Linux Sucks 2021

Linux Sucks 2021

It’s worth watching. Haiku is also mentioned :slight_smile:

This post is also not here to stir controversy or debate whether Linux sucks or not. I’ve been a fan of Linux myself for a longtime.


Each man to his own…
Opinions are like arse holes…every body has one and it stinks

1 Like

probably the most honest discussion about the future of Linux I’ve seen. I expect fuchsia to deplete most of the consumer space, and that will cause forks of linix to put it back into the high performance compute amd daya center useage.

Could you try being a bit less crude?


I can only agree that Linux sucks.

The best operating system ever made was Windows 2000. W2K was stable and had full driver, software and game support. It was complete but without unnecessary feature creep. If I could, I would use it to this day. When W2K was no longer usable, I reluctantly downgraded to Windows XP. After it was no longer possible to use XP, I again had to reluctantly downgrade to Windows 7. In between, XP and W7 I used Mac OS X as my primary OS, but Apple has totally destroyed its OS á la 1984. Worth noting is that I have been a BeOS and System 7-9 user, but never as my primary OS.

I have used Unix and Linux a lot, both on servers and desktop, but never liked it. Unix and Linux is broken by design. When I can no longer use W7 as my primary operating system, I truly hope Haiku is ready to take the helm. The only piece missing for me is the web browser. If not, I will be forced to use either a BSD or Linux. Windows 10+ and Mac OS are not an option.

1 Like

Them’s fighting words (on a Haiku forum) :sunglasses:


BeOS was an excellent operating system, but it did not reach technical maturity and was not even close to obtaining a good ecosystem.

1 Like

@Null I think I agree almost with all points made. I had used Windows 2000 at work back then and I can confirm your opinion. I am currently using Linux at work without problems at all although I couldn’t say I am an expert. I just can find my way around and get help from Internet when needed. And since I always was an OS tinkerer I have installed maybe hundreds of Linux distributions in my life. Certainly there is no way for me to go back using either Windows or MacOS X. (Used Mac OS X some years at home). Linux is much more better than both of them and everyone can configure it they way they want or find the distribution that suit them. Having said all the above, since I am here, of course it means that I would like to use Haiku for everything ! But I will be very happy if I could use it for personal reasons with better browsing experience :slight_smile:

This is the reason why Linux sucks and why it is broken by design. It is not coherent. I do not want to spend hours upon hours to configure the operating system, only to have it break on the next update, or just break by itself. Haiku is coherent and reasonably configured by default and is currently my only hope for the future.


Linux fans cite the number of different Linux distributions as one of its strengths. They are in fact a very visible symptom of its biggest weakness - a lack of coherence, as you say.
To me, one of the major attractions of Haiku is that there is just one version. Plus it is simple, fast, elegant and not trying to make money out of me, by fair means or foul.


@sebrof certainly I am not a Linux fan, I am a Haiku fan :smile:
But the truth is that for some years now I have no problem at all using Debian based distributions for work (Ubuntu, Pop!_Os, Debian) and I am not spending time on configuring them other than installing some gnome extensions. Things tend to just work. I did spent time once to configure a weird printer. Of course I am not saying anything about underlying OS design, architecture and other stuff. I am just saying that I can do my job very decently with Linux compared to Windows and Mac OS having a UI/UX experience that suites my needs. I can’t have Haiku’s one but I can dream I can have it one day!

Without taking a view on if the “end of linux is nigh” I must disagree with most of the points made in the video and point out some errors.

For starters, Richard Stallman was reinstated at the FSF months before this video came out.

Next… yes Red Hat got bought by IBM, and Red Hat do a lot of work on a huge number of Linux based software projects. But IBM do a whole ton of work for Linux based software projects too. They pay people to work on the kernel and on projects like QEMU. I’m not especially worried about their attitude towards open source and I don’t think they will dismantle or destroy RH. IBM CPUs do not have closed source firmware components like Intels and they created the OpenPOWER initiative partially to cement this. He has overlooked the fact Linux holds the largest market share in embedded and server markets, with thousands of other companies backing it, as well as its use in government, armed forces etc. The NSA developed SE Linux, China is all Linux, and the biggest Red Hat user is the US armed forces. The majority of smart cameras, routers, tablets, printers, web servers etc all run Linux, with huge industrial backing.

About growth rate being too fast… the vast majority of the 2 million lines of code added to the kernel every year are for device drivers, which don’t add to the complexity of the kernel itself, and only a small fraction of the total drivers are in use on any one system. Linux also adds support for new architectures very quickly (e.g. RISC V) and these add lines of code. Moreover, the Linux kernel is known as a pioneer and exemplar of how to run very large scale software engineering projects with thousands of contributors, and do it well. The kernel development model is studied by undergraduates in University courses. Linux is well placed to deal with the growth. He doesn’t mention that they also deprecate or mark as unmaintained large amounts of kernel code every year, that remain in-tree in case someone picks up maintainership, but are not used.

Lunduke states that security vulnerabilities will increase in number but I note that he does not show in an evidence based way that there is an upward trend in the number of vulnerabilities over time. Vulnerabilities are inevitable, but Linux has a great track record for having fewer of them and patching them fast, so this point is purely speculative.

In person Linux events cancelled? Has he been in a cave for the last two years? Because in the UK and much of the rest of the world most people were not allowed to leave the house for large parts of that time, never mind visit a conference.

About Fuscia. Well, FreeBSD and OpenBSD can run Linux binaries. It hasn’t caused any noticeable impact on Linux, and the fact that Fuscia can do it is not exactly some ground breaking news. Haiku being able to run BeOS binaries, and wine being able to run windows binaries are far more impressive to be honest. Look at the investment made by big companies into Linux there (steam, play on linux etc), it’s insane. I don’t think Fuscia has shown anything ground breaking that I’ve heard about so far.

“Other operating systems have died so Linux will die… list of closed source commercial operating systems”. Is this causation?

I actually agree with many of the points made in this thread. Linux is not cohesive or consistent and Haiku is leagues ahead in that regard. Maybe Linux will stall and die, I don’t know. And I am not trying to bash Lunduke as I have enjoyed every other video of his that I’ve watched (not a great number, but a few). But I don’t think this video was very well thought through.


I’m not watching the rest of this video because this guy refuses to get to the point and drones on for 45 minutes. I will say that if Fuchsia has a stable ABI and a Linux compatibility layer in a microkernel, Linux will have its lunch eaten. End of story.

1 Like

consider that Google deploys more linux installs than any other org out there.

If Google goes Fuchsia, so does the entire Android ecosystem etc.

As Munchausen says, Linux may not have made significant inroads into the desktop market, but it pretty much owns the Web. I do use Linux for some things, and I agree that it works, but I hope to replace it (and Windows) with Haiku in due course.
Software has a habit of staying around long after anybody thought it would. I believe parts of the airline industry still use Cobol, and OS/2 lives on as ArcOS. The idea that Linux will disappear in the next ten years seems preposterous.
IBM has long supported Linux, and was the target of a lawsuit by a gentleman called Darl McBride, who claimed Linus infringed intellectual property rights owned by his company, SCO. I am sure most people will be familiar with the story. For once IBM was the good guy, and the good guy won.


Heh, the only instance when the notorious Naz’gul (IBM lawyers) were actually deployed for good.

Anyways, there’s a continuation of the SCO lawsuits with the successor company Xinuos:

I don’t really think Linux needs android or google. The rest of the (gigantic) industry built around Linux is not dependent on android or google in any way, if they change it will not mean others have to. And android is quite different in any case; some driver APIs are very different, the rest of the kernel is quite heavily modified, and the userland is very different… I don’t think moving it to Fuscia can really be considered a significant end to Linux.

If we do see current Linux distributions moving to a fuscia kernel with a mostly unchanged userland, that is also not really the end of Linux that Lunduke and many people in this thread appear to be suggesting. The distributions will essentially remain the same, just without the Linux kernel at the core. I can see many individuals and organisations being very untrusting of moving to a google kernel, in any case.


most of the $$$$ coming into Linux, is via Andriod and other applications us8ng linix for IOT, which, fuchsia looks to murder, quickly.

Linix will live on, as a server os and some workstation use.

1 Like

Pedant alert: Technically the operating system proper is GNU/Linux - the GNU operating system running on the Linux kernel. If it moved to Zircon, then it becomes GNU/Zircon, which would indeed be the end of linux. Before the Linux kernal came along, GNU was envisioned to have a microkernel.

1 Like

GNU also /has/ a microkernel, it is called gnu hurd.
The notion of “just switch kernels” is a bit wierd tbh, and apart from linux I don’t know any OS that thinks that would be a good idea. (linux distros have tried ripping out the freebsd kernel for instance and running linux userland on it kinda, but never really got far).

I’d also like to add that there are a couple of linux desktop/server distros that don’t have any GNU software, Like alpine linux, Haiku has more GNU software. So it seems a bit silly to try to pounce on the gnu/linux marketing name, also with other pieces of software becoming more and more important for linux but don’t get their own mention either.