory, type make install, and magic starts: the box finds out what it needs, gets it thru ftp, and builds a brand new binary. Magic, or not ?
As far as emulation goes, let's see...
- I've been running Linux's quake.
- svgalib works as well, even though it's somewhat insecure.
- libggi is mostly working.
- my netscape is a bsdi binary, and I use Maple for linux every day.
Icky stuff such as OSSaudio calls works pretty well under OpenBSD (xanim runs, complete with sound)... it just becomes a question of how far you're willing to push linux support. One reason to run OpenBSD in the first place is so that you don't have to finnagle with all those libc5/libc6/glibc2.1/egcs fun that Linux seems to have these days.
Oh, and I also know that Mathematica runs, or that WordPerfect has a complete functional port... but I don't use these.
www.ixsoft.de, www.lob.de
--
"Just believe everything I tell you, and it will all be very, very simple."
Wow, what a bold statement. "The main difference is that Linux is a success and *BSD is a failure" and "Most BSDs will be gone within a year or two due to fincancial problems".
Hmm, what "facts" do you have to support such a bold statement? (Personal feelings, etc?)
As for *BSD dists dieing, HAHA. There are many corporate/commercial sites that only run FreeBSD for their "live" web, mail, and DNS servers.
Sarcasm On
Gamespot, Yahoo, the Apache Group, etc... They must be insane!!! for using FreeBSD; educate them before it's too late!!!...
Sarcasm Off
Ahh, anyway, why would I want to buy, for my purposes, a FreeBSD CD if I can:
So, CD publishers are not a big concern for me, and maybe the other users of the *BSD's as well, although I don't have any data to back that statment up. :)
"But only a fool would switch to this BSD sinking ship"
Oh man, you top it off by insulting the *BSD users! Clearly yours is the attitude of a person who has not used a BSD variant (correctly) before. As for me, and I can ONLY speak from personal experience, *BSD is here to stay. It has been truely "Rock Solid" for me, and has met and surpassed any requirements we have thrown it's way. Linux is great as well.
Use what you like. Use what works best for you.
Well, someone appears to be pissed off that his favourite Linux distro didn't get the offer from Usenix!
NetBSD will be pressing its own CD-ROMs, actually, though it will likely wait for the 1.4.1 release, since it will be out in the fairly near future (a couple of months) and it seems a bit of waste to press a huge pile of 1.4 disks only to have them go obsolete a month later. And I doubt that Walnut Creek is lacking in the resources to press FreeBSD CDs for the new release, either.
cjs
The world's most portable OS: http://www.netbsd.org.
The thing that I don't get, is that the Linux people are always saying that BSD is a waste of time and that there's too many versions to make a difference.
This is coming from a community where there are approximately 12 different distributions of the same operating system. Not to mention, there is a new kernel patch released every other day, then Alan Cox makes his own patches of the same thing. Doesn't that sound a bit hypocritical?
As far as the different free BSD's go, they all have different goals for what they want their operating system to do, but they also cooperate with each other. The Linux kernel hackers have been known to borrow from the BSD's from time to time also.
So quit your quibbling over which is better. If you look in comp.unix.bsd.freebsd.misc and comp.os.linux.advocacy, almost all the threads are a BSD vs. Linux vs. Windows, or a GPL vs. BSD license argument. Both sides are trying to convince the other that they're better, while the other side isn't listening at all. It's like a catholic priest trying to convert a buddhist monk.
The moral of the story, kids, is use what you like, and don't worry about which is better.
Binary emulation under OpenBSD (at least Linux emulation on i386) works more or less exactly like that of NetBSD.
/dev/dsp instead of /dev/audio will not sound too good. Just hack the specific program and recompile at it will work fine.
My only problem under OpenBSD is that linux programs using
/G
How can a mutlithreaded application using userland threads take advantage of SMP? The kernel will only schedule the process on one processor and then all the userland threads must take turns running on that single processor. Kernel threads let one process run threads on multiple processors simultaneously.
cpeterso
FreeBSD at least has very robust Linux emulation. Civ:CTP will probably work with very little effort (Wordperfect and Q3Test both run on FreeBSD using the Linuxmulator.)
I believe that binary swapping between the BSD's is also possible, though being a FreeBSD user, binaries for NetBSD and OpenBSD are usually already available for FreeBSD, making emulation of them unnecessary most of the time. As for the other two, I will defer to users of those.
Cheapbytes has it.
May 19th? Hmmm...I think waiting in line for Star Wars tickets and downloading a new version of a un*x are pretty equal, but still....
Uh... "a taste of honey"
when Push Comes to Shove
"Some would say that the absence of the GPL is one of the features of BSD".
Some other would say that the absence of the BSD license is one of the feature of Linux.
It depends of what you think about both licenses. I don't like to think that the BSD license allow people to take the software you have written and are just obliged to say "thanks to x for programming this great software that will allow us to make big bucks on it without giving anything back to the community".
The GPL isn't perfect but at least she infect the code so anybody can still make money out of a GPL'd program but he must give something back to the community if he want to improve the program (or to keep the improvment for himself otherwise).
"The obvious mathematical breakthrough would be development of an easy way to factor large prime numbers." Bill Gates,
What's wrong with "cvsup -g -L 2 -z /etc/cvsupfile" followed by a "make world"?
But don't hesitate to buy the CDROM if you'd like to support FreeBSD.
hroi@sorbated ~ > uname -a
FreeBSD sorbated.prunes 3.2-BETA FreeBSD 3.2-BETA #0: Tue May 11 13:35:35 CEST 1999 root@sorbated.prunes:/usr/src/sys/compile/SEXTANT i386
Is there some sort of connection between Usenix and the BSD release dates?
All of the free BSD's are releasing just in time for Usenix it seems. Hmm.
The NetBSD folks will have NetBSD 1.4 running on an IMac at Usenix.
I have always been curious about portability between the various versions of FreeBSD versus the portability between distributions of Linux.
For example, the new game "Civilization: Call to Power" should be able to run on any distribution of Linux, as long as the kernel is 2.0.x or 2.2.x.
Are binaries usually portable from FreeBSD to NetBSD to OpenBSD?
-- UOZaphod
"The unicode stuff in the latest version is working fabulously well. My russian mafia friends are ecstatic."
Last time I ordered, you could buy any of the three BSDs for $5 apiece from www.cheapbytes.com or all three together (3 cdroms) for $10. These aren't the "official" releases though. Just everything you'd get with an "official" release.
A problem still exists with international distribution (at least of the NetBSD CD) because the crypto is included on the CDs.
I would consider shipping a NetBSD CD with the crypto directory removed (it's not very big and you can download that part yourself) to a few individuals if needed. (but I only have a slow CDR drive so I'm NOT a publisher by any means).
An interesting, and to some developers, central point of the difference between Linux and the BSD unices is that BSD is not under the GPL. This can become a very political issue if pushed to the wall. There's an interesting article about these issues in the latest issue of Daemon News , which is a publication well worth reading if you're interested in BSD.
Some would say that the absence of the GPL is one of the features of BSD.
Hmmm....
If you are going to buy an OpenBSD cd, please do The Right Thing(TM) and buy an official cd. It may cost a few more dollars, but those dollars are going to support the project.
While Open BSD is free to use and distribute (as in beer and speech), it is not free (beer) to develop.
Myself, I will be buying an OpenBSD cd and couple tshirts when they are released on may 19 (I can just drive to the distribution point in Calgary).
But since OpenBSD is free software, you are free to make your own decision.
*BSD is the continuation of a failed idea -- that quality matters. The lesson of the last 20 years has been that marketing is more important. Unix has always had superior technology to anything Microsoft. Microsoft has always out-marketed Unix vendors. Unix vendors (and this includes the *BSD people) have responded by creating Yet Another Unix which will have higher technical quality than the opposition. But the opposition is not the other Unices, it is Microsoft (or any other proprietary OS vendor).
Me? I use Linux, because I want to make a difference in the world. I will give the *BSD folks an ounce of credibility when they reconcile their differences.
-russ
Don't piss off The Angry Economist
At the moment I run a SuSE-Linux at home. Since you can never learn enough, I'd like to try out one of the BSD-Clones on some DOS-infected Hardware.
Where can I buy a CD-ROM with all I need to setup a *BSD-System? I only have a 33'600 Baud dialup-connection to the internet and therefore do not want to download the whole stuff. Probably there would be two newer versions of each, before I finished downloading. =:-)
Does anyone know if there is some shop selling a distribution here in switzerland/europe?
--- If OS were buildings, then the first woodpecker to come around would erase 95 % of civilization.
I've been using linux since 1996 and am happy with it. Never tried any BSD but I'd like to. What are the differences between linux and BSD? Is BSD worth a try?
So you can install it on your laptop while waiting for the movie.
And this gets score 1??? Is Slashdot now condoning FUD?
There is no ongoing financial difficulties with any of the BSDs. Usenix is not funding the pressing of BSD/Debian cdroms, they are funding the cds they will distribute to Usenix visitors.
(8-DCS)
Yeah, I was wondering these days what was the deal with new releases of SuSE, Red Hat, the new kernels, etc... :-)
I know that some people *did* object to a FreeBSD one month ahead of schedule, but with *everyone* else doing it, there was really no option.
(8-DCS)
I don't know about NetBSD and OpenBSD binaries, but I'd be surprised if the Linux Civ:CTP binaries don't work on FreeBSD... :-)
(8-DCS)
It is not true that the performance is similar. The performance of userland threads is much higher than that of kernel threads. On FreeBSD, up to 3.1-RELEASE (but not 3.1-STABLE) that has been obscured by a performance bug in the pthreads library, but feel free to compare an application with a significant amount of threads (try 100) on any kernel threads implementation of any OS versus the FreeBSD pthreads library on 3.2-RELEASE.
(8-DCS)
Unlike what many people think, the GPL does not permit freedom, it forces openness, if you use it or any of the software under it. The BSD license and thus the BSD OSes are a better choice for commercial businesses in that it allows them to patch sources with their own code and sell the binaries, provided that the authors and contributors get credit for their work. I consider this to be an amazingly good and often overlooked advantage of the BSD License over the GNU GPL.
FreeBSD is mostly based on 4.4BSD Lite2 developed in the University of California in Berkeley. FreeBSD only barely has to do with GNU, in that it uses gcc (egcs in 4.0-CURRENT) and a few other of the binutils (as, nm, etc.), however, RMS has no right to say that it should be called GNU anything, since GNU is a minority under FreeBSD.
How soon until RMS makes us call it GNU/FreeBSD? Or just GNU/BSD?
I'm getting so confused with these proclamations!
DT
--
Free Pittsburgh!
Is this thing on? Hello?