GNOME on FreeBSD page is useless/dead

sidetone

Aspiring Daemon

Thanks: 313
Messages: 915

#2
Gnome development on FreeBSD is not dead. The Gnome project maintains a few ports.
I've noticed that some of them are board games, which they do update.
 

Oko

Daemon

Thanks: 763
Messages: 1,620

#5
What is the purpose of keeping outdated GNOME (3.18) in the 'main' Ports tree instead of importing working and stable '3.26' from here:
https://github.com/freebsd/freebsd-ports-gnome/tree/gnome-3.26

... as people report its stable?
The same reason other 10000+ obsolete ports are kept in the ports tree. Lack of man power of political will to prune obsolete crap. It makes FreeBSD project appears bigger than actually is. Very low standards when to be a port committer. In particular very few people use FreeBSD on the desktop and even fewer of those people use Gnome (possibly nobody). Recently interesting discussion of Distro Watch Weekly regarding Debian vs FreeBSD approach in porting software (don't get me wrong I hate with a passion Debian installation scripts which try to guess what I am going to do and are starting daemons for me). I would also like to see much smaller FreeBSD ports tree with well maintained and well tested software (QA, regression testing not just to test if it is building or not). I would like to see lot less code in FreeBSD proper and more developers focused on deleting stuff instead of importing pre-alpha quality crap.
I'd like to clear something up here. Yes, both Debian and FreeBSD have official package committers and it takes a lot of time/effort to get that official status, however their roles are quite different. In the Debian camp a committer is responsible for packages they enter into the distribution which means committers usually only work on packages they personally find useful or interesting. I have submitted a handful of packages to Debian committers and they were all rejected. Not because of errors or licensing, but because the committer did not want to take on more work, looking after a package they had no interest in.

On FreeBSD official committers also look over submitted packages and test them, but the FreeBSD commiter is not responsible for maintaining the package and keeping it up to date, the submitter is. As a result, FreeBSD committers rarely reject packages (unless there there is a security issue or they fail to build properly). They can accept any valid/secure package without worrying about the extra burden of looking after the package in the future.

This means FreeBSD can have relatively few committers (who check security/syntax) but a lot of submitters who maintain the packages. Debian tends to have a lot of committers, but few submitters outside of that core group because submitted packages are often ignored/rejected.

I'm not comparing the process to becoming an _official_ committer in my article above, I'm comparing the process for submitting and getting new packages accepted.
 

shepper

Aspiring Daemon

Thanks: 240
Messages: 711

#6
With each release, Gnome is becoming more and more difficult to port to the BSD's. Antoine Jacoutot and Jasper Leivisse maintain Gnome for OpenBSD and if you look at the shear size of the port, it is massive. The developers have to deal with systemd, pulseaudio, network managers, cups, geo location and soon wayland.
 

Preetpal

Active Member

Thanks: 27
Messages: 118

#7
In particular very few people use FreeBSD on the desktop and even fewer of those people use Gnome (possibly nobody).
I am using Gnome 3 on FreeBSD as a desktop right now with no issues. Why bash FreeBSD by spreading misinformation on a FreeBSD Forum?

You're also wrong about having "Obsolete" ports in the ports tree. They are useful, as they work on the system and can be used as a basis for importing later versions of the software into the system. I am not against further testing but to remove old ports simply because they are old is ridiculous and unjustified.
 

Oko

Daemon

Thanks: 763
Messages: 1,620

#8
I am using Gnome 3 on FreeBSD as a desktop right now with no issues. Why bash FreeBSD by spreading misinformation on a FreeBSD Forum?
What misinformation? That the number of FreeBSD desktop users is minuscule. I challenge you to check the percentage of FreeBSD developers at a FreeBSD DevSummit or one of BSD conference. You will be surprised how low is the percentage of developers using FreeBSD as a desktop. Ask wblock@ he was giving stickers to people who use FreeBSD on their laptops 2 years ago at BSDCan if you don't trust me. Meeting a regular folks like me who run FreeBSD is even more challenging. Check the number of FreeBSD related jobs on Dice.

You're also wrong about having "Obsolete" ports in the ports tree. They are useful, as they work on the system and can be used as a basis for importing later versions of the software into the system. I am not against further testing but to remove old ports simply because they are old is ridiculous and unjustified.
Software which is unmaintained upstream is obsolete and should be immediately removed from the ports. Full stop. Here is a simple example for you. LPRng related ports

https://www.freebsd.org/cgi/ports.cgi?query=LPRng&stype=all

It is dead, upstream! While I concur that Gnome is not in that group, as shepper correctly pointed out upgrading Gnome on any BSDs is becoming increasingly challenging. If the port fails significantly behind the upstream it should be removed and put into the attic. People who want to use it can always check out the attic as a foundation for their porting effort. QA and regression testing should be must for the port to remain in the port three not an option.
 

sidetone

Aspiring Daemon

Thanks: 313
Messages: 915

#9
I wish the Gnome project and everything else that chooses to use systemd, pulseaudio, etc. would fork from the ports tree. There there would be a repository setting to enable it, and that's fine if FreeBSD hosts it on a forked repository. There should be an improved ports framework, where everything runs on or is in transition to use oss, rc, sndio, etc...

A big part of the problem is, the Handbook section on Xorg is obsolete, and when a bug is submitted to improve it, it sits there for months. The section on Gnome, XFCE and KDE needs to split off into an article. After the FreeBSD Handbook is fixed, it needs to be referred to in the Required Reading subforum.
 

Snurg

Aspiring Daemon

Thanks: 326
Messages: 794

#10
Desktop FreeBSD is not dead, period.
Distrowatch atm shows ~1450 users of FreeBSD and its derivatives
This is 10x the number of OpenBSD (144) there.
FreeBSD is no mainstream thing and never will be.

It is dead, upstream! While I concur that Gnome is not in that group as shepper correctly pointed out upgrading Gnome on any BSDs is becoming increasingly challenging. If the port fails significantly behind the upstream it should be removed and put into the attic. People who want to use it can always check out the attic as a foundation for their porting effort.
I strongly disagree. If one would do like you propose, Gnome2 and KDE4 would have to be removed also.
But there are quite a few people who either prefer these over the newer versions or there are no current ports due to the simple fact that , as for example shepper mentioned, it is a challenge to sort out all these poetteringisms, etc.

Thus I'd like to say a big, big kudos to the Gnome, KDE etc porters.
Their work gets highly appreciated.
 

Preetpal

Active Member

Thanks: 27
Messages: 118

#11
What misinformation? That the number of FreeBSD desktop users is minuscule. I challenge you to check the percentage of FreeBSD developers at a FreeBSD DevSummit or one of BSD conference. You will be surprised how low is the percentage of developers using FreeBSD as a desktop. Ask wblock@ he was giving stickers to people who use FreeBSD on their laptops 2 years ago at BSDCan if you don't trust me. Meeting a regular folks like me who run FreeBSD is even more challenging. Check the number of FreeBSD related jobs on Dice.
I think your overly negative about FreeBSD. Not every user of FreeBSD is going to publicly say they use it. Not every user of FreeBSD engages in public events about OS development.

Consider Linux as an example (since its community is so "healthy"), many people who develop for it (or deploy to it (including me for example)), do not actually use it as a desktop (and that's fine). Their installation numbers are also probably inflated (many people just use it via virtualization only; also not all downloads result in installs).

For example, even the Linux Foundation director uses macOS:

You cannot judge a community by its size. The only reason many are using Linux is because it's cheap (they're essentially freeloading no matter what anyone wants to believe). The FreeBSD community (from what I've experienced) is great and very welcoming/helpful.

Also many people are using FreeBSD via the Nintendo Switch and the Sony Playstation consoles (there's over a 100 million end users). Before you discount those numbers, realize that the Android operating system (often used to inflate the Linux user count) itself is far from being vanilla Linux and is very locked down as well.

On a side note, there are some people who are (IMO unjustifiably) very negative about the continued existence of non-Linux operating systems. Take Minix for example, it has an immense install base as it's embedded in recent Intel processors.
 

stratacast1

Active Member

Thanks: 20
Messages: 183

#12
Honestly, if GNOME starts to see some love on FreeBSD again I'll switch from Fedora to FreeBSD on my workstation, like I've been wanting to do for a while now. I'm no coder, but I would love to at least help with bug reports n what not. All that would be needed then is Visual Studio Code for my daily workflow and I'd be set, FreeBSD has everything else I use :D
 
OP
OP
vermaden

vermaden

Son of Beastie

Thanks: 1,042
Messages: 2,694

#13
With each release, Gnome is becoming more and more difficult to port to the BSD's. Antoine Jacoutot and Jasper Leivisse maintain Gnome for OpenBSD and if you look at the shear size of the port, it is massive. The developers have to deal with systemd, pulseaudio, network managers, cups, geo location and soon wayland.
Maybe its time to join forces in an OpenZFS fashion, one GNOMEonBSD team to share similar work ... just a thought.
 

stratacast1

Active Member

Thanks: 20
Messages: 183

#14
Maybe its time to join forces in an OpenZFS fashion, one GNOMEonBSD team to share similar work ... just a thought.
That'd be an interesting thought...with the systemd takeover of Linux, even some non-systemd distros have troubles. Not sure what that solution would look like, but I could only predict duplicated efforts. But maybe there could be something to collaboration with say..Gentoo dev and others who don't use systemd to fork KDE Plasma/GNOME3/whatever so non-systemd OSes could have a modern DE
 

Snurg

Aspiring Daemon

Thanks: 326
Messages: 794

#15
To avoid redundant work, this would be great indeed.
There is some work ongoing on systemd shims, for example.

This would be an area where systemd-free Linuxes and BSDs could cooperate...
 
OP
OP
vermaden

vermaden

Son of Beastie

Thanks: 1,042
Messages: 2,694

#16
IMHO They should not limit their 'colabiration' to GNOME, something like OpenCORE initiative seems more sensible for the systemd refugees.
 

Trihexagonal

Daemon

Thanks: 525
Messages: 1,037

#17
Why bash FreeBSD by spreading misinformation on a FreeBSD Forum?

*snip*

You're also wrong about having "Obsolete" ports in the ports tree. They are useful, as they work on the system and can be used as a basis for importing later versions of the software into the system..
Oko isn't bashing FreeBSD. He doesn't sugar coat things, is a long time BSD user and I know him to be speaking the truth in regard to his statement about the BSD conference and block@ distributing stickers. The part about a good portion of developers not using FreeBSD on their boxen is common knowledge.

It doesn't matter who I tell IRL I use FreeBSD, they don't have the slightest idea what I'm talking about if its not Windows.

I use multimedia/xmms on all my machines and would hate to see it vanish (I have it saved in case it does), but it's an abandoned project in favor of audio/xmms2 and most if not all links on its page are dead.
 

Preetpal

Active Member

Thanks: 27
Messages: 118

#18
Oko isn't bashing FreeBSD.
Come on, read what he said again.

The same reason other 10000+ obsolete ports are kept in the ports tree. Lack of man power of political will to prune obsolete crap. It makes FreeBSD project appears bigger than actually is. Very low standards when to be a port committer. In particular very few people use FreeBSD on the desktop and even fewer of those people use Gnome (possibly nobody). Recently interesting discussion of Distro Watch Weekly regarding Debian vs FreeBSD approach in porting software (don't get me wrong I hate with a passion Debian installation scripts which try to guess what I am going to do and are starting daemons for me). I would also like to see much smaller FreeBSD ports tree with well maintained and well tested software (QA, regression testing not just to test if it is building or not). I would like to see lot less code in FreeBSD proper and more developers focused on deleting stuff instead of importing pre-alpha quality crap.
He insulted the ports committers which are the people who are responsible for anything being available to install on FreeBSD (many of them actually work to get their changes merged upstream).

He basically insinuated that the FreeBSD desktop userbase does not exist.

He stated that he would like to see the FreeBSD project downsize in terms of what it offers end users. He literally wants the developers to delete code that people are relying on.

He literally wants the FreeBSD project to do the job of upstream maintainers which is an impossible task (no one does this, it's the job of the upstream maintainers to ensure that their software runs properly on different platforms such as FreeBSD).

He accused the developers of FreeBSD of producing pre-alpha quality crap.

IMHO, Oko is bashing FreeBSD in a toxic way.
 

sidetone

Aspiring Daemon

Thanks: 313
Messages: 915

#19
He literally wants the FreeBSD project to do the job of upstream maintainers which is an impossible task (no one does this, it's the job of the upstream maintainers to ensure that their software runs properly on different platforms such as FreeBSD).
This is not possible with Gnome. I respect their right to do as they want, but it will be a clunky system with attempts to hardwire bloat within its codebase. FreeBSD porting guidelines should be improved and be made the standard here.
IMHO, Oko is bashing FreeBSD in a toxic way.
I wouldn't say bashing, but most recently his attitude towards FreeBSD has become a little toxic.
 

Minbari

Well-Known Member

Thanks: 117
Messages: 250

#20
I know I'll be a little off topic but why port a software full a systemd (thanks Lenntard for your crap init) when devs could improve/support Xfce which by the way is BSD licensed.
Look at those morons (Moore brothers) they try to reinvent the wheel instead to improve FreeBSD. Why don't add parallelization to rc init instead of use openrc (which is itself a sysv clone, more or less) and create that abomination Lumina Desktop because we don't have a BSD licenced desktop which is a false statement. Wasted time and human resources like Ubuntu do it with the "might crap" Unity which now they abandoned.
 

Preetpal

Active Member

Thanks: 27
Messages: 118

#21
I know I'll be a little off topic but why port a software full a systemd (thanks Lenntard for your crap init) when devs could improve/support Xfce which by the way is BSD licensed.
You know what, if Gnome is not supporting FreeBSD upstream, then that's a sufficient reason (for me at least) to eventually switch over (I cannot do so right now due to time constraints) to XFCE on FreeBSD.
 

Snurg

Aspiring Daemon

Thanks: 326
Messages: 794

#22
I have to second Preetpal.
TBH, I don't know what is going on with you, Oko.
Large part of your posts are rants, however these were usually backed and justified.

But denigrating the efforts of the many porters and maintainers is all but constructive.
Exactly this stuff makes people lose motivation to invest time and effort to make things work on *BSD. :(
What do you intend with this?

Minbari you made me laugh :) "Lenntard" :D And "Moorons" just comes into my mind :D
 

sidetone

Aspiring Daemon

Thanks: 313
Messages: 915

#23
You know what, if Gnome is not supporting FreeBSD upstream, then that's a sufficient reason (for me at least) to eventually switch over (I cannot do so right now due to time constraints) to XFCE on FreeBSD.
That is more than likely not the fault of FreeBSD, that included bloat is incompatible with FreeBSD's tendency to try to keep things unbloated.

Upstream can have the implication that, they want dependencies to be required that FreeBSD already has efficient alternatives to. Fixing that, will be a difficult task, because once it is fixed, upstream can then change it, like a fast moving game of catch up, causing plenty of wasted FreeBSD resources.
 

hukadan

Active Member

Thanks: 153
Messages: 244

#24
those morons (Moore brothers)
I do not think insulting people makes your post more convincing. I used TrueOS for quite few months with Lumina as Desktop (did you try it yourself ?). It is not the best Desktop I have ever used, but it is far from an abomination. I am now back to OpenBSD and XFCE and my opinion on TrueOS was mixed to say the least. From what I know, XFCE is already well supported on both FreeBSD and OpenBSD. As for OpenRC, when they develop a new Desktop they are wrong because they reinvent the wheel (statement with which I agree at least in part) and when they use an existing project that fits their needs, they are wrong again because they should have taken a project that did not and modify it ? If you are interested, the reasons of the OpenRC choice were detailed somewhere else on this forum IIRC.
 

Trihexagonal

Daemon

Thanks: 525
Messages: 1,037

#25
Come on, read what he said again.
Personally, I prefer someone be brutally honest with me rather than BS or lie to me about something. I have much more respect for that and deal with it better.

There are good and bad points to everything. Some perhaps more valid than others. I'm not always on the mark looking back on things I have said, can be very opinionated and not afraid to speak my mind, even if at times to my own detriment.
 
Top