Is there a real interest in pushing FreeBSD on the desktop space?

[getopt]

Anyway I believe to make everyone happy the foundation had better to ...

You cannot "make everyone happy". Nobody can do that.
Foundations serve a purpose. It is not to make someone "happy".

 

[ucsdboy]

GhostBSD is nice but it is a one man vision project, perhaps a desktop spinoff made by the FreeBSD foundation would be more in tune with FreeBSD values and goals, without even mentioning that you can't ask for support for anything else but FreeBSD itself.

People may have different expectations and different reasons to use FreeBSD instead of Linux for instance, if those do not align with majority of the FreeBSD pro and power users doesn't mean that must ignored, in fact in that video one point that is mentioned is to lower down barriers for people that come from other experiences, and I believe this is good, for all.

Anyway I believe to make everyone happy the foundation had better to create a separate project, so nobody will feel hurt cause its toy has been touched. FreeBSD will continue to be the same and the other project would be whatever the foundation would consider the right approach.

I can appreciate what you're saying, but it's time to point out that it's very vague, contains no actionable items, and is basically you telling me that a third party (Foundation) needs to tell a fourth party (Core team) to 'make something' happen. This is almost certainly the least effective way to make what you want to see happen.

I can only speak for myself but I think you're better off writing to the Foundation yourself and addressing your concerns to them.

This may be the 'Forums', but IMHO, I don't usually see it as a place where people just 'hang out' a la Clerks and wax philosophical about 'what's wrong with FreeBSD."

As far as 'one man vision' projects are concerned, I'd hazard that many of the more successful projects on the internet started out that way including Unix itself, the C language, Linux, Emacs, Vi, BSD itself. In recent years Git, Rust, are just a few. How many successful projects were created by committee? The C++ standard? Multics? the OSI model? GNU/Hurd? Taligent?

 

[LibreQuest]

Don't forget about MidnightBSD. It's literally made to contend with a standard Linux distro. It just may be the right place for those who are wishing they could make FreeBSD into a typical Linux distro experience.

 

[rmomota]

There are plenty of Desktop Environments and Window Managers to make whatever desktop one needs with FreeBSD as the OS. (here it runs well, several years unchanged, only updated).

Main challenge seems to be for those who want something out of the box like most Linuxes give, Wadnows seems to give. Pre-set 'freedom of choice', 'enhanced user experience', however within pre-set boundaries -- title bar colour, y'r own wallpaper (wow-factor!), taskbar bottom or disappearing, that kind of superficial stuff. I forgot 'dark mode' as a huge asset...

Most illustrative examples IMHO are questions like 'what distro does Linus use' (it must be best), and 'I want to listen music, what is the best <fill in>'. Probably the lot doesn't realize that they can change DE or WM with a few keystrokes, install all Kali software even in Hannah Montana Linux (and have the pink UI).

Here choice is yours, which is difficult in this world of brands, fanboys and home-made-pancakes-that-just-to-have-to-be-warmed-in-the-microwave.

Choice can be personal, the rest can be copied.

Distro's with commercial interests might have the incentive to serve to those needs to attract and keep customers. FreeBSD however isn't for the intellectually lazy.
You'll have to try, crash and reboot. Such is Life

I loved that Hannah Montana Linux part

 

[bakul]

Is there a real interest in pushing FreeBSD on the desktop space?
I have the impression the answer is: not really!

Here's how I would approach this if I was sufficiently interested in this:

1. *Define* what this really means. The phrase "Desktop space" is too vague and means different things to different people so it is not actionable. The more specific & precise the definition, the more likely people can help.
2. Right here one can spend a bunch of time and iterations/refinements to come up with a definition that is good enough and doable by a small team. This would also be a good point to ask if this is really what I want, estimate time and see if I can commit to doing this. Also figure out *why* you want to do this and for what user base.
3. Avoid getting distracted by all the available choices as well as by armchair experts and random people who will opine of everything. That is not going to change, nor can one control it, so best to focus on what is under my control and on the project itself.
4. Formulate an initial plan. Try to organize things so that it can be done in small chunks. Create an initial list of tasks. Create a github repo and populate it with the plan, task list and any notes made during the initial stages.
5. Tell the world what I am doing. Not wait for anyone's permission or blessings or participation. I'd just start!
6. If the project is sufficiently interesting and people can figure out how they can help, they will want to join. So I'd be open to such participation and willing to even switch from writing code to helping others.

Many projects start out this way, where *one* person does enough interesting work which then draws other people in.

 

[Beastie7]

Seeing as how a (project sponsored) desktop build of FreeBSD is already in the wild, and the committers seem keen on sticking to it as a development environment; I think an official release of desktop FreeBSD is imminent. If that happens; I'll have my bucket here ready for all of your tears. Qt/KDE as a main desktop focus for whatever development use case... it's only a plus IMO. Note this won't stop you from building your own environment as I'm sure a vanilla ISO will still be provided; so all of the mUh cHoiCe detractors can stop being paranoid.

 

[Sevendogsbsd]

My take on things is normally not in line with others so I’ll just say it. FreeBSD has no desktop, in terms of the core project anyway. All of the desktops/window managers are third party. Not sure why the foundation has to worry about that. Linux is completely different: there is no “base OS” and third party apps: it’s all rolled together. Not really a comparison there.

FreeBSD makes a great desktop but it takes a little work. I don’t feel that work is the responsibility of the foundation. My opinion.

 

[LibreQuest]

Seeing as how a (project sponsored) desktop build of FreeBSD is already in the wild, and the committers seem keen on sticking to it as a development environment; I think an official release of desktop FreeBSD is imminent. If that happens; I'll have my bucket here ready for all of your tears. Qt/KDE as a main desktop focus for whatever development use case... it's only a plus IMO. Note this won't stop you from building your own environment as I'm sure a vanilla ISO will still be provided; so all of the mUh cHoiCe detractors can stop being paranoid.

Tears of joy in a bucket is a strange analogy. Whatever the devs decide to do I'm sure will be done with great care to detail and testing. That would be very interesting to see.

I would be curious to see the adoption rate of FreeBSD users and devs of a system with a default gui or GUI based installer.

The first Linux I ever used, Mandrake 7.2 Complete, came with an installer that would ask you what you wanted to install. It had categories of software , in a gui installer, and you could choose one desktop or all of them. If there were an official gui based FreeBSD I would expect to see a similar approach of not limiting the users options in regard to the gui.

I don't feel that there are imposed restrictions with the current system and I appreciate that very much.

 

[richardtoohey2]

FreeBSD makes a great desktop but it takes a little work. I don’t feel that work is the responsibility of the foundation.

I don't use FreeBSD for desktop work, but I think what is being asked is more FreeBSD support for desktop *hardware*.

So not so much the window managers but the plethora of Wifi options, the USB devices, Bluetooth, Thunderbolt, (multiple-)displays, GPUs, DRM, etc., etc., etc. that underpin a "rich" GUI experience.

But personally if I want all that I'll fire up a Mac or Windows or a shiny Linux desktop or an iOS tablet where most stuff will work "out the box" (even though a part of me dislikes me for giving into the walled gardens.)

 

[Alexander88207]

Is there a real interest in pushing FreeBSD on the desktop space?​

I would say, in some areas more in some areas less.

 

[hruodr]

Many projects start out this way, where *one* person does enough interesting work which then draws other people in.

It would be wonderful if those that complain of FreeBSD on Desktop make a new project,
but that is not what they want. They want to make FreeBSD as one Desktop "OS" more,
although there are a lot of Linux and even FreeBSD variants for "Desktop". That is why all these
discussions are so ugly. They come and tell stories about "Desktop", it is not even clear what that mean
and why is the system one is using for many years or decades as Desktop is supposedly a bad Desktop.
They come, tell stories, criticize, but never something concrete, never ask for help for dealing with the
supposedly insufficient system (what for me is the purpose of being in the forum).

 

[bakul]

They come and tell stories about "Desktop", it is not even clear what that mean
and why is the system one is using for many years or decades as Desktop is supposedly a bad Desktop.
They come, tell stories, criticize, but never something concrete, never ask for help for dealing with the
supposedly insufficient system

To complain about something is human nature. It doesn’t bother me! Better to try to move the discussion to doing something useful or at least interesting. Encouraging new users in a constructive way can be useful in the long term.

In this particular case, just thinking through and clarifying their own ideas about a desktop freebsd can be a valuable exercise.

 

[hruodr]

but I think what is being asked is more FreeBSD support for desktop *hardware*.

Drivers are main components of an operating system. Drivers for desktop may be important, but other
drivers are also necessary, and there is a finite numbers of developers. I think, it is done, what can be
done. If they mean drivers for desktop, then they may perhaps collaborate developing them. But in reality
we do not even know what they want, and perhaps they also do not know.

 

[drhowarddrfine]

a (project sponsored) desktop build of FreeBSD

I don't see where FreeBSD sponsors this.

 

[getopt]

Drivers are main components of an operating system. Drivers for desktop may be important, but other drivers are also necessary, and there is a finite numbers of developers.

Looks like this thread burned it’s way to the core problem. It’s about drivers and about allocating resources (financial,human) for making device drivers for FreeBSD/all the BSDs/Linux.

Drivers are a bumpy road for any open source OS just because they are still not made open source by hardware vendors. We all know what that means in terms of human resources and knowledge required when it comes to hack this.

How problematic drivers can be in terms of OS security can been seen when looking at MS-Windows when proprietary drivers get hacked. FreeBSD must be careful in respect to driver security.

I’m not so sure how the FreeBSD Foundation is doing when it comes to collaboration with hardware vendors. Some may sponsor FreeBSD if it is in their interest, but there are other roads too.

It’s the Foundation’s job to make vendors learn that they can improve their business if they also engage with FreeBSD as markets can change their directions. There is a lot of room for improvement and I frankly cannot see much efforts, may be because of lack of transparency in communications.

Praising all the time how great we are doing is causing cognitive dissonances when a need for improvement is obvious, isn’t it?

 

[kpedersen]

Seeing as how a (project sponsored) desktop build of FreeBSD is already in the wild

The sponsored port of FreeBSD to CHERI is *not* a desktop build but just like any FreeBSD, it *can* run a desktop; but this is far from the goal of exploring 128-bit CHERI architecture.

But as you can see, just because desktop isn't a specific focus, it doesn't mean developers are disallowed from doing so anyway! Keeping FreeBSD flexible and "not focused on one thing" means everyone is happy. Sysadmins, embedded, developers, users, etc.

I think one thought exercise is if FreeBSD did go the desktop route and chose to focus on CDE. That wouldn't really please you would it? (you prefer KDE?). It would end up trying to add unnecessary Motif cruft to your install since you will be installing KDE anyway. This kind of feeling is what many people who don't like full desktop environments get if FreeBSD focused on any DE.

 

[freezr]

You cannot "make everyone happy". Nobody can do that.
Foundations serve a purpose. It is not to make someone "happy".

I believe that sometimes it is smart do not taking the meaning literally...

 

[Beastie7]

But as you can see, just because desktop isn't a specific focus, it doesn't mean developers are disallowed from doing so anyway! Keeping FreeBSD flexible and "not focused on one thing" means everyone is happy. Sysadmins, embedded, developers, users, etc.

I understand this, and I agree; hence the last part of my previous post. Go back to the dev summit video and listen to what the committers were saying. The goal was to lessen the barrier to entry for users; for any type of development use case. Providing a desktop is simply making the experience of getting things up and running less complicated. Think of universities, schools, offices, industrial environments, SMEs, etc.

I think one thought exercise is if FreeBSD did go the desktop route and chose to focus on CDE. That wouldn't really please you would it? (you prefer KDE?). It would end up trying to add unnecessary Motif cruft to your install since you will be installing KDE anyway.

CDE? Blasphemy.

 

[kpedersen]

The goal was to lessen the barrier to entry for users; for any type of development use case. Providing a desktop is simply making the experience of getting things up and running less complicated.

I think this may be where we fundamentally disagree. I find desktops often get in the way (think automounting before a dd, opaque errors, it is so difficult to help a newbie diagnose Xorg issues if i.e gdm crashes and they have no idea why). So much development and tools are CLI only that the face validity of "DE makes things easy", is not always quite so straight forward. But yes, this is just my opinion from one side of the camp.

But our opinions aside, I don't feel it is even actionable. The foundation would have to start considering getting Xorg in base which alone is probably going to be a decades worth of work to clean up. The modesetting drivers were even removed from base, so it seems like we are going the opposite way.

If it just means that they will sponsor someone to work on the gnome3 port, then certainly that is fine. No complaints from me. Actually, if they instead decided to rip out the Gnome 3+ or KDE 4+ ports and instead wanted to focus on (and potentially fork) Gnome 2, KDE 3.5 or any other viable DE, I would even help out!

 

[mer]

I think this may be where we fundamentally disagree. I find desktops often get in the way (think automounting before a dd, opaque errors, it is so difficult to help a newbie diagnose Xorg issues if i.e gdm crashes and they have no idea why).

I agree with this 100%. Most people get by just fine with a window manager and running startx.
CDE would be a fine option in my opinion. Automounting: never liked it.

 

[Jose]

GhostBSD is nice but it is a one man vision project, perhaps a desktop spinoff made by the FreeBSD foundation...

Anyway I believe to make everyone happy the foundation had better to create a separate project...

You seem to think the Foundation should care about your ideas. There's a super-easy way to get them to care. Donate $1 million dollars, and they'll start working on whatever "desktop" experience you want.

Doesn't even have to be your money, you could organize a fund-raising drive, contact wealthy benefactors, etc.

 

[vienuolis]

Sorry, I did not quite catch your desire:

The goal was to lessen the barrier to entry for users; […] Providing a desktop is simply making the experience of getting things up and running less complicated.

But that is what NomadBSD is for:

NomadBSD is a persistent live system for USB flash drives, based on FreeBSD®. Together with automatic hardware detection and setup, it is configured to be used as a desktop system that works out of the box, but can also be used for data recovery, for educational purposes, or to test FreeBSD®'s hardware compatibility.

I am new to the Unix environment, a refugee from MS Windows, but I installed NomadBSD on my own without anyone's help (except for handbook, manual, and forum). First on a USB stick, then on my home computer, and finally I started my own FreeBSD server. I have never had to go back to MS Windows, and I still have not missed it at all.

 

[Alex Seitsinger]

gnome & kde are bloated. That's why some linux distro's only come in xfce4 or lxde version.

Real men use i3 and such. Be a real man. Do it.

I really do not understand this recurring theme "desktop".

I use FreeBSD as Desktop, I always used as Desktop a BSD version. If you do not want it, use as desktop
what you like, linux, windows, mac.

Does Desktop mean gaming? Then perhaps a game console is what you need. Then go playing and leave us
in peace.

I think the idea that most of us eventually arrive at begs the question "why isn't this widely distributed?" to which I say, "who cares".

I guess you're pointing out that FreeBSD is lean by default, and you've got to enable stuff by your own. Well, for large scale deployment that's not such a big thing IMHO, because people would use stuff like Ansible anyway.

What could more of a blocker for some people putting FreeBSD on a server is the lack of Docker; then again for some this might be an advantage.

Free also as in be careful, some safe defaults may also be unimplemented.

 

[astyle]

My comparison is gonna have holes in it, but still:

Comparing FreeBSD to Microsoft/Apple is a little like like comparing these:

• A plane built by rank-and-file community members
• Something built by Boeing/Airbus.

Rank-and-file community members are not exactly organized, they each want to do their own thing. Even if they somehow manage to put together some quality components (because there are a few smart people among rank-and-file community members), they're not gonna be able to put together something that even gets off the ground, let alone be airworthy. Just looking for the right expertise to get anything done takes forever. Everybody's happy, in spite of not having the results. Best results that rank-and-file community members can hope to achieve is a Cessna.

Boeing/Airbus, on the other hand, have the money and organizational skills to make things actually happen. Yeah, the big companies will work people to the bone, and make decisions that don't make sense to the people assembling the plane. The plane itself will be huge, be awful for the environment, complex to fly, expensive to maintain, and lots of people will be unhappy. But the end result is something like a 747 or an A380.

 

[Alex Seitsinger]

Just looking for the right expertise to get anything done takes forever. Everybody's happy, in spite of not having the results.

Poetic.

Yeah, the big companies will work people to the bone, and make decisions that don't make sense to the people assembling the plane.

Boo. Boo. Boo.

Most things are ruined after they're forced over the cliff of inclusion--all for a dumb buck. "Eventually, you wake up one day, rub your eyes, and realize you're staring at 7 years of driver development for hardware we'll likely never use."

Whatever happened to the days of enthusiastic financiarism and such: The days of numerous large sums of unaccounted-for contributory donations generously offered from the many enthusiastic supporters--who happen to have an eye for talent--seem to be lost to time. Why I remember, all they wanted was the dream that their hardware support would eventually manifest itself out of the ether and into the software they prized so much.