Is there demand for a "FreeBSD Kommunity Edition"?

  • Yes, sure

    Votes: 19 18.6%
  • Likely

    Votes: 11 10.8%
  • Maybe

    Votes: 10 9.8%
  • Doubtfully

    Votes: 10 9.8%
  • No

    Votes: 43 42.2%
  • Don't know

    Votes: 9 8.8%

  • Total voters
    102
Heh, don't feel bad. I provide loads of feedback that isn't valuable. I just keep going (much to the despair of the other forum members ;))
Hey thanks. I am new here and wanted to be more friendly oriented. My first post was not to helpful so I reposted. I would love to see FreeBSD grow, so I donated. Which is funny it's something I never felt like doing with Linux. I also help a lot on another forum which currently has a app that runs on FreeBSD. Not sure if I am allowed to mention that or not. Anyway thanks not wanting to hijack this thread..
 
I recommend contributing and promoting the FreeBSD/KDE Initiative. That's really what the project is for; community collaboration. In terms of spitting out an ISO for others to use, someone in that initiative should do it.

Why there's no WiFi/Bluetooth/ACPI hardware probe and setup in bsdinstall is beyond me. That'll definitely help with the experience of transition to FreeBSD.
The sheer variety of hardware makes installation after the base system difficult.
Unlike Windows and Linux, there is no big hardware test laboratory available.

A community project (i.e. driven by users, not by employed professional programmers) could compensate for this.

Imagine sort of "user contributed plug-in modules" for /stand/sysinstall bsdinstall.

These could serve various specialized purposes:
[ul]
[li]detect video card(s), install and configure console and xorg + drivers, make auxiliary settings for common configurations (for example, deactivating Nvidia Optimus on laptops with Intel graphics), correctly select and configure console (for example, to avoid activating vt in cases it will predictably fail, like with Nvidia cards)
[li]Wifi: wpa_supplicant could be configured quite easily by a interactive dialog() script using a questionnaire.
Bluetooth: Very difficult to configure. Basically the problem is lack of information about working configurations, so there is massive repetitive redundant researching.
[li]ACPI is a real must to handle, as suspend/resume depends on that. Sadly FreeBSD's ACPI AML implementation is rudimentary compared to Windows and Linux just because of lack of test hardware. But this would be a kernel team task, not for external contributors.
[li]Suspend/resume configuration: There are a number of suspend/resume breakers. These can be searched for, found and fixed by scripts. On newer UEFI systems, it is even possible to add hibernate, which takes quite some steps and should be scripted, too.
[/ul]
Such additional install/configure options could save users many hours setting up a system.

P.S.: How do I make lists on this forum?
 
Like a QA community for hardware support? We should have a documented procedure all of us can follow for random x86/ARM laptops. Then bug reports and fixes can be exchanged and pushed to committers. We’d probably need an HCL of community “certified” FreeBSD laptops. Including low level stuff like special function keys, backlighting, etc. I’m probably piggybacking off the Wiki list though.

If I’m missing something, my apologies. I blame the jazz cabbage.
 
After doing a fresh install of FreeBSD/KDE on my T480; there's a lot hardware probe/setup improvements to be had in FreeBSD, whilst integration improvements to be had on the KDE side.
IIRC that topic was listed in the Ideas page in the wiki (ideas for new projects, from large & complicated to beginner's tasks). So maybe s/o with the right skills took it & what you see are the 1st results of that work.
Why there's no WiFi/Bluetooth/ACPI hardware probe and setup in bsdinstall is beyond me. That'll definitely help with the experience of transition to FreeBSD.
Let's just guess & hope that this is work in progress. From my limited understanding, that's very hairy, because many hardware manufacturers are lazy in following the standards & from an economical viewpoint they're forced to comply to Windows & Mac, and Linux in case of server hardware. FreeBSD's market share is just too low to be treated special or test on.
P.S.: How do I make lists on this forum?
Click Help -> BB Codes there:
--------------------------------------------------------------------\
|
|
\/

STTBD (Scroll to the bottom, dude)
 
Like a QA community for hardware support? We should have a documented procedure all of us can follow for random x86/ARM laptops.
Yes, this is the difficult thing.
We need to develop a procedure tree for every of these components, similar to these used for plant identification, disease diagnosing, and the like. As a starting point we can use what is already described in the wikis.

We’d probably need an HCL of community “certified” FreeBSD laptops. Including low level stuff like special function keys, backlighting, etc.
Definitively! And not only for laptops, also for mobos etc.

Recently I built a computer with Gigabyte mobo, AMD graphics and AMD CPU. I planned to use to learn how to configure xorg on AMD and to finally update my cpupdate program so it can also update AMD processors' microcode. So I was thoroughly frustrated that I had to find out that FreeBSD panics when trying to figure out the mobos' ACPI configuration while booting, while Linux and Windows just work. If there were a blacklist of mobos, I would have saved a lot of time and money by buying another, "user certified" board.

I’m probably piggybacking off the Wiki list though.
The general problem with the FreeBSD wiki is its restrictive maintenance.
Very few people have writing rights.
So it is a big effort to get things into the wiki.
For this reason most FreeBSD Wikis I saw are more or less obsolete, and lack detail.
For example, the suspend/resume wikis' test procedures are for FreeBSD up to version 9 or 10, and if you do not know the differences on contemporary FreeBSD, you might not have much delight with that wiki.

Could it be a good idea to set up some etherpads or the like, so an editable wiki could be made accessible to the community, and serve as sort of "inbox" of updates that the official wikis' maintainers could integrate in the official wiki?
Could this help getting the FreeBSD wikis so well-made and detailed like the Arch Wiki?
What do you think?
 
We need to develop a procedure tree for every of these components, similar to these used for plant identification, disease diagnosing, and the like. As a starting point we can use what is already described in the wikis.

Could this help getting the FreeBSD wikis so well-made and detailed like the Arch Wiki?
What do you think?

This is precisely what I was pondering on for a while. Etherpad approach would work, but we would need a way to facilitate some kind of accountability. Like objective proof that it works (ie. This X webcam works with Y app, or This X PCIe device works with Y mobo for Z peripheral, etc.) I'd say we may need a separate Freshports-like database that can curated and searched. Hence the "certified" part.
 
Regarding the Nvidia Optimus problem, maybe there is a possibility to make a small driver, just for deactivating it. Maybe this could be integrated in the i915 driver?
There is bug 192617 and a proposed new port. It worked for me when I checked the status of things over the last year or so. Maybe a community initiative can help find someone to finish it and and many other patches which have stalled over the years when the original submitter lost interest and those like me who used them for a while didn't have enough motivation or skills to take them up.
 
Could this help getting the FreeBSD wikis so well-made and detailed like the Arch Wiki?
What do you think?
In general I find Linux distro infrastructure pretty flaky but the Arch wiki is surprisingly substantial. I do feel they are doing something right (or that every other distro is doing it so very wrong!)

One thing that I think we could improve on though is make a wiki that can be exported as a PDF from the very start. None of this "online only" stuff. Surely we could set something like this up?

Since the X11 DRM drivers have moved closer to Linux, they have been a nightmare in terms of documentation and mismatched versioning. This wiki idea could really help here.
 
Thoughts on Arch Wiki:
1. Linux circle claims Gentoo Wiki was the best wiki
2. Gentoo Wiki burned to the ground. Rise of Arch!
3. Arch's psyche: CLI, RTFM, STFW. Sounds familiar?
4. Wiki is to Arch Linux, as /usr/doc is to FreeBSD.
 
Wiki is to Arch Linux, as /usr/doc is to FreeBSD.
Not entirely. Arch has a lot of third party setup tutorials. It's very handy for things like building your own desktop, or setting up a FAMP stack. A lot of my understanding of the workings of Xorg came from the Arch wiki. FreeBSD should have a third party handbook or wiki for configuration. I had to hunt through Google and this forum just to curate a list of steps to ensure all the ACPI functions of my T480 work. That's not exactly helpful for potential newbs.
 
Not entirely. Arch has a lot of third party setup tutorials. It's very handy for things like building your own desktop, or setting up a FAMP stack. [...] That's not exactly helpful for potential newbs.
You are correct. But you have also consider, they have the numbers.
More users (esp. desktop) -> more contributes -> better Wiki/docs.
 
In addition to this, FreeBSD /usr/doc is
a) not installed by default, and
b) is useless and thus worthless just because it cannot be accessed on a machine that yet needs to be configured. The hen-egg question.

Furthermore, as Beastie7 said, the FreeBSD handbook only concentrates on the core.
Applications are not covered. But getting applications work is the real part of getting a computer work.

BTW, I doubt that Arch has considerably more users than FreeBSD.
What they do have, though, is more openness to willing co-workers and cooperators.
On FreeBSD the number of people who get actual write/commit permission in is very small.
This small circle of people decides what gets ignored and what gets cared about.
I have myself experienced the apparently sometimes very unwelcoming attitude against people who willingly offer their contributions.

There is bug 192617 and a proposed new port. It worked for me when I checked the status of things over the last year or so. Maybe a community initiative can help find someone to finish it and and many other patches which have stalled over the years when the original submitter lost interest and those like me who used them for a while didn't have enough motivation or skills to take them up.
Things like that are what annoys me most on FreeBSD.

Willing contributors supply solutions.
But those who have write/commit rights just ignore the issues, for unknown reasons.
In the example mcit mentioned, this is already going on for seven (!) years.
(Edit: just saw another example thread, a user asks for a patch/fix supplied by another user 12 years ago to be finally integrated.)
This way problems will be guaranteed to permanently stay unfixed, making the load of technological debt bigger and bigger.

If this underlying problem, the often-unwelcoming attitude against willing new contributors, cannot be improved, then the only way to get these issues solved might actually be joining a more open distro.
I feel I begin to understand why things like PC-BSD and GhostBSD were born out of pure necessity, just to have things work out-of-the-box.

I never thought I would try out GhostBSD, as I would prefer less fragmentation in the BSD world.
But maybe that could be actually a solution, to get the notebook-and-desktop-use-related issues solved which are persistently being ignored by the FreeBSD team.
So I will download and install GhostBSD on a test notebook, just to see how easily it could be adapted to add a KDE configuration.
 
Following the Press link on the FreeBSD website, from the It's FOSS interview (Feb 2020) with Deb Goodkin from the FreeBSD Foundation: "Q: Let’s talk about desktop. [...] A: [...] Regarding supporting new hardware, we’ve stepped up our efforts to get FreeBSD working on more popular newer laptops. For example, the Foundation recently purchased a couple of the latest generation Lenovo X1 Carbon laptops and sponsored work to make sure that peripherals are supported out-of-the-box.
Q: Are there any plans to make it easier to install FreeBSD as a desktop system? The current focus seems to be on servers.
A:
The Foundation is supporting efforts to make sure FreeBSD works on the latest hardware and peripherals that appear in desktop systems, and will continue to support making FreeBSD easy to deploy, monitor, and configure to provide a great toolbox for building a desktop on top of it. [...]"

So I'd say if we are a little bit insistent about willing to help improving FreeBSD towards a better desktop experience, we can reference these words. Of course, that would imply to have the neccessary skills, or at least the will to improve these.

And I'd strongly suggest to do it the BSD way: discussing topics 1st, e.g. on a mailing list (or freebsd-kde if it's specific to KDE) and/or here in the forum, think before type, and test test test, instead of quick hacks which have to be reverted & redesigned after only a few years or month, like is common on some other OSs. Of course, spike solutions/PoCs clearly marked as such are perfectly ok.

OTT, it's obvious that there are tasks common to all DEs, and since manpower is a bottleneck as usual, it would be beneficial to have a common layer (for FreeBSD-specifics). I.e., find out wether the various FreeBSD DE distributions already have such a common abstraction layer, or are willing to agree on & contribute to a common base concerning (Free)BSD specific tasks, à la FreeDesktop.org for the common DE parts.
 
So I'd say if we are a little bit insistent about willing to help improving FreeBSD towards a better desktop experience,
Insistent that other write what we want? Unfortunately that does not work.

There is a finite number of developers, there is a main direction in the development. Better they
continue to do good and better what they do. Otherwise we get neither this nor that.

For wonderful Desktop experience there is Windows, MacOS and many flavours of Linux.

I really do not understand what people miss on FreeBSD regarding Desktop. I do not understand
why this topic continuously arise.
 
In addition to this, FreeBSD /usr/doc is
a) not installed by default, and
b) is useless and thus worthless just because it cannot be accessed on a machine that yet needs to be configured. The hen-egg question.

To be fair, a lot of third party ports evolve and change over time; which could make maintenance a challenge for something like DocBook. I like the community wiki idea with federated reviews and whatnot. Or the moderators of the forums could uphold some sort of formatting (for topic titles and structure) standard for how-to tutorials. Then the community can submit input, fixes, etc for that particular tutorial thread.
 
But those who have write/commit rights just ignore the issues, for unknown reasons.
I share your frustration but to be fair any new thing that gets committed adds to the long term maintenance burden which seems to be very high already relative to the available resources. In this specific case, the maintainer of nvidia-driver also maintains 220 (!) other ports.
 
FreeBSD is welcoming enough:
Welcome to FreeBSD! This handbook covers the installation and day to day use of FreeBSD --- quote from FreeBSD Handbook
FreeBSD is supporting enough:
It has src, man(1), mailing-list, and one introductory handbook. More is bonus. Build site and write tutorials.
 
Insistent that other write what we want? Unfortunately that does not work.
Insistent e.g. to get write access to the wiki. I wrote an e-mail several month ago asking kindly for an account, and I didn't get a reply until then. Hopefully I do not mix that up with an account for the phabricator, which I got very quickly. I'll have to check my past e-mails.
For wonderful Desktop experience there is Windows, MacOS and many flavours of Linux.
Contra. Neither of these supplied me with a wonderful desktop experience ever, to be more precise, the desktop itself is ok but not the underlying junk (except on Mac), which also causes trouble on the desktop itself. In the case of Mac, I know a few guys who canceled it because of Apple's shift towards restrictions on 3rd party software & against user's privacy. So where do they go? One choose OpenBSD, all others L*x.
I really do not understand what people miss on FreeBSD regarding Desktop. I do not understand
why this topic continuously arise.
A noob can not easily install & use FreeBSD, because s/he is confronted with terms s/he doesn't understand. But every noob has every right to get a free & open source, performant & easy to use OS + GUI. Maybe s/he is doing s/th wonderful with the help of it, where we nerds benefit? Besides that, today's modern life is nearly incomplete w/o access to a computer + GUI. Since the internet has become an integral part of so many basic aspects of our lifes (communication via e-mail, (video-) chat & messaging, so-called "social media", news, etc.) one can even argue that it (a computer usable for a noob) has become a human right... Seriously, I'm not kidding here.
FreeBSD is welcoming enough: FreeBSD is supporting enough:
It has src, man(1), mailing-list, and one introductory handbook. More is bonus. Build site and write tutorials.
See above. For you & me, yes. But for my sister? No.
 
A noob can not easily install & use FreeBSD, because s/he is confronted with terms s/he doesn't understand.
Systems easy to install, like Ubuntu, are very difficult, if not impossible to install in a Computer that is not
the newest. FreeBSD is also uninstallable in many Computers. OpenBSD is the easiest to install.

What is easy or difficult to install is a relative issue. We should be happy that there is Windows, MacOS
and Linux, and continue use what we want: FreeBSD, OpenBSD, etc. We should be happy that we can
choose among different operating systems, insetad of expecting that all be identical.
 
Systems easy to install, like Ubuntu, are very difficult, if not impossible to install in a Computer that is not
the newest. FreeBSD is also uninstallable in many Computers. OpenBSD is the easiest to install.
Ummm, that's not my experience. Ubuntu and Mint have been fantastically easy to install, find Wifi, install the kitchen sink and start in a graphical environment, regardless of machine age.

I like OpenBSD and use it for my desktop environment, and the installer is very simple and quick - "easiest" if you know what to do, but Wifi and desktop environment setup not "easy" as such (in comparison to the the "main" Linux distributions).

FreeBSD I've not tried much in terms of Wifi or desktop so can't comment on that side of it. Nice and easy for a base install, though, and hardware support for server hardware seems better than OpenBSD.

Anything with a "rich" graphical environment will need newer machines with lots of RAM and fast storage for a usable experience.

Last Windows 10 install I did was horrible - everything works but it's so s-l-o-w compared to other OS install experiences.

We should be happy that we can
choose among different operating systems, insetad of expecting that all be identical.
But definitely agree with that.

The biggest hurdle that many people find trying a new OS is grasping the concept that it's NOT the same as Windows or Linux or iOS or whatever. A lot of the concepts are the same, but the implementation might be different. And usually that's not wrong or needing fixing. The sooner you ditch your hang-ups about how OS Y does something compared to OS Z, the happier you'll be. Or maybe you should stick with OS Z.
 
Following the Press link on the FreeBSD website, from the It's FOSS interview (Feb 2020) with Deb Goodkin from the FreeBSD Foundation: "Q: Let’s talk about desktop. [...] A: [...] Regarding supporting new hardware, we’ve stepped up our efforts to get FreeBSD working on more popular newer laptops. For example, the Foundation recently purchased a couple of the latest generation Lenovo X1 Carbon laptops and sponsored work to make sure that peripherals are supported out-of-the-box.
Q: Are there any plans to make it easier to install FreeBSD as a desktop system? The current focus seems to be on servers.
A:
The Foundation is supporting efforts to make sure FreeBSD works on the latest hardware and peripherals that appear in desktop systems, and will continue to support making FreeBSD easy to deploy, monitor, and configure to provide a great toolbox for building a desktop on top of it. [...]"

So I'd say if we are a little bit insistent about willing to help improving FreeBSD towards a better desktop experience, we can reference these words. Of course, that would imply to have the neccessary skills, or at least the will to improve these.
I wasn't trying to be disrespectful toward you or your proposal earlier and am sorry if you took it that way. I could have got the same point across sans poking fun at the flock, but I was on a roll and not sorry for that.

However, I worked hard, and struggled harder than I would have had to if I had read the Handbook, and taught myself to use FreeBSD through trial, error, laughable stupidity and dogged determination.

So I wrote a Beginners Tutorial that is posted here for people who had never used the commandline and possessed basic English Language skills should be able to follow and get a fully functional FreeBSD using x11-wm/fluxbox and all the doodads I want on mine to be waiting for me there at first boot.

For the large part most people have been successful over the past 2-3 years. The Tutorial was featured in freebsdnews.com with a cover screenshot of ILUXA's desktop. Twice actually, the first time under my bots name Siseneg and their article was picked up by the English and Arabic Facebook pages of bsdmag.com. Happy Happy Joy Joy for everybody.

It's actually a glorified Task Analysis and I've written them for such mundane things as making a peanut butter and jelly sandwich in excruciating detail for every step from start to finish including materials needed and taking into account if the person learning the skill is right or left handed.

Not for the people learning the skill. If they can't make a peanut butter and jelly sandwich chances are they can't read the TA. For the people teaching it so they could teach the skill in a consistent manner.


If I could have made it any easier I continue to try by thinking of ways to make it happen. I just updated it so people with older nVidia chips would know what to do others don't who use a different GPU. I'm happy with my Win7 Era Thinkpads and that's all I run but a Gateway/Acer clone from the same Era.

I get the same results every time and know in advance what I'm going to end up with. Though I had to step in and pick up where ports-mgmt/portmaster failed on a different program in 3 of 4 builds this time, once I manually installed it portmaster picked up where I left off and on we went.

I used make install clean to install the problematic programs just like they installed portmaster but would lack the skill to figure out what needed done so it could continue. People with newer machines might encounter issues I don't cover, because I don't experience them. There are always people willing to answer questions and have gracefully answered them for me in my absence.

Thank you very much.
 
I wasn't trying to be disrespectful toward you or your proposal earlier and am sorry if you took it that way. [...]
I didn't read it that way. Despite the fact that you tell newbies to use ports-mgmt/portmaster, which you must not do (and I'm shure you know why), I appreciate your efforts. Even if I think a "full" DE like KDE or XfCE meets a noob's demands better than x11-wm/fluxbox, it might be ok for those who want to learn some nerdy stuff.
 
my 2 cents,the idea of make more documentation,howtos,tutorials,
maybe a wiki,etc
based in particular tasks,hardware
are good
but without interfering or modifying the oficial handbooks or resources

why? , beyond that the user who create or add a chapter for example,the oficial handbook
or wiki
today is one,tomorrow are more and more,until the original "spirit"
and methods of the mantainers of FreeBSD are buried and all become bloated
I mean this because like many users I love the simplicity of the comunity and their sources/documentation/way to make things

so,is a very good idea,but outside of official sources,the mantainers had their (closer,not bad) actitude for one reason and for my that reason make FreeBSD not follow the actual tendencies or errors like "lets make a super desktop" "change the init system" "a have a
unstable system in cost of beauty and make everything easy for the new user"

note: I refer "users" for experienced users that can contribute with very good things
like this forum and mailing lists
 
Back
Top