Does Desktop have a future on BSD?

[YYY__]

Philosophy will ruin your lives: If you need more than one reason to do something, you probably shouldn't. Great strength requires great control. ?

 

[Deleted member 30996]

If you don't think it has any future why you are using it and why you are here? Or you are here only to rant and actually not using it? Why don't just go with Linux since it's the ultimate operating system?

I know who he is. I looked back in his posting history.

In Feb. 2019 I was at DistroWatch and some Gentoo penguin squealing in Ecstatic Display, gawked an erroneous statement regarding Firefox built from ports that ended with "been there, done that, got the T-shirt" in reference to his expertise with ports in a ritualistic display to declare ownership of the area typical of the species.

I replied in a post titled "Clothes don't make the man" that if he knew as much about ports as his T-shirt led him to believe he would know all the dependencies were built first and Firefox was last in the build.

The next day fryshke and his penguin pal ekingston followed me here. To test through trial their tater tot troll two-timing teamwork thought trailer trash terrifyingly tough thereby thoroughly tasking The Talker.

Then the titanium-tongue terror Trihexagonal towered tall, treacherously throwing thunderous temporal torrents tastelessly toward two timid tuxies too tired to tango thus turning them to tears, triumphantly taking the tenure that turgid Titan to Tautology tenaciously took to totality.

I will have to give fryshke some credit though. He's the one who gave me the idea of a new way to play this game, which I have since refined and used recently:

fryshke said:
You have mental problems.

Please detail them for me, I'd be most interested in hearing your opinion on my mental state.

And, please, be brutally honest...

 

[kpedersen]

FreeBSD is becoming a framework, a scaffold. You wouldn't write your own OS, so you take framework - FreeBSD - and build upon it. Build software, drivers that support your product and release. Like Playstation 5, like Switch, like iOS, like macOS, etc.

... so you are saying FreeBSD does have a future then?

This seems very similar to how it is now to be fair. You install the base OS (you call this a "framework") and then add desktop stuff built from ports, including drm-kmod/nvidia drivers (you call this "Build software, drivers").

Only the PS5 (and 3,4) uses the FreeBSD OS (i.e CellOS, OrbisOS, etc) . The other examples use only small parts of it (along with GNU components). The Switch apparently uses the same OS as the DS (a much smaller custom OS).

macOS only has remnants of BSD as part of its network layer and BSD subsystem. It is almost like saying that Windows runs on Linux because WSL exists. Or Windows runs on FreeBSD because Winsock is based on Berkeley sockets. It also has more of a GNU userland these days. For example it has Vim, Screen and until recently it used Bash as default too.

I do almost agree with you though. A decent OS will never be suitable as a user-friendly consumer desktop. FreeBSD has very little future in this space because doing so would ruin it. Where our opinions differ is that I believe all open-source consumer centric operating systems will regress to the point of unusability. Making FreeBSD competitive by default. With a recent influx of Linux escapees, we are already starting to see symptoms of this.

 

[fryshke]

It's easy - conspiracy theories. Like, I was on DistroWatch (I was not) and I... followed... you here (I did not) and actually teamed up with someone? Fruitcake now, fruitcake then, noticed and said now, noticed and said then. Easy peasy. )

Edit: Holy shit, even called you paranoid back then - https://forums.freebsd.org/threads/the-tragedy-of-systemd.69621/page-2#post-417990

Hehehehe. Being egoistic is another problem - you think others care about your conspiracies. I posted that and forgot about you after 5 minutes. I did not think about you for 2 years, like, come on, be realistic. You're commenting, I'm responding, that's about it, I don't track random people on the internet, I really don't care, it's just how forums work - someone comments, someone responds, deal with it, or use thicker tinfoil hat.

If you don't think it has any future why you are using it and why you are here? Or you are here only to rant and actually not using it? Why don't just go with Linux since it's the ultimate operating system?

Because hope dies last. I don't want to use Linux. Linux is crazy, FreeBSD is sane. I'd gladly use FreeBSD on servers, but - Docker. It's here. It's quite good. Everyone's using it. I like it. FreeBSD - left behind. .NET - I love it. We have zero issues deploying our .NET software for a couple of years now on Linux. FreeBSD - left behind. On desktop - no Visual Studio Code, no jetbrains IDE support, no Davinci Resolve. Yea, some software might work at the mercy of Linux compatibility. Might break later. I want stable OS, not emulation crap, and if it's running on Linux compatibility anyways - just use Linux and have a calm state of mind that your tools will not break, because Linux compatibility does not ensure 100% Linux software will work. ZFS - like, one of the killer FreeBSD features, also moved primarily to Linux, oof, because, again, not enough people care about FreeBSD.

And that makes me sad. That doesn't make me idealistic and go "ooooh freebsd is the best", it makes me go "freebsd is irrelevant. af". And I express my grief here. But others are in denial, or just simple minded "well everything I need works".

It is time to be educated. I recall recently with the Threadripper 3xxx series processors, Linux having a bear of a time with compatibility, not even booting without tweaks and patches, but FreeBSD worked without any issue or needing any additional configuration right out of the box.

https://www.phoronix.com/scan.php?page=article&item=3990x-freebsd-bsd&num=1

Don't listen to FUD or hype, look at reality. FreeBSD "fell behind" in some areas, but has greatly caught up in most of those. FreeBSD never "fell behind" in other areas and continues to be a premier choice for some applications.

It's time to get educated - ARM is here, every user and tech journalist fucking loves ARM - https://macdailynews.com/2020/12/04/ars-technica-apples-m1-macbook-air-is-hilariously-fast/ it's super efficient with crazy battery life, it's crazy fast (for 4 core cpu). Linux - already working on apple M1 distro and can boot and run. FreeBSD - ARM is tier 2 support. Nice. x86 on desktops is dead to me, I'm not spending any more money on x86, ARM is so much better now than x86, FreeBSD, again - ARM is tier 2.

If you are right, I hope you agree that it's a sad/bad thing that our platform choices are getting more and more limited.

It is sad. In ideal world - I'd like more choice. But can our one planet support four desktop OSes - Windows, macOS, Linux and FreeBSD? It can barely support three.

 

[matt_k]

I do almost agree with you though. A decent OS will never be suitable as a user-friendly consumer desktop. FreeBSD has very little future in this space because doing so would ruin it. Where our opinions differ is that I believe all open-source consumer centric operating systems will regress to the point of unusability. Making FreeBSD competitive by default. With a recent influx of Linux escapees, we are already starting to see symptoms of this.

exactly what I said in some similar thread here some time ago. It was like "why is FreeBSD so user unfriendly". What are you talking about. FreeBSD is user friendly by being simple, well-documented and well-behaved. Can't get any more user friendly than that. If some people are looking for point-and-click bloatwareOS, then there is *buntu, look no further, it works and it is exactly what you need. For god's sake, don't try to bend FreeBSD to this type of mindset, there is enough *buntu clones already.

What I don't get is, why do we keep having these same types of forum posts over and over again? The sticky should be enough

 

[Snurg]

FreeBSD is becoming a framework

It was always, and that is the reason why it can be used so universally.
This makes it so competitive, without actually trying to "compete".

I believe all open-source consumer centric operating systems will regress to the point of unusability. Making FreeBSD competitive by default. With a recent influx of Linux escapees, we are already starting to see symptoms of this.

The more constricted and bloated a system becomes, the less freedom it leaves to its users.
A vicious cycle.
Users demand more bloat, and then they suddenly decide that it got too much bloat and demand an overhaul.
So, new distros appear and older ones disappear.

Not worth participating in this game of losing.

 

[fryshke]

It was always, and that is the reason why it can be used so universally.
This makes it so competitive, without actually trying to "compete".

Yes, competitive as a framework OS. And it's winning here over Linux. But not because of technical superiority, but because of license, let's be real. That's the only reason. Topic was about "desktop os" - and here freebsd is losing hard.

 

[kpedersen]

Topic was about "desktop os" - and here freebsd is losing hard.

Absolutely, and Linux is also losing.

And yet, we can hardly say that macOS or Windows are suitable. Part of a desktop OS involves privacy and they can simply not comply with that requirement. They are also both propritary so are instantly out of the running for open-source desktops.

So the next best are Linux and FreeBSD. They are basically 1% apart in terms of "desktop superiority". Pick whichever has the coolest wallpaper. Honestly at this point, we are so far at the bottom of the bucket that you could probably justify "dirt on a stick" as a competitive desktop.

My advice to anyone hung up on "desktops" is make do without, get on with your life (and projects). There is *much* more to computing than fancy buttons and pictures.

 

[jardows]

Yes, competitive as a framework OS. And it's winning here over Linux. But not because of technical superiority, but because of license, let's be real. That's the only reason. Topic was about "desktop os" - and here freebsd is losing hard.

Again with they hyperbole and FUD. What do you mean by "losing hard?" Do you mean market share? Because if that's your metric, nothing is going to "win" unless it comes out of Redmond. At its best MacOS can only manage about 9% market share of desktop computing. You mentioned Linux trying to get to boot on an M1 - I imagine less than 1% - probably closer to less than .1% of the Apple computers with an M1 processor will ever end up with anything other than MacOS - maybe when the original M1 is EOL and cannot get MacOS updates anymore, you might see a few more people trying to tinker with the old hardware. I mean, did you see the big announcement the other day that the Linux kernel now finally supports installation on the Nintendo 64? Hurry up FreeBSD developers - gotta catch up to Linux here!

You are asking the wrong questions and looking at things from the wrong perspective, and the spreading FUD based on this. You should not look at FreeBSD and ask the question - "can this OS gain mass adoption among the computer simpletons (It's the year of the 'FreeBSD Desktop' finally!)?" Because that will never happen. It won't happen with any 'BSD, Solaris, Linux distro, Haiku, RedOx, ReactOS, you name it. The question should be - "Can I use FreeBSD as a Desktop computer Operating System without significant trouble, and will it continue to grow in its support and development?" The answer to that is YES! YES! A thousand times YES!

I like FreeBSD as both a server operating system and a desktop operating system. Is it always the best tool for the job? No. Does it work for the vast majority of my personal use cases? Yes. Am I willing to put in some of the work to build an operating system environment that is a pleasure for me to use, not just the same boring thing? Yep, sure am.

As I see it, there are only two things that would make FreeBSD "better" for Desktop usage - 1. Improved hardware compatibility, 2. Increased application support. Neither of these are a fault of the FreeBSD operating system, and even at this, these areas are not so far behind as one might think. These areas will not improve by some radical shift in thinking of the core team, or a new direction for the project, they simply require more people willing to work on it. I have seen numerous times where hardware support or application porting happened in a relatively short period of time by only one or two talented programmers taking the time to do it.

By focusing on what FreeBSD does well (not necessarily the best, but well, and there are lots of areas we can point to), we can get more people educated and interested in the project, meaning more people that may be willing to help in driver or application compatibility.

 

[zirias@]

I don't care for all that jabbering. Typical "Desktop" WILL become a niche for sure, no matter which OS, maybe needed in some offices, probably needed by software devs. Given that, I need it, and to me personally, FreeBSD is the *perfect* dektop OS. It really depends on your expectations, mine isn't having all sorts of bells and whistles OOTB but a simple, clean and logical design leaving me in control.

 

[Snurg]

As I see it, there are only two things that would make FreeBSD "better" for Desktop usage - 1. Improved hardware compatibility, 2. Increased application support.

Both 1. and 2. are good.
I guess what people are actually complaining about is the "CLI firewall" that keeps ubuntized (*) people out.
Probably not a bad thing.

By focusing on what FreeBSD does well ( ... there are lots of areas we can point to), we can get more people educated and interested in the project, meaning more people that may be willing to help ...

+1


(*) The GUI fancy is often called "Klickibunti" in German. Hard to translate, it derogatorily says about "colorful things to click around". This is what I think when I hear/read "Ubuntu". Imho we don't need more KlickibuntiBSDs. Why don't the Klickibunti people just go to GhostBSD?

 

[YYY__]

I risk guessing that the initial thread question is the result of the fear of wasting time and effort in learning a system that was not desktop-oriented, with the underlying technical limitations it entails supported by all the FUD around, and that it's more of a personal choice of a new user who dropped around here.

It's a legitimate question without a right answer, so it gets so many looks. Life is full of daily micro-choices, accepting it and moving on. There is no technical aspect that resolves the issue except your use case. The survival of the desktop project will be fixed by the users and the commercial push around. Time will tell.

Don't take it seriously and enjoy it. —Jeff Bridges' wife

 

[zirias@]

I risk guessing that the initial thread question is the result of the fear of wasting time and effort in learning a system that was not desktop-oriented, with the underlying technical limitations it entails supported by all the FUD around, and that it's more of a personal choice of a new user who dropped around here.

Then what MUST be made clear is that FreeBSD, although not particularly "desktop-oriented", isn't (only) focused on servers either. It might have been in the past (?), but mission statement clearly is a "general purpose" OS nowadays, which is already reflected in the very first sentence on the project website:

FreeBSD is an operating system used to power modern servers, desktops, and embedded platforms.

 

[Mjölnir]

Where did I read that Intel recently enhanced/increased their support for FreeBSD, concerning just about everything from CPU, GPU to LAN & WLAN network drivers? It was here on this forum, couldn't find it again. Or was it in vermaden's weekly RSS feed, also posted here in Feedback IIRC? Well, if FreeBSD is sentenced to death, then why would they do that? Besides that I'd like to know about the plan B of those companies that build their products upon FreeBSD.
You know my mantra: the yea free BeaSD makes up for a very sound, secure & performant foundation of a modern desktop system (usable for ordinary non-techies). It's just that a (large?) team has to assemble the missing links & plug them together. It boils down to manpower & organisation.

 

[zirias@]

You know my mantra: the yea free BeaSD makes up for a very sound, secure & performant foundation of a modern desktop system (usable for ordinary non-techies). It's just that a (large?) team has to assemble the missing links & plug them together.

What I also meant with expectations above: I probably wouldn't want to use that.

It's not rocket science to install and configure software you'd want for your desktop, and whatever you might want to tweak in the base system isn't too much either. I'm totally willing to do that (it can be done quickly) to get the result I want: A clean, well-structured system, one that I know and understand (at least on a "building blocks" level) and can easily manage, maintain and fix any problem.

Back in the days, when the ecosystem around Linux was still somewhat sane (not talking about the missing defined base system of course, but the times before ALSA, pulseaudio, systemd, you name it...), you could get something remotely similar with e.g. Debian.

 

[drhowarddrfine]

...and 20 years later...(thread continues below. Different trolls. Same useless comments.)

 

[Deleted member 30996]

exactly what I said in some similar thread here some time ago. It was like "why is FreeBSD so user unfriendly". What are you talking about. FreeBSD is user friendly by being simple, well-documented and well-behaved. Can't get any more user friendly than that. If some people are looking for point-and-click bloatwareOS, then there is *buntu, look no further, it works and it is exactly what you need. For god's sake, don't try to bend FreeBSD to this type of mindset, there is enough *buntu clones already.

You make good points.

It can't be that "user unfriendly" if I could teach myself to use it without looking at the Handbook.

Not having done that, I've noticed over time I ended up doing quite a few things in a way other people do differently that seems a lot more complicated than the way I learned to make it work for me. That being most recent in memory.

And it works very well for me. I don't want to see anything changed in the base system and please continue to allow me to choose what programs I install from that point on.

 

[drhowarddrfine]

Ooh! Oooh! I know!! Just like the thread about introducing yourself, let's keep this thread for all the trolls to feed their line of BS! It will keep their garbage out of the useful posts! Problem solved!

 

[Snurg]

You make good points.

it can't be that "user unfriendly" if I could teach myself to use it without looking at the Handbook.

Well, my impression is that it was actually the existence of PC-BSD and its quick and easy installation, which made you recognize what qualities FreeBSD has to offer.

I have to admit that I often used PC-BSD when I had to set up another computer.
Just because its installer saved me a few hours of doing the boring steps between bsdinstall and an usable desktop system.

And this is one of the reasons why I work on something similar, which can save me (and maybe others) this chore.
I have to admit that one of my motivations to make it easy to get set up an usable system quickly is make people say "ohhh FreeBSD is different, it is unique, I want to use this because nothing else offers me such great things!".

 

[Deleted member 30996]

Well, my impression is that it was actually the existence of PC-BSD and its quick and easy installation, which made you recognize what qualities FreeBSD has to offer.

I already knew I wanted to use FreeBSD but the installer looked beyond my skillset at the time and inconceivable an itty-bitty incubus dare invoke it.

I was browsing Live CD's available for different Operating Systems when I first saw PC-BSD among the Linux discos listed. That came with KDE3 installed and all I needed was to get to the desktop.

I joined the PC-BSD forums as Weixiong in June 2005 and the rest is Black Sheep Sorcery.

 

[Mjölnir]

YMMD (once again) ROTFLMAO

 

[mickey]

As I see it, there are only two things that would make FreeBSD "better" for Desktop usage - 1. Improved hardware compatibility, 2. Increased application support. Neither of these are a fault of the FreeBSD operating system

As for 1 this is not entirely true, as I just had to experience myself. Trying to overcome the "no WiFi" situation on a friend's notebook (due to unsupported builtin QCA9377 WiFi module) after having spent a great deal of time into researching what should be supported, she ordered an ASUS USB-N10 Nano WiFi adapter that supposedly had a RTL8188CUS chipset, supported by FreeBSD's rtwn_usb(4) driver. When that thing arrived, FreeBSD would not assign the driver to it, because - suprise, surprise - that thing is Rev. B1 of the device that uses an RTL8188EUS chipset instead. Not that this should generally be a showstopper, as RTL8188EUS should be supported as well, but finding out that there is a hardcoded list of USB vendor/product IDs that does simply not include this specific device's ID really made my day. Saying WiFi support in FreeBSD leaves things to be desired would be an understatement.

 

[Mjölnir]

[...] that thing is Rev. B1 of the device that uses an RTL8188EUS chipset instead. Not that this should generally be a showstopper, as RTL8188EUS should be supported as well, but finding out that there is a hardcoded list of USB vendor/product IDs that does simply not include this specific device's ID really made my day. Saying WiFi support in FreeBSD leaves things to be desired would be an understatement.

1. Did you research if that chipset is supported in stable or 13-beta?
2. It's open source. Unless that chipset needs special instrumentation & initialization, you can easily add that device ID to the list & build the module from source. Did you send a patch or the missing device ID to FreeBSD's bugzilla, to help to improve this unsatisfying situation?

 

[Snurg]

Realtek is a low-quality nuisance to avoid if possible, even worse than ATI.

Their chips need firmware updates to work.
There are many different series that are not or not easily discernible from naming/numbering scheme.
Even worse, completely different chips share the same PCI ID.
Just using the driver disk that came with your device is no solution here, of course.

So, to support particular Realtek chips, every chip has to be verified to work with a particular driver, and for new chips the firmware also needs to be extracted from the Windows drivers.

If you could determine a driver work with that chip after its PCI ID/sub ID has been added to the driver's list, it would be extremely helpful if you could post a PR with the necessary data so the maintainers can update the FreeBSD driver.
Your feedback would help others much.

P.S.
Sadly it is difficult to officially warn against particular hardware manufacturers.
But imho this is necessary, as it is not worth supporting low-quality low-performance hardware.
This only fires back.

 

[kpedersen]

Ooh! Oooh! I know!! Just like the thread about introducing yourself, let's keep this thread for all the trolls to feed their line of BS! It will keep their garbage out of the useful posts! Problem solved!

Heh, perhaps not a bad idea.

This topic comes up enough that it is worthwhile having a thread we can just merge it all in with. Safer than deleting in case there is ~1% useful content.

But this same topic comes up so often that it is far too much pressure on the mods. Maybe we need a mod employed specifically for this *one* purpose of quelling the "Desktop FreeBSD topic".

I would offer my services but I believe I am too biased (and argumentative) to do an appropriate job. I sometimes just can't help myself but get involved. Especially when we get a feisty one. I am trying to improve here, but... babysteps