Should There Be A Unified BSD Operating System?

[BitRot]

And an identical kernel and packaging system.


PC-BSD uses pkg and PBIs.
DragonflyBSD uses pkgsrc, too.

So now what?

Well, that's easy to imagine.
Your point was that BSDs share kernels. That's not true, as the BSD being talked about in the article all use different kernels.
Also, you said they share packaging systems. That's true to a much lesser extent than you implied, it's just that DragonflyBSD uses netpkgsrc (and maybe OpenBSD too, I don't know).
PC-BSD is just a distribution of FreeBSD, not another BSD.

[BitRot]

I don't mean by distro, I mean by identifiability and userbase. The BSD devs should join the linux fold, wherever they end up. So too should disparate Solaris devs, and devs of other disparate OSs.

We need their help! They're clearly gifted and capable developers.

Last I heard, BSD was losing users, not gaining them. They should consider writing software where there is a growing userbase.(even if realistically, that user base growth is slow. It's still growth)

As an aside: I do think there's an argument to be made that there's too many linux distributions as well, but on a whole having multiple distros and a choice of them is a good thing.

You're implying that the goal of the BSD devs is to be popular. It's not.
The devs have ideals. That's why they are doing what they are doing.
Other than that, if everyone would only develop for Linux, it would be pretty boring.

[halfmanhalfamazing]

Why should elitists couple with ordinary persons?

LOL Oh come on now. Macs are computers for elitists, let's be honest here.

Anyways, the BSD crowd has a lot in common with the Linux crowd. It's not like moving to Windows.

[ryao]

I don't mean by distro, I mean by identifiability and userbase. The BSD devs should join the linux fold, wherever they end up. So too should disparate Solaris devs, and devs of other disparate OSs.

We need their help! They're clearly gifted and capable developers.

Last I heard, BSD was losing users, not gaining them. They should consider writing software where there is a growing userbase.(even if realistically, that user base growth is slow. It's still growth)

As an aside: I do think there's an argument to be made that there's too many linux distributions as well, but on a whole having multiple distros and a choice of them is a good thing.

The main foci of major BSD-derived operating systems have always been technical excellence and the ability to share their innovations others. A major strength of BSD-derived operating systems is that their developers consider the long term usefulness of changes to the platform as a whole. That is in contrast to various Linux distributions, which often accept short-sighted decisions of other groups without question, even when they conflict. If BSD-derived operating systems gain market share, it should be because they are doing things well, not because they abandoned principles that much of the Linux community has never had.

With that said, the only significant things that the BSD community appears to lack at the moment are KVM, Network Manager, some wireless drivers and some graphics drivers, all of which could easily be ported if key Linux developers were to join the BSD community. If Linux developers joined the BSD community to do that, end users would be better off.

[halfmanhalfamazing]

You're implying that the goal of the BSD devs is to be popular. It's not.

I didn't have popularity in mind. "Hey, that BSD is soooooo unpopular...." That's a silly notion.

What I'm saying, is that there's strength in numbers.

http://www.freebsd.org/doc/en/articl...and-linux.html

The opening line:

For the average user, the difference is surprisingly small

Then we should work together, instead of working separately.

[disi]

The main foci of major BSD-derived operating systems have always been technical excellence and the ability to share their innovations others. A major strength of BSD-derived operating systems is that their developers consider the long term usefulness of changes to the platform as a whole. That is in contrast to various Linux distributions, which often accept short-sighted decisions of other groups without question, even when they conflict. If BSD-derived operating systems gain market share, it should be because they are doing things well, not because they abandoned principles that much of the Linux community has never had.

With that said, the only significant things that the BSD community appears to lack at the moment are KVM, Network Manager, some wireless drivers and some graphics drivers, all of which could easily be ported if key Linux developers were to join the BSD community. If Linux developers joined the BSD community to do that, end users would be better off.

/rant
Drivers! What they need are drivers or at least don't break working ones with an update.
Old SATA I controllers are not supported, newer ones are not supported and the one I found in the middle doesn't work properly since 8.3 any more.
rant/

Since I got a bloody nose with FreeBSD, it worked 1 year, then upgrade to 8.3 and tried ~6 months to get it working again (you cannot downgrade back to 8.2 since zfs is not compatible -.-). I am back on Gentoo Linux with that box and no data loss in the zfs pools.

[halfmanhalfamazing]

The main foci of major BSD-derived operating systems have always been technical excellence and the ability to share their innovations others. A major strength of BSD-derived operating systems is that their developers consider the long term usefulness of changes to the platform as a whole. That is in contrast to various Linux distributions, which often accept short-sighted decisions of other groups without question, even when they conflict. If BSD-derived operating systems gain market share, it should be because they are doing things well, not because they abandoned principles that the Linux community has never had.

With that said, the only significant things that the BSD community appears to lack at the moment are KVM, Network Manager, some wireless drivers and some graphics drivers, all of which could easily be ported if key Linux developers were to join the BSD community. If Linux developers joined the BSD community to do that, end users would be better off.

As a matter of principle, I always thought BSD had the upper hand. But that's not how it's all shaken out. Intel, for example, is doing a lot of work on Mesa and x.org to improve OpenGL compliance and so forth. Sure, BSD can benefit from this, but that's not why they're doing it. The large OSS user base is a linux user base.

What I'm saying is probably somewhere between the law of diminishing returns, and the law of marginal utility.(considering where the OSS user base is at) As I said just a moment ago, there's strength in numbers.

But even though I think that BSD has the upper hand based on some of the principles you laid out, the notion(and even the original article states this in the opening paragraph) that it's "bsd vs linux" is weird to me. It's OSS vs Microsoft and/or other corporations who seek to either make us dependent upon their sub-par(or even faulty) wares, over charge, or other such nonsense.

[ryao]

/rant
Drivers! What they need are drivers or at least don't break working ones with an update.
Old SATA I controllers are not supported, newer ones are not supported and the one I found in the middle doesn't work properly since 8.3 any more.
rant/

Since I got a bloody nose with FreeBSD, it worked 1 year, then upgrade to 8.3 and tried ~6 months to get it working again (you cannot downgrade back to 8.2 since zfs is not compatible -.-). I am back on Gentoo Linux with that box and no data loss in the zfs pools.

Did you file a problem report with FreeBSD? As one of the Gentoo BSD developers, I would appreciate it if a problem report were filed for the SATA issue that you encountered. As the Gentoo Linux ZFS maintainer, I am very happy to hear that you enjoy ZFS on Gentoo Linux.

As a matter of principle, I always thought BSD had the upper hand. But that's not how it's all shaken out. Intel, for example, is doing a lot of work on Mesa and x.org to improve OpenGL compliance and so forth. Sure, BSD can benefit from this, but that's not why they're doing it. The large OSS user base is a linux user base.

What I'm saying is probably somewhere between the law of diminishing returns, and the law of marginal utility.(considering where the OSS user base is at) As I said just a moment ago, there's strength in numbers.

But even though I think that BSD has the upper hand based on some of the principles you laid out, the notion(and even the original article states this in the opening paragraph) that it's "bsd vs linux" is weird to me. It's OSS vs Microsoft and/or other corporations who seek to either make us dependent upon their sub-par(or even faulty) wares, over charge, or other such nonsense.

Intel's work has been ported to FreeBSD. I tried the port on FreeBSD 9.1-RC3 a few days ago and it works well. It currently requires that you add WITH_NEW_XORG=true and WITH_KMS=true to /etc/make.conf and build xorg and libdrm from their ports. However, I expect that they will have appropriate binary packages in time for the FreeBSD 9.1 release. I plan to merge this into Gentoo FreeBSD after we get our X server working on Gentoo FreeBSD.

[BitRot]

I didn't have popularity in mind. "Hey, that BSD is soooooo unpopular...." That's a silly notion.
Sorry, I misread then

What I'm saying, is that there's strength in numbers.

http://www.freebsd.org/doc/en/articl...and-linux.html

The opening line:



Then we should work together, instead of working separately.

From the end user perspective this might be the most logical thing to do.
However, most devs realize their own visions on the various BSDs. That's why there is not one single BSD OS.
For example, Matt Dillon wanted a diferent approach to SMP on FreeBSD, but he wasn't allowed to(or whatever), so he forked FreeBSD and created DragonflyBSD.
You won't get him or anyone working on it to develop for Linux.
Same with OpenBSD, you sure as hell won't get Theo de Raadt to give up his vision on having the best security, which he can only realize with an own OS.

[Cthulhux]

LOL Oh come on now. Macs are computers for elitists, let's be honest here.

Macs are for girls and gays.

Back to the "Linux and BSD should work together" part: They do. They are inspired by each other.