Meet UbuntuBSD, UNIX For Human Beings

prisoninmate writes: What's ubuntuBSD? Well, it's not that hard to figure out yourself, but just in case you're not sure, we can tell you that ubuntuBSD promises to bring the power of the FreeBSD kernel to Ubuntu Linux. The best part of using the FreeBSD kernel is that you'll end up using the famous Z File System, or ZFS. Xfce is also included along with the popular Firefox, LibreOffice, and Ubuntu Software Center apps. ubuntuBSD is inspired by the Debian GNU/kFreeBSD project, it is hosted on SourceForge, and has been created by Jon Boden.

Comments

Why would anyone want this?

[Anonymous Coward]

I've tried to run FreeBSD in the past. Hardware was poorly supported with frequent system crashes. Software was poorly supported; many packages and ports simply wouldn't install or build. The ports collection, often touted as a great benefit of BSD, often failed to properly build software. Because ports are built as root so they can be installed, I once had a port try to build and install a rootkit in the process. There are far fewer BSD users, so bugs are infrequently reported and the lack of developers means that bugs often go unfixed. Ubuntu isn't perfect, but why would anyone want to bring this experience to Linux?

Re:Why would anyone want this?

[Anonymous Coward]

That's funny. I'm an OpenBSD user and I'm saying this seriously -- you've just summarized my experience with Linux. I tried using a Linux distro and after lots of pain came back to OpenBSD which has been so much easier to use (once you get past the bare bones installation. I'll admit).

Re:Why would anyone want this?

[0100010001010011]

Hardware was poorly supported with frequent system crashes.

This is the trick I've figured out with FreeBSD: They don't support crap hardware. The best hardware support comes from companies that pay developers to make FreeBSD drivers. (See RealTek vs Intel ethernet drivers). If you look at who the core users of FreeBSD are and look at who sponsors development it's mostly servers.

For most of my desktops I've returned to server hardware anyway. Whitebox builds were fun when I was poor and my time was free. But after the N'th time of dicking around with figuring out why my Motherboard and RAM won't play nice or the monster heat sink I added scrubbed off traces I'd rather just buy a machine that's supported.

Buy good hardware. Get good results.

so bugs are infrequently reported and the lack of developers means that bugs often go unfixed.

I've had the opposite experience. Bugs are so infrequent that if it's not a PEBKAC error then the FreeBSD guys can usually drill down to the bottom of it quick. If it's a problem someone else has experienced then the fix is most likely out there already anyway. Google will return results with the error I'm having instead of threads of noise from the Ubuntu forums.

often failed to properly build software

[Pics or it didn't happen].

Even if you somehow screwed the system up so bad that ports wouldn't build, there's always pkg.

I once had a port try to build and install a rootkit in the process.

[Citation Needed]

Re:Why would anyone want this?

[convolvatron]

Why would anyone want this?

FreeBSD is hardly a perfect system. But why would I want to cripple it further by making it look like Ubuntu?

Re:Why would anyone want this?

[Bruce Perens]

They don't support crap hardware.

:-)

It would help me believe this if you could point out "crap" hardware, explain why it's crap, and show that BSD developers decided not to support it because it was crap, with reference to mailing list messages.

Otherwise, it might be better to say "the developers support what they had at a point in time and now only buy what is supported.

Re:Why would anyone want this?

[Anonymous Coward]

Infidel! Wash your mouth out with soap immediately! Didn't you know that BSD is perfect and not to be criticized?

Re:Why would anyone want this?

[DNS-and-BIND]

Ah, I know this one. We had the same problem when Solaris came out with an x86 version back in the early 2000s. These drive-by linux users downloaded the ISO as if it were the distro-of-the-week, attempted to install it on whatever random crap hardware they had, and then loudly declared that Solaris sucked because they couldn't get something or other to work properly. The problem isn't the OS, the problem is between the keyboard and chair.

When you install a serious OS, you first must do something very unfamiliar and disconcerting. First, you must look at the list of supported hardware. The importance of this cannot be underestimated. If your hardware is not on the list, then the OS will not support it. Most linux fans come from Windows, which supports just about every bit of random junk Taiwanese hardware in the world. Other OS's aren't like that.

After you look at the hardware list, you then must do another thing that is totally counter-intuitive and out of most people's experience - go and buy supported hardware. They won't do this, they install the OS anyway, and surprise surprise when something doesn't work because the documentation clearly states it's not supported, guess what happens? They get frustrated, feel insulted, and begin screaming loudly online that the OS sucks ass. They wonder why they heard so many great things about the software. If it's so fucking great, why doesn't their wireless trackball and webcam from 1997 work seamlessly like under Windows? This happened so often back in the day with Solaris x86 that we had a slang term for the people who did it, though I forget what it was.

Re:Why would anyone want this?

[Antique Geekmeister]

To avoid systemd, I suspect. Since systemd only works with the Linux kernel, and so far has produced a great deal of difficulty in return for its very aggressive re-engineering of the entire Linux back end infrastructure, it seems very very reasonable to try simply replacing the kernel to get a clean divorce from the systemd infrastructure.

Re:Why would anyone want this?

[drinkypoo]

I wanted to use OpenBSD on my netbook for secure mobile computing, but they denied the patch that someone submitted for my wifi hardware on specious grounds so that they wouldn't have accept his patch — they claimed that getting some values from a Linux driver was a copyright issue when it has been conclusively proven that it hasn't. So I can't use it without buying additional hardware, so I installed Linux.

No joke, someone actually submitted a working patch, and they denied it on bullshit grounds. They don't want decent hardware support.

Last time I built a box just for OpenBSD it choked hard on "supported" NICs and failed to route my packets, almost costing me a job. Admittedly, that was over ten years ago. Still not going down that road again.

Re:Why would anyone want this?

[Anonymous Coward]

Are you fucking serious? You're suggesting he basically maintain his own spin of the kernel because some idiots without any sense of how people actually want to use computers don't think the patch is useful enough to integrate upstream? Fuck me, people here really do live in their own bubble.

Re:Why would anyone want this?

[0100010001010011]

if you could point out "crap" hardware, explain why it's crap,

If the vendor doesn't think it is worth their time to develop drivers for it then it must not be good enough to develop drivers for.

I already gave the RealTek vs Intel Ethernet drivers.

Another is Nvidia vs AMD/ATI. I hand Nvidia money. Nvidia hands me a working video driver. No reverse engineering needed. No "here's a bunch of specs, write your own driver".

Yet another is Supermicro and Intel motherboards vs consumer boards. Turns out Intel and Supermicro designs and develops motherboards with full driver support. Most consumer boards I've seen will toss in Realtek drivers and a bunch of other cheap chips to save a penny or so with the end result being no driver support in Linux or FreeBSD. Windows drivers are still hit or miss.

What's wrong with PC-BSD?

[0100010001010011]

It has everything 'pretty' that Ubuntu could have, minus the weirdness of a kFreeBSD/GNU userland. It sets everything up from install. You can even select what window manager you want to use during install.

Personally bare FreeBSD itself is just fine if the command line doesn't scare you. It takes less time to install FreeBSD and the 5-10 packages needed to get to a 'normal' desktop than it does to install any version Windows. I'm just kicking myself for not making the switch earlier.

Additionally. Describing projects as "For Humans" must stop. FreeBSD is already for humans. I could teach a middle schooler to install it and get to Facebook. It's honestly not that hard.

Re:What's wrong with PC-BSD?

[Xtifr]

The "weirdness" of the *nix userland that most people in the *nix world have standardized on these days. Yeah, can't think why they'd possibly want that.

Regular FreeBSD is indeed just fine, but it has dozens of minor incompatibilities with both GNU and SysV that will trip you up if you're used to those others.

GNU/kFreeBSD is a really nice compromise if you're used to either GNU/Linux or SysV. GNU is designed to be as compatible as possible with both BSD and SysV. (So it's also a nice option if you're a BSD guy forced to work on a SysV system. Which is actually how I got started with it.)

If you're a BSD guy using BSD, then yeah, there's little point in putting GNU on it. But for everyone else, it makes a lot of sense.

Frankly, I try to stick GNU on everything. I've run it on BSD, HPUX, Solaris, SunOS, Windows, and, of course, Linux.

Right now, I'm running GNU/kFreeBSD in a vm to port some software to BSD. When I get it running, on that, I'll install regular BSD to make sure everything still works, but I'm interested in the porting effort, not in mucking around learning all of BSD's quirks, so GNU/kFreeBSD is a godsend. I'm not scared of the command line. I just don't want to waste my time remembering all the little incompatibilities of the BSD userland while I'm trying to get some work done. In recent years, every single time I've tried to use straight BSD, I end up getting frustrated at some little thing that just won't work the way I want it to. The way I'm used to.

Now, for someone who's new to *nix, I agree that BSD is a very fine option. Heck, I started with the BSD-based version of SunOS, and still have fond memories of it. But at this point in time, I am so used to the GNU userland that I just don't want to deal with something that doesn't have it. And I'm pretty sure I'm not the only one who prefers GNU—after all, not only has the Debian GNU/BSD survived for many years, but now there's a second one! :)

Thanks summary

[orledrat]

Thanks for answering the question "What's UbuntuBSD?" seeing as this is Slashdot it would have been more useful to explain "What's a human being?". You insensitive clod.

Re:Thanks summary

[Kjella]

Thanks for answering the question "What's UbuntuBSD?" seeing as this is Slashdot it would have been more useful to explain "What's a human being?". You insensitive clod.

It would, but all we got is an undocumented blob written in quad-bits (ACGT) and the original developer can't be reached. Not that it'd do much good, the code is constantly morphing through forking off new child processes while old ones come to a halt so there's probably little of the original left. There was an instruction manual too, but it's equally cryptic like "Rule over the fish in the sea and the birds in the sky and over every living creature that moves on the ground". Source control seems completely absent so there's billions of versions in production. While we're making some progress on reverse engineering to fix the most critical bugs it's mostly a black box project full of unexplained and absurd behaviors. Like making dry jokes on a nerd website.

What?

[Anonymous Coward]

"[...] bring the power of the FreeBSD kernel to Ubuntu Linux."

WTF?

What's the point?

[Anonymous Coward]

I've heard this talked about before, on bsdnow.tv - about people wanting the stability and reliability of the FreeBSD kernel, but with a GNU userland. We use Ubuntu Server at my work extensively, and I use FreeBSD at home, and to be honest, I really don't think it's hard to pick up one if you know the other. Moreover, Linux does have a kernel module now for ZFS, so what's the point? You're almost certainly better off using stock Ubuntu, if you're a Linux guy - it's got actual dollars going into maintaining it. If you're into FreeBSD, you're also better off going with stock FreeBSD.

I don't even know why people use PC-BSD, to be honest, and I've tried it several times. FreeBSD with an XFCE desktop is easy and as quick as sin to set up, and I'm not just a FreeBSD fanboy. People should just go ahead and check it out. Even GhostBSD and DesktopBSD don't get that much traction these days - it's because FreeBSD is pretty straightforward, if you're willing to put an evening or two into finding your way around.

Re:What's the point?

[dbIII]

Moreover, Linux does have a kernel module now for ZFS, so what's the point?

Because ZFS on linux is not yet ready for serious use. It is catching up and is fine for the home PC I'm using now but it still lags a long way behind other systems with ZFS on a lot of points, especially performance.

Re:Hosted On SourceForce?

[cfalcon]

There's nothing wrong with Sourceforge. That's a 2015-era complaint, get modern!

Re:As opposed to?

[cfalcon]

OS X is for dolphins, Windows is for monkeys, standard BSDs are for superior alien beings, and Linux is for robot overlords with laser eyes.

Z File System?

[xbytor]

It's just ZFS. It used to stand for "Zettabyte File System" but that was dropped years ago.

Then you're doing something wrong.

[Ungrounded Lightning]

Are you sure it was FreeBSD, not Net or Open? If so you were either on some oddball hardware where support was just coming into existence, or you were doing something wrong.

BSD forked three ways and the three branches are specialized for three purposes:
  - FreeBSD is about running on as many kinds of hardware and peripherals as possible. If it runs under BSD it runs under FreeBSD. Once the driving code is solid it might get imported into other branches, or ported to other things (like Linux). Meanwhile, maybe the code for your device is still new and still flaky.
  - NetBSD is about being a reference platform for developing, and pushing the envelope on, networking technology (at the expense of only bothering to be guaranteed to run on a limited number of platforms and configurations).
  - OpenBSD is about being reliable and secure - at the expense of being limited to devices and code that are open enough to be audited, and being developed and maintained outside the US and its "encryption is a weapon" export controls.

(Or at least that's how I understand it. I'm not following it closely right now because I'm not running it - though I'm considering switching from Linux to Open and may soon be giving it more attention.)

Re:Then you're doing something wrong.

[Anonymous Coward]

Not even close. NetBSD's focus has always been on portability and running on as many platforms as possible. FreeBSD is a general purpose operating system. DragonflyBSD is a fork of FreeBSD, originally for the purpose of improving performance for symmetric multiprocessing and threading. OpenBSD is focused on security and reliability.

Re:Then you're doing something wrong.

[drinkypoo]

NetBSD is about being a reference platform for developing, and pushing the envelope on, networking technology (at the expense of only bothering to be guaranteed to run on a limited number of platforms and configurations).

Before Linux, Netbsd was THE distribution for running on weird platforms. Since Linux, Netbsd has no particular reason to exist. Linux runs on dramatically more hardware than netbsd now. Several ports have withered and died since Linux took on all the weird architectures.

Re:Yea, but systemd...

[Eunuchswear]

SystemD trolls don't know their history.

Upstart, you idiot,

You know, an init system that runs everything under ptrace.

Re:Hosted On SourceForce?

[DrXym]

There's nothing wrong with Sourceforge except for the fact that they hijacked "abandoned" open source projects in order to bundle crapware into the installer, tried to persuade popular projects to bundle crapware, placed fake download links on project download pages, created fear distrust in their community, and failed to keep up with rival hosting services.

So nothing wrong at all. Even if they're trying to make amends or correct these things, it's too late. The trust is gone. Most active projects have decamped to the likes of Github.

Re: Hosted On SourceForce?

[Anonymous Coward]

So, last month?