What is the future of FreeBSD?

I know how my car works, I can even do some basic maintenance on it. But if there's something wrong with it I take it to dealership and have a professional take a look.

The "casual" user is non-technical, I dare say the majority of computer users have absolutely no idea how a computer works. It's all some sort of dark magic to them. I find it rather interesting open source projects need to be made so the casual user can install and configure it. What percentage of computer users actually install and configure an operating system by themselves? It's not going to be the "casual" user, it'll be the computer enthousiast, someone who knows a bit about computers. The casual user is going to buy something in the shop, and if there's something wrong with it they'll take it back to the shop or hire someone to fix their problem.
 
reasonsandreasons said:
Having keyboards without hardware scroll lock isn't unheard of, as that thread attests. Read the messages, of course, but "you should have another device to look up your unique key combination so you can read a message to get your desktop set up" isn't the world's most user-friendly response.
Click to expand...
Then the correct solution (as that thread mentions) is to then fall back to a pager (i.e less(1)) or terminal multiplexor (i.e tmux(1)) or redirect to a file (i.e tee(1)). UNIX offers literally so many options to read console messages that citing "user-friendly" only goes so far until the user is outright sabotaging themselves (or being a "concern troll" on forums...).

Using incorrect hardware as an excuse to not read warning messages is quite bizarre for a beginner. An i.e smartphone without a keyboard will not be a good reason to add touch events to the FreeBSD virtual terminal. The opposite would be like asking on the Windows forums how to use the OS without a pointing device. Its possible but they will simply suggest you buy one for £3 until you become more familiar with the OS.
 
SirDice said:
I know how my car works, I can even do some basic maintenance on it. But if there's something wrong with it I take it to dealership and have a professional take a look.

The "casual" user is non-technical, I dare say the majority of computer users have absolutely no idea how a computer works. It's all some sort of dark magic to them. I find it rather interesting open source projects need to be made so the casual user can install and configure it. What percentage of computer users actually install and configure an operating system by themselves? It's not going to be the "casual" user, it'll be the computer enthousiast, someone who knows a bit about computers. The casual user is going to buy something in the shop, and if there's something wrong with it they'll take it back to the shop or hire someone to fix their problem.
Click to expand...
I think this is a bit of a false dichotomy. There's a middle group of people who are interested in FOSS and Unix-like operating systems but are not preternatural graybeards. Having a straightforward entry point to desktop use (again, the request here is mostly for an option in the installer to set up a desktop environment) helps onboard those folks at very little cost to anyone else and hopefully makes them more likely to stick with the OS long-term and give back to the community. There's a reason this is a Foundation priority!
 
SirDice said:
That's not going to work, you can't install a Windows driver on MacOS, or a Linux driver on Windows.
Click to expand...
I believe that UEFI to be the hypervisor of the computer and all devices are exposed as strictly standardized (about interfaces) {runtime|boottime} services should be way to go.
 
drhowarddrfine said:
Next up: Why doesn't Plan9 cater to new users and have a graphical installer?!!!
Click to expand...
Yeah, or be functional without a mouse.

The up arrow key going to the line above rather than recalling the last command is infuriating but as per their design goals, they certainly won't be making that change just for my definition of "user-friendly".
 
drhowarddrfine said:
Next up: Why doesn't Play9 cater to new users and have a graphical installer?!!!
Click to expand...
It does have a graphical installer! Its default interface is graphical as well. And it caters to new users by being far simpler. In many ways it is better than unix but because it is not unix it has hardly any 3rd party apps.

And its C is ancient and doesn't pay attention the latest C standard violence^Wfashion. Example: earlier today FreeBSD-CURRENT buildworld broke because some !#$!#$ added this:
672 | size_t res [COLOR=rgb(226, 80, 65)]__maybe_unused[/COLOR];
A later commit shows this:
sys/cdefs.h: Introduce __maybe_unused attribute

The __maybe_unused attribute should be used for variables which may or
may not be used, such as when their only use is in an assertion. This
attribute is functionally identical to __unused, suppressing compiler
warnings for particular variable if it remains unused.
Click to expand...
Now why in the world is Freebsd tracking such C inanities so closely that hardly adds anything useful? And now buildworld will take an extra hour. Compiler should be far far more stable and should be on its slow schedule. Further, the kernel & world should use C standard (and not the latest) and not be tied to any particular C compiler. Let users who want more modern compiler use a package. But I digress!
 
bakul said:
It does have a graphical installer! Its default interface is graphical as well. And it caters to new users by being far simpler. In many ways it is better than unix but because it is not unix it has hardly any 3rd party apps.

And its C is ancient and doesn't pay attention the latest C standard violence^Wfashion. Example: earlier today FreeBSD-CURRENT buildworld broke because some !#$!#$ added this:
672 | size_t res [COLOR=rgb(226, 80, 65)]__maybe_unused[/COLOR];
A later commit shows this:

Now why in the world is Freebsd tracking such C inanities so closely that hardly adds anything useful? And now buildworld will take an extra hour. Compiler should be far far more stable and should be on its slow schedule. Further, the kernel & world should use C standard (and not the latest) and not be tied to any particular C compiler. Let users who want more modern compiler use a package. But I digress!
Click to expand...
Never took time to try it out. Is it still worth it? Looking at the system requirements of 32MB RAM and 300MB space, that's like Win98. Running native or in a vm would barely make a difference...
 
Ahh, another bikeshed discussion.

yeah, it does take 10 minutes to do a barebones installation.

yeah, that barebones installation does provide a lot of tools.

But - those tools are only useful if you know about them. Most rank-and-file user cannot be expected to know how to use dd, tcpdump, ssh or vi.

That's kind of like giving somebody a high-quality engine and expecting them to assemble a car around that engine. Most people don't even know how to change a tire on an existing, prebuilt car (I don't, for example), let alone know how to find a shop that sells aftermarket seat covers. For me, an engine by itself is useless, even if it's the best engine in the world. And no, I'm not gonna take the time to learn how to assemble a car, I have better things to do with my time.
 
Ilovehotdog said:
What is the future of FreeBSD?
I used to see FreeBSD as a base system much like Arch Linux—one that gives users all the choices but doesn't provide an official desktop environment. Recently, however, FreeBSD 15 announced (and subsequently delayed to 15.1) that it would ship with an official KDE desktop. On the other hand, the KDE project stated this year that it will be dropping support for FreeBSD.
So where does this leave FreeBSD going forward? Will it remain a foundational system that leaves everything to the user, or is it moving toward becoming a more standard, desktop-integrated system like Debian? Could it even evolve into something more akin to Ubuntu, Zorin, or Deepin, with official support for new technologies such as AI?
On a related note, I’d also like to clarify: does FreeBSD actually have a plan to officially support a KDE desktop, rather than leaving users to install or compile it themselves as is currently the case?
Click to expand...

I've always been one to be correct in my doom and gloom prophesies. In the future I see a world where computing devices are not "owned" by their users, but instead are restricted black boxes that can only be managed by the "service providers" who provide them. Eventually the major CPU manufacturers will design chips that cannot be programmed using an open-source model. Instead they won't even boot without a secure key exchange with a licensed software provider.
 
MG said:
Never took time to try it out. Is it still worth it? Looking at the system requirements of 32MB RAM and 300MB space, that's like Win98. Running native or in a vm would barely make a difference...
Click to expand...
IMHO it is worth studying and playing with it. Prob. a better intro to OS design than the old Lions book for v6 Unix.

I run 9front (a fork of orig. plan9) in a VM. The orig plan9 has run on pretty much every model of raspberryPi, including pi5[1] and you can get a runnable SDcard image. You do need a 3 button mouse. But note that it can be frustrating if you assume it will work just like Unix. It is far simpler than Unix but more composable and discovering & using its power will require some careful studying.

[1] plan9 was ported by Richard Miller to Pis (with a tiny bit of help by yours truly). The same Richard who also did the orig Unix v6 port to a new processor arch. back in 1977!
 
kent_dorfman766 said:
Eventually the major CPU manufacturers will design chips that cannot be programmed using an open-source model. Instead they won't even boot without a secure key exchange with a licensed software provider.
Click to expand...
Funny, AMD's Ryzen AI 445 runs FreeBSD just fine. I guess I can always look for Loongson and Baikal processors to play with, and get those to run FreeBSD.
 
kpedersen said:
Indeed. Luckily this is a FreeBSD forum so assuming using FreeBSD's console to scroll is fairly safe.

... and certainly not an excuse to ignore messages / warnings that come up when installing 3rd party software from ports.
Click to expand...
Ah, now we're back to having to know everything before you can do anything.

I wasn't talking about ignoring them, I was talking about not even seeing them.
 
kent_dorfman766 said:
I've always been one to be correct in my doom and gloom prophesies. In the future I see a world where computing devices are not "owned" by their users, but instead are restricted black boxes that can only be managed by the "service providers" who provide them. Eventually the major CPU manufacturers will design chips that cannot be programmed using an open-source model. Instead they won't even boot without a secure key exchange with a licensed software provider.
Click to expand...
well, i hear so many answer in forums "if you want a simple way to use bsd without any reading, you can use ghostbsd".
guess what?
i see the article.
and i really don't know what it is meanning to normal guys who like freedom.
not only bsd, but all open source project.
i realize all of them are meeting the same situation.

Addressing XLibre Change and GhostBSD Future

Why GhostBSD is Moving to XLibre GhostBSD is not Linux. It's a FreeBSD-based system for the desktop. That said, we use desktop environment software from the Linux world. On FreeBSD and GhostBSD, there are things that don't exist, like systemd, which a lot of software depends on. MATE is starting …
ericbsd.com

another news, the NanUI project is poweroff because its author turn into selling steels work for living.🥶


i realy realy don't know what will be in the future.
 
Ilovehotdog said:
i realy realy don't know what will be in the future.
Click to expand...
To get rid of anything related, we have to prohibit semiconductor-containing electronics. Silicium and germanium must be illegal as well as research into binary logic. I will go underground with the people creating a mechanical computer of gears and valves. 😁
 
The "works out of the box" OS is something you pay for. I can't see where the incentives are for FOSS developers to work on something like that, unless there is some corporate funding or steering going on in the background - i.e. why is someone going to devote their own free time to making everything "easy", which will in turn add complexity, to cater to those who don't want to learn the basics, instead of working on e.g. a wifi chip driver?
 
msplsh said:
Ah, now we're back to having to know everything before you can do anything.
Click to expand...
Indeed, but again, people are happy to learn how to scroll up on a web browser. Why not a terminal window?

msplsh said:
I wasn't talking about ignoring them, I was talking about not even seeing them.
Click to expand...
If someone allows messages to scroll off the screen before reading them (and choose not to use a pager or any of the many solutions to avoid this), then they are ignoring them.

Sure, ignoring i.e compile output as an unprivileged user is generally OK (if process returns success) but ignoring messages from files being sprayed over the filesystem (as root) is a little careless, regardless of operating system.

Pkg-messages seem to be batched until the end of the install, so a potential compromise could be to default to an -I (interactive) mode that blocks execution displaying them until a user presses enter.
 
blackhole said:
The "works out of the box" OS is something you pay for. I can't see where the incentives are for FOSS developers to work on something like that, unless there is some corporate funding or steering going on in the background - i.e. why is someone going to devote their own free time to making everything "easy", which will in turn add complexity, to cater to those who don't want to learn the basics, instead of working on e.g. a wifi chip driver?
Click to expand...
A lot of people genuinely enjoy working on UI. They're often people with a different skill set to those who work on WiFi drivers anyway - in fact the ability to hack WiFi drivers plus access to suitable hardware for testing is very much a minority affair, one of the causes for FreeBSD lagging so far behind. But re funding or steering: bear in mind that a large proportion, I've seen it reported as the majority, of base system FreeBSD commits are funded (in contrast, most work on the FreeBSD ports tree is volunteer-led). From memory it's mentioned in this interview: https://softwareengineeringdaily.com/2026/03/31/freebsd-with-john-baldwin/

GUIs from Apple, Microsoft, Chromebook, Android, Samsung etc have a very polished look and feel: they're prepared to spend big money on teams of experienced designers and UX experts, extensive usability testing (professional researchers watching a range of real users interacting with the system, noting pain points etc) and a chain of command that can enforce a consistent style across multiple projects. Doesn't mean they always get things right but decisions do have logic and data behind them - a lot of people still object to MS Office introducing the "ribbon", but Microsoft made that choice on the basis of $$$$ of usability research proving the change improved discoverability and users were incorporating advanced features in their documents they'd never tried out when they were just menu items. In the FOSS world a lot of these decisions rely on the gut instincts of devs plus extensive bikeshedding on mailing lists - usually by power users who aren't representative of the typical user. Compromise to give everyone what they want can result in extreme customisability, which is nice for ricers but also increases complexity and reduces consistency.

Even CLI aspects of FOSS projects could really benefit from some usability research but it's expensive and very professionalized. Still, there's no point having instructions or messages that are misleading or harder to understand than necessary, even if it's a point of pride to be an OS for "technical people". I've read or watched at least 3 stories of first-time FreeBSD installs where people have interpreted "Invite $username into other groups? []" as a yes/no question and been confused by what happens when they type "yes" or "no". Even if you do know it really wants a list of groups, the way it's phrased gives no hint as to how they should be separated. That's not because those users are non-technical and failing to follow a simple instruction on their screen - it's the kind of gotcha that usability research would have discovered from watching a dozen newbies work through bsdinstall. That one comes from https://github.com/freebsd/freebsd-src/blob/main/usr.sbin/adduser/adduser.sh

Another area FOSS generally lags a long way behind the paid-for OSes is accessibility. I know Alfonso Siciliano has been doing a lot of work on it but FreeBSD has been very bad for visually impaired users - I'm not sure most devs realise what happens when a screen reader tries to interpret a TUI. Again this is just a lot easier to do to a consistent level if you have a professional accessibility team, consistent style guides written with accessibility in mind, regular UX testing with users with a range of impairments, etc. It's the sort of "making things easier for people to use" that feels like a moral imperative (and in a paid-for product would be legally required in some jurisdictions), but without those resource advantages or legal compulsion it's unsurprising if accessibility becomes an afterthought. If the future of FreeBSD is greater "professionalization" - which might require reversing the loss of many of its corporate users - then I'd expect this to be an area where FreeBSD catches up a bit. But I don't think everyone's going to like the fruits of that. Case in point: expect to see the option of a GUI frontend for the installer, since that's better for accessibility (screen readers are designed to understand GUIs, not TUIs). That will be a most unwelcome development for many users. https://alfonsosiciliano.gitlab.io/posts/2025-01-27-ideas-installer.html
 
blackhole said:
why is someone going to devote their own free time to making everything "easy", which will in turn add complexity, to cater to those who don't want to learn the basics, instead of working on e.g. a wifi chip driver?
Click to expand...
Unfortunately, compared to writing wifi drivers, people think UI and usability is "easy" because they believe they understand it based on their own personal experiences and biases. Technically drawing glorified boxes is easy, so it opens the doors for a large variety of people "having a crack at it".

But ultimately, this is why most open-source UI environments are still pretty trash and inconsistent, even after many decades of thrashing around. The last true usability study of a FOSS desktop was done by Sun Microsystems in ancient times.

We do need i.e. wifi drivers written but I do not believe the sorts of users that FreeBSD will attract by making it "easy to install" by non-technical users will ever provide the skills to get this complex work done. OpenBSD is doing extremely well in terms of hardware support and in one of the interviews Theo outright stated that he doesn't even want to attract users. Obviously the companies behind the FreeBSD foundation have quite different motives for attracting consumers.
 
I think you should pass a standardized technical test before being allowed to install FreeBSD. Currently, this is chaos. Users like me are using FreeBSD without ever consulting any log file. It's outrageous.
 
You must log in or register to reply here.
Back
Top