This new release is based on FreeBSD 14.0-STABLE. Update Station got a significant change to upgrade to a major FreeBSD version, allowing upgrading GhostBSD from 13.2-STABLE to 14.0-STABLE. Also, a major change to the installer is the user created is an admin, and the root user gets the same password as the admin. If the admin password is changed after the installation, the root password will not change.
↫ GhostBSD’s website
GhostBSD is a user-friendly, desktop-first ‘distribution’ of FreeBSD – a project which, in my humble view, should be part of the FreeBSD project-proper. With some old-time Linux feeling a sense of disenfranchisement towards Linux due to things like Wayland and systemd, FreeBSD could serve as an excellent alternative, and an official desktop-first ISO could play a role in that.
Of course, that’s not exactly core to FreeBSD’s mission, and they really shouldn’t be listening to idiots like me, but I think it’s an idea worth pondering.
I love the idea of a desktop-focused FreeBSD spin, but every time I’ve tried GhostBSD on real hardware, from a humble Intel based mini PC to an AMD Ryzen workstation to my current workstation, a NUC 11 Core i9 with an AMD Radeon Pro GPU, it always has major hardware incompatibilities. That’s probably more my fault for not trying to run it on hardware blessed by FreeBSD, but at the same time, it will never be ready for general use if it’s limited to FreeBSD’s hardware support.
By contrast, I have been able to run OpenBSD on nearly every piece of hardware I’ve ever come across, including a 23 year old Dell Latitude laptop along with the same Intel mini PC, Ryzen workstation, and NUC 11 workstation mentioned above. In fact the only hardware I’ve failed to get OpenBSD to boot on was an HP Stream Mini Desktop with an obscure Braswell era Celeron CPU. It also wouldn’t boot Haiku, NetBSD, or any other non-Windows, non-Linux OS.
OpenBSD is easy to use as a desktop workstation, and I have always heard that this is because OpenBSD devs run it as their bare metal workstation and effectively “dogfood” the OS, whereas FreeBSD devs typically use Macs or Windows PCs and remote into FreeBSD running on server hardware to do OS dev. I don’t know if the latter is true (if so, how does any real hardware get attention from them?) but regardless OpenBSD is my desktop BSD of choice because it just plain works.
How does another kernel, running xorg (presumably) or Wayland, address “a sense of disenfranchisement towards Linux due to things like Wayland”? I’m confused by that statement. I assume this must be running some version of xorg – and the situation is the same, no matter what kernel you run that on. Xorg has been maintained by the same people who have developed Wayland, and they don’t want to maintain xorg any more. Using freebsd would seem to not address that in the slightest. Am I missing a piece?
Perhaps Thom meant that Wayland, like systemd, is a Linux-first/Linux-only project born on Linux and meant for Linux, and so the BSDs get snubbed. The BSDs and their respective userlands were not even on the radar for the original Wayland developers, and from my understanding there is little (FreeBSD, OpenBSD) to no (NetBSD, MidnightBSD, DragonflyBSD) support from the Wayland team for any *BSD devs looking to implement Wayland on their OS. Another issue with regard to Wayland on FreeBSD specifically is that Nvidia provides the best GPU driver support on FreeBSD, but Wayland on Nvidia hardware is a joke right now.
And with all of that said I’m not defending Xorg/X11. It’s bad, really really bad, and the OpenBSD devs forked it long ago into Xenocara so they could fix all the bad stuff and keep it maintained for OpenBSD users and devs who need a desktop. But Wayland, as far from ready as it is on Linux — and it is FAR from ready! — is even further from ready on the BSDs.
I see, they forked xorg – that makes sense. Are they maintaining it? Pushing it forward?
Just going to push back gently on this one thing – Wayland is not “far from ready.” I’d say it’s completely ready – the support around it is lagging, but that makes sense. I can use it successfully today, for most things. It’s just some legacy support – I can’t use things like Barrier, not because “Wayland isn’t ready” but because Barrier hasn’t been ported to Wayland yet – that’s not Wayland’s issue. Third party support may be missing, but it’s largely ready to go. It takes time for ecosystems to adapt.
Maybe I could accept an argument that the ecosystem around Wayland is not ready to go, but that’s a VERY different statement, and the distinction is important.
This narrative that Wayland is far from ready just isn’t true. It’s ready, and it’s necessary to switch to it – soon, no more foot dragging. The only way to help it, is to start moving to it more quickly. OSS is generally slow at letting things go. That’s not necessarily a bad thing, but it’s time to let xorg go.
Yes, it gets updated along with the rest of the OS, as it is a core part of the project.
https://github.com/openbsd/xenocara
Those are conflicting statements and that is the kind of thing that makes me doubt anyone who says “Wayland is ready”. If you can’t even agree with yourself on the matter, it’s not ready.
I do get what you are saying though: Most of the issues with Wayland today are with the rest of the software world playing catch-up. However, part of that is due to the unwillingness of the Wayland devs to help or even work with developers of other projects. NIH syndrome is real.
Morgan,
It’s more of a semantics argument. Some wayland devs could say it’s ready, but from a user perspective wayland won’t be ready until the wayland compositors support their needs.
Personally I’m open to switching to new remote desktop software, but until wayland is actually ready for the desktops and hardware I use, the point is mute. I hope it gets done sooner rather than later, but we’ll see.
On sites like this, we will be hearing that Wayland is not ready for the next 5 years ( somebody will be saying it for longer ). At the same time, almost every Linux user will be using it.
It would not surprise me if more than 50% of Linux desktops are Wayland already. It is the default on GNOME and most people just run with the default. GNOME is the default desktop on almost all the “big” Linux distros.
Most people still using X11 today do so because of their DE choice more than their windows server choice. That includes me by the way as most of my systems use a DE that is still X11 only. We will leave 2024 with GNOME, KDE, XFCE, Cinnamon, COSMIC, Budgie, and others supporting Wayland, Some will default to Wayland. Some may even drop X11. In 2024.
Some other fraction avoids Wayland because of GPU compatability. I don’t have numbers on that but see below. NVIDIA may hold Wayland back but I expect good progress in 2024. If the 545 driver is an indicator, they may actually begin to priority Wayland over X.
It is becoming my impression that it is a fairly small fraction of users that avoids Wayland because of the “not ready” issues beyond GPU. This is not to disparage that few. They are mostly the “power users”. I am just talking numbers and they are not the majority.
Look at this poll from half a year ago:
https://www.gamingonlinux.com/users/statistics/
In that poll, a third of the users are on Wayland. That number is being dragged down by the fact that almost 50% of them use KDE ( which will be defaulting to Wayland in literally a couple of days ). 40% of them use NVIDIA. The Proton Beta adds support for Wayland. I am sure Valve and Crossover want to make it the default in 2024.
My guess is that we exit 2024 with at least 70% market share for Wayland. It will be 90% in 3 years. People will be saying it is “far from ready” even then but what will it even mean at that point ( valid criticisms or not ). There are new use cases every day that are Wayland only. With Linux market share growing ( new users with no bias ), it may not be long before “X is unusable” or even “what is X11” is more common than “Wayland is not ready”.
tanishaj,
My prerequisits are very reasonable my KDE and XFCE linux desktops need to work on my hardware with remote desktop. There’s always someone claiming wayland is ready, and yet every single time I’ve tried it wayland was still broken for me. Naturally I’ve been pushing out attempts farther in hopes that given more time wayland will eventually work with my requirements met. I also hope 2024 is the year it happens.
I find such categorizations misleading because remote desktop and screen sharing aren’t the exclusive domain of power users. Less so after returning to work, but many millions of regular users are impacted when things like this don’t work. Rather than blaming users, team wayland need to make sure it’s working everywhere!
BTW this feature fragmentation around desktops is exactly why some of us were critical of wayland’s design. We knew that this would cause problems, and sure enough it is causing problems today. Unfortunately this means they have to reimpliment the same features across several different compositor code bases. Regardless, this is the path taken by wayland so in order to get the job done they can’t just finish one compositor and say it’s done, they have to finish all the compositors… hopefully in 2024.
I haven’t heard anyone say “it is far from ready”. It’s actually really close but the broken parts still need to be fixed. Once that’s done the pragmatic among us will switch. Simple as that.
I really should give BSD another try. I first/last tried in 1995 haha. 100% Linux user since 2004.
Was wondering about the Xorg/X11 situation myself for BSD. Thanks for sharing the info on the situation.
Wayland… I’m on Debian 11.9 on DELL XPS 15. Keep giving Wayland a go on KDE-Plasma. When the whole session goes black whilst I’m working and returns me to SDDM, I select Plasma/X11 and get on with my day with a few expletives.
FreeBSD itself is good, something to tinker with.
The FreeBSD community, namely their forums, are very cliquey. I got permanently banned as I had an opinion that was my own. Forum thread exchanges between a couple of users can’t be left alone as their admins must involve themselves to be seen, or to feel, all powerful – very sad.
The forum mods/etc are not affiliated with the FreeBSD Foundation, if you ever wondered.