Slashdot Mirror
FreeBSD 6.0 to Target Wireless Devices
BSDForums writes "FreeBSD is hoping to move beyond the server and desktop market by providing expanded wireless support. FreeBSD developer Scott Long said that 'one of the primary reasons for improving wireless support is to give companies the tools to put FreeBSD into their wireless devices. The guy at FreeBSD who is adding wireless support is under contract from wireless companies to do the work.'"
As if they are trying to tap into the market of Linux pda's and such...
Microsoft is like...no, it's much worse.
This is an interesting development. Companies have been using Linux in their wireless boxes due to the lack of any viable alternative. Due to the GPL, these companies were forced to publish their changes to the kernel, which has allowed the number of cool hacks we've been seing. Clearly, those companies would rather keep their changes proprietary, so BSD based systems are much more attractive to them. While it's nice to see improved hardware support to another free operating system, this might bring adverse consequences in the long run. We'll see...
The filesystem is the package manager
even in death, BSD still seems to come up with more great ideas and interesting features. don't mind me, im sticking to my cadaver operating system!
as someone who works for a company that uses many variants of linux in their products, i can tell you that most companies dont have a clue about the GPL and the parts about giving away _their_ code comes as a shock to them.
i see more and more companies turn to BSD licensed stuff, such as the *BSD OSes
(no linux vs BSD flame, please)
There's still issues with FreeBSD that make it unsuitable for desktop use (I realize that's probably going to be modded a troll, so I'd like to point out the following example: http://support.daemonnews.org/viewtopic.php?p=995
). These need to be addressed before it can even really come out of the server corner, if that's their intention.
(Note: The page might be down at the moment, but try again and it should eventually come up)
The beautiful part is that OpenBSD is NOT in a competition with anybody despite what your post asserts. The OpenBSD developers have goals that are in line with their ideals and they haven't strayed from them and the developers certainly aren't out to "corner wireless development."
You make it sound like 2 companies competing against each other when it's just FreeBSD's developers losing sight of their ideals and doing thing the easy, non-OpenSource way.
This guy is way out there
Seems like a bad thing to me. Imagine FreeBSD being used on new routers. Who benefits from this? The manufacturer! No one else. BSD license is bad for this. With GPL, everyone benefits.
You mean like secure, reliable wireless devices?
That would be a damn shame.
Lost what? Have you checked usage statistics for FreeBSD vs. OpenBSD lately?
And yes I use both, but zealotry with no backing other than "because its free" is a lost cause.
that's a very good point. i think that's a crucial reason why freebsd has already lost this 'market'
imo, freebsd's developers (or maybe it's core@) has, and i hate to use the cliche, but, 'sold out'
otoh, openbsd developers develop because they want to. they want quality software that's *free*. freebsd developers seem to have gone the "we want something that works" route, and that's too bad
vodka, straight up, thank you!
Well, it works in WAP!
Yes, God forbid we have working software. Who'd want that? Probably some Microsoft fanboy. They always want stuff that "works" and doesn't break when you upgrade your system, etc. What is wrong with these people? I'm glad we have people like you who can keep the remaining sane folks from straying from the path of broken, hard to use software.
In other words, the use of the Linux kernel allows free software users can pressure these would-be proprietors into helping them maintain our software freedom for derivative works so long as one leverages their laziness. That is, as long as one doesn't distribute proprietary kernel modifications.
Given FreeBSD's willingness to include proprietary software (see discussions between FreeBSD and OpenBSD developers and advocates surrounding technical specifications for cards -- FreeBSD is happy to include whatever the proprietor delivers, OpenBSD wants specs so they can write and maintain their own drivers), this "improved" hardware support may end up being nothing more than a means to deliver more proprietary software to users.
Digital Citizen
If you look in FreeBSDs mailing list archive, their is a message explcitily saying we need to be more like NetBSD.
There is also con-currently somebody ranting about the "off-site" development of FreeBSD features:
Myself, I'd rather FreeBSD sort out it's current problems out before thinking about introducing New problems.
*
What are you talking about? Oh, a flamebait AC, that's right. Probably one who's never installed FreeBSD or OpenBSD either.
This guy is way out there
I see a number of companies switching to BSD so they don't have to give away their code like they do with the GPL
that wireless is dying now?
Jonathanjk.com
Juniper uses FreeBSD (they call it 'JunOS'). Their routers have become quite popular for very high traffic installations, due in no small part to the efficient networking code of the FreeBSD kernel. Also, don't forget that the f-root name server (actualy a distributed network of servers) is exclusively FreeBSD.
And I'm sure that someone will think that this just isn't pure. Like the Olympics, which was once ruled that only self-supporting people who could do it for "The Love of the Sport" were worthy.
Personally this is a great move for OSS, and a vote of confidence in the value of freely available Unixs by the companies. I hope it becomes a model for each part of the industry to do more to support their devices (graphic cards, USB plug in devices, you name it) to the free and OSS communities!
I will be disappointed if there is a single negative comment about this aspect of how the work is being funded and getting done!
"It's the height of ridiculousness to say for those 9 lines you get hundreds of millions."
Well most the computers that use wireless are laptops (not that desktops don't, just laptop take most the market) and this is a large step im make BSD a good solution for laptops, but The OS is still lacking (last time i check) in many part that are more vital to laptops. For example, when I tried 5.3 (or was it .4) there was no support for my battery, my AMD CPU did not go down from full speed and there was no ACPI, standby or hiberantion. Now I heard they put in something that allow Pentium M to lower CPU speeds (not sure if the voltage changes, if it does not that would take away alot from the point of change the cpu speed). I an very greatful for all the work of the freebsd team, but it just seems to me that they are saying "we finished the front door, now all we have to do is build the house that goes around it."
Like many articles about BSD, this one will surely have a ton of comments along the lines of "Oh no, companies don't want to give away their code, so they're all going to use BSD licensed software and the world is going to end!". This arguement is, in almost all cases, bullshit. Why? Because usually it's not the operating system that matters so much as the software on top of it, and Company X has just as much control of their own program with Linux as with BSD (or Windows, etc.). Most products do not require significant changes to the OS, if any at all, and even in those products that do, far more of the products value comes from the company's own software.
So stop it people, the sky isn't falling.
This is not intended to be a flame as I really like FreeBSD as well. FreeBSD could learn a lot from the OpenBSD project in this area. I have been absolutely amazed at OpenBSD's out of the box wireless detection configuration. I installed OpenBSD on my laptop over my WPC11 wireless NIC without effort. I also had the same results with the WMP54G.
and deserves to die.
Linux will always beat the crap out of BSD
Long live FreeBSD!
Sorry, my deadpan humor is getting out of control
Get your Unix fortune now!
Is this related to the fact that NetBSD is in so many of my PSP games? I'm not a *BSD guy, so I'm not really sure if they're related. But the pages of the manuals that deal with wireless functions always have the word "NetBSD" mixed in with the Japanese text.
-- I'm old enough to have lived through six different meanings of the word "hacker."
Umm... I think you got the point reversed. The poster was talking about the fact that you can take BSD licensed code, make changes to it to support your product, and not be forced to give your work to anyone else (for example your competition).
While we all understand the idea behind the GPL, many businesses will simply not even consider using OS's based on it because of the forced nature of it. FreeBSD stands to get a lot of users because of this in the embedded space.
Now, the gamble with the BSD license is that people might use the code without ever contributing back. But the bet is that the big companies will give some sort of kickback to the projects, even if it is not the complete solution.
An example of this is practice would be Apple and KHTML. While Apple has not completely given everything it could have given, the KHTML project has benefited from Apple using a derivative of KHTML. We can argue about whether it is enough, but it is benefit that Apple would not have contributed if KHTML were GPL rather than BSD. Management would not have touched it with a 10 foot pole.
Actually, KHTML is LGPL.
I agree with your comment, but you mention that the bet is that companies will foster further development of the projects even if they're not forced to provide code back [I'd quote but I'm posting this from links]. I have my doubts.
The filesystem is the package manager
I think you're responding to this:
otoh, openbsd developers develop because they want to. they want quality software that's *free*. freebsd developers seem to have gone the "we want something that works" route, and that's too bad
(If you're not responding to that, please ignore this post.)
There's been a lot of heat of this, and not much light. One can argue about whether goal X is better served by building something, or using a tool with a different license, certainly. And I'd agree that, all other things being equal, free is the way to go. I might even say that long-term, free will beat expedient (I'd like to believe that's always the case, but it isn't.).
But all other things are not equal, and I tend to trust the fBSD folks. They've made more than one good decision in the past that has pissed people off, and the OS still rocks. (I've had my laptop running it for a while, after having moved for the most part to Debian for a few years for business reasons. I'm planning on moving my main desktop back to fBSD when I have some time - I hadn't realized how much I missed the feel of it until I started playing with it again.) I don't think it is a bad thing to be pulled by commercial interests, so long as you keep your long term goals independent of your short-term goals; hell, how is this different than any (non-independently wealthy) poster on this thread?
I forget what 8 was for.
Well, it might be for some, but I think that most companies would be fine with publishing any minor kernel modifications made to Linux. The real reason why companies are a little afraid of the GPL is that there is always the potential for lawsuits. Granted, it's a remote possibility, but a possibility all the same; most companies would rather not be bothered with the GPL if given a choice.
And this is where FreeBSD can give people a choice. And assuming it's just as good as Linux, it's a better choice due to legal issues.
One last thing about the GPL is that most companies don't see distributing changes as a free endeavor. Someone has to be in charge of overseeing the process - and labour isn't free. In addition, distributing the changes requires other resources (like bandwidth) which, while not very expensive, just add to the complexity of using Linux. While this might be a minor issue, it's still an issue that companies would rather avoid.
And what happens when the slashdot community burns a company for forgetting to post something, or posting modifications that are difficult to utilize. Remember Apple and Konqueror?
Oh well, but to respond to your last point, I don't think there will be many adverse consequences. The work currently being paid for by these companies is under the BSD license and I don't see them being that protective of the kernel. The GUI however is another story.
Willy
When your big deal for a release is pretty much just the incorporation of stuff from another project it just seems, I dunno, less newsworthy to me.
But I must agree about Scott Long and Poul-Henning Kamp, those two have earned a set of solid kicks in the balls for the nonsense they've been pulling. Poul-Henning's nonsense about breaking the law with reverse engineering at an OpenBSD talk in Ottawa, if I'd been there you can bet he would have been needing a surgeon to remove my foot.
I'm sick of following my dreams - I'm just going to ask them where they're going and hook up with them later.
Remember all the big talk about how great 5.0 was going to be? And yet the 5.x series has been a monumental embarrassment to the FreeBSD community. FreeBSD has been incredibly long on talk, and shamefully remiss in its ability to deliver.
This has all the markings of another round of grasping at straws, a desperate hope that someone, anyone, will pull FreeBSD's fat out of the fire.
Apple's support alone will keep FreeBSD going. The video chip guys (ATI and nVidia) release object code drivers under closed licenses and anyone who wants can link it into Linux's GPL world. They aren't breaking the GPL, since they don't do the linking. Are the Linux users breaking the GPL? - I don't think so, but who's going to bust them even if they were.
This case is different since these wireless companies want bundle an entire OS. Using non-GPL'ed code seems to be to their advantage. And I mentioned before Linux and the GPL hasn't forced video chip drivers out to the light of day... I don't think the GPL has levered that many hardware vendors.
"squirl away"
Bwaaahhh haaahhh haaa!!!
squirl away!!!
Ha ha ha, that's great.
Is that some new bastardisation of the English language you yanks have come up with? Or is it just a side effect of this poor fool trying to phonetically spell your hilarious pronunciation of "squirrel"?
Keep up the good work!
Doesn't matter if he's installed any OS at all. You're skirting the point, which is that broken but free code is not useful. Working, non-free code is.
I have actually installed FreeBSD, but currently use Gentoo (so sue me). But broken software is broken software. Not being able to use the hardware on your machine is annoying, especially if it's because of somebody's political axe to grind.
Wrong. The hardware works with 2 methods. Either OpenBSD's method or FreeBSD's method. 2 minutes of Googling will let you know what card is supported for your OS before you go shopping next.
This guy is way out there
...is Sam Leffler, who was involved with BSD back when it was Free/Open/Net/DragonFlyBSD, but the one original BSD back at UC Berkeley. That's dedication. McKusick also makes some major contributions as well.
I have seen many clients where I used to work switch from FBSD to Linux after 5.x came out. Even with Linux threads in mysql its still buggy.
I used to be a BSD zealot for years and now slowly switching back to Linux.
It would be nice if the kernel designers would fix the many problems with the locks and threading and use a simplier design like DragonFLY's BSD model.
5.x is a mess and it looks like 6.x will be based off it.
http://saveie6.com/
There was talk early in the year about Linux forking if you read slashdot.
The issue was non GPL binary drivers in the kernel and module support.
First off its Powell and the FCC and not greedy capitalists making the decisions to stay closed with wifi. Its required infact to be a licensee of the FCC to have permission to sell your product.
Now the greed has spread to all markets in computers as the FCC could change its rules for any product that produces EMI. Also greed and the length it takes to file a patent makes closed source drivers attractive to protect their IP. Their shareholders demand it.
Its a mess but at least wireless companies are legally obligated to stay closed source for that reason. Someone hacking on a wifi router could wreck havoc for things like airplanes and other equipment utilizing radio waves.
http://saveie6.com/
It's a lot easier to develop for FreeBSD since it has one consistent version controlled set of user space and kernel code with timed regular releases.
It is stable and companies don't have to worry as much about keeping their own specially forked version to support their device,
How many people have ever looked at sourcecode released by vendor? Even better - how many has SEEN any contributions released by vendor that have any significant impact?
Has Realtek ever released source for their wireless driver (for 8181, for example)? Nope. Has Broadcom ever released source for their wireless driver? Nope again.
You want to use Atheros in your AP? Fine. All companies, so far, either cash out $20,000 and get the sourcecode from Atheros, or they subcontract the job (to someone from http://www.atheros.com/partners/AADC.html list). No GPL here.
Not to mention that many Atheros based APs are running VxWorks, not Linux.
Who cares if vendor released a patch that changes some allocation routines (specific for hardware platform their AP runs on) or whatever.
Vendors only care to rollout product ASAP. Nothing else. Most of the time, they roll it out even before it's actually production ready ("we'll fix the bugs as we go" approach).
If FreeBSD allows them to rollout products faster - they'll go with FreeBSD. But something tells me (and it's not part of *BSD vs Linux argument) that they'll stick with Linux, since there are countless wireless-ready development kits/boards available.
Don't think they'll take another route, just because of BSD license.
- There is no uClibc equivalent for BSD. Thus, you have to use the standard BSD libc, which is much, much more bloated.
- There is no BSD-licensed equivalent of Busybox, thus you currently have to include the standard, much larger tools to accomplish most of the same tasks.
- The current net80211 stack itself is fairly bloated. I'd like to see some of the functionality of the stack stripped out so that it is smaller. It is currently a ~230kb+ driver on X86 architectures, which is rather large.
These are very real issues since manufacturers would much rather buy 2MB flash pieces than 4 or 8MB flash pieces for their routers/APs. It can literally save them 2-8 dollars *per unit*. For this reason, I don't see many vendors switching to *BSD any time soon until BSD can come up with busybox- and uClibc-type projects that don't have a GPL license.Surely the biggest stumbling block for FreeBSD on small, cheap devices isn't support for wireless, so much as the neglected port to ARM? Oh how I long to run FreeBSD on an NSLU2! And I don't think they are doing much better with the other embedded CPU architectures, either ...
These are very real solutions to your imaginary issue, since it's damn hard to buy flash that small, most companies don't make smaller than 16 MBs unless you are special ordering massive bulk. Often, this kind of special ordering has additional costs involved. I don't see any problem with crunching the system down into flash and using virtual memory filesystems.
OpenBSD wireless is nothing short of phenominal.
Of course that probably means the OBSD drivers will get ported to the other two. That's another great thing about the BSD world -- they all end up benefitting each other.
Excuse me, broken? If OpenBSD is "broken" why is FreeBSD importing all of its ideas? Is PF broken? Is OpenSSH? Guess where those came from...
The problem with "software that works" is if you write software to "just work" then you're missing out. You want it to do more than work -- you want it to work well and actually be robust. OpenBSD has that philsophy, FreeBSD does not. (I'll give you a hint: it has to do with more than drivers.) That's the context of the quote that you and other trolls are parroting. Software developers should know exactly what he's talking about. Good ones, anyway.
One thing I have noticed in the recent year is the upsurge in Ralink (RA) WiFi devices. Taiwanese motherboards manufacturers are bundling 802.11 cards by the thousands with high end motherboards with RA cards. ASUS, MSI and Gigabyte all bundle RA cards with their "deluxe" motherboards.
Since RA is a very Taiwanese component, and motherboards are - of course - *very* Taiwanese components, it would be excellent if FreeBSD took advantage of the opportunity to embrace RA in the same manner that Linux has. For bonus points, PLEASE backport to 5.x and 4.x since many of us (particularly DJB and myself) refuse to move off 4.x unless it is absolutely needed.
HJ
You might want to look into PicoBSD. It has an amazingly small footprint, yet it is fully BSD.
!ERR: Signature not found.
Yes, but who says the whole system needs to be BSD licensed? Or GPL licensed? If you use BusyBox what does it matter wether thats GPL or BSD licensed when you have not made any modifications? Same for uClibc.
uClibc may work with FreeBSD. The Debian GNU/KFreeBSD guys have the FreeBSD kernel working with a non-BSD userland space. This would mean one could use (parts) of the BSD licensed kernel, write modifications, and license them under whatever although i'm not sure uClibc also works.
WE DON'T NEED NO BLOG CONTROL.
Does this mean that Linux will come up with an OpenBSD compatibility module to take advantage of the OpenBSD drivers?
On the other hand GPL could actually be beneficial for business. Watch how Carmack chose GPL instead of BSD for the q3 source license. Relseasing their changes as GPL also ensures that their competitors wont be able to simply appropriate the code.
US-UK-Israel: The real Axis of Evil
Does this mean that Linux will come up with an OpenBSD compatibility module to take advantage of the OpenBSD drivers?
Haven't quite got your head round this "Open Source" stuff yet, have you?
You may have your doubts but history simply proves that they do (not all of them, but quite a few still)
FreeBSD has a thing called netgraph, from its manpage:
HISTORY
The netgraph system was designed and first implemented at Whistle Commu-
nications, Inc. in a version of FreeBSD 2.2 customized for the Whistle
InterJet.
Then, from man jail:
AUTHORS
The jail feature was written by Poul-Henning Kamp for R&D Associates
http://www.rndassociates.com/ who contributed it to FreeBSD.
Of course PHK is a core member of the fbsd team, but that doesn't change that it was written and payed for by a commercial user of the system and then contributed to it.
There is a simple very good reason for companies to contribute their changes, given that they get accepted:
It saves them the cost on maintaining such a component and keeping up with the development of the system.
if you write software to "just work" then you're missing out.
I am sure MS would agree with that, specifically, their bank account is really missing out..
You may not like it, but good enough will do the job in 99% of the cases. OpenBSD and its approach are more suitable for the remaining one percent.
You want it to do more than work -- you want it to work well and actually be robust.
Yeah, the workstation I am at now was just upgraded from FreeBSD 5.3 patchlevel 20 to FreeBSD 6.0 beta 2.
An extremely painless upgrade )from source no less). It works very well, andwas only upgraded because of wanting wireless support.
OpenBSD does not even manage to finish a compile of the world on that machine (and after quite a few discussions in the last 4 years about build problems I have given up even arguing about it on the OpenBSSD lists. If they have nothing else to say then "your hardware must be broken" while it works perfectly fine with everything else, then they are not anzwhere close to working, let alone working well for as far as I am concerned)
OpenBSD has that philsophy, FreeBSD does not. (I'll give you a hint: it has to do with more than drivers.) That's the context of the quote that you and other trolls are parroting. Software developers should know exactly what he's talking about. Good ones, anyway.
Software developers who actually have to get some result instead of trying to achieve software utopia know that good enough is good enough.
If you want something that is designed well, works well, and lacks some of the uncleanness of FreeBSD then look at NetBSD.
There is one and only one thing in which you have a bit of a point, OpenBSD and even more so NetBSD have a cleaner design for the things they added to the BSD codebase. On the other hand, neither runs extremely well on most modern hardware and can take good advantage of that hardware. FreeBSD does in many cases.
I run quite a bunch of servers, some of which happen to have a SATA based raid 1 configuration.
Now, I have had some disks break, and in cases where the primary disk in a raid 1 fails, Gentoo refuses to boot untill you manually change things. FreeBSD on the other hand just boots. Stick in a new disk, tell it that it has a new disk in its raid 1, and it syncs it nicely in the background. With SATA this is fully hotpluggable.
On my workstation on the other hand, I do run both Linux (Gentoo) and FreeBSD. I run both because neither supports all the hardware I have, both support a different subset.
Yeah, come back when OpenBSD can consistently build a world (even when it is for a different version then what you are currently running)
flashboot fits an OpenBSD system that includes wireless, IPsec, firewalling, bgpd, ospfd, ntpd, ssh and a bunch of standard Unix utilities into a 5Mb ramdisk kernel without trying. If you are embedding then you could reduce this easily (you wouldn't need most of the extra commandline utils or all of the half-dozen wireless drivers for a start). uclibc is only required for Linux because glibc is so horribly bloated.
I didn't get anything backwards, but you misstated the reason the new BSD license (and similar non-copylefted free software licenses like the MIT X11 license) are considered to be a "gamble".
Private derivatives are derivative works which are not distributed; changes are made and not sent back upstream. Private derivatives allow people to help themselves by improving their programs or getting someone to do it for them. Not everyone is in a situation where they can effectively send back changes (people who live in repressive countries, and people without email and web access, for instance). Surely we should not work to stifle anyone's freedom to improve their copies of programs.
The gamble with the new BSD license is that a popular incompatible proprietary derivative will be made and the programmers of the new BSD-licensed program will have to compete with what is, in part, their own work. This risk is thought to be tradeoff--trying to leverage a network effect for maximum popularity by trading away most of the power copyright law grants. The Vorbis and Theora codecs are purposefully distributed in such a way that will maximize popularity so that more people can have free codecs and reduce or eliminate their reliance on proprietary and/or patent-encumbered codecs.
Some, particularly those who champion free software freedoms for derivative works, stress that this tradeoff can effectively treat a business like a charity, or deny any user of that proprietary derivative their software freedom. They also cite that it is not the free software community's job to look out for the interests of proprietors. It is the free software community's job to look out for the interests of software freedom. Thus, they use a license that stipulates that distributed derivative works must distribute complete corresponding source code as well.
Finally, there is no "force" in the GPL and merely executing software doesn't fall under the terms of the GPL. The GPL is a distribution license; its terms, particularly its implementation of copyleft, don't kick in until distribution happens. One chooses what software to distribute, and so long as copyright is the only barrier to contend with (as opposed to patent law), one has the freedom to reimplement whatever software one wishes to reimplement and the power to license that software any way one wishes. Proprietors know this very well, which is one reason why we have so many programs that broadly do the same thing.
Digital Citizen
Maybe one of those days you should take a peek at something called picobsd, it is included with the FreeBSD source tree.
Of course PHK is a core member of the fbsd team
Poul-Henning is not a member of the core team, and hasn't been for years.
The "guy at FreeBSD who is adding wireless support", by the way, is none other than CSRG veteran Sam Leffler.
DES
Hmm, no more indeed. I thought he was at the time of adding jail support, but I could be mistaken of course :)
Woudl the guy who's working on this work on it if he wasn't getting paid?
heh.
It's good for Carmak because he wrote the code and the GPL allows him to restrict what his competitors do with it. It's bad for everyone else who might have used the Quake 3 souce but now has to re-invent the wheel.
*BSD experienced moderate success about 15 years ago in academic circles. Since then it has been in steady decline. We all know *BSD keeps losing market share but why? Is it the problematic personalities of many of the key players? Or is it larger than their troubled personas?
The record is clear on one thing: no operating system has ever come back from the grave. Efforts to resuscitate *BSD are one step away from spiritualists wishing to communicate with the dead. As the situation grows more desperate for the adherents of this doomed OS, the sorrow takes hold. An unremitting gloom hangs like a death shroud over a once hopeful *BSD community. The hope is gone; a mournful nostalgia has settled in. Now is the end time for *BSD.
You could have just said from the start you were just spreading FUD and I wouldn't have wasted my time. You managed to provide a whole whapping 0 examples of openbsd's security features limiting functionality, and instead pointed out two things that have nothing to do with it, and then told me to check the archives. Guess what, I have been subscribed to the list for years, tech and misc. This is why you can't lie to me about this and expect me to believe it.
Autorun is not supported because this is not windows. Wtf do you want it to run? Setup.exe? Automounting is supported just fine though.
Your usb device is simply a driver issue. Nothing to do with security at all. Have you actually tried it in openbsd, or are you simply assuming that it works the way you think it does? If it doesn't work properly, it simply needs a driver, this has nothing to do with security at all.
Selinux is not useful as simply adding access controls does not provide security. All the local root exploits in the linux kernel are still there, and access controls will do nothing to prevent someone from exploiting them.
As for debian, if you think reacting to exploits after the fact in a timely matter is the same as providing a secure OS, then you are simply crazy.
And the apache question has nothing to do with openbsd, its a general unix question, which applies the same to every unix system. Run multiple apache instances, or better yet don't run apache at all, use a better webserver. Are you seriously saying you expect openbsd to have some magic feature that alters the way one particular application works to suit you?
It ate my comment, so here's a short version. You said apache, therefore its an apache question, pretty simple. And I didn't interpret it as an attack on anything, I interpreted it as a remarkably stupid question.
OpenBSD doesn't claim to have forked apache, they simply maintain an older, free version. If they had forked it, then it would be significantly different, and they would not be allowed to call it apache. Its not being developed, just maintained.
And finally, I did answer the question you apparently meant to ask, go back and look. Based on your new statements where apache doesn't matter anymore, I don't even know if PHP is still required, or if you just need some language, so its hard to give you a real answer. None the less, give each user a web app running as a unique user id. PHP with fastcgi works fine, although I think PHP sucks donkey cocks. I would suggest lighttpd if it has all the functionality you need, its not as featureful as apache, but its got most of the typical stuff you would need.