FreeBSD with Linux Kernel

Status
Not open for further replies.
Not because of "FreeBSD didn't goose-stepped Linux". PR and marketing are alien to FreeBSD Foundation. That's the reason.
No. The root cause is not the foundation. FreeBSD developers & traditional user-base refuse to integrate
  • a modern graphical installer for standard PC (currently x86-64, but that seems to change).
  • there's a strong notion that FreeBSD is oriented towards server & not desktop.
  • methods for automated configuration. pkg/ng does not include automagic configuration intentionally. The package managers on Linux have it.
  • For these reasons, young nerds grow up with Linux, thus that's what they choose in their professional career.
 
Things like wayland come about more because of the NIH syndrome that afflicts systemdOS than any deficiencies of the old system it seeks to replace.. eventually.

Hmm, fair point. You’d figure all these years they would’ve fixed all those deficiencies. Unless X is so fundamentally broken that a new re-design is warranted. Perhaps, they’re lazy?

No. The root cause is not the foundation. FreeBSD developers & traditional user-base refuse to integrate
  • a modern graphical installer for standard PC (currently x86-64, but that seems to change).
  • there's a strong notion that FreeBSD is oriented towards server & not desktop.
  • methods for automated configuration. pkg/ng does not include automagic configuration intentionally. The package managers on Linux have it.
  • For these reasons, young nerds grow up with Linux, thus that's what they choose in their professional career.

Great point. I’ve wondered what the reluctance is with the developers on this. I’ve watched every single FreeBSD Want/Need/Have session from various conferences. Only 10% of the entire discussion addresses desktop related issues. Are graphics kryponite to kernel developers? Or does our developer base not have enough experience in that domain? From an implementation perspective; one blocker could be API/ABI compatibility with the rest of base; which we couldn’t do with something like X.org/Mesa, etc unless we fork it.
 
Hmm, fair point. You’d figure all these years they would’ve fixed all those deficiencies. Unless X is so fundamentally broken that a new re-design is warranted. Perhaps, they’re lazy?

I think the story went somewhat like this:

- We need something better than X. Let's not fix what we have let's invent something. It's Not Invented Here but it must be Invented Here.
- X is in need of some bug fixes, we can't be bothered we're working on this new concept called Wayland.
- Those using X need to step up and fix it; we're not.
- X slowly dies as a 'generic' platform.

It's not fundamentally broken, it's a little flawed but so is TCP/UDP. It maintains backward compatibility for everything back into the '80s, and there's still a lot of hardware out there using it that will never use wayland or its compositor.

I should add, NIH applies to Redhat who basically owns Linux nowadays. I guess it's the default for redhat now?
 
No. The root cause is not the foundation. FreeBSD developers & traditional user-base refuse to integrate
  • a modern graphical installer for standard PC (currently x86-64, but that seems to change).
  • there's a strong notion that FreeBSD is oriented towards server & not desktop.
  • methods for automated configuration. pkg/ng does not include automagic configuration intentionally. The package managers on Linux have it.
  • For these reasons, young nerds grow up with Linux, thus that's what they choose in their professional career.
And now explain why Linux has 100% market share in supercomputers, and probably 90% among servers that are in the large cloud datacenter (with the remainder being Windows, and very little BSD, AIX, MVS/VM and others).

Do you think that people who build $100M supercomputers or data centers with millions of machines care about graphical installers? Shouldn't they love an OS that is "oriented towards servers"? And automated configuration? The argument with "mindshare of nerds" is particularly pointless.

I think trying to make BSD in general and FreeBSD in particular into a mainstream success is pointless. That ship has sailed. It is a niche operating system, for a set of niche users. I greatly enjoy using it for my little server. But I admit that my personal use case doesn't scale to other people or other uses.
 
  • Like
Reactions: a6h
I should add, NIH applies to Redhat who basically owns Linux nowadays. I guess it's the default for redhat now?
RedHat is owned by IBM since July 2019. They commited 2nd most changes to the Linux kernel, after Intel. Who "owns" Linux? Here's the list of members of the Linux Foundation. Compare to that to the FreeBSD Foundation...
And now explain why Linux has 100% market share in supercomputers, and probably 90% among servers that are in the large cloud datacenter (with the remainder being Windows, and very little BSD, AIX, MVS/VM and others).

Do you think that people who build $100M supercomputers or data centers with millions of machines care about graphical installers? Shouldn't they love an OS that is "oriented towards servers"? And automated configuration? The argument with "mindshare of nerds" is particularly pointless.
Well, from a technical viewpoint FreeBSD is better in many aspects, equal in some, and worse in a few. Most Open Source software is developed on Linux, just because that's what the developers (nerds) have on their machines. That's a criterium (availability of software (-stacks)). Thus, yes, I do think in the end it's this profane reason. In general, people do not like to try & learn something new, when they know what they're used to & know about is sufficient for the task. If it would be only for technical reasons, I guess DragonFly BSD should have a 67% market share on super-clusters.
 
RedHat is owned by IBM since July 2019. They commited 2nd most changes to the Linux kernel, after Intel. Who "owns" Linux? Here's the list of members of the Linux Foundation. Compare to that to the FreeBSD Foundation...

Well, from a technical viewpoint FreeBSD is better in many aspects, equal in some, and worse in a few. Most Open Source software is developed on Linux, just because that's what the developers (nerds) have on their machines. That's a criterium (availability of software (-stacks)). Thus, yes, I do think in the end it's this profane reason. In general, people do not like to try & learn something new, when they know what they're used to & know about is sufficient for the task. If it would be only for technical reasons, I guess DragonFly BSD should have a 67% market share on super-clusters.
Ok, I'll rephrase: linux is run/steered/directed by redhat, regardless of who owns redhat.
 
a modern graphical installer for standard PC (currently x86-64, but that seems to change).
But I admit that my personal use case doesn't scale to other people or other uses
There're 4 delays in FreeBSD installer: before "Welcome | Install", after "DHCP | Network configuration", "Commit" and "Final configuration | OK"
If I could ignore them, aka let's suppose no delay and seamless installation , I can literally close my eyes and successfully go through FreeBSD installation.
I'm not a nerd and I don't what to be, but I like it. Some people hate it. Changing FreeBSD installer? It would be a shame. NAY!
 
Ok, I'll rephrase: linux is run/steered/directed by redhat, regardless of who owns redhat.
Doesn't make it true, no matter how often you rephrase it. Have a look into that impressive list, and decide yourself if all these big players would allow a single entity to own a system that is mission-critical for them.
There're 4 delays in FreeBSD installer: before "Welcome | Install", after "DHCP | Network configuration", "Commit" and "Final configuration | OK"
If I could ignore them, aka let's suppose no delay and seamless installation , I can literally close my eyes and successfully go through FreeBSD installation.
I'm not a nerd and I don't what to be, but I like it. Some people hate it. Changing FreeBSD installer? It would be a shame. NAY!
And if you could choose between a graphical & CLI installer, which both call the very same backend-scripts? Where the only change I propose is to add a language selection screen at the very beginning, which also pre-sets the choice of keyboard layout & timezone (which can be changed by successive steps)?
 
  • Like
Reactions: a6h
Doesn't make it true, no matter how often you rephrase it. Have a look into that impressive list, and decide yourself if all these big players would allow a single entity to own a system that is mission-critical for them.

And if you could choose between a graphical & CLI installer, which both call the very same backend-scripts? Where the only change I propose is to add a language selection screen at the very beginning, which also pre-sets the choice of keyboard layout & timezone (which can be changed by successive steps)?
Um, look at systemd, flatpak and wayland and tell me redhat isnt steering that ship. Of course I clarify this is the generic 'linux' not just the kernel - you know the one with hundreds of 'distributions'.
 
Why should I? I'm not using these, and I even do not know what flatpak is ;) Maybe your point is that many contributions come from RedHat. Yes. More came from Intel in the last few years. So then you'd have to correct your view to: "Intel steers that ship". I don't know about more current numbers. Anyway, Linux is, like FreeBSD, a joint effort of many people. Hard to tell if there's a single entity that leads all others.
 
I'm not counting contributions, but who is setting the direction/running of 'linux'; it's certainly not set by intel. Name your intel projects that are changing the very way 'linux' works? I've given you 3 that I can think of the top of my head, I'm sure I could probably find more should I care to look but I don't care to. I've spent more time than I care thinking about an OS I don't have interest in.
 
Nowadays, Linux is the standard itself! Sadly, but true.

The proprietary companies doing the development set the "standard". Linux is just along for the ride. It is literally going to be torn apart by those messy guys. Just wait and see. It is not even UNIX-like anymore. The next step is to tightly integrate the display system into the OS (Wayland) and finally remove the terminal. SystemdOS is basically dying right in front of your eyes. Enjoy! ;)

BTW, I don't think if Linux switch to Wayland, you could free from worry. It only means FreeBSD has to adopt Wayland,
Wayland is basically Mir. I don't think we need to worry unless we run Gnome 3 on FreeBSD. And most people don't. FreeBSD has 100's of window managers to choose from. Linux now only has 3!
More damaging is that Linux has basically lost its ability to do effective remote desktop. I suppose with the direction Linux is going, it certainly won't need that anyway haha.

OpenBSD is always a joke (or a toy?).

If you check out their funding... You will see that the companies guiding the industry certainly don't think so.

I believe OpenBSD receives more in funding than many Linux distros combined. Perhaps have a think why. Also, without OpenSSH, I don't think Linux would even be around today.
 
I think trying to make BSD in general and FreeBSD in particular into a mainstream success is pointless

I think the most of us just want FreeBSD as a viable, functional alternative to mainstream desktops. I am so done with Apples idiotic direction (with their hardware, and now recently, software), political bullshit against users (anti repair), and draconian behavior to the point where I’m starting to despise them. If marketing is a means for FreeBSD to get all the driver support it needs, so be it. If FreeBSD had it’s own in-kernel GPU driver support and it’s own display server to match; I’d be a VERY happy clam.
 
I think the most of us just want FreeBSD as a viable, functional alternative to mainstream desktops. I am so done with Apples idiotic direction (with their hardware, and now recently, software), political bullshit against users (anti repair), and draconian behavior to the point where I’m starting to despise them. If marketing is a means for FreeBSD to get all the driver support it needs, so be it. If FreeBSD had it’s own in-kernel GPU driver support and it’s own display server to match; I’d be a VERY happy clam.
Observation: there're some Intel fellows actively contributing to FreeBSD, their GPUs are supported well, you can download the latest drivers e.g. for em(4) from ports/packages. AFAIK, their competitor is not that cooperative, and other GPUs are reported more often to have issues, at least here in the forum.
 
  • Like
Reactions: a6h
I think the most of us just want FreeBSD as a viable, functional alternative to mainstream desktops. I am so done with Apples idiotic direction (with their hardware, and now recently, software), political bullshit against users (anti repair), and draconian behavior to the point where I’m starting to despise them. If marketing is a means for FreeBSD to get all the driver support it needs, so be it. If FreeBSD had it’s own in-kernel GPU driver support and it’s own display server to match; I’d be a VERY happy clam.
Some who benefit of the server Marketing, are not interested in FreeBSD's default graphical desktop extended and have opposed, dare to call other nerds when I think that nobs or average users have no idea what a nerd is. Most, by not say everyone, use some graphical desktop environment or graphical manager of their hardware or device.
 
I think the most of us just want FreeBSD as a viable, functional alternative to mainstream desktops.
I don't know about "most of us". I certainly have zero interest in running FreeBSD on the desktop. I used to use Linux on the desktop, and I don't want to go back to that either, it's just too painful. I'm quite sure that any *BSD would be harder than Linux, so: no thank you.

Personally, I use Apple as a desktop (both at work and at home), and I'm quite happy with that. Not perfectly happy, but better than Windows (which I've also used for many years).
 
I've started using a Freebsd desktop after years of using mostly Mac and some Linux. Hasn't been too bad yet, but I'm not very picky.

Apple is clearly trying to unify their operating systems. I'm not interested in having a "laptop" that's just a big phone.
 
If you check out their funding... You will see that the companies guiding the industry certainly don't think so.

I believe OpenBSD receives more in funding than many Linux distros combined. Perhaps have a think why. Also, without OpenSSH, I don't think Linux would even be around today.
Disagree with you again. Should I remind you that OpenBSD is not OpenBSD alone but a bunch of Open[Something] projects? Yes, you are right. These companies paid them to maintain OpenSSH. The money didn't pour into the development of OpenBSD as an OS itself. On their latest release, I see very moderate improvement to vmd. All of them are just minor improvements.
 
Well yeah... You think the Linux Foundation gets donations to work entirely on the kernel?
Of course I know.
The Linux Foundation hosts many of the most important open source projects in the world, including Linux. With more than 1,000 companies backing tens of thousands of active developers, our projects harness the power of open source development to fuel innovation at unmatched speed and scale.

But this, doesn't change a single thing. Linux Foundation is relevance because of Linux. OpenBSD is relevance, because of OpenBSD based tools (OpenSSH, OpenSMTPd, ...), not OpenBSD itself.
 
My request to SirDice/Crivens: can this be closed? Enough of linux, already.
In the other hand, I don't understand why the FreeBSD forums Mods and community members could tolerate trolls and morons too much (it's not to insult, because I think they indeed not stupid, they present themselves like that in order to you... feed them, the troll, when you yourselves don't know you are being used and they laughed in front of the monitor) but treat me too harsh (or too bad). If I'm in authority, any thread like that would be deleted from the beginning and the OP banned immediately, not a simple thread close. But I know the reason they don't do that. The SJWs. They will accuse them of being dictators and spread hate. No one want to be dictators, though. But the forums rules could be used to protect them from these SJWs. They could invoke the forums rules to delete the thread and ban the OP without being dictator.

p/s: if the Mods think that the OP is not a troll but only lack of knowledge, they could simply delete the thread and send a private message to him to address the problem politely: his idea do not work and not supported here. Done.
 
Status
Not open for further replies.
Back
Top