OpenBSD 3.7 Reviewed

busfahrer writes "Jem Matzan has written a review of OpenBSD 3.7 for Newsforge. He talks about their licensing issues, network features, upgrading packages and the new supported architectures."

Comments

Declare your bias, why don't you?

[Anonymous Coward]

Opening line:

The operating system world has been blessed by another regular release of OpenBSD.

And, no, it doesn't get any more objective further down. Nor does he talk about the licensing issues or new architectures in any detail at all - less detail, in fact, than he talks about the theme tune.

Re:Declare your bias, why don't you?

[millahtime]

licensing issues

You have issues with BSD licensing? How much freeer do you want it?

Re:Declare your bias, why don't you?

[Urkki]

• You have issues with BSD licensing? How much freeer do you want it?

Fact 1: BSD license isn't free enough to allow merging in GPLed code.

Fact 2: GPL isn't free enough to allow merging in BSD licensed code.

Wether this restriction of freedom is justifiable or not, wether it is unavoidable or not, the fact is that IMHO this is a significant restriction on freedom (both for the programmer and for the code). It forces people to "re-invent the wheel", and therefore it goes against one of the basic motivations and justifications for FOS software.

Re:Declare your bias, why don't you?

[Anonymous Coward]

You have issues with BSD licensing? How much freeer do you want it?

RTFA. The issue mentioned is that OpenBSD folks object to the Apache 2 licence, and so OpenBSD won't get Apache 2.

In fact you won't even find Apache 2, because its license is more restrictive than its predecessor. OpenBSD 3.7 includes a heavily modified version of Apache 1.3.29 instead.

Re:Declare your bias, why don't you?

[Anonymous Coward]

Fact 1: BSD license isn't free enough to allow merging in GPLed code.
Fact 2: GPL isn't free enough to allow merging in BSD licensed code.

No, the modified BSD licence - which everyone uses nowadays - allows you to mix BSD and GPL code. The result is always GPL.

But that's not the issue here - RTFA.

Re:Declare your bias, why don't you?

[Urkki]

• No, the modified BSD licence - which everyone uses nowadays - allows you to mix BSD and GPL code. The result is always GPL.

Thanks for the clarification. No wonder I was modded troll :-)

Re:Declare your bias, why don't you?

[compass46]

I'm not sure he was complaining about the BSD license so much as the article not expanding on the license fights OpenBSD has had with hardware vendors. There were only a few sentences scattered throughout.

Re:Declare your bias, why don't you?

[Metteyya]

He talks about the licensing issues. Which drives me to the question: what's the logic behind throwing away Apache 2 (because of too restrictive license) and distributing closed-source wireless drivers at the same time?

Re:Declare your bias, why don't you?

[compass46]

The drivers are open source. The board firmware is closed source. They got permission to distribute the blob for the card firmware to make wi-fi setup easier for users so they didn't have to jump through hoops to get their cards to work.

Re:Declare your bias, why don't you?

[Caligari]

Who said anything about closed source wireless drivers?

The whole point of the recent OpenBSD wireless developments are that the drivers are completely free!

Stallman gave Theo de Raadt the 2004 FSF award in Febuary as recognition for crying out loud!

Re:Declare your bias, why don't you?

[SquadBoy]

Yeah. But people who really use OpenBSD generally know better than to want Apache 2 anway. So it all works out.

Re:Declare your bias, why don't you?

[RdsArts]

Fact 1: BSD license isn't free enough to allow merging in GPLed code.

Fact 2: GPL isn't free enough to allow merging in BSD licensed code.

Fact 3: The purpose of a BSD license is to flip out and kill processes.

BSD licenses can `kill -9` anyone they want! BSD licenses cut off threads ALL the time and don't even think twice about it. These licenses are so crazy and awesome that they flip out ALL the time. I heard that there was this BSD license who was eating at a diner. And when some dude dropped a packet the BSD license killed the whole town's connection. My friend Mark said that he saw a BSD licenses totally uppercut some kid just because the kid installed GRUB on their router.

And that's what I call REAL Ultimate Power!!!!!!!!!!!!!!!!!!

Re:Declare your bias, why don't you?

[LooseChanj]

As much as I like Openbsd, you're right. The review might as well have been a cnet review of a microsoft release.

Re:Declare your bias, why don't you?

[itsybitsy]

I'll bite. How isn't BSD "free enough"?

Are you talking about the so called "freedom" that code has under the GPL that "keeps" it open? That's not really freedom as it comes at a cost, a large cost, the authors "give up" their "rights" is the cost. Now they might want to do that and pay that price (which is perfectly fine if they choose to do so), but afterwards they are no longer free to do what they want with the code, and neither are users who might choose to use the code as the "many rules" of the GPL will keep you in line with the "commune of the GPL".

Freedom isn't the right word for the GPL'd code. It's too bad that Richard Stallman usurped that word. Yes, you get some freedom but it's more like a restricted freedom only if you obey the party line.

The BSD license with its minimal terms gives authors and users maximal choices including the freedom to modify the code and not release it and sell such modifications! That's simply not an option if you want to stay within the terms of the agreement with the GPL! So the GPL isn't "free" in a way that the BSD is free.

These two licences aim at different audiences and use different methods (minimal v.s. wordy) to achive their goals.

The BSD is about "freedom of choice" of BOTH authors and users.

The GPL doesn't care about users or authors. It simply cares about the code and will impose whatever restrictions by having users and authors surrender their natural and legal rights to the commune of the GPL.

Which do you want? The freedom of choice or the rules of the GPL commune you wish to live under! It's your own personal choice until you commit your code to one of them (or another licence scheme of your choice). Choose wisely and after consideration is my best advise. Be free, stay free.

Re:Declare your bias, why don't you?

[itsybitsy]

Are you talking about the so called "freedom" that code has under the GPL that "keeps" it open? That's not really freedom as it comes at a cost, a large cost, the authors "give up" their "rights" is the cost. Now they might want to do that and pay that price (which is perfectly fine if they choose to do so), but afterwards they are no longer free to do what they want with the code, and neither are users who might choose to use the code as the "many rules" of the GPL will keep you in line with the "commune of the GPL".

Freedom isn't the right word for the GPL'd code. It's too bad that Richard Stallman usurped that word. Yes, you get some freedom but it's more like a restricted freedom only if you obey the party line, and that's not true freedom at all since your freedom is restricted. Restricted-freedom-at-a-high-cost is more like it... or Freedom-with-legirons.

The BSD license with its minimal terms gives authors and users maximal choices including the freedom to modify the code and not release it and sell such modifications! That's true freedom of choice for authors and users! That's simply not an option if you want to stay within the terms of the agreement with the GPL! So the GPL isn't "free" in a way that the BSD is free.

Re:Declare your bias, why don't you?

[setagllib]

What makes you think it's a war? BSDs harmonize with other projects. While it's rare that anything is given back to the projects, it does happen - Darwin gave back to FreeBSD in a few places.

So I highly doubt a BSD would care if, say, Windows inherited its own OpenSSH-based SSH daemon (though it might need a different SSL library). I quote from Theo, "Their security is our security", and if Windows machines are given boosts to security capabilities, the whole world benefits. There's no point in starving other projects of good code: in fact, the open source spirit is about reducing redundancy where possible! And despite occasional technical inadequacies, the BSDs are much closer to this spirit than GNU/Linux (which has much less code sharing, and a license that makes it difficult to import its code).

But I'm expecting a lot of 'omgtrol!1!!' in response to this.

Re:Declare your bias, why don't you?

[QuantumG]

Yeah, and we didn't actually hear Theo's reply to RMS but you can be sure it was only two words.

Re:Declare your bias, why don't you?

[ostiguy]

HP ships a windows SSH service with some of its management software. Of course, have we seen HP donate to OpenBSD?

Re:Declare your bias, why don't you?

[nacturation]

... and the second word is three letters long, but I haven't determined whether it begins with an 'o' or a 'y'.

Re:Declare your bias, why don't you?

[jbplou]

you don't get objective reviews on slashdot.

Re:Declare your bias, why don't you?

[Brandybuck]

...mix BSD and GPL code. The result is always GPL.

Sort of like mixing champagne and sewage. The result is always sewage.

Re:Declare your bias, why don't you?

[Brandybuck]

Too bad people don't understand this. I spent the last couple of months developing a driver for a new board we're making. As a software developer I don't even get the source for the firmware, but have to get the firmware binary from the hardware department. And I work for the same company! Less then fifty feet away from the guy writing the firmware!

This isn't unusual, so people need to stop acting like they've been raped if they don't get complete verilog sources. Full specs for the hardware is much more important.

Re:Declare your bias, why don't you?

[jd]

I'm not sure that's the point being made - the article posting states that they DO discuss BSD licensing, so if they don't, then mentioning that fact is entirely relevent, whether the BSD license is OK or not.

That's it, I'm switching

[Anonymous Coward]

I liked linux back when nobody knew what it was. Now my mom even knows what it is.

Good bye linux, hello obscurity, er, OpenBSD!

Re:That's it, I'm switching

[Ledskof]

You're mom knows that linux is a kernel?
Probably 2 or 3 people (counting me) in my IT dept can answer "what is linux" properly.

I did ask someone once if they knew what BSD was, and they said "linux".

Re:That's it, I'm switching

[filesiteguy]

What's wrong with that? My mom runs it as her primary OS. (SuSE 9.2)

Everyone is biased....

[NXIL]

but in their own way. Everyone is unique--we are all alike in that respect. There is no spoon.

Re:BSD, the history

[oKtosiTe]

BSD is not dying at all. Most of the major operating systems are based on BSD, or have borrowed code from it.

Why Can't We All Just "Get Along"?

[snookerdoodle]

"...an effigy of a crown-wearing penguin."

Sheesh. The prez in "Mars Attacks" said it best:

"Why can't we work out our differences? Why can't we work things out? Little people, why can't we all just get along?"

Mark

Re:Why Can't We All Just "Get Along"?

[nurhussein]

Then the aliens kill the prez and plant a flag on his body and laugh manically.

Not a good image to evoke.

Re:Why Can't We All Just "Get Along"?

[snookerdoodle]

Nah - you understand perfectly.

Just picture a bunch of little BSD Daemon guys killing Stallman and... ;-)

Mark

Re:Why Can't We All Just "Get Along"?

[Rick the Red]

Just picture a bunch of little BSD Daemon guys killing Stallman

Hey! That's my "happy place"! I saw it first! Thanks for blabbing about it on Slashdot. Now everyone will want to go there, it'll get all crowded, then I'll have to find some other place to go in the middle of meetings...

Actual information

[slavemowgli]

Is there *any* actual information in this article at all (useful information, anyway)? The only tidbits I could find boil down to things like "my on-board controller didn't work", "I couldn't compile KDE myself", and "this and that specific option to this and that program gives a warning when you use it".

Outside of these things, the only pieces of information I could find boiled down to "there's two new ports", "it still doesn't include Apache 2.x", and "you get daily (in)security reports mailed to you". If it wasn't for the irrelevant fluff mentioned above, I'd assume the author of this article hasn't even installed OpenBSD and instead just looked through the website and maybe Google'd for some extra information.

I really hope the author didn't get payed too much for this, because no matter how much he got, the article wasn't worth it.

Re:Actual information

[LWATCDR]

I have to admit I am disappointed. No benchmarks. No list of new features, No detail on setting up a server running it.
Heck it was more of a bad press release than a review.

Re:Actual information

[molnarcs]

Once I criticized this guy (in a comment) for an even more shallow review FreeBSD 5.3 - because I firmly believed (and I still believe) that he didn't even bother to install it :)) He wrote a review on 5.2 a few months before that, which was very critical (and rightfully so, 5.2 was a quite flaky release) but at the same a really good review. In his 5.3 review he basically recirculated the points he made for 5.2, and even got some really weird factual mistakes. His response was to put me on his foe list ... now that's a pretty childish behaviour, isn't it?

Later he wrote an article on newsforge about "Being Free is Hard to Do" on free software, than he submitted this article to slashdot under his nick (ValourX), describing it in the following terms:

What is more important to you -- the four freedoms of Free Software, or the ability to maximize the value of your computer? It's a question that comes up on Slashdot often, but rarely is it so well argued as it is in this NewsForge article. Link.

What shameless self promotion! It is a pity, for once this guy wrote excellent reviews and articles, but what he does lately is prostitution, not journalism.

Re:Actual information

[ulib]

I wish I could mod you up.
After his FreeBSD 5.3 "review", it's amazing that a piece written by this guy - who rarely does argue so well as himself :D - can make news *anywhere*, let alone on Slashdot.

Thanks for revealing another one of these trollish characters - you really seem to have a talent for spotting them. ;)

(Yes, I want to be on his s**t list too. Does it show? :D)

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

Re:Actual information

[weicco]

Benchmarks? Why do you want benchmarks from OpenBSD? Everybody knows it sucks compared in speed to other OSes so there really isn't need for benchmarks. OpenBSD was never about speed but security and all the "benchmark" I need is this: http://secunia.com/product/100/

Re:Actual information

[Argon]

I felt the same way. Where's the review exactly? It read more like a personal blog entry than a "review" of a new OS release.

Re:Actual information

[molnarcs]

Now only if I could learn to spell correctly ;)))

Re:Actual information

[LWATCDR]

1. Computers need to be fast
2. Computers need to be secure.
Without knowing how fast and how secure it is how do you know if it will suit your needs?
Is it faster than the last version?
What tuning options do you have?
How is it on disk access.
How does it do an large reads?
How does it do on small reads?
What raid drivers does it have?
How fast are those?
How fast is the TCP/IP stack?

You say it sucks but is secure? How bad does it suck then? Am I going to need a 4 way SMP box to run a firewall for my house on it? Or will it take a 300 mhz PII to handle a small office firewall?

yea a Quake or POVRAY benchmark would be less that useful but benchmarks are of some value even for OpenBSD.

Re:Actual information

[the morgawr]

Slightly more technical answer:

1. micro-benchmarks arn't relaible use application benchmarks (i.e. the best thing to do is install it and see before putting it into production)

Routing - OBSD is really fast and uses almost no resouces on my small network (~30computers, 1.5Mbit connection that stays at about 90+% useage;); i'm using an old K6-II 400 /w 64MB of ram and a 2GB hdd. CPU usage hasn't topped 3%, disk isn't used except durring boot up and memory stays at about 20%

Multimedia and Development- For reference my render times, encoding times, and compile times are only maginally different(to the point of statistical noise) between Linux, FreeBSD, and OpenBSD.

Disk speed - I don't notice a difference, I can still cap out a 100MB/s network with my OBSD server when I transfer large files

Encryption - OBSD has the best hardware encryption support, if you need crypto, OBSD is the fastest.

Misc details:

OpenBSD 3.7 uses an N:1 threading model. So Linux and FreeBSD clearly beat it here. There is work going on at the hack-a-thon to transition to a 1:1 model based on Plan9 rfork, so expect this to change in the future.

SMP - OpenBSD still uses biglock so for applications that use a bunch of system calls OpenBSD will be slower.

The threading is mostly noticed when you try to run XMMS and then scroll throught the playlist, or when you try to run a SQL server. The devs know about them and are working hard to fix it. The big-lock is surprising less troublesome then it was with linux and FreeBSD, I don't know why. I presume they intend to remove it in favor of another option in the future.

I use OBSD exclusive now and am VERY happy with it. I've used almost every major Linux disto and the three major BSDs and OBSD is my clear favorate. The documentation is superb, the system is well integrated, and everything just works. It is a lot less stressful to use and admin than any other system I've ever used. If you have a spare computer I'd suggest trying it out. Note, the biggest "downside" to OBSD is that users are expected to actually learn. The documentation is top notch but it doesn't help you if you don't read it.

As a side point, I've had a ton of general OpenSource advocacy success with OBSD. For some reason I have much better results(i.e. people are happy and keep using OpenSource software) when I help them install OBSD vs. when I install Linux. Take from that what you will.

Re:Actual information

[Brandybuck]

He's up to his old tricks. Yesterday he put out a 5.4 review. Deja vu! It reads just like his 5.3 review! Including wierd factual mistakes!

The jemreport has a widespread reputation, and it's not a good one.

Re:Not dead?

[millahtime]

I guess BSD isn't dead after all.

It's now a zombie. Neither dead or alive.

Re:BSD

[Big Mark]

Oh come on. Real leeter-than-thou people wouldn't dream of using anything that runs on x86.

Standard OS Review Questions?

[Adelph]

I've seen crap reviews like this all over the place recently. Perhaps some standardized questions to answer about the OS as it is being reviewed would make the reviews more informative and comparable?

I expected more, but of course...

[dayid]

As someone who replied to the original article (see bottom of original link) mentioned, it would've been interesting to have seen a true comprehensive analysis of OpenBSD, rather than a lot of "I think" and "I liked".

I would have appreciated the article more if it were a lot more in-depth, but perhaps that would've ward off others. I would like to see him not just talk about the install process (initially), but also how easy it was to install applications (and not just "I had to type too much"), configure them (interface-configuration, or purely text-editing), and finally - how well they all interacted. Now, I know that sounds more like an analysis of the individual applications rather than the operating system, but what is an operating system if not a platform that you use to interact with applications?

We also hear about the "new wireless" stuff... where was that? Test with multiple cards? USB-Wireless perhaps? PCMCIA Wireless? To tout such things (even in the review) and then not do anything with them is rather disappointing.

Re:I expected more, but of course...

[TheLetterPsy]

A lot of the information on the wireless work can be found in the Changelog. Basically, they got a lot of work done and are continuing to work on 802.11g chips and HostAP protocols for many wireless devices.

Also, as noted in a previous story, the Sharp Zaurus port allows you to (with a CF ethernet adapter) set up a handheld, on-the-go wireless AP.

A much better source of information can be found at this ONLamp Interview With OpenBSD Developers

Re:Not dead?

[matt-fu]

It's more machine than man.

Re:BSD, the history

[/ASCII]

What are you saying??? Other people are eating BSD alive, and you say it is not dying? Microsoft bit down hard and tore BSDs left foot right of! It is decaying in the fould beasts belly as we are speaking! Of course this is killing BSD. BSD is just limping along, trying against all odds to outrun the pack of wild hyenas that are stealing more of it's codebase every day. But where are they going to run with only one foot? Where I ask you?

Note to moderators and BSD zealots - this is a _joke_.

The shot at RMS for the day

[The Slashdotted]

From the article:

The theme of the OpenBSD 3.7 CD set is The Wizard of Oz, and the cute little CD jacket cartoon strip shows the OpenBSD mascot and friends on a journey to achieve better wireless card drivers. Their adventure takes them to the Emerald City to meet the great and powerful Wizard of OS himself -- an effigy of a crown-wearing penguin. The man behind the curtain turns out to be a Richard Stallman-like character with GNU horns. The characters are disappointed because the Wizard ends up being "all talk -- no action!" So they decide to code the wireless driver by reverse-engineering the device.

Re:The shot at RMS for the day

[Slashcrap]

Just curious, when was the last time that RMS reverse engineered a useful device and posted the driver source?

Just curious, when was the last time Theo wrote a compiler suite capable of e.g building OpenBSD?

That's what I love about the OpenBSD community - the total lack of arrogance.

Re:The shot at RMS for the day

[Megaweapon]

Arrogance or not, at least they are actively persuing hardware manufacturers to get specs opened up a bit. None of those corps are going to listen to RMS while he preaches GNU philosophy from his pulpit. And hey, why bring up the compiler? It's free in all the appropriate senses, so Theo and crew are simply using available tools.

Re:The shot at RMS for the day

[Megaweapon]

And the BSDs wouldn't be where they are now without GCC. Trying to argue otherwise is simply revisionist history.

That assumes that another compiler wouldn't have matured along with the BSD progression. Granted, GCC was the first solid one on the block, so it got used. Nothing against GCC, but you can't just state that without it various "open source" projects would not be where they are today. To argue otherwise is simply pretending to be all knowing.

Besides, GCC should be called K&R GCC, because where would GCC be without the C language?

And I can point out your arrogance.

*shrug*

Answer this - what is the greatest contribution to the Free Software community? GCC or being able to distribute the firmware for a few wireless cards?

GCC has certainly been a boon to the "free/open/whatever-word" communit(y/ies). However a compiler by itself isn't terribly useful if you don't have much operating system source to compile it against. Hence the importance of device drivers, firmware API's, etc. But I'll try to remember your point when I'm setting up a high availability GNU/Hurd box with all the latest hardware...

Re:The shot at RMS for the day

[Megaweapon]

Quit your whining. My question was valid.

Re:The shot at RMS for the day

[Megaweapon]

No, the point is that Theo and crew are taking an active approach to get what they need. To make a claim that BSD and related projects wouldn't be where they are without GCC implies that only the GNU project was/is capable of producing a good C compiler. It just happens that the GNU project made a good C compiler under palatable licensing terms, so the various BSD projects used it. So what?

As far as "slamming" RMS, I think the cartoon in question is basically just a "joke", and I think some people should take a chill pill before trying to scream foul.

Re:The shot at RMS for the day

[sp0rk173]

Wow. What a lame argument, "If GCC didn't exist, BSD wouldn't be where it is today, and you can't say otherwise because otherwise didn't happen!"

The simple fact of open source isn't what gets built, it's the spirit behind it - a spirit that exists without some fat bearded douche bag writing PART of a compiler (which isn't JUST WRITTEN BY RMS, IT'S WRITTEN BY MANY MANY OTHER PEOPLE TO REACH IT'S CURRENT, USEFUL FORM!!! Jeebus people). That spirit would move other people to develop their own compiler, much like the anti-GPL spirit in the BSD community has fostered increasing support for TenDRA. So, to answer your question, what is the greatest contribution to open source? Motivation to write open source software. If GCC didn't exist, that motivation would push coders to develop a different free compiler. To say otherwise is to speak with dogmatic blinders.

It's not as though RMS is the only one who can start a compiler and get it to attain self-sustaining momentum. Any good college CS program involves a class in compilers. A compiler is not some great mystery of comptuers, it just takes a lot of work to get one that works well enough for production use. Once TenDRA becomes stable and feature-rich enough to be used in production, BSD wills switch over to it in droves.

Re:The shot at RMS for the day

[Simon Lyngshede]

The man behind the curtain look more like Ty Semaka (http://www.tysemaka.com/). He's guy behind the lyrics and art work, well most of it anyway. It doesn't look anything like Stallman.

Re:The shot at RMS for the day

[Slashcrap]

To make a claim that BSD and related projects wouldn't be where they are without GCC implies that only the GNU project was/is capable of producing a good C compiler.

Right, so presumably you think that there is an infinite amount of development resource behind OpenBSD? If Theo and his crew had needed to put resources into developing a suitable compiler before they even started the project, that would have made no difference to it's current status?

It just happens that the GNU project made a good C compiler under palatable licensing terms

The last time I checked, GCC was licenced under the GPL. According to Theo the GPL is the work of Satan. Do you not read the OpenBSD mailing list?

As far as "slamming" RMS, I think the cartoon in question is basically just a "joke", and I think some people should take a chill pill before trying to scream foul.

I wasn't exactly screaming foul was I? Merely pointing out that certain people have an over-inflated sense of their own importance which isn't entirely based in reality.

The fact that this is apparently considered trolling doesn't do much to refute my argument that the OpenBSD community suffers from it's arrogance.

Anyway, feel free to mod this down. You can carry on living in your ivory towers confident in the knowledge that no-one really gives a shit.

Re:The shot at RMS for the day

[evilviper]

Just curious, when was the last time Theo wrote a compiler suite capable of e.g building OpenBSD?

Tendra is more-or-less on his TODO list.

The parent's post was too specific. Let me generalize it... When was the last time RMS wrote useful code of any kind, rather than just blowing hot air?

Re:The shot at RMS for the day

[QuantumG]

No, a BSD licensed compiler would never have matured as everyone who made a new cpu would have released a proprietary version of that compiler ported to their platform instead of giving the port back to the community. The GNU license is the only reason why GCC became as widely used as it was.

Re:The shot at RMS for the day

[Rick the Red]

Are you really claiming that the chip manufacturers are the ones who port GCC to the latest CPUs? How many times has Intel ported GCC? AMD? Anyone?

More to the point, do you seriously think that if your scenario played out the BSD folks could not port their compiler to the new CPU as easily as the GCC folks port thiers today?

Re:The shot at RMS for the day

[QuantumG]

We were talking about ancient history where you could actually start a chip company and compete in the UNIX server market on the strength of your architecture. Many many companies did this and the first thing they did was port GCC to their architecture, then compile their AT&T licensed UNIX to it.

Yes, that's right, I'm not talking about some wild theory about why licensing GCC with GNU was a good thing, I'm talking about documented history. There were (and still are) other alternatives to porting GCC but every one of them was inferior solely because anyone who put any work into them refused to give that work to others. The fact that RMS demanded that you give up your changes if you want to distribute GCC meant that all these little companies contributed a little bit of work each and the whole got built.

Re:The shot at RMS for the day

[Rick the Red]

Many many companies did this

Name some. I've used UNIX workstations from Apollo, HP, IBM, Sun, and SGI. They all had proprietary compilers, not GCC. I know because our code had to compile on all of them.

Re:The shot at RMS for the day

[QuantumG]

You just did, every single one of them (except possibly Apollo which I've never heard of) has contributed changes to GCC. The fact that they have properietary compilers as well really proves nothing. Meanwhile, go look at the length of the contributors page.

Re:BSD

[Anonymous Coward]

Actually, from an admins point of view, the BSDs are easier to learn than any linux distribution. That is not to say they are easier to use, but they are simpler when it comes to managing the system:

Less complicated init;
MUCH better documentation;
Less painful filesystem management (though LVM2 is really nice);

The downsides are significant however:

Bad support for esoteric hardware;
Less vendor support;
Fewer eyes looking over the code (though, to be fair, there is MUCH less code for them to look over)

Re:Not dead?

[snookerdoodle]

Heh heh - yesterday in Sunday School at church, my 9-year-old son learned about a guy named "Lazarus" who, in the Bible, is raised from the dead by Jesus.

His teacher was a little flummoxed because my son kept referring to Lazarus as "re-dead", which refers to some zombie like characters in some of his Zelda games on his Gamecube.

For some reason, she didn't get any happier after I "educated" her about who Zelda was... Believe it or not, we really only let him have 30 minutes of total screen time (TV + Computer + Gamecube) per day. I'm not sure *she* would believe me. ;-)

Mark

Re:Mandatory Access Controls?

[tim_mcc]

the default install is essentially useless, as there have been holes in the services that most people would want to enable...

Well, this depends really on what you think *most* people want. The system contains a number of tools from the default install, including:

ntpd
pf
bgpd
isakmpd
spamd
OpenSSH
X.Org
Gcc
Perl
Apache
OpenSSL
Groff
Sendmail
Bind
Lynx
Sudo
Ncurses
Heimdal
Arla
Binutils
Gdb


Although I may have missed few...

As you can see from the apps mentioned, there are a number ways you could put a default install box to use. Basic web server, firewall, mailhost..?

There is less reason to review OpenBSD.

[emil]

When I ran RedHat, there were some pretty annoying things that got changed from release to release (inetd disappears, two different C compiler installs because of kernel problems, etc.).

This kind of stuff doesn't happen in OpenBSD. From an administration perspective, my first 3.2 install is very similar to the 3.5 that I run now, which itself is similar to 3.7. There are no large architecture changes (perhaps because things are well-thought-out from the start).

Because of this, you pretty much know what you're getting when a new OpenBSD release comes out. The installer is practically identical, and the running system yeilds few surprises. There will always be new features, but there won't be lots of things to unlearn.

So no, I don't really pay much attention to the reviews. The list of new features on the OpenBSD web page pretty much tells me all that I need to know.

Re:There is less reason to review OpenBSD.

[evilviper]

There are no large architecture changes (perhaps because things are well-thought-out from the start).

Or perhaps because releases are frequent enough that changes are spread out over different releases.

How many times has OpenBSD completely broken binary compatibily in the past ~2 years? 3 times or so?

Personally, it's the tiny useless changes that really bother me. In your config files, one option will be known by a slightly different name from release to release.

Re:There is less reason to review OpenBSD.

[evilviper]

Upgrading RedHat is far simpler than upgrading OBSD. With RH, you just stick the CD in and choose upgrade.

Now that's pure BS. Upgrades with OpenBSD are far simpler by any account. It mainly has to do with the underlying OS being simpler (elegant, whatever), but no question it is simpler.

Instructions for upgrading OpenBSD are . Now please tell me which is more administration work.

Those instructions are quite verbose, and really talk a lot about borderline cases that most everyone can ignore. Files in /etc rarely change, and you only really need to delete all your packages if there's been some really major architectural change (like the one-time switch from a.out to elf).

Okay Troll, I'm done with you. Go away.

Re:There is less reason to review OpenBSD.

[ArbitraryConstant]

"two different C compiler installs because of kernel problems"

OpenBSD had GCC 2.9 and 3.x for a while. I think 3.7 is the first release in a while with just one C compiler.

"There are no large architecture changes"

What, like the IPF to PF transition? Or the a.out to ELF transition that broke all the old binaries?

OpenBSD is my OS of choice, but let's not kid ourselves. It has uphevals from time to time.

Re:There is less reason to review OpenBSD.

[emil]

OpenBSD had GCC 2.9 and 3.x for a while. I think 3.7 is the first release in a while with just one C compiler.

Each platform had only one compiler, and the kernel was built with that same compiler. Red Hat 6 came with two different compilers on each platform.

What, like the IPF to PF transition?

I started using OpenBSD at 3.2, and this was already done, so it was no pain to me. Implementing PF was a sound choice, it caused a major version upgrade (2 to 3), it removed a big licensing problem, and it was obviously the correct way to handle the situation. I would much rather have the IPF to PF transition than the ipfwadm to ipchains to iptables mess.

Or the a.out to ELF transition that broke all the old binaries?

I knew that this was coming, and I planned for it. Yes, a major version upgrade was probably called for, but it didn't really change anything from an administration perspective.

I switched from Linux

[Anonymous Coward]

Last night I switched from Debian unstable to OpenBSD 3.7, on account of the better wireless support.

So far it's been a good deal. I copied my $HOME from Debian, installed a bunch of stuff from the ports tree, and I can hardly tell the difference now, other than better wireless support, and probably a cleaner userland.

OpenBSD's base system is great, and though the ports tree is nowhere near as massive as Debian, it still contains nearly 100% of the relevant tools that I use every day, packaged in a very clean manner. I'm satisfied!

Re:I switched from Linux

[jruschme]

Let me begin by saying that I want to like OpenBSD 3.7, I really do, but I find that I can't... not enough to switch to it.

Last week, my dual-boot Win98/Ubuntu laptop (an Inspiron 3700) ate itself, so I decided to use the opportunity to try OpenBSD 3.7, given the allure of the new wireless drivers.

The install went well and I got X running as soon as I realized that I needed to use 'xorgconfig' rather than 'xf86config3'. I downloadeed a snapshot of the ports tree and did the obligatory builds of Gnome (2.8), Firefox (1.0.4), and Mozilla (1.7.8); the latter in an attempt at getting a Gecko SDK. I even built the native JDK.

After all that, though, I came back to the same issue which has always kept me from running a BSD as a fulltime OS- limited ports. I understand the licensing issue, but compared to the NetBSD package tree and the even-more vast FreeBSD ports tree, the OpenBSD ports tree is downright spartan. Many packages are not there (mplayerplug-in, w32codecs, OpenOffice)while others seem hopelessly out of date (Netatalk, Wine). While it is likely that many of the missing ones would build from source anyway, this removes most of the advantages of the package system.

As for the rest, this is perhaps the legacy of the Linux vs. BSD wars. In the old days, some software would run on BSD, some on System V, some on both. Back then, it always seemed like the "interesting" ones would run on BSD. Today, it seems most of the OpenSource development centers on Linux with BSD, at best, an afterthought.

That said, perhaps it's time to try a FreeBSD iso... I need to wipe the drive anyway. :-)

Re:I switched from Linux

[ulib]

If the number of ports/packages is one of your main concerns, FreeBSD currently has more than 12.000 ports.
--
Being able to read *other people's* source code is a nice thing, not a 'fundamental freedom'.

Re:I switched from Linux

[sp0rk173]

Definitely go for FreeBSD Then. The only two OS's i settle into after 5 years of extensive experimenting and shuffling are FreeBSD and Gentoo. If you do try FreeBSD, my word of advise is to install all third party apps from the ports tree, and as few from the install CDs as possible. Things just end up cleaner that way.

Oh, and don't listen to the bullshit about portupgrade being a sad excuse for an upgrade mechanism. I kept a production server running for 2 years using cvsup and portupgrade. I actually ran a portupgrade on my box last night after a month or so of stagnation and only had two problems, one of which was solved by deleting a partially downloaded .bz file, while the other was solved by simply running portupgrade again.

FreeBSD is indeed a solid OS, with LOTS of ports in it's ports Tree.

Re:I switched from Linux

[TheRaven64]

I understand the licensing issue, but compared to the NetBSD package tree and the even-more vast FreeBSD ports tree, the OpenBSD ports tree is downright spartan

You can, of course, use NetBSD's pkgsrc on OpenBSD - and the OpenBSD FAQ even recommends you do this in some situations. If you find software you want in pkgsrc but not in OpenBSDs ports (by the way, you should really use packages unless you have a very good reason for compiling from ports), then this might be a better solution for you.

Re:I switched from Linux

[the morgawr]

*sigh*

As I always tell people: OpenBSD is great, provided you actually read the doucumentation! Since you seem to want OpenBSD to work exactly like Linux so that you don't have to read anything, I'm going to suggest that you just stick with Linux. This isn't intended an insult; for some reason, some people just don't want to read the fine manuals. Those people need to avoid OpenBSD like the plauge. Put another way, your post makes it obvious that you didn't read the instructions...

>I downloadeed a snapshot of the ports tree

Two things:

1. Use packages whenever possible
2. Ports snapshots are for -current, if you must use ports, use the tgz from the same version of the OS. Unlike Gentoo and FreeBSD, the OpenBSD 3rd party software is synced with the OS; in general -current ports and packages will not work correctly with -release and -stable

> Many packages are not there Or you just didn't look for them when they wern't were you expeced them to be; for example, w32codec is in graphics/win32-codecs/.

Re:I switched from Linux

[Triumph The Insult C]

ports are not audited nearly at the same level of base. auditing is usually up to the maintainer. patches are sent upstream and are hoped to be incorporated, but they aren't always

Re:I switched from Linux

[toadlife]

I would recommend anyone who uses FreeBSD to give portmanger (/ports/sysutils/portmanager) and portsnap (/ports/sysutils/portsnap) a try.

I never had any big problems with the portupgrade/cvsup method myself, but portmanager is MUCH easier to use.

I use portmanger in place of portupgrade and portsnap in place of cvsup now, and updating has never been easier.

Re:Mandatory Access Controls?

[tim_mcc]

You have a point, although as far as memory serves sendmail *is* enabled by default too.

At least, I don't remember enabling it on my boxes. But then I'm not known for my amazing memory.

Forgetting one tiny detail...

[ulib]

>Most of the major operating systems are based on BSD, or have borrowed code from it.

Some of the major operating systems *are* BSD. FreeBSD, NetBSD, OpenBSD.

Re: That's it

[ulib]

>Now my mom even knows what it is.

Your mom knows a lot of other things you wouldn't suspect.

Oh, and btw BSD rules. So, you *please* stay away from it. Thanks

(No, I didn't forget the "post anonymously" blah blah)
--
Requiem for the FUD

No troll, I'm dead serious and love OpenBSD

[LM741N]

When will OpenBSD finally boot above cylinder 1024 or whatever? I am very serious about this because I love OpenBSD and would like to see it on more desktops. It has progressed much in the last 10 years.

OpenBSD is not open to the typical install process, ie 10Gig of Windows then no possible booting for OpenBSD.

Do we have to wait for version 5.0 before Theo "gets it?"

Re:No troll, I'm dead serious and love OpenBSD

[Anonymous Coward]

It has done this for some time now.

Re:No troll, I'm dead serious and love OpenBSD

[nacturation]

I just installed OpenBSD 3.7 on Sunday. I have a 120GB SATA drive with an 80GB Windows partition. So I used OpenBSD's fdisk to allocate 20GB for OpenBSD *after* the Windows partition. I marked the wd0a partition (OpenBSD) as active, rebooted, and it worked just fine. I now need to figure out what boot manager I'm going to use, but that's a different story! :)

If you're still experiencing those kinds of issues, the cause is likely to do with an outdated BIOS or other hardware issue rather than a deficiency of OpenBSD.

FUDster.

[Anonymous Coward]

I know this is slashdot, but please stop spreading FUD.

http://slashdot.org/comments.pl?sid=127944&cid=106 91304

Re:Free, but not quite

[Nimrangul]

But it is not proprietary, it is being sold off. It's is not exclusively owned, it is exclusively copyable.

Re:BSD

[PapaZit]

Fewer eyes looking over the code

I don't know if this is really true. There are three major BSD "distributions" with subtle differences. Fans of each routinely look over the code for the others looking for good, "stealable" code. Not only does that mean that people are looking at the code, but informed "outsider" coders are looking at it with a critical eye. So, even if the code is reviewed by fewer people, it's reviewed by people who are more likely to notice, report, and fix bugs.

Re:Mandatory Access Controls?

[Nickus]

Sendmail is enabled by default but it only listen on the localhost interface.

Apache2

[paulwalker]

Open BSD does not support Apache2. This affects me personally because I love Apache and can't do without it. I don't know how the rest of you feel...I really don't see why any1 should take up such an OS. Windows rocks!!!

Re:Apache2

[TheRaven64]

Apache 2 has a license which the OpenBSD team feel is too restrictive to include in the base system. Their fork of Apache 1 is very secure, and runs in a chroot jail be default. Unless you have something that requires Apache 2 specifically, I would suggest you keep it. If you really do need Apache 2, then it is in the packages collection, and so can be installed with a single command.

Re:Mandatory Access Controls?

[sp0rk173]

Also, not trying to start a war, but I fail to understand why the six years without a hole in the default install is used so often to promote it..., the default install is essentially useless, as there have been holes in the services that most people would want to enable...

Well, the philosophy is this, "Hey, here's an OS that will be secure out of the box. You don't have to scramble around the default install to lock it down. You are now free to build upon it with the degree of security you wish to implement." So...whats wrong with that? I think it's a good selling point for admins that don't want to deal with bullshit. Whether it's practical or not is moot. They do it because they want to do it, period.

This then ties into your other question - by auditing all of the code in their base system (which another poster has replied with details on) you can be fairly certain that by enabling openbsd's BIND, Xorg, etc you will have a system that runs well-tested, combed-through software.

As far as a comparison betweeen SELinux and RSBAC, i can't go there as i don't have that much expertice.

Re: That's it

[Directrix1]

Just out of curiosity do you like FreeBSD or OpenBSD better. And why?

Another review of OpenBSD

[Anonymous Coward]

there's a review that looks at openbsd from the perspective of its ancestor, netbsd:

http://www.feyrer.de/NetBSD/openbsd-comparison.htm l

looking at openbsd commit logs, the apm issue was solved shortly after that review came out, but without mentioning the review.

Re:Another review of OpenBSD

[Triumph The Insult C]

article was written april 2005

hackathon was last week

there was lots of communication between developers on what people might be working on in the time leading up to the hackathon

oh, nevermind. i am wrong. developers communicating just isn't possible! the little dickwad who has something in for openbsd is just being completly subjective

Re: That's it

[ulib]

FreeBSD because of the number of ports, i386 optimization, availability of features that one day could turn out handy (like jails).
There are also other reasons related to the goals of the projects - I like FreeBSD for emphasizing the "tool" aspect of software, keeping policies/politics completely out of the door. But it's not that I don't respect OpenBSD activism, as a matter of fact I do, they have a point (and by pressing hardware vendors they've already got amazing results). It's just that the FreeBSD point of view happens to be closer to mine. (Btw this last issue influences which one I like better, not which one I use, since IMHO this is not a reason to use one over the other.)
Anyway, I think that what the *BSD projects have in common is far more important - that is, the academical spirit of the BSD license.
--
Being able to read *other people's* source code is a nice thing, not a 'fundamental freedom'.

Re:Requiem for the FUD

[Adam Avangelist]

I notice your post is titled Requiem for the FUD. A requiem as defined by dictionary.com is...

# Requiem Roman Catholic Church.

1. A mass for a deceased person.
2. A musical composition for such a mass.

# A hymn, composition, or service for the dead

The common theme among the definitions is death.

Quite ironic because we all know the connection between this and the less than funny joke on BSD... Nevermind.

Re: That's it

[Directrix1]

So the difference between OpenBSD and FreeBSD (besides the maintainers of course) is more along the lines of politics (OpenBSD only allowing software in which meets their definition of free). Do these two share between each other? Is there a common BSD kernel or anything like that?

Re: That's it

[ulib]

I simply told you which one I like better and why (i.e. what you asked for). I didn't tell you what are the differences between them, so your deduction is wrong.
In fact, the main differences are technical, in their very goals: while FreeBSD focuses mainly on features and i386 performance, OpenBSD focuses mainly on code correctness and security.

>Do these two share between each other?

Sure they do - and massively.
For example, one little jewel that came from OpenBSD to the other *BSDs is pf (packet filter), that has an excellent reputation for its being very clean and easy to use.

>Is there a common BSD kernel or anything like that?

No.
The *BSDs are developed like OSes, not "distros". So, while they massively share code, they maintain their own kernels.

To better understand the differences, it helps to notice that OpenBSD was born as a NetBSD fork, 8 years ago - and even today, it shares more code with NetBSD than with FreeBSD.

But to understand even better, well.. FreeBSD and OpenBSD are renowned for their excellent documentation, that is well worth having a look at.
http://www.freebsd.org/doc/en_US.ISO8859-1/books/h andbook/index.html
http://openbsd.org/faq/index.html
--
Requiem for the FUD

Re: That's it

[ulib]

I linked to the FreeBSD Handbook, that is the user documentation. But for more general info about the FreeBSD project you might want to have a look at the FAQ first
http://www.freebsd.org/doc/en_US.ISO8859-1/books/f aq/index.html

I also forgot: this is a very nice place to find competent and informative answers to BSD and Unix related issues
http://bsdforums.org/forums

Btw, one little thing that the forum above has, and IMHO Slashdot is missing *badly*, is the ability to edit comments to add things you forgot (damn it).
--
Requiem for the FUD

Re:Mandatory Access Controls?

[Homology]

Both Sendmail and OpenSSH is enabled by default, but Sendmail only listen on localhost.

The last remote hole was in OpenSSH, if I remember correctly.

Server OS

[eraser.cpp]

OpenBSD is really more of a server OS. Sure you /can/ use it as a desktop, but there are better alternatives. I think its strongest point is how in /etc/rc.conf one can simply change say named_flags=N to named_flags="" (command arguments could go inside the quotes) and bam! BIND is up and running. Many of these services are available in /etc/rc.conf from the default install. Thus with OpenBSD it is possible to bring up a reliable and secure server fairly quickly.

Re:Server OS

[RazzleDazzle]

Of course anyone who actually USES OpenBSD knows that /etc/rc.conf should not be modified and you should really modify /etc/rc.conf.local instead. That way, when you upgrade, changes you make are not lost since /etc/rc.conf might have new options and /etc/rc.conf.local will still work as expected.

Also, what alternatives are so much better than OpenBSD for a desktop? Unless you are trying to play 3D rendered games, OpenBSD can do pretty much anything FreeBSD or Linux can do. Have you not heard of emulation? A minor to zero noticeable performance hit and blammo you can run Linux binaries. Where all this negative FUD comes from towards OpenBSD as a desktop is beyond me? If you are looking for a SECURE desktop, how is OpenBSD not appealing? If you are NOT looking for a SECURE desktop, you should be?

Also their licensing will never prevent you from using any of their software in any way you choose. They think the GPL is too restrictive and segregate any GPL or less free licensed software from the rest of the OS to minimize contamination. They do their best and have made, IMO, a wonderful OS that works for more than a server/firewall.

Re:BSD

[evilviper]

Typical troll. Say a few nice things, then slip in factually incorrect BS in there...

Bad support for esoteric hardware;

Really? Have you heard of NetBSD? Do you not realize that FreeBSD has been not only on-par with Linux, but ahead of it in some areas for many years now?

Less vendor support;

That's debatable. There are fewer vendors for BSD-based OSes, but you get perfectly good support from those that do exist.

Fewer eyes looking over the code

Saying that this is a disadvantage is dishonest. The idea is that more eyes makes code more secure... While that theory has been dismissed many times, even if you do believe it, you can't possibly deny that all BSD distros have a much better security tract record than Linux.

Re:Free, but not quite

[kernelpanicked]

Only the format of the cd is Theo's copyright, which I don't see an issue with. Especially since he gives away the format file anyway. Am I to take it that it's only free if it's handed to you on a silver platter with absolutely no effort of your own. That is pretty lame.

Re:Mandatory Access Controls?

[TheRaven64]

OpenBSD has a number of things that help insulate the user from security holes in programs, such as Propolice stack protection, W^X memory protection, randomised malloc, etc.

One of the most useful of these is systrace - access control on a system call level. You can explicitly grant access to certain system calls to processes. The privilege escalation feature of this allows you to grant a process the ability to call a certain system call with certain arguments as root, so you can (for example) allow a daemon to bind to a privileged port without requiring that it run as root.

Unlike SELinux, systrace is very easy to use and adapt - and it is far easier for a system to be secure if the system administrator understands it.

Re: That's it

[MikeBabcock]

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

I'm not a big RMS fan, but here's where I'd disagree with you -- how GM engineers their engines and why may be none of my business, but being allowed to tear it apart and replace parts myself for my own use should be a basic right. If I then want to tell others what I did to my GM engine to make it better, so they can do it to theirs, that should be fine too.

Re: That's it

[ulib]

how GM engineers their engines and why may be none of my business, but being allowed to tear it apart and replace parts myself for my own use should be a basic right. If I then want to tell others what I did to my GM engine to make it better, so they can do it to theirs, that should be fine too.

I think we're talking about two different things: the right to see the software source code is like the right to have the *blueprints* of that engine (i.e., the right to have their knowledge and know-how, the "recipe" of their product).
OTOH, I think your example is comparable to be allowed to *reverse engineer* the software (according to the philosophy, with which I tend to agree, that since I bought it I should be allowed to do with it whatever the heck I want).

Anyway, it ultimately depends on the license/EULA that the software comes under. I think (and here's where my favouring BSD over GPL/GNU/Linux comes out) the ideal scenario is the one in which any vendor can choose for his code the license/EULA that he wants, with no GPL-like restrictions, and the *market* is the one that decides if he's gonna succeed or fail.
After all, according to history, this is the scenario that yields the best results for the economy - and not only in software production.

(Needless to say, all other things being equal, I prefer software that comes with no anti-reverse-engineering EULAs - and *much* more, of course, I like Free/Open software that comes with the complete source.)

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

Re:BSD

[nxtw]

Less complicated init is also less powerful, and less painful filesystem managment is highly subjective. And, from an admin's point of view, BSDs being easier to learn is highly subjective also... I find Debian significantly easier to maintain, and I don't have to deal with the inconvenience of compiling software or the inconvenience of poorly compiled packages.

Re:BSD

[nxtw]

No, I think you're the troll...

Bad support for esoteric hardware;

Really? Have you heard of NetBSD? Do you not realize that FreeBSD has been not only on-par with Linux, but ahead of it in some areas for many years now?

Try PC hardware support. That's the concern of most people. There are a lot of drivers for Linux exclusively. While supporting a bunch of different platforms is a good thing, it's not something most people will choose one operating system over another for if they're using the most prevalent platform in the world.
And on a personal note, the lack of a journalled filesystem has kept me away from FreeBSD ever since XFS was available on Linux.

Less vendor support;

That's debatable

No, it's not debatable. There are much fewer hardware manufacturers that support BSD. There are much fewer companies that sell and support BSD. There are much fewer software companies that support BSD.

you can't possibly deny that all BSD distros have a much better security tract record than Linux

And you're saying that without context; there are many different distributions available and many more Linux users. Since more people are interested in Linux, more people will find vulnerabilities. There's also more code to find problems with
But at the end of the day, I wouldn't really be too concerned with the security of either operating system.

Re:BSD

[evilviper]

Try PC hardware support.

I already did. And I quote: "FreeBSD has been [...] ahead of it in some areas for many years now"

There are a lot of drivers for Linux exclusively.

Really, now? Start listing them, and I'll be happy to challenge them, one-by-one. Unless of course you're just trolling and have no evidence behind your claims...

There are much fewer companies that sell and support BSD.

The number of companies is completely and totally irrelivant. Microsoft is only one company...

Re:BSD

[nxtw]

"FreeBSD has been [...] ahead of it in some areas for many years now"

In what areas? I see a lot of Linux-only drivers out there, many binary-only. A lot of drivers are developed on Linux and later ported to other open source operating systems after they become stable on Linux...

Really, now? Start listing them, and I'll be happy to challenge them, one-by-one. Unless of course you're just trolling and have no evidence behind your claims...

So if I can't make a list of drivers, I'm automatically trolling? That sounds like a trollish challenge -- "if I can't prove my claim by factual evidence and the other person won't either, then I'm right and the other person is wrong". There are lots of Linux drivers out there that are supported on Linux and Linux only...

The number of companies is completely and totally irrelivant. Microsoft is only one company...

The number of companies is not irrelevant. (is that word that hard to spell?) "Less" is a quantitative measure and there are truly less companies that support BSD than Linux or Windows. Microsoft is only one company, but there are many, many companies that sell Windows or offer support for Windows. The same applies for Linux. No matter how you look at it, there's less vendor support for BSD than Linux or Windows.

Re: That's it

[ArbitraryConstant]

"So the difference between OpenBSD and FreeBSD (besides the maintainers of course) is more along the lines of politics (OpenBSD only allowing software in which meets their definition of free)."

No. There are many technical differences in addition to the political differences.

In general, FreeBSD picks features, compatability and performance first, while OpenBSD picks security and robustness first. Over time, that drift has left them in completely different positions.

The freedom of 3rd party software is somewhat a secondary issue. There are plenty of other problems with Apache, they have refused many security patches. It would basically be a fork either way.

"Do these two share between each other?"

Yes. All the BSDs use OpenSSH and PF from OpenBSD, OpenBSD got much of its SMP code from NetBSD, NetBSD got UFS2 from FreeBSD, and so on.

"Is there a common BSD kernel or anything like that?"

No. All the BSDs maintain their own kernel.

The closest two are probably OpenBSD and NetBSD. NetBSD performance optimizations make their way into OpenBSD pretty frequently, and security enhancements flow the other way just as often.

FreeBSD and NetBSD's common ancestor is way too far into the past for code to move that freely between their kernels. OpenBSD is a fork of NetBSD so it's in the same position.

DragonFlyBSD is a fork of FreeBSD, but it has drifted very quickly because the they've been making huge changes. AFAIK they can still port stuff from the FreeBSD kernel, but after a certain point "porting" becomes "using the other code as a guide for a rewrite" and I think they'll get there sooner rather than later.

Re:BSD

[evilviper]

In what areas?

USB, Firewire, NICs, 802.11, etc. Look it up. I'm not going to give you a lecture on all the developments of FreeBSD and Linux over the past several years.

I see a lot of Linux-only drivers out there, many binary-only.

Oh, you see "a lot" eh? "A lot" meaning 2? 3? Which ones? Provide some real evidence, or quit making noise.

A lot of drivers are developed on Linux and later ported to other open source operating systems after they become stable on Linux...

Now that is absolutely ridiculous. The only thing I can think of where that is actually the case is Nvidia's drivers. This is really a blatantly untrue, and shows that you're either trolling, or (more likely) don't have any knowledge of the subject you're talking about.

"if I can't prove my claim by factual evidence and the other person won't either, then I'm right and the other person is wrong"

It's not possible to prove a negative, only a positive. I can't prove there aren't Linux-only drivers out there, but you have the possibily to prove there are. I, however, can go through your list, and disprove them individually.

Saying something with no truth to it is trolling. Since you claim there are lots of Linux-only drivers out there, it's up to you to try to prove it.

There are lots of Linux drivers out there that are supported on Linux and Linux only...

Bullshit. Try and prove it.

The number of companies is not irrelevant. (is that word that hard to spell?)

Oh great, you're a Spelling Nazi too. You have so many wonderful redeeming qualities. Consider yourself on my foe list... I get tired of idiots quickly.

"Less" is a quantitative measure and there are truly less companies that support BSD than Linux or Windows.

"Less support" has absolutely no relationship to "Less companies".

Re: That's it

[laffer1]

FreeBSD does not focus on i386 performance. That was an old goal.. its not true anymore. FreeBSD 5.x and greater now focus on SMP. SMP designed kernels are slower for single cpu cases which is why Solaris always gets a bad rap as VERY slow.

NetBSD arguably is now the fastest BSD for single cpu i386 boxes. They have "benchmarks" to prove it.

Here's the new goal list for the *BSD community:

FreeBSD: threads/SMP (5.x)

NetBSD: Performance (2.x)

OpenBSD: Security and now replacing the entire
userland with BSD stuff.

DragonFly: FreeBSD did it wrong.. i can fix it... Note that matt D. wrote a lot of the freebsd 5.x code he makes fun of now! Seriously, his message approach reminds me of a Mach style.

OpenDarwin: Some consider this a BSD others do not.. based on Apple's Darwin/OSX releases, it has a Mach kernel with monolithic memory management from FreeBSD. Mach descends from 4.3/4.4 BSD releases so its a BSD derivative at least.

PC BSD: New kid on the block (sorry).. trying to be GNU GPL licensed, freebsd distro/fork? focusing on Desktop market.

Then the countless BSD distros emerging.. most are freebsd based and focus on live cds. Frenzy is one example. PicoBSD is another... etc.

Like it or not the BSD community is going through what the GNU/Linux community went through in the mid to late 90s. I don't have to say its not dead anymore.. its quite alive. While I don't agree with Theo De Raadt, many BSD people feel the Linux camp sold out with most of the big players taking the helm.. IBM, Novell, Sun, Redhat, etc. Its a shame linux has gone corporate.. at least we have a few grassroots distros like Gentoo left.

Just for clarification, I am a FreeBSD, OSX, Gentoo, Solaris, and Windows user. Its helpful to know what my bias is. :)

Re:Free, but not quite

[DashEvil]

Whether or not it is hypocritical...

"Note that only the CD layout is copyrighted, OpenBSD itself is free. Nothing precludes someone else from downloading OpenBSD and making their own CD."

You are greatly exagerating an otherwise small deal. It takes a small amount of time to make your own ISO. There is nothing complicated about it. If this bothers you that much, well, I don't know what to say. Sorry. I don't find anything hypocritical about it. To me it's like calling a fisher a hypocrite because he also lobbies against the death penalty.

Re:Why I use OpenBSD.

[Ezdaloth]

I'll give you another reason not to use it: it lacks features i'd expect from an OS i feel compatible with.

The openbsd guys are for as i understand it mostly concerned with security and making a nice and integrated system. I feel their kernel work is lacking a little (no SMP until recently, no real threading, no unified buffer cache, etc)

Taking the previous paragraph to the extreme, why don't they make openbsd an improved distro of some other BSD replacing that BSD's userland with their own, since that is what they excel at.

(I know i'm skipping over important things like wireless networking support they worked very hard for)

Ridiculous arguments

[vmalloc_]

It never ceases to amaze me how many people still cling on to the theory that BSD is "dieing" and that OpenBSD is "bad" and has "no driver support". I've been a sysadmin since before Linux was a corporate buzzword, and I've used just about every operating system out there, and I can say that OpenBSD is the best server operating system, hands down.

Sure, there are speciality corners where other operating systems might work better, but I can't see any good reason why security shouldn't be the priority of every single server on the internet. OpenBSD is the only operating system that is actually taking a serious look at security (whereas most Linux distributions are so bad with security right now, that they are less secure than Windows Server). While OpenBSD's "security audit" of the system code is the major point in its favor, some of the real advances in security are sometimes the subtle hidden ones that aren't even being talked about, such as the Pro Police stack, W^X memory protection (which Linus dismissed for not being a "silver bullet"), and randomized malloc. They might be stupid, little changes, but each one means that it is that much harder for somebody to remotely break into the server. The point is, OpenBSD is thinking about all this, while the other Linux distributions are thinking of ways to stuff more dangerous programs into their default startup.

OpenBSD driver support is far cleaner, plug-n-play's far better, and in some areas is actually more complete than Linux (OpenBSD supports a ton of networking devices right now, including wireless).

Once more difficult than Linux, it is now becoming signifigantly easier, because the core system is clean and concise, and there is extensive documentation both in the form of the manpages and in published books. Where most Linux distributions have opted for bloated, unmanageable monsters that are immediately susceptible to remote buffer overflows the second they are installed, OpenBSD has opted for cleanliness and system simplicity, which I think is a far more enlightened and managable approach in the long run.

When software isn't good enough for the team, they make their own. It is this no-compromise position that led to OpenSSH, which every unix system in the country is now using. It also led to PF, the packet filter that absolutely blows away ipfilter (though I'll admit could be a bit easier to use, I'm a moron when it comes to filter rules). The fact that they can make these incredibly powerful, landscape-changing programs on a whim should be a hint to just how talented the developer team really is.

There are some areas where the system needs work, but those areas are quickly focused on and resolved (after last week's hackathon we'll probably have better RAID device support, despite driver snubs by many of the companies).

I probably sound like a hypocritical ideologue in this post, but I'm more interested showing off the good side of OpenBSD so I can drive this point: People that blindly use Linux for everything are missing out on the real advances in practical server technology, which aren't happening in penguin country.

Re: That's it

[nacturation]

Btw, one little thing that the forum above has, and IMHO Slashdot is missing *badly*, is the ability to edit comments to add things you forgot (damn it).

Likely so that people couldn't change what they wrote after replies were written, but it would be good to have it eBay-style, so you'd see something like:

--- On May 31st, 2005 11:01 AM PST, nacturation added:

Oh, forgot to mention that... blah blah.

So you could only add, not remove.

Re:Mandatory Access Controls?

[Triumph The Insult C]

and, when you install apache in redhat or debian or gentoo or whatever, it's fully customized and configured to your liking? during the os install?

please ... find a valid argument next time. the "but the default install is useless" is old