It is very tortuous install FreeBSD desktop

Status
Not open for further replies.
But it's the core committers (and somewhat the Foundation) that influence the direction of the Project. By all means, use what you want to use. I'm not suggesting not to use FreeBSD as a desktop. In fact, I wouldn't mind if the Lumina Project in particular received a little more love and support from the greater community. I'm just giving credence to why things have been the way they are. People who lurk the forums and complain (and don't do anything either) aren't seeing the forest for the trees. For example, just look at all the Foundation sponsored projects within the last 5 years; very few of them has to do with improving FreeBSD as a desktop.

If you truly want an open source desktop, for the sake of it being open source; you're better off (hate to say it) with Linux. GNU owns the open source desktop, no matter how you slice it. If you want a good BSD Unix desktop, such one already exists.

I'm not trying to be an ass here, it's just annoying seeing "FREEBSD SUCKS! WHY NO WORK ON DESKTOP?!"-like posts when the situation isn't just "Linux and vendors focusing on Linux only". It's simply not a focus, and for good reason.

It would be nice to see the Foundation show a little bit more interest in the desktop area, at the very least, graphics support. Given the large difference in corporate involvement, developer, and monetary resources between the FreeBSD project and Linux though, I can understand to a certain extent. After all, you need to make the paying customers happy first, and by paying customers, I mean the companies who donate both through code as well as monetary donations to the project that help keep it going. This is of course how I've understood things and certainly not an official position of the project or Foundation in any way. Work is being done on the desktop though, albeit slower than some would like. Anyone that can help is encouraged to get involved. :)

I can only speak for myself, but I'm not an open source purist. I merely want a good operating system that has good documentation, allows me access to change any part of it as I see fit, and use it how I wish. For me, FreeBSD meets that criteria. Linux is also an option, but changes too much and too often for my liking and OS X is locked down tighter than Edward Snowden's laptop besides Apple's hardware being much too expensive for my taste.

Keep in mind everyone was green once and posts on forums now and again that voice frustrations and the like are to be expected in any similar project. Linux forums have their share, if not more, of these posts as well. It can be quite frustrating to spend your weekend trying to get an operating system installed on your laptop or server only to find out after two days of constant reading and tinkering that your hardware is just not compatible. It's always a smart move to match the hardware to the OS and not the other way around if you can. This goes for server and desktop use alike. Windows is really the only operating system exempt from this as for the most part nearly all consumer hardware, barring some hardware in the mobile space, is supported by Microsoft.
 
A lot of GTK dependencies are redundant, and unnecessarily bloated, because a Linux style wants to be kept in ports. Qt dependencies seem to be a lot cleaner. There is a lot of good in GNU and BSD licenses together, and they are both beneficially mutual. I wonder, what if BSD makes the leap, then the next obvious code is copyrighted by a more restrictive license based on FreeBSD's efforts, but alternatively, there may be an infinite amount of ways to achieve the same result. I think the base system of FreeBSD is efficient, with a few minor exceptions, so this goes a long way into desktops even if they are full of dependency madness. Many may hate to hear this, but I think the ports tree itself needs a fork, for those who insist on using Linux style, dependencies or scripts without having a BSD/MIT default when it exists. GNU is perfectly good, but I see some of its related sub-projects as bloated, when many times as much ports with their full features have to be installed to get just 1 library dependency or feature. When this isn't the case, I can see why a port wants certain features, because it fits a niche. This is related to the desktop environment, because this is mostly about graphical tool-kits and desktop related dependencies.

It would be great if Wayland Thread wayland-moves-to-mit-licence.26502 were brought in, not as the base system, but readily available. Then Xorg optionally be ported on top of it with less hardware bloat. It makes no sense to bring a display server into a base system, since many servers are meant to be set up and left to do their work, without a direct display.
 
A lot of GTK dependencies are redundant, and unnecessarily bloated, because a Linux style wants to be kept in ports. Qt dependencies seem to be a lot cleaner. There is a lot of good in GNU and BSD licenses together, and they are both beneficially mutual. I wonder, what if BSD makes the leap, then the next obvious code is copyrighted by a more restrictive license based on FreeBSD's efforts, but alternatively, there may be an infinite amount of ways to achieve the same result. I think the base system of FreeBSD is efficient, with a few minor exceptions, so this goes a long way into desktops even if they are full of dependency madness. Many may hate to hear this, but I think the ports tree itself needs a fork, for those who insist on using Linux style, dependencies or scripts without having a BSD/MIT default when it exists. GNU is perfectly good, but I see some of its related sub-projects as bloated, when many times as much ports with their full features have to be installed to get just 1 library dependency or feature. When this isn't the case, I can see why a port wants certain features, because it fits a niche. This is related to the desktop environment, because this is mostly about graphical tool-kits and desktop related dependencies.

It would be great if Wayland Thread wayland-moves-to-mit-licence.26502 were brought in, not as the base system, but readily available. Then Xorg optionally be ported on top of it with less hardware bloat. It makes no sense to bring a display server into a base system, since many servers are meant to be set up and left to do their work, without a direct display.

Wayland is a good start. That alone would convince me that they care.
 
It is very tortuous install FreeBSD desktop, you can not install some applications like Skype or Flash player support nor in Linux. In 2008 there was the DesktopBSD system based on FreeBSD, it was very comprehensive. Hopefully FreeBSD extends to GUI, the other derivatives are so incomplete. :)

How torturous is pkg install xorg mate mate-desktop...

I would consider lack of Flash to be a feature of FreeBSD.
 
How torturous is pkg install xorg mate mate-desktop...

I would consider lack of Flash to be a feature of FreeBSD.
I think you have to consider who your user is. Sure, you run pkg install xorg mate mate-desktop, but after that you have to change configuration files. And Flash is still used by many sites. I personally find Flash still helpful. And people also use things like Skype. The availability of applications can be important.
I am not sure FreeBSD is intended for the "average" user, and I don't know to what extent FreeBSD derived systems solve the problem, but to have a leap in terms of user base it would be nice to solve it, like Linux did. That would also mean, among other things, better hardware support.

Personally I am usually fine with FreeBSD since I stopped using ports, and there are a few reasons why I would prefer it, the last one being systemd morphing into a system of its own (add this to a better licensing model, a less chaotic distribution landscape, and a venerable history). But I am still using mostly my Debian testing exactly because of its wider hardware and software support.
 
I am not sure FreeBSD is intended for the "average" user, and I don't know to what extent FreeBSD derived systems solve the problem, but to have a leap in terms of user base it would be nice to solve it, like Linux did. That would also mean, among other things, better hardware support.

FreeBSD is hardly a good choise for an "average" user who expects those bits to be available out of the box. Both flash and skype are going to be a lost cause very quickly unless something changes drastically, not on FreeBSD's end but in the way the vendors of those pieces of software mentioned treat the more obscure platforms. Even on Linux flash is very much dying because it's completely closed source and Adobe killed the NPAPI plugin version of flash. Skype is now in the hands of Microsoft and they have zero interest in supporting Linux.
 
I am not sure FreeBSD is intended for the "average" user
FreeBSD is a professional operating system for professionals. It's target is not Windows users as Linux has turned into.

I am still using mostly my Debian testing exactly because of its wider hardware and software support.
FreeBSD's hardware support is mostly not an issue except for one networking vendor and one graphics vendor. Software support generally is not an issue either. While one can cherry pick things that are missing on FreeBSD, most everything people use on Linux is also on FreeBSD.

Whenever I read someone complaining about such things, it's cause Linux supports 10,000 variations of something while FreeBSD only supports 9624 of the same category.
 
FreeBSD is hardly a good choise for an "average" user who expects those bits to be available out of the box.
Okay, so this also means that when someone says that it is very tortuous to install FreeBSD the only answer is "learn your way or choose another system". It is a respectable answer, although maybe not what I was hoping for, unless the other system is PC-BSD or equivalent. Too bad.

FreeBSD's hardware support is mostly not an issue except for one networking vendor and one graphics vendor. Software support generally is not an issue either. While one can cherry pick things that are missing on FreeBSD, most everything people use on Linux is also on FreeBSD.
Well, the thing is that this is not an issue for you. I personally didn't install FreeBSD 9 because it had a problem with USB 3. And many people would like to use Skype. A friend of mine doesn't even use Linux, because he needs AutoCAD. He's stuck with Windows.
As I said, you can say this is the choice FreeBSD is taking. Fine, I think I can live with that. I'll say "too bad" and I'll walk away.
 
Even on Linux flash is very much dying because it's completely closed source and Adobe killed the NPAPI plugin version of flash.

Well, even on [insert platform here] Flash is effectively dead. It still exists, but nobody worth a damn is actually developing new websites or web applications with it.

As for Skype, every time I see someone chalk its absence up as a loss for *nix, I can't help but consider that I've never met anyone in meatspace who's actually used it.
 
Well, even on [insert platform here] Flash is effectively dead. It still exists, but nobody worth a damn is actually developing new websites or web applications with it.
Flash has been pretty much dead everywhere for a long time. The problem is that Flash doesn't know it. It's not about new websites or web applications, but old ones. There are still many of them.

As for Skype, every time I see someone chalk its absence up as a loss for *nix, I can't help but consider that I've never met anyone in meatspace who's actually used it.
Unfortunately I can't found really up to date figures, but this should give an idea:

http://blogs.wsj.com/digits/2014/01/15/skypes-incredible-rise-in-one-image/

I have in this moment 14 online contacts on Skype, even if many are afk. I don't really trust this information, to be honest, but again, just to give an idea.

And those are just examples, only part of the whole issue. Anyway, I reread the thread a bit, and I saw quite a few posts making the point apparently better than me (I would like to add videogames to the bunch, but maybe it's just me; which actually is surprising). I feel like I'm beating a dead horse; I'll see if I can refrain from posting in the future, unless those posts are brought into context.
 
As I said, you can say this is the choice FreeBSD is taking.
You are mistaken. Not having those is not a choice on FreeBSD's part. Some of it is the lack of effort on the part of the software or hardware vendor. Others are the lack of people available to work on such things. In both cases, complaining doesn't do anything but reiterate what we already know. Becoming proficient enough to contribute is a far better thing to do.
 
You are mistaken. Not having those is not a choice on FreeBSD's part. Some of it is the lack of effort on the part of the software or hardware vendor. Others are the lack of people available to work on such things. In both cases, complaining doesn't do anything but reiterate what we already know. Becoming proficient enough to contribute is a far better thing to do.
Ah, okay. I thought there was more a general desire to keep things "user unfriendly", because it's always possible to hack the system somehow to solve whatever problem.
This is very understandable; it is unthinkable to pretend everything is solved overnight. I saw Linux for the first time in 1992, but I waited about 13 years to use it as main system. These things go slowly.
In that case I will be happy to stay. Who knows, if I get into it enough I might contribute. It's probably a very steep learning curve though.
 
Okay, so this also means that when someone says that it is very tortuous to install FreeBSD the only answer is "learn your way or choose another system". It is a respectable answer, although maybe not what I was hoping for, unless the other system is PC-BSD or equivalent. Too bad.


Well, the thing is that this is not an issue for you. I personally didn't install FreeBSD 9 because it had a problem with USB 3. And many people would like to use Skype. A friend of mine doesn't even use Linux, because he needs AutoCAD. He's stuck with Windows.
As I said, you can say this is the choice FreeBSD is taking. Fine, I think I can live with that. I'll say "too bad" and I'll walk away.

Well I agree with your AutoCAD statement. I use it and Solidworks and must maintain a Windows box as there are no viable alternatives...
 
I am surprised how small an install my FreeBSD desktop is. I am have set up Mate, web-browser, email and Anjuta and Glade all with a 5Gig Install (Sources and Ports Too). I am using an old 16GB SSD.

I was daunted at first after being dumped to a command prompt, after the DVD install, but after trying PC-BSD, I decided to hit the books and figure it out a FreeBSD desktop. 2 small file mods are a small price to pay. It helped that I am an pfSense user and have a motivating factor to learn more so I can contribute back.
 
I'm guessing half the effort of learning MS-Windows since you won't need a law degree. :)
I meant more in terms of knowledge of the code, not to mention the commitment. I don't think you can just say "I'll write two lines of code". Diving into the existing code is probably not easy, especially if you want to do things like Linux compatibility, which to me is quite important.
 
My first FreeBSD was version 6.?? (I forgot) and I have to used Fluxbox and KDE. Later with version 10 I installed FreeBSD on my iMac and use Fluxbox, Lumina, KDE and later remove KDE and installed GNOME 3. And as a not computer educated person I didn't have problem with installation. And I use FreeBSD as a desktop from version 6 (before I had OS/2 and Linux).
 
Diving into the existing code is probably not easy, especially if you want to do things like Linux compatibility...

Yeah, I'm a perpetual novice when it comes to coding, but I'll bet Linux compatibility would be the among the most confusing, if not the most confusing, parts of the codebase to actually get involved with (short of perhaps Sendmail :p). *BSD devs tend to pride themselves on code cleanliness--see style(9), for instance--but if your focus is on a very specialized layer of abstraction, then you'll probably hurt yourself in the long run if you don't look at the rest of the code before jumping head-first into the more complex, obscure, possibly less-used parts.
 
True enough, but merely being conscientious about code and documentation standards counts for something, and code that's easy to read and patch into a larger base is easier to maintain. But now I guess I've dragged this off-topic...
 
Yeah, I'm a perpetual novice when it comes to coding, but I'll bet Linux compatibility would be the among the most confusing, if not the most confusing, parts of the codebase to actually get involved with (short of perhaps Sendmail :p). *BSD devs tend to pride themselves on code cleanliness--see style(9), for instance--but if your focus is on a very specialized layer of abstraction, then you'll probably hurt yourself in the long run if you don't look at the rest of the code before jumping head-first into the more complex, obscure, possibly less-used parts.
Absolutely. I never get those people who just write code without even looking at the consequences. Perfect way to implement bugs. Hackers. I need the bigger picture.

True enough, but merely being conscientious about code and documentation standards counts for something, and code that's easy to read and patch into a larger base is easier to maintain. But now I guess I've dragged this off-topic...
Maybe a bit, but it's worth it :)
Documentation is very important, and yet too often overlooked. More than the standard is the content that matters; you can have a great standard and terrible documentation. It's about an important aspect of code quality: readability and understandability. Having people you can ask helps too, but in that case I wouldn't say the code is good.
 
It would be nice to see the Foundation show a little bit more interest in the desktop area, at the very least, graphics support. Given the large difference in corporate involvement, developer, and monetary resources between the FreeBSD project and Linux though,

Work is being done on the desktop though, albeit slower than some would like.

FreeBSD to lost a lot of ground to other systems, betting only to servers and that has a quota low in the world for their service. Because follow pawned in the old rhetoric of other systems with desktop? Even OS X does not give eating to FreeBSD, because it is a closed source system, and depends on a minimum of BSD.

Therefore the FreeBSD system, had constant problems of financing, and the earnings for the service of the servers have been insufficient. What strikes me, is that FreeBSD can not evolve to default for graphical desktop, as did Debian along with their derivatives.
 
"What strikes me, is that FreeBSD can not evolve to default for graphical desktop"

I really don't care about that. If I want a good desktop graphic environment I use Windows (eventually Mac OSX), I really don't care about using Linux or BSD just for the pleasure of aping what Windows does very well. I am not a *nix fascist, all systems have their own qualities and especially since Windows 7. Personally I have never had serious problem with virus and spywares under Windows because I am a clever guy and I know what I am doing. Windows or not WIindows this is not the question to me.

I can use Linux/BSD for desktop in some particular situations :
- Money is a matter for the user, he can't afford to pay.

- The hardware platform is quite old and last version of Windows doesn't support it very well.
For example I have collected an old Pentium 4 Prescott x86_64 platform. First, I must switch to Win 10 x32 because Win 10 x64 needs now the virtualization instruction (you know that Windows 64 is quite different of Windows 32, it uses the kernel of Windows server), second "Wintel" dropped the support of the integrated i915 graphic chipset so Windows 10 is miserably displaying under the generic and slow Microsoft VGA card driver with No possibility to add a graphic card... so Linux/BSD is a magic solution here. The i915 is still well supported and I can install the x64 version, the lack of virtualization instruction won't stop the installation.. and all is working dramatically well. A Pentium 4 6xx has still enough power to do basic things as browsing, playing music and video, this is surprising.

In this case I install Linux for its better ability to support various hardware. I know that automount of usb key is working so the client won't come back to me with some disatisfaction. Making automount working under BSD is something terrible, I have never succeed in... and i think I am not the only one.

But in some cases I will choose FreeBSD in the following situations
- a lack of memory, BSD is incredibly good in the memory management
- embedded systems through pico BSD

I use FreeBSD for what it is tailored to : server. It is still great to have a graphic help, I am not crazy I won't refuse that, but I don't need here something perfect as I want to for my regular desktop computer.

Complaining here for a lack of graphic environment is useless and is a nonsens because FreeBSD is very clear on this point : developers don't care about ease of use, about graphic aesthetic, they work on the core system to implement full stability.

But in fact you have misunderstood something : FREEBSD ALREADY LISTENS TO YOU.
Desktop environment is a PC-BSD's concern. PC-BSD is not a fork of FreeBSD, this is a complementary project. This means that FreeBSD will never get involved in graphic desktop environment because this is the job of the PC-BSD staff. They have done a great job, the graphic installer works nearly as well as the Mageia installer. All is not perfect, there is a lot of work to do... but they have recently launched the Lumina DE desktop environment project. This project is just answering to your question : a graphic interface perfectly tailored for BSD that would replace KDE (which seems to get closer to Linux), the horrible GNOME, a graphic interface fullfilled of "linuxism" which raises many issues.

Many people just install PC-BSD to get inspiration regarding some strange settings and switch back to FreeBSD to apply what they learned... because FreeBSD gives the incredible "flexibility" to compile the way you want, to choose and discard the components you want to install or not. You must choose : you want something out of the box... ok go to Linux of PC-BSD, you want the flexibility... stop complaining, flexibility means that you need to challenge all kind of issue.

But work is still on... the Lumina project is quite "young".
 
When I say "desktop" I mean something like "does run the usual GUI stuff, can run Office software, GIMP, browsers, etc." In other words: I come from the server and/or console side and "desktop" for me is "GUI stuff on top. FreeBSD can do that, too. Cool!"

I think you hit on something important there - input method. It would seem that some people insist on using a mouse and tick boxes because to them it is comfortable and not because it is either useful or required. I learnt to type very many years ago and find it infinitely easier than using a mouse. My tired eyes are just not up to the task of searching all over a screen for some clue about a relevant clickable item any more. I even still use DOS because it is so damned comfortable and does now what it did when I first started using it. That is perhaps why when I talk about a "desktop" in my world it is only for the purpose of doing graphical things - i.e. things that involve pictures. Like you say, Gimp, Office, and browsers. If something can be done by plain typing, then of course (to me) that is the way to go.

So, I guess that you and I think of a "desktop" as kind of a hybrid affair, whereas there are others who religiously avoid typing commands. I recently had the fortune of being asked to do some file management on a Windows 7 computer owned by someone who has used MS-Windows for many years but has yet to understand about files and directories. It was a steep learning curve for me and a frustrating experience because I had to keep doing web searches for things like "how to get to a terminal", and "how to open two file browsers", etc. In the end I got the work done, but not without screaming (silently) that Windows was not ready for the desktop. That, of course, is because "desktop" means something different to me.
 
Status
Not open for further replies.
Back
Top