I see Linux being more and more widely adopted in Europe, for one. I imagine there will be an 'EU standardized' distro forked to prevent any issues of interoperability between government systems. Hopefully, such a widespread adoption for PCs encourages more development for Linux from software devs.
So a couple of weeks ago I finally had enough of windows and put linux mint on my laptop instead. I haven't used linux since the early 90s and couldn't remember much at all, was concerned I'd have trouble with drivers etc.
In fact the whole experience was a piece of piss. So easy. There were no urgent-seeming pop ups with arcane terms, no crashes or problems, I didn't need to use the command prompt.
Then I was able to find and install the programmes I use easily too. Slack, steam, etc. Within maybe an hour I was able to do my work on the computer again. I kinda feel like I got my laptop back - windows was getting so buggy and aggravating that I had been avoiding using the machine.
If Linux wants a future in which it continues to grow, it needs to do more of this, appealing to the casual, non-technical user. Because we probably represent most of the market.
Since I first learned about Linux I have never envisioned a future where Linux didn't eventually take over essentially all operating system spaces and I still don't. The question is how long will it take to get there.
But as others have said, I think the overall decline of desktop PC use combined with the just pure overall quality of Linux compared to Mac and Windows PCs in 2026 implies that the x86 PC space will become majority Linux within the next 10 years if not less.
Bright. As it continues to grow, and MS gets more terrible I think desktop share will always increase.
I seriously worry about what happens when Linus decides to end his involvement though. I don't keep up with the happens of the kernel devs but the thought of a split, takeover or power vacuum is a concern for me. Hopefully someone can assure me.
My exact concern too. Don't see it as an imminent problem, but would like some sort of succession plan or even to develop some sort of framework to ensure the kernel endures after Linus stops being involved.
What amazes me is, that a formal official plan (this) is just done recently a few months ago. How was there no official plan before? I don't understand...
WTF we are talking about Linux, the biggest and most important piece of software. What are you even talking about will and adults? Sorry if this was meant to be a joke. But having a plan like this is crucial important.
The point, is that people don't like thinking about this subject, as evidenced by the fact that half of people don't put plans in place even when it's something as comparatively simple as a will unless they get more imminent prior warning. This is not a Linux kernel specific issue, Bram and vim being an obvious example. Something as potentially complicated as the Linux kernel succession? I can see why it wouldn't be a subject the community was keen to tackle.
In future there will be a "next generation" Linux,
not microkernel, but also not so monolithic.
More modular subsystems, drivers will be easier to write.
Native sandboxing.
Deprecation of classic POSIX permissions, file ACL by default.
Rewritten network userspace tools,
everything could be declared in one file (like in the network equipment).
Yet another desktop server.
Desktop will respond near real-time.
Better audio experience, low latency by design, no more xruns.
And more...
Depends where hardware will go. Like another architecture?
Like today, just better? It's likely still going to power most of the servers, 70% of smartphones, a lot of the embedded devices... And maybe desktop marketshare is going to rise a bit above the current 4%.
I think it's going to start skyrocketing in global desktop use. Maybe not in the US, but globally lots of other countries have good reason to migrate away from US based software companies
Linux runs the entire internet, a good part of corporate-level servers and it dominates the supercomputing space, you can't get more mainstream than that.
As for desktop use, most people around the world don't know what an operating system is, some even believe the browser IS the internet.
With such tech-illiteracy there's no way Linux can be widely adopted, because no opensource project can beat Microsoft commercial power in "convincing" PC vendors to offer it preinstalled, yes a few do, but they're very scarce compared to those offering Windows, some even claim you will break the warranty if you install something different than what comes preinstalled (Windows), such a claim is illegal in many countries but some people don't know any better.
If you don't know what an OS is, you have no reason to change, unless you're lucky and have a friend or family member doing the work for you, but you can't rely on that for the entire world.
What I hope is every institution and company has a costum (or bought license to a costum) linux distro which is reliable and stable and fullfill their needs, without having to pay to bug tech nor share all their data with them, but instead a ecosystem where the companies has their own linux devs or IT-department for their linux instead of current SD departments with focus on microsoft environment. Smaller businesses or institutions should be able to buy IT equipment with preinstalled and supported distros to their need.
I tried it with Sway thinking the interface would work like Hyprland. Eventually I just switched to Parrot since security tools are useful and have not made an opinion yet. If it doesn't work out, Bazzite or Endevour are in the list. There are just so many.
I doubt it will over take Microslop offerings anytime soon, but I also think the rise of the Linux desktop has only just started. I think it will come from non-US countries where government departments make the switch. People start getting comfortable using it at work, which helps build confidence in the alternatives to Microslop shit. Also Valve still pushing hard in the gaming space. I think 10% to 15% market share in 5 years is possible. I doubt it'll go beyond 20% without some fundamental changes, like laws forcing hardware OEMs to support Linux equally as windows and Mac, and better DRM support.
More popular. More users. Higher percentage of desktop/laptop PC users
Flatpak permissions handled in a very easy to use way. No silent failure. No need to go to flatseal and users understand why something didn't work how they expected and what they need to do to fix it
Growing Linux userbase eventually results in great day one support for new products from Qualcomm, ARM mali GPUs, PowerVR, etc. They'll want to be able to compete year after year with Intel and AMD someday
Someday native Linux games rather than WINE/Proton will become the norm
Popular media software categories continue seeing open source software gain mainstream/professional viability. Talking like Blender, Godot, Krita today. Someday stuff like Kdenlive, Scribus, Inkscape, Ardour, GIMP, Darktable, etc will breach some line of good enough functionality, interface design. Someday the user base will grow enough and enough will make it into industry with their experience and opinions
Someday more normal Linux phone OS's like PostmarketOS will become a solid piece of the mobile pie. Like ~5%. Like how desktop Linux is today. Good usability but still working up to streamlined. That'll be way better than today. In what I imagine would be well over a decade when a Linux phone is as popular as desktop Linux is today, it'll actually be pretty easy to use like desktop Linux is today
I see everything through the lens of the difference in user experience and mainstream penetration of 2010 compared to today. Like Kdenlive of 2010 compared to today. 2010 Blender vs today's Blender. 2010 OpenOffice compared to 2026 Libreoffice. Gaming with WINE in 2010 to today with Proton/WINE/Steam. Unity/KDE/GNOME/etc of 2010 compared to today.
Asshole authoritarians spread their disease throughout the world. First they pass laws requiring age verification options and successfully pull off the non-optional bait and switch. Then they require more and more mandatory shit until open source becomes impossible. Only paid distros like Red Hat survive and we've all got cameras in bedrooms and slave gays, lesbians, blacks, and non Christian people in concentration camps that operate on the latest Microslop Winblows data-miner OS.
I think it's going to be higher than that. I think a lot of counties will start rapidly migrating away from American software companies, and the only alternative is Linux. China will soon really start pushing out their own fully home-grown cheap PCs to the world with some flavor of Linux as the OS. American software companies won't be able to compete.
Globally, I bet the desktop marketshare for various flavors of Linux is pushing 90% by 2040.
I think it'll keep on running most of the world's servers and embedded devices.
People who think desktop Linux is dead because of spicy OS level age declaration laws that only apply in Brazil and a fraction of the US clearly weren't around when SCO was trying to sue Linux into oblivion and it shows.
Linus is not the only one involved in Linux. Most work and decisions are done without Linus' involvement at all. There are candidates (especially Greg) who are as important as Linus themselves and probably will take over the project. This is not a single user driven project.
I think it's more likely that soon China will start heavily pushing out a forked version of Linux on cheap fully home-built PCs. Then it's basically game over for American software companies.
i'm aware. however there's a lot of money involved in linux, without linus i'm not certain linux will be able to be mostly free from corporate influence and such.
Unfortunately, I don't see Linux ever gaining widespread adoption. Not for general desktop use.
In order for that to happen, Linux would have to gain enough market share to actually be a threat to Microsoft's business, and when that happens, Microsoft is going to respond one of two ways: actually making their products more appealing, and regain their lost share, or (far more likely) REALLY ramp up their efforts to suppress Linux's growth. And I don't think Linux has the strength to withstand the full weight of Microsoft's legal and financial power.
But Hurd is too far behind and will probably never be a real alternative to Linux. Also the Linux Kernel is not what adopts the Age Verification, its the distribution. Therefore instead changing the Kernel, they just need to change on the distribution what they do not like. Or create the same distribution with Linux, they would create with GNU Hurd.
Brighter than ever, thanks to Steam Deck and Microsoft's general idiocy.
Linux would look great even if Microsoft didn't help, but oh my god, does Microsoft help.
I see Linux being more and more widely adopted in Europe, for one. I imagine there will be an 'EU standardized' distro forked to prevent any issues of interoperability between government systems. Hopefully, such a widespread adoption for PCs encourages more development for Linux from software devs.
Just `man future` .
Seriously, RTFM.
/s jk, this comment brought to you by parody.
So a couple of weeks ago I finally had enough of windows and put linux mint on my laptop instead. I haven't used linux since the early 90s and couldn't remember much at all, was concerned I'd have trouble with drivers etc.
In fact the whole experience was a piece of piss. So easy. There were no urgent-seeming pop ups with arcane terms, no crashes or problems, I didn't need to use the command prompt.
Then I was able to find and install the programmes I use easily too. Slack, steam, etc. Within maybe an hour I was able to do my work on the computer again. I kinda feel like I got my laptop back - windows was getting so buggy and aggravating that I had been avoiding using the machine.
If Linux wants a future in which it continues to grow, it needs to do more of this, appealing to the casual, non-technical user. Because we probably represent most of the market.
Also Microsoft convinced the people that system updates are a bad thing.
Mint, Manjaro, Fedora, Bazzite… there are plenty of beginner friendly distros these days!
Since I first learned about Linux I have never envisioned a future where Linux didn't eventually take over essentially all operating system spaces and I still don't. The question is how long will it take to get there.
But as others have said, I think the overall decline of desktop PC use combined with the just pure overall quality of Linux compared to Mac and Windows PCs in 2026 implies that the x86 PC space will become majority Linux within the next 10 years if not less.
I mostly agree, except I think there could be a new version of something that isn't Linux if it gets that big.
Bright. As it continues to grow, and MS gets more terrible I think desktop share will always increase.
I seriously worry about what happens when Linus decides to end his involvement though. I don't keep up with the happens of the kernel devs but the thought of a split, takeover or power vacuum is a concern for me. Hopefully someone can assure me.
My exact concern too. Don't see it as an imminent problem, but would like some sort of succession plan or even to develop some sort of framework to ensure the kernel endures after Linus stops being involved.
There is a continuity process at least https://github.com/torvalds/linux/commit/102606402f4f5943266160e263c450fdfe4dd981 Although it is mainly, here is the process to figure out the plan.
What amazes me is, that a formal official plan (this) is just done recently a few months ago. How was there no official plan before? I don't understand...
Have you got a will?
Estimated 50% of adults don't. There's always a more pleasant task on the to-do list to be getting on with.
WTF we are talking about Linux, the biggest and most important piece of software. What are you even talking about will and adults? Sorry if this was meant to be a joke. But having a plan like this is crucial important.
The point, is that people don't like thinking about this subject, as evidenced by the fact that half of people don't put plans in place even when it's something as comparatively simple as a will unless they get more imminent prior warning. This is not a Linux kernel specific issue, Bram and vim being an obvious example. Something as potentially complicated as the Linux kernel succession? I can see why it wouldn't be a subject the community was keen to tackle.
That's easy, I just take a gander into my Linux Orb, and it reveals itself to me
In future there will be a "next generation" Linux,
not microkernel, but also not so monolithic.
More modular subsystems, drivers will be easier to write.
Native sandboxing.
Deprecation of classic POSIX permissions, file ACL by default.
Rewritten network userspace tools,
everything could be declared in one file (like in the network equipment).
Yet another desktop server.
Desktop will respond near real-time.
Better audio experience, low latency by design, no more xruns.
And more...
Depends where hardware will go. Like another architecture?
How could the subsystems be more modular?
My understanding is almost everything in the kernel is modular outside of the schedular
So what your saying is 40 years of bike shedding. The more things change the more they stay the same
Wow. Every word of that is frightening.
My money is on 2027 as The Year Of The Linux Desktop.
My money is on 1999 as the year of the Linux desktop. I mean, that's when I switched to Linux full time 🤷
Same here. Debian since Slink.
Similarly, 2025 wasn't the year of Linux desktop... but it was the year of Linux on my desktop.
Like today, just better? It's likely still going to power most of the servers, 70% of smartphones, a lot of the embedded devices... And maybe desktop marketshare is going to rise a bit above the current 4%.
I think it's going to start skyrocketing in global desktop use. Maybe not in the US, but globally lots of other countries have good reason to migrate away from US based software companies
Through a monitor.. probably customizable Vr based DE's or something if I'm being real
Linux runs the entire internet, a good part of corporate-level servers and it dominates the supercomputing space, you can't get more mainstream than that.
As for desktop use, most people around the world don't know what an operating system is, some even believe the browser IS the internet.
With such tech-illiteracy there's no way Linux can be widely adopted, because no opensource project can beat Microsoft commercial power in "convincing" PC vendors to offer it preinstalled, yes a few do, but they're very scarce compared to those offering Windows, some even claim you will break the warranty if you install something different than what comes preinstalled (Windows), such a claim is illegal in many countries but some people don't know any better.
If you don't know what an OS is, you have no reason to change, unless you're lucky and have a friend or family member doing the work for you, but you can't rely on that for the entire world.
What I hope is every institution and company has a costum (or bought license to a costum) linux distro which is reliable and stable and fullfill their needs, without having to pay to bug tech nor share all their data with them, but instead a ecosystem where the companies has their own linux devs or IT-department for their linux instead of current SD departments with focus on microsoft environment. Smaller businesses or institutions should be able to buy IT equipment with preinstalled and supported distros to their need.
Eventually someone will make it 100% idiot proof.
I second this. Am idiot
I tried it with Sway thinking the interface would work like Hyprland. Eventually I just switched to Parrot since security tools are useful and have not made an opinion yet. If it doesn't work out, Bazzite or Endevour are in the list. There are just so many.
THE FUTURE IS NOW!
I doubt it will over take Microslop offerings anytime soon, but I also think the rise of the Linux desktop has only just started. I think it will come from non-US countries where government departments make the switch. People start getting comfortable using it at work, which helps build confidence in the alternatives to Microslop shit. Also Valve still pushing hard in the gaming space. I think 10% to 15% market share in 5 years is possible. I doubt it'll go beyond 20% without some fundamental changes, like laws forcing hardware OEMs to support Linux equally as windows and Mac, and better DRM support.
More popular. More users. Higher percentage of desktop/laptop PC users
Flatpak permissions handled in a very easy to use way. No silent failure. No need to go to flatseal and users understand why something didn't work how they expected and what they need to do to fix it
Growing Linux userbase eventually results in great day one support for new products from Qualcomm, ARM mali GPUs, PowerVR, etc. They'll want to be able to compete year after year with Intel and AMD someday
Someday native Linux games rather than WINE/Proton will become the norm
Popular media software categories continue seeing open source software gain mainstream/professional viability. Talking like Blender, Godot, Krita today. Someday stuff like Kdenlive, Scribus, Inkscape, Ardour, GIMP, Darktable, etc will breach some line of good enough functionality, interface design. Someday the user base will grow enough and enough will make it into industry with their experience and opinions
Someday more normal Linux phone OS's like PostmarketOS will become a solid piece of the mobile pie. Like ~5%. Like how desktop Linux is today. Good usability but still working up to streamlined. That'll be way better than today. In what I imagine would be well over a decade when a Linux phone is as popular as desktop Linux is today, it'll actually be pretty easy to use like desktop Linux is today
I see everything through the lens of the difference in user experience and mainstream penetration of 2010 compared to today. Like Kdenlive of 2010 compared to today. 2010 Blender vs today's Blender. 2010 OpenOffice compared to 2026 Libreoffice. Gaming with WINE in 2010 to today with Proton/WINE/Steam. Unity/KDE/GNOME/etc of 2010 compared to today.
Asshole authoritarians spread their disease throughout the world. First they pass laws requiring age verification options and successfully pull off the non-optional bait and switch. Then they require more and more mandatory shit until open source becomes impossible. Only paid distros like Red Hat survive and we've all got cameras in bedrooms and slave gays, lesbians, blacks, and non Christian people in concentration camps that operate on the latest Microslop Winblows data-miner OS.
Linux itself is an important part of tech today, llms, servers, Iot etc, it will be like that in the future too.
The linux desktop is improving a lot and receiving more attention , im excited to see gog galaxy on linux personally.
KDE makes great things too, and im looking forward to see the Vr mode for Plasma and Plasma bigscreen.
As for linux on mobiles, i hope it gets better, the thing about google closing down android worries me!.
Linux always gets better. Thats something i really love about linux. There are a lot of communities doing great work on a lot of aspects everyday.
10% desktop usage by 2040, possibly 2035 if Microslop keeps being Linux's #1 marketer.
I think it's going to be higher than that. I think a lot of counties will start rapidly migrating away from American software companies, and the only alternative is Linux. China will soon really start pushing out their own fully home-grown cheap PCs to the world with some flavor of Linux as the OS. American software companies won't be able to compete.
Globally, I bet the desktop marketshare for various flavors of Linux is pushing 90% by 2040.
Yeah, this is a lower bound estimate. I wouldn't be surprised to see 20-30% by 2040.
I think it'll keep on running most of the world's servers and embedded devices.
People who think desktop Linux is dead because of spicy OS level age declaration laws that only apply in Brazil and a fraction of the US clearly weren't around when SCO was trying to sue Linux into oblivion and it shows.
Time-limited.
The recent changes to how the digital world operates are not conducive to open operating systems and I imagine this trend will accelerate.
Only if we allow it.
not sure if it'll be okay after linus is gone tbh..
Linus is not the only one involved in Linux. Most work and decisions are done without Linus' involvement at all. There are candidates (especially Greg) who are as important as Linus themselves and probably will take over the project. This is not a single user driven project.
I think it's more likely that soon China will start heavily pushing out a forked version of Linux on cheap fully home-built PCs. Then it's basically game over for American software companies.
i'm aware. however there's a lot of money involved in linux, without linus i'm not certain linux will be able to be mostly free from corporate influence and such.
Unfortunately, I don't see Linux ever gaining widespread adoption. Not for general desktop use.
In order for that to happen, Linux would have to gain enough market share to actually be a threat to Microsoft's business, and when that happens, Microsoft is going to respond one of two ways: actually making their products more appealing, and regain their lost share, or (far more likely) REALLY ramp up their efforts to suppress Linux's growth. And I don't think Linux has the strength to withstand the full weight of Microsoft's legal and financial power.
It will last well into the 24th century
Ah yes, that will be THE year of Linux!
Bad.
Linux will become the new compromised OS
not far removed from Android,
while GNU Hurd will become the new Linux.
[edit]
Go ahead and vote me down!
It doesn't change the way Ubuntu and Systemd have operated.
I feel like this comment was a sarcastic joke and people didn't pick it up.^^ Especially after the Hurd line, this has to be a joke.
No, this is a response to the mass adoption of systemd by Linux and the Age Verification Law.
It's GNU philosophy or owning a compromised OS.
But Hurd is too far behind and will probably never be a real alternative to Linux. Also the Linux Kernel is not what adopts the Age Verification, its the distribution. Therefore instead changing the Kernel, they just need to change on the distribution what they do not like. Or create the same distribution with Linux, they would create with GNU Hurd.