Opensource alternatives to *BSD and Linux? (Unix)

Hello,

Both *BSD (e.g. FreeBSD) and Linux are outstanding. Nothing against *BSD and Linux at all. If some day, Linux and *BSD might change a lot. What would you do? Another Unix-Branch?

Which opensource alternatives to *BSD and Linux might still exist?
/* which would be really new (e.g from original bsd/scratch/unix/....) */
/* free, opensource, opensource for education and studying, Unix */

So far, a possible third possibility would be Android, which is actually Linux-inside. Well....

What else might be still a good alternative today (2018)? Happy to know more Unix and discover fantastic Unix operating systems.

Best regards,
--
More reading about Unix:
https://en.wikipedia.org/wiki/Unix
https://en.wikipedia.org/wiki/Devuan
https://www.redhat.com/
https://www.gnome.org/
Apple might control BSD (often heard)
Apple dev
Linux: Future of the Linux kernel will eventually be subsumed into systemd/wayland/...
Anyhow there is still in any case: http://www.freedos.org/ (x86, PC), the fantastic graphics library http://grx.gnu.de/ running on ARM Raspberry! (GRX PI), DJGPP
 
Ubuntu controls Debian
For this reason a fork "Devuan" has been created recently for people not satisfied with systemd, and for people who are fed up of the influence of Canonical.

But Apple doesn't control BSD, this is TOTALLY wrong
You are not experienced enough with BSD, but with time you will see that between FreeBSD and Apple... there are very few common "gateways".

If you were more experienced with FreeBSD, you would know that in fact this is probably iXsystems and Juniper, thoses companies who may have THE major influence on FreeBSD, absolutety not Apple.

FreeBSD is not the background of OSX, OSX borrows various elements to different BSD flavors.
Apple uses Packet Filer... this is an OpenBSD firewall, OpenSSH... this is also OpenBSD
The OSX kernel is a kernel of his own different from FreeBSD, OpenBSD... where all Linuxes flavors share exactly the same kernel.

Due to the particular BSD licensing, Apple OSX is in fact a closed source operating system, you seem to forget that, and for this simple reason, even if we wanted to, FreeBSD could not imitate so easily Apple without their authorisation.

Did FreeBSD develop a native support of HFS ? HFS support is only available with fuse
This simple point shows you that OSX and FreeBSD live their own life, nearly ignoring each other.

Please keep in mind the very strong difference between the BSD licensing and the GPL licensing.
Steve Job has been the most genius man of the universe, understanding far before the other what we could do with BSD licensing ... so using "free software", modifying it, to make our own "closed software" with no counterpart... and this point makes the Free Software Foundation completely crazy

If Sony has choosen FreeBSD and not Linux for their playstations, this is for the same reason.
Sony doesn't have to share any BSD licensed code.... Sony doesn't tend to "influence" the BSD world, they just use BSD to create their own closed source system... this is a very big difference with Linux, based on a collaborative works.

This is what we call a "permissive" license.
If you modify a GPL licensed code YOU MUST SHARE IT PUBLICLY.
 
Last edited:
I first came in contact with Unix-like open source systems thanks to OpenSolaris. Lately I've been re-embracing the Solaris World once gain, and trying Illumos spins: OpenIndiana and Tribblix are quite good and solid.

Amid the niche stuff instead, you have:

- RedoxOS a new emerging, promising, independent Unix-like system

- Another one, not as popular, is TaruOS

- MINIX3.4 has now Xorg built in base system; not a desktop alternstive,but good for learning, or developing on embedded

- same thing as minix mostly goes for 9front too, though not exactly Unix-like
 
I first came in contact with Unix-like ipen soyrce systems thanks to OpenSolaris. Lately I've been re-embracing the Solaris World once gain, and trying Illumos spins: OpenIndiana and Tribblix are quite good and solid.

Amid the niche stuff instead, you have:

- RedoxOS a new emerging, promising, independent Unix-like system

- Another one, not as popular, is TaruOS

- MINIX3.4 has now Xorg built in base system; not a desktop alternstive,but good for learning, or developing on embedded

- same thing as minix mostly goes for 9front too, though not exactly Unix-like

Why not a full rewrite from scratch, without using endlessly the common libraries (Xorg/ X11, SDL, OpenGL, ..)?
 
Why not a full rewrite from scratch, without using endlessly the common libraries (Xorg/ X11, SDL, OpenGL, ..)?

Because it takes time and manpower =P; I think however RedoxOS resembles what you're talking about the closest ;)
 
Because it takes time and manpower =P; I think however RedoxOS resembles what you're talking about the closest ;)


An OS can be written in Python too ;) (of course, I am joking)
https://github.com/wfxpanisa/pythonix
https://biosbits.org/scripting/

thank you... "...written in Rust,..." But, C may offer many advantages (surely).
A microkernel is actually a good way to go. Simpler to implement. https://en.wikipedia.org/wiki/File:OS-structure.svg

Minix

It takes time, but keep to minimum, step by step
 
As soon as OpenVMS's port to x86-64 is complete and shipping, one could run it on commodity hardware, using the Hobbyists license. For a home computer, that would make an excellent solution. Note that this is not open source, but it is free.

I have no idea what level of hardware support for desktop usage (GUI) one would get; I don't know whether OpenVMS has had any serious GUI use in the last 15 years, since the demise of the last Alpha-based workstations.

The other crazy idea would be to run a free mainframe OS (MVS version 8 is freely available) on an IBM 370 instruction set emulator (such as Hercules). A friend of mine does that on a Raspberry Pi: With the high-end RPi model having four cores, he emulates a 4-CPU mainframe (the IBM model 3084), and runs a 4-processor version of MVS on it. The problem with that is: (a) one needs an operating system to run Hercules on. (b) This very old version of the MVS is not useful as a home server, since everything needs to be emulated.
 
Solaris up thru v4 was heavily influenced by BSD. Redox is still pretty embryonic. OpenVMS - ah' c'mon! If we're talking about an OS to replace the one we're using, it's gotta have some modern umph-pa-pa. So, Minix3 and Plan9 are the truly different items in this thread so far. Relative to what I want from an OS, Plan9 misses and Minix3 doesn't strongly impress (although the microkernel is intriguing for some things).

HaikuOS is truly different than any of the *nix stuff, as was Syllable (now defunct). Haiku is usable for some/many of the requirements dictated by modern OS terms, at least in my opinion. But it's alpha/beta ATM.

There is one other OS in development by Google, called Fuchsia, which is based on a new microkernel designed by the same guy who wrote the kernel that was forked to Haiku. That could be interesting, but it's currently only at a "could be good someday" level. It seems there ought to be more, given how many creative minds there are in the world ...
 
Nobody knows whether Fuchsia should replace Chrome OS, Android, both or none just yet. In general, microkernels are nice though.
 
I think the OP should refine his question, and make it in terms of desktop, server, or both. Most of the things I mentioned would never be candidates for server machines IMO.
 
Redox, which is written in Rust, has no ability to use Internet/ethernet, last time I checked.
Minix lacks plenty of drivers, including basic USB, but they were working on this. Maybe the next release will have that.


http://distrowatch.org/search.php?o...uage=All&defaultinit=All&status=Active#simple

Here is a list from Distrowatch, with the selection of "Other OS". Some have mentioned ReactOS and Haiku. There's RISC OS that has a small following, but it is for ARM architectures. KolibriOS is written in FASM assembly language, but it is from Russia. The list is lacking Redox.

* Edit -

MenuetOS is mentioned (from which KolibriOS is based on), but not on that list. There are some details on it here, https://www.informit.com/articles/article.aspx?p=768380&seqNum=5 that it is small enough it has fit on a floppy disk. From this full Informit link, Contiki is even smaller, that its OS size is in Kilobytes made for embedded systems. Syllable is also mentioned from that parent link.
 
Just some notes on a few threads:

1. illumos. I did read that Garrett D’amore who started illumos just before Oracle closed OpenSolais has said he is unhappy with some directions because of decisions by Joyent who is the biggest contributor and player in illumos-land.

I’d have to search for his remarks and I thought he mentioned forking it.
http://garrett.damore.org/2014/11/a-better-illumos.html?m=1

2. OmniOS is now OmniOSce (community edition) since OmniTi stopped sponsoring development.

3. Plan 9. 9front is actively maintained and the Abaco web browser supports JavaScript and CSS. It won’t display pages nicely, but it works on https sites too.

Plan 9 takes getting used to, but Brantley Coile who started the storage company Coraid uses Plan 9 for their products. He also developed the PIX firewall and used Plan 9.

Again, I’d have to find his remarks, but he has said he’d always use Plan 9.

If you want to study the operating system, then Plan 9 or Inferno are your best choices because you can wrap your head around the source code and understand it.
 
If you want to study the operating system, then Plan 9 or Inferno are your best choices because you can wrap your head around the source code and understand it.

Thank you, danh for the very interesting information. When I saw the group of people that started illumos, I wondered how such an eterogeneous group would have been able to continue working together. Unfortunately, it seems D'Amore's fork, illumos-core is not too active.

Still, OmniOSce seems quite healthy, and OpenIndiana (i.e. "new" OpenSolaris) is quite healthy too.

Thank you also for mentioning Plan 9 and 9Front. Indeed, that would seem to be at the forefront of serious OS development.

If you have time, would you point to Brantley Coile's remarks that you mentioned?

Thank you again.

PS: I find 9Front's c.o.c. refreshingly humorous. Every time you refresh that page it will offer something additional.
 
Yes, there is. The illumos distributions usually import the FreeBSD WiFi drivers, so if your WiFi works in FreeBSD, chances are that illumos will give you the same experience.
 
Wi-fi works fine on my ThinkPad w541. It does not work at all on one of ours custom X399/AMD Threadripper workstations.

As usual, it depends on the hardware, possibly how common that is.

Personally, I feel hardware compatibility is not a good criteria to value the worth of an Operating System.

Some of our engineers embrace macos, and that is very hardware specific.

Some others choose the other way around: they are ready to have a workstation built that will work on OpenIndiana. OpenIndiana's ZFS with TimeSlider is too important for some of them.
 
Hi Sensucht94 ,

how is, or is there any WiFi support on OpenIndiana, Tribblix?

Kindest regards,

M

I second Cthulhux/priyadarshan affermation, Illumos has ported a lot of drivers from *BSD; might still lag behind, expecially OpenBSD, but my NICs (quite old ones ~ 8 years old) were supported out ofthe box with ral and ath drivers; see Wifi-OpenInidiana for a list of supported chips ;)
 
Back
Top