“By the time you read this, the 2004 holiday season will have dissolved into memory. I spent the last days of the year going through my list of “things I want to check out if I ever get the chance.” As usual, I found some interesting tidbits.” Read more here. Elsewhere, Scott Long of FreeBSD release engineering team describes some of the finer points where FreeBSD continues to innovate and display its mature development environment.
Dru Lavigne’s articles are an excellent addition to the existing FreeBSD documentation, especially for newcomers such as myself. I’ve always been wondering when she had the time to write all of them, since I can’t find enough time to read them all . Thanks Dru!
I agree with Scott Long wanting FreeBSD to take a leading role in the Open Source O/Ses. It is a very solid and stable Operating system with cutting edge features.
Media has focused solely on Linux as the Open Source option. When ever you see people comparing Windows to an Open Source OS it will most probably be Linux. Though I am not against Linux and often use it myself, I maintain that FreeBSD deserves a better share of the media attention.
For those who think FreeBSD isnt meant for desktop use, I use FreeBSD extensively as a desktop/workstation on a 450 mhz p3 198 MB RAM. I use it to write code, play music, IRC, word processing with OpenOffice and run a local webserver and database off it too. I have never had any performance issues. I think FreeBSD should no longer remain silent.
My P3-650 384MB laptop runs great on FreeBSD 5.3, as do my larger machines. Its a little slow on the builds, but everything runs quick once compiled. I think the best advocacy is to demonstrate to people that they can accomplish their normal computing tasks on FreeBSD better than on other osen. What impresses my Windows friends is that I *never* have to purchase software for *anything*, and can still accomplish the same tasks they do. One balked because I do not have the firefox flash plugin functioning, but then I pointed out that he had them disabled, heh.
Media has focused solely on Linux as the Open Source option.
That might be so, but I consider FreeBSD the most successive free OS to date. What Apple made with it, proves my thinking. There’s no way they would do the same with a GPLed OS. And don’t forget there are more people using FreeBSD (yes I consider Mac OS X FreeBSD with makeup) than Linux (khm, Google zeitgeist).
yes I consider Mac OS X FreeBSD with makeup
It’s just a partial BSD userland. The kernel architecture is different, as are the file system and authentication, and the biggest part of the GUI is proprietary. And then even some standard tools have been modified (eg: top(1)). It isn’t quite FreeBSD…
Scott Long wrote:
Most recently there were a series of articles posted my
Chritos Zoulas describing why NetBSD is relevant and why it’s a better choice than either FreeBSD or OpenBSD.
Could anyone point out where I can read those articles?
I believe Scott was referring to the introduction made by Christos at the NetBSD annual report:
http://www.netbsd.org/Foundation/reports/2004.html
“[i]Media has focused solely on Linux as the Open Source option.[i]”
The media focuses on whatever the enterprises are interested in and whatever is adopted by large corporations like IBM, Novell, etc. as one of their products and provide support for. FreeBSD’s licensing scheme allows people to take what they want without even declaring it to the rest of the world. Not so with Linux, since it’s under the GPL.
The BSDs have always been more of an educational and research project. A bunch of hobbyists writing an OS that they KNOW might be used by people to make money off without revealing the source code, and most of us BSDers are very aware of that and comfortable with the fact.
FreeBSD is used in some of the busiest network sites. It is used and implemented in a lot of enterprises. Not only is it relatively easy to use, it is very stable as well.
I agree with you on the licensing part, but a lot of good code does filter back into the community. My point was it is very easy to maintain and run a FreeBSD desktop and a lot more people should be made aware of that. People are daunted when they hear of installing a BSD but once they use it they seldom go back to anything else. Features like the ports and ease of updating are strong points for desktop users who like to keep current.
I also agree with BSDs being mainly an educational and research project but if proper support is offered a lot more enterprises will be willing to migrate to them.
The usual misconception:
FreeBSD’s licensing scheme allows people to take what they want without even declaring it to the rest of the world.
No, that’s not true.
If you use any BSD code in your software, you MUST give credit to the author by distributing the BSD license along with your software, because that license is *still* covering the code you imported.
Sorry for the bold, but it’s amazing how widespread is the misconception that you can just strip away the BSD license.. No, you can’t! That’s copyright infringement, like this guy did (maliciously, and not simply because of ignorance):
http://feyrer.de/g4u/g4l.html
… that they KNOW might be used by people to make money off without revealing the source code
I disagree. If somebody modifies FreeBSD and then tries to make money off this modified version, he will succeed only if the extra code he wrote is *worth* the price he’s charging. Otherwise, everybody would get FreeBSD for free instead.
So, I really think he wouldn’t make money off FreeBSD, like you’re saying, but off *his own* modifications.
I disagree. If somebody modifies FreeBSD and then tries to make money off this modified version, he will succeed only if the extra code he wrote is *worth* the price he’s charging. Otherwise, everybody would get FreeBSD for free instead.
So, I really think he wouldn’t make money off FreeBSD, like you’re saying, but off *his own* modifications.
Which is basically why there aren’t 5000 flavours for BSD while Linux contains 5000. Simply because you don’t need to do anything to make it your “own” release. This is one of the main reasons I like BSD =)
I never understand why to much worry about BSD license: Isn´t the most of the userland license over GPL ? (KDE,GCC,MONO,…).
One second question: What’s the freebsd part cover by bsd license ? what development tools ? where can we find out ?
1) Thanks Dru!!
2) FreeBSD does make an excellent desktop. With the standards like: Fluxbox, Gnome/KDE OO.o, Gaim, Thunderbird/Slypheed, Mozilla/Firefox.
3) Updates to the kernel or userland is a snap (3 ways).
a) Just CVS and rebuild.
b) You could also do a binary update
c) You could patch and rebuild the application.
4) BSD init system. There is one config file, see Dru’s article “FreeBSD for Linux Users”.
5) 12,000 ports (App’s/Lib’s). Ports would be the source code for said App’s/Lib’s. Simple to install, “make install clean”. Or you could fetch the binary “pkg_add -r program”. Yes it also has dependancy resolution.
6) Most app’s are current releases: Mozilla 1.7.5 or you could build the beta’s that are out (if you choose).
Just my humble opinion: FreeBSD is a combination of (an analogy):
Debian for apt vs pkg_add (binary)
Gentoo for Portage (Source based)
Slackware for basic configuration (simple and clean).
As always, documentation is top notch. Just in case anyone is curious:
FreeBSD Documentation:
http://www.freebsd.org/handbook (handbook)
http://www.freebsd.org/cgi/man.cgi (man pages)
http://www.freebsd.org/projects/newbies.html (newbies)
Ports/Packages:
http://www.freebsd.org/ports
i think its a testament to the achievements of netbsd that scott long feels that freebsd has to be defended in such ap ublic way.
Well, after all it was just a mailing list post. Maybe he just got a little pissed because the NetBSD advocacy turned a bit too FreeBSD-bashing lately, considered the historical cooperation and the heavy *code sharing* that there is among the BSD projects – things that Robert Watson, in that same thread, is reminding:
http://lists.freebsd.org/pipermail/freebsd-current/2005-February/04…
Your right ulib.
I have only been using FBSD since 4.5/4.6 and there has always been a very co-operative spirit between all the BSD’s (users perspective).
Recently, it seems that there has been some infighting between various BSD’s. All the projects draw from each other IE: features/enhancements and so on. Hopefully people will realize that co-operatition will help all the BSD’s shine rather than bickering who’s OS is better.
1) Use the OS that best suits your needs. All have their
strenghts and weakness.
2) Promote your OS to gain users/developers/commercial
support.
3) Bashing helps no one, it will reflect on the person or
worse, the project itself. It doesn’t matter if your a
developer, user or whatever, people can walk away with a
perception of the project, by what you convey.
4) Remember, there is a fine line between Zealots,
Advocates, Bashers.
Ok, I will turn off the pom-poms.
One second question: What’s the freebsd part cover by bsd license?
True much of the userland is covered by a variety of licences different than the BSD. On the other hand, the OS itself is BSD licensed, and provides a fully functional system. The kernel (obviously), libarchive, libc, sshd,fetch, etc. What can you do with these? You can provide virtual hosting via freebsd jails. You can set up a dns server (mit licence – similar to bsd), ftp, ssh, mail, etc. When it comes to ports, you do have to use gcc (but you don’t need ports for running the above) – or intel’s c compiler. Obviously, only those 3rd party programs will be BSD (or similar to bsd) licenced where the developers chose to do so: postresql, enlightenment desktop, xorg, etc.
Well sheesh, I don’t see why ya went and modded me down here. I mean, the second article is all about FreeBSD vs. NetBSD. And I really do see the Dragonfly team making some great progress and design decisions, and it seems it may possibly be the dominant BSD in the future.
Whenever I read that FreeBSD 5.x is faster than 4.x I always wonder if that is true only if I use more than one processor?
Just my humble opinion: FreeBSD is a combination of (an analogy):
Debian for apt vs pkg_add (binary)
Gentoo for Portage (Source based)
Slackware for basic configuration (simple and clean).
You realize that FreeBSD had those features first, right? The difference is simply that each of those Linux distros took that one idea in a different direction.
apt is simply pkg_add with far more up to date binaries.
Portage is a direct copy of ports (hence the name), with the addition of support for optimizations above -02 and testing of various cflags.
Slackware is stated, by the author, as being an attempt to bring a BSD type installer, configuration, and simplicity to Linux.
Whenever I read that FreeBSD 5.x is faster than 4.x I always wonder if that is true only if I use more than one processor?
It’s faster from code rewrites if you get all the debug code out of the way, but in general even programs that have nothing to do with FreeBSDs core code are faster for one simple reason:
The upgrade from gcc 2.x to 3.x. GCC 3.x produces much faster code, and on a system scale that fact alone would give a noticable speed increase.
That isnt true.
On mailing list even developers stated, that 5.x is slower on UniProcessor than 4.x That is because they have rewritten a lot of things since 4.x, but their goal was stability in the first phase . The next thing to do is to fine-tune these parts, that’ll make maybe 5.x series faster than 4.x. But that didnt happen yet.
Anyway, if one is reading mailing lists on a daily basis (sure it needs a lot of time) especially -CURRENT or -QUESTIONS there seems to be developers waste lot of time/energy into flaming about meaninglesses, instead of enhancing our favorite OS
..or for example the hundreds of open-bugs at
http://www.freebsd.org/cgi/query-pr-summary.cgi
Maybe something should be made to improve the efficieny of solving bugs/making proper crash-dump analyzing easy for newbies (step-by-step walkthrough)