Xorg metaport extremely bloated with linux crap

Freedesktop is basically overall good, that the licensing used is permissive, or LGPL2.1 that is used is meant for other programs to use it. It's also good, that you can pull anything off of it, and make it the BSD way. However, there's nothing big enough to maintain such variant.

At least Freedesktop.org maintains projects, which is better than not having any organization maintain or follow a group of projects. They use licensing which allows freedom for use. A lot is in the flavor of Linux, however. BSD needs its own, for its own style.

In Xorg, at least they could have used LGPL2.1, instead of GPL code. Because LGPL doesn't try to leak into absorbing other code. LGPL isn't so bad in itself.
 
Freedesktop is basically overall good, that the licensing used is permissive, or LGPL2.1 that is used is meant for other programs to use it. It's good, that you can pull anything off of it, and make it the BSD way.

At least Freedesktop.org maintains projects, which is better than not having any organization maintain or follow a group of projects. They use licensing which allows freedom for use. A lot is in the flavor of Linux, however. BSD needs its own, for its own style.

In Xorg, at least they could have used LGPL2.1, instead of GPL code. Because LGPL doesn't try to leak into absorbing other code. LGPL isn't so bad in itself.
I agree. The type of copyleft I hate is strong copyleft. I think weak copyleft is perfectly fine. I like the CDDL, and I am ok with it being in base. I would also be fine with MPL or EPL in base, and I would tolerate LGPLv2.1-Only in base. FooGPLv3 is a pile of trash and I think it benefits proprietary dual-licensers more than it does the open-source community.
 
FooGPLv3 is a
I wouldn't say that, but it goes too far. It needs an updated version. They may do that when they realize the benefits for both them and other opensource. Motif took a long time to get to LGPL: first it moved to an for use open license, but not opensource. Nothing wrong with Motif being proprietary at first, later they moved to LGPL for the best for their project.
I think it benefits proprietary dual-licensers more than it does the open-source community.
I agree with that part. Depends on the purpose of why authors choose that, or if it becomes clear when an improved GPL version update comes along, and which ones that have the ability move on to it.
 
I wouldn't say that, but it goes too far. It needs an updated version. They may do that when they realize the benefits for both them and other opensource. Motif took a long time to get to LGPL: first it moved to an for use open license, but not opensource. Nothing wrong with Motif being proprietary at first, later they moved to LGPL for the best for their project.

I agree with that part. Depends on the purpose of why authors choose that, or if it becomes clear when an improved GPL version update comes along, and which ones that have the ability move on to it.
Sadly, RMS's GPLv4 series would be even more restrictave. I have read the GNU website, and they make this clear. They also hate FreeBSD as ports allow you to obtain software under "open-source but nonfree" licenses like the Artistic License. They don't want you to be in control of your own computing, they want to be in control of your own computing. If you have read all of gnu.org you realize that RMS goes so far into free-software extremism that he loops back into the world of proprietary software ;)
 
How was it killed by Red Hat? I know the creator of Xlibre is A BIT paranoid, given his crap about vaccines. I just think it was killed because the Linux world is collectively idiotic, and yet they are at the helm of everything open-source. The death of Xorg reeks of Linux Being Linux.
I believe you were on a long vacation for quite some time.

Let me give you the shortcut to get back to reality:

-
View: https://youtube.com/watch?v=iwaaSatk0pI

-
View: https://youtube.com/watch?v=ujJCyXfWpOo

-
View: https://youtube.com/watch?v=ONdcalK5JLQ

-
View: https://youtube.com/watch?v=ONdcalK5JLQ

-
View: https://youtube.com/watch?v=W4o1uXe6cVc

- https://youtube.com/watch?v=XsbietQFf1w
- https://youtube.com/watch?v=ehEoRkMtoT8
- https://youtube.com/watch?v=Y0l4W-LYYdc
 
What's wrong with Happy Pride Month? This guy who made these videos is just a right-wing pundit in the tech world. Why do we need politics in our space whatsoever? Everyone should just put aside their political differences and biases and just enjoy an open-source world free of left-right politics. Who gives a fuck about anything? Let's not make big issues over small things such as who someone likes, or if you belive in social security. A community divided cannot stand. If people split the OSS community, MICROSOFT WINS.
 
Sadly, RMS's GPLv4 series would be even more restrictive. I have read the GNU website, and they make this clear.
There's no GPLv4, but maybe you're predicting.

They also hate FreeBSD as ports allow you to obtain software under "open-source but nonfree" licenses like the Artistic License. They don't want you to be in control of your own computing, they want to be in control of your own computing. If you have read all of gnu.org you realize that RMS goes so far into free-software extremism that he loops back into the world of proprietary software ;)
FSF likes permissive licenses, because it allows GPL to eat them up. Even though they can use BSD licenses, they don't like it as much, because it requires them to do minimal preservation work of license terms to use it. They like MIT, because they can use it and don't have to acknowledge it at all.

While I bought the arguments in the beginning, they fundamentally didn't make sense. It made me wonder how permissive licensing could exist, when they allowed another license to eat them up. Then, it seemed like a paradox between permissive and GPL. In between is better. Their terms still don't make sense, the rationale for what they want to do, and how it's used makes sense, but that doesn't make sense to the terms of the wording of GPL terms. They're justifications made to be able to use other libraries, that it even restricts on LGPL. I had to question it, and read the explanation, and make it work, and say ok, well, it's the only option. When getting around the rationale, it's not even close to the best way to do it.

LGPL2.1 is ok, except it also allows GPL to absorb it. LGPL also limits protections for it, as the licensing terms can't be updated to have a patent retaliation clause. It's just there for compatibility with everything, including both major GPL versions.

Depending on his motives. If they're like that, then it's time to displace the GPL, as can be done, as there's more libraries under licenses under Apache, MPL and enough CDDL, while they can use LGPL. There's more libraries to gain from, than to lose. Eventually, they'll either bend to having a version which allows it to use libraries dynamically linked, so that GPL doesn't have to force BSD, ISCL, Apache, even LGPL to dual license. So it doesn't make the authors of MPL code have to dual license as well. To have a library which is allowed use by both major versions of GPL and everything else, you'd have to use permissive licenses which don't have patent clauses, or dual license with LGPL2.1 and Apache 2.0.

I'm not sure about Artistic License, but licenses do need to have clear terms, if that was the case. Artistic License 2.0 is accepted. FSF recommended that Artistic License 1 be dual licensed, which doesn't seem like going against their wishes, however making anything dual license either is good for clarity, or it causes an impediment, depending on the terms of that license.

I just think it was killed because the Linux world is collectively idiotic, and yet they are at the helm of everything open-source.
Arch Linux does a few things right. Without them, we wouldn't have BSD like programs like sxhkd and be able to use their documentation for programs. Arch is almost like the Linux which does things the BSD way. They use gmake, Linux documentation software, and Bash, but that's their being Linux. I heard of Alpine Linux too.

If people split the OSS community, MICROSOFT WINS.
When it was Bill Gates against Linus, Microsoft was this way. Microsoft actually has a good Opensource license. They have another open license, which isn't opensource. Also, Microsoft put DOS under MIT license. Maybe they don't see opensource as a threat to their business model now.
 
There is so little GNU software in base that it is almost nonexistent. There is gcov and one or two other things.
Of course that's the case. BSD has its own utilities in base, they are different from GNU. If you have a problem with that, you might want to dig up the lawsuits that ATT filed back in the day, make sense of THAT, see what ATT's beef was, and be prepared to pony up a bit of money, like which US court you will file your appeal in, and the like. Then please post back here with results!

Xorg silently killed by Red Hat and FreeDesktop.org is not a future solution.

FreeBSD - as a project - same as GhostBSD does - needs to migrate to X11 solution that has a future - that solution is XLibre at the moment.
How was it killed by Red Hat? I know the creator of Xlibre is A BIT paranoid, given his crap about vaccines. I just think it was killed because the Linux world is collectively idiotic, and yet they are at the helm of everything open-source. The death of Xorg reeks of Linux Being Linux.
Wayland is making pretty good progress on FreeBSD, and is probably outpacing the Linux camp at this point. In fact, this may shame the Linux camp into the 'standardize or die' mode ;) Oh, and I've been helping the matters along 😈
 
Today, many things have swelled up like yeast.
For example, compare the size of MS Office 97 Standart Edition (134MB) / MS Windows 98SE (175MB) with analogs in FreeBSD.
I have 3 GB for sway with all dependencies = 133 packages (400 MB downloaded; 3 GB - took up under 133 packages).
sway + basic programs-settings-tunings = 48 packages (50 MB downloaded; 269 MB taken up under 48 packages).
Libreofice - 100 packages (466 MB downloaded; 1 GB taken up by Office out of 100 packages).
FreeBSD + Wayland(sway) + Office = 4.8GB + 3GB + 1GB = 8.8GB.
I would gladly not pull this GIGA-slow LibreOffice, if I found a program that does not glitch and does not crash from every sneeze,
like AbiWord/Gnumeric. But this AbiWord/Gnumeric combination is just awful and has been falling apart for the entirety of its existence.
 
vmisev, Fedora doesn't only come with Wayland. The workstation version which runs Gnome, might only come with Wayland, but at present, it's quite easy to set up any X window manager. RH and clones (Alma, Rocky), only have Wayland, but X is still quite available in Fedora. I just do a custom install, their new name for minimal, then run dnf install -y xorg-x11-server-Xorg and a few other bits and pieces, like xorg-x11-xinit so startx works, xorg-x11-drv-libinput, and so on.
 
vmisev, Fedora doesn't only come with Wayland. The workstation version which runs Gnome, might only come with Wayland, but at present, it's quite easy to set up any X window manager. RH and clones (Alma, Rocky), only have Wayland, but X is still quite available in Fedora. I just do a custom install, their new name for minimal, then run dnf install -y xorg-x11-server-Xorg and a few other bits and pieces, like xorg-x11-xinit so startx works, xorg-x11-drv-libinput, and so on.
Fedora KDE Plasma Desktop 42 is definitely Wayland only OOB, I installed XLibre form custom COPR, and then plasma-workspace-x11 and kwin-x11 from official repo; also had to install/reinstall few more things and to edit sddm.conf to make it work, probably it is simpler with official xorg-x11
 
Wayland is making pretty good progress on FreeBSD, and is probably outpacing the Linux camp at this point. In fact, this may shame the Linux camp into the 'standardize or die' mode ;) Oh, and I've been helping the matters along 😈

If You want to run Wayland - run Wayland - I do not want to.

I want X11 - and X11 works perfectly for me.

Red Hat and FreeDesktop tried (and still tries) to kill X11 just to 'force' Wayland. This is not competition - this is betrayal of the core open source values.
 
RedHat may have done a lot for open source, long ago. However, while I agree a company should be allowed to make money, I think that RH betrayed open source awhile ago, with their stance on various clones.
To show how smart I am, I remember when RH hired an important developer for Scientific Linux, and thinking, Uh oh, they're going to try a take over--which they did several years later when they made CentOS CentOS steam (or is it stream?) and began to make it hard for anyone to use their binaries. With all the lawyers RH and IBM have, I'm sure it fit the letter of GPL, but it sure as heck violated the spirit of it.
Oh....kay, I'm ranting about Linux on a FreeBSD forum, I apologize, with the feeble excuse that I find it hard to resist an opportunity to badmouth RedHat.
 
Not sure what you mean there. I have noto-cjk fonts and /usr/local/share/fonts/noto is 368M. I think those are the only noto fonts I have. Were you serious that they take up so much room or being hyperbolic? Or, am I missing some other place where noto fonts take up space.?
 
Not sure what you mean there. I have noto-cjk fonts and /usr/local/share/fonts/noto is 368M. I think those are the only noto fonts I have. Were you serious that they take up so much room or being hyperbolic? Or, am I missing some other place where noto fonts take up space.?
I have the upstream version of noto installed since the FreeBSD ports one is outdated, but even with only the stuff from ports noto is still supposedly 2GB:
Code:
└─[$]> sudo pkg install noto-2.0
Password:
Updating FreeBSD repository catalogue...
Fetching data.pkg: 100%   10 MiB   3.6MB/s    00:03
Processing entries: 100%
FreeBSD repository update completed. 36506 packages processed.
Updating FreeBSD-kmods repository catalogue...
FreeBSD-kmods repository is up to date.
All repositories are up to date.
The following 57 package(s) will be affected (of 0 checked):

New packages to be INSTALLED:
        noto: 2.0 [FreeBSD]
        noto-extra: 2.0_7 [FreeBSD]
        noto-hk: 2.004 [FreeBSD]
        noto-jp: 2.004 [FreeBSD]
        noto-kr: 2.004 [FreeBSD]
        noto-kufi-arabic: 2.109 [FreeBSD]
        noto-music: 2.003 [FreeBSD]
        noto-naskh-arabic: 2.019 [FreeBSD]
        noto-naskh-arabic-ui: 2.017 [FreeBSD]
        noto-sans-arabic: 2.012 [FreeBSD]
        noto-sans-arabic-ui: 2.011 [FreeBSD]
        noto-sans-armenian: 2.008 [FreeBSD]
        noto-sans-bengali: 3.000 [FreeBSD]
        noto-sans-cham: 2.005 [FreeBSD]
        noto-sans-cherokee: 2.001 [FreeBSD]
        noto-sans-devanagari: 2.006 [FreeBSD]
        noto-sans-ethiopic: 2.102 [FreeBSD]
        noto-sans-georgian: 2.005 [FreeBSD]
        noto-sans-hebrew: 3.001 [FreeBSD]
        noto-sans-hk: 2.004 [FreeBSD]
        noto-sans-jp: 2.004 [FreeBSD]
        noto-sans-kannada: 2.006 [FreeBSD]
        noto-sans-khmer: 2.004 [FreeBSD]
        noto-sans-kr: 2.004 [FreeBSD]
        noto-sans-lao: 2.003 [FreeBSD]
        noto-sans-math: 3.000 [FreeBSD]
        noto-sans-myanmar: 2.107 [FreeBSD]
        noto-sans-runic: 2.002 [FreeBSD]
        noto-sans-sc: 2.004 [FreeBSD]
        noto-sans-sign-writing: 2.005 [FreeBSD]
        noto-sans-sinhala: 2.006 [FreeBSD]
        noto-sans-tamil: 2.004 [FreeBSD]
        noto-sans-tc: 2.004 [FreeBSD]
        noto-sans-telugu: 2.005 [FreeBSD]
        noto-sans-thai: 2.002 [FreeBSD]
        noto-sc: 2.004 [FreeBSD]
        noto-serif-armenian: 2.008 [FreeBSD]
        noto-serif-bengali: 3.000 [FreeBSD]
        noto-serif-devanagari: 2.006 [FreeBSD]
        noto-serif-display: 2.009 [FreeBSD]
        noto-serif-ethiopic: 2.102 [FreeBSD]
        noto-serif-georgian: 2.003 [FreeBSD]
        noto-serif-hebrew: 2.004 [FreeBSD]
        noto-serif-hk: 2.003 [FreeBSD]
        noto-serif-jp: 2.003 [FreeBSD]
        noto-serif-kannada: 2.005 [FreeBSD]
        noto-serif-khmer: 2.004 [FreeBSD]
        noto-serif-kr: 2.003 [FreeBSD]
        noto-serif-lao: 2.003 [FreeBSD]
        noto-serif-myanmar: 2.106 [FreeBSD]
        noto-serif-sc: 2.003 [FreeBSD]
        noto-serif-sinhala: 2.007 [FreeBSD]
        noto-serif-tamil: 2.004 [FreeBSD]
        noto-serif-tc: 2.003 [FreeBSD]
        noto-serif-telugu: 2.005 [FreeBSD]
        noto-serif-thai: 2.002 [FreeBSD]
        noto-tc: 2.004 [FreeBSD]

Number of packages to be installed: 57

The process will require 2 GiB more space.
1 GiB to be downloaded.
 
That was several decades ago. Today, the most common operating system in use at Microsoft is Linux. Thinking that Microsoft is the enemy of open source is outdated.
Who is the enemy now then? I would say Oracle. Except for their good job with Java, they ran everything else into the ground.
 
Who is the enemy now then? I would say Oracle. Except for their good job with Java, they ran everything else into the ground.
Oracle is running even Java into the ground. Ever since the pandemic ended, Oracle managed to drive even HP up the wall with their Java EE license harassment. I was on the sidelines for active discussions of what alternatives to Java EE would be the least disruptive option for HP's R&D to migrate to.

In the arena of rank-and-file consumers like us, Oracle is not worth our attention. Maybe a bit of entetainment, if that. 😏
 
Why do we need an enemy?

Open source is succeeding beyond any expectation. About 95-99% of all servers in the world run Linux. Exactly 100% of all supercomputers on the TOP500 list run Linux. Looking for the enemy that has destroyed open source, or is trying to do so, seems somewhat foolish.
But if there is no enemy how do we know who to blame? :)
astyle Oracle is good for one thing though: Rivaling SCO in the ability for a law firm to put on the mask of a tech company ;)
Ha. I thought I was done hearing about SCO....

Circling back to XLibre, it seems like it has a definite niche, should probably be a port (at least) or a group project between the *BSDs.
I'm saying this based on technical goals, not political, because there is zero chance I will politically agree with every developer on every opensource product I use.

If XLibre becomes "last version of Xorg/whatever we call it, with obsolete features removed, with new desired features added, with bugs fixed and CVE/security practices applied and removes all hooks for Wayland/systemd", I'd be happy.
 
I'll try XLibre as soon as it's available in pkg :p (I'm interested in trying it but Xorg still works fine)

It might go something like freenginx: I first heard about it on FreeBSD, tried it for a bit, but noticed after a while the pkgs were outdated compared to upstream. I went back to nginx, but I'm still curious about freenginx. If XLibre works comparably and drops-in easily, I'd switch to it primarily and check out changes between it and xorg. Anything that maintains this low-latency or somehow makes it even lower sounds interesting :cool:


I looked at xorg vs xorg-minimal, and minimal wanted to install more packages (not sure the specifics but it was 30 vs 40 packages). https://docs.freebsd.org/en/books/handbook/x11/#x-install mentions:
A smaller version of the X Window System suitable for experienced users is available in x11/xorg-minimal. Most of the documents, libraries, and applications will not be installed. Some applications require these additional components to function.
I didn't feel like guessing what some apps might be and went with the smaller non-minimal meta:p
 

Seems to drop in easy!

Xorg.0.log has a flood of this though:

Code:
[2025-08-29 14:51:58] input-thread: InputThreadDoWork waiting for devices
[2025-08-29 14:51:58] input-thread: InputThreadDoWork waiting for devices
[2025-08-29 14:51:58] input-thread: InputThreadDoWork waiting for devices
[2025-08-29 14:51:58] input-thread: InputThreadDoWork waiting for devices
[2025-08-29 14:51:58] input-thread: InputThreadDoWork waiting for devices
[2025-08-29 14:51:58] input-thread: InputThreadDoWork waiting for devices
[2025-08-29 14:51:58] input-thread: InputThreadDoWork waiting for devices
[2025-08-29 14:51:58] input-thread: InputThreadDoWork waiting for devices
[2025-08-29 14:51:58] input-thread: InputThreadDoWork waiting for devices
[2025-08-29 14:51:58] input-thread: InputThreadDoWork waiting for devices
[2025-08-29 14:51:58] input-thread: InputThreadDoWork waiting for devices
[2025-08-29 14:51:58] input-thread: InputThreadDoWork waiting for devices
[2025-08-29 14:51:58] input-thread: InputThreadDoWork waiting for devices
[2025-08-29 14:51:58] input-thread: InputThreadDoWork waiting for devices
[2025-08-29 14:51:58] input-thread: InputThreadDoWork waiting for devices
 
Back
Top