The Year of the Linux Desktop

Many have been arguing that almost every year since 2016 or so might become “The year of the Linux desktop.” But this year, it might actually happen, because something has changed. Previously, Linux was always lacking something — support from large vendors, gaming support, usability for non-technical users. Don’t get me wrong, there are still many issues that will prevent Linux from becoming mainstream this year. But in this article, I will argue that Linux has come far enough to start its ascend from the fringes of the operating systems into the mainstream.

That is what I think the real “year of the Linux desktop” will be — not the one where Linux supersedes Windows, but the one where enough pieces are in place that its ascend can begin.

What Makes A Popular OS?

To start off, I first want to outline what I believe makes an operating system popular in the first place. And there are essentially three large components to this: corporate support, simplicity, and software support.

Both Windows and macOS fulfill these criteria. Both are backed by their manufacturer — Microsoft and Apple respectively. Then, both are pretty simple to use: You turn on your computer, and you can get started on your work without having to configure them first. And lastly, they have widespread support from companies who build their software specifically for Windows and macOS.

All of these reasons lead to widespread adoption and popularity. There are also feedback loops where adoption at some point begets software support, and where corporate support can manufacture adoption (Windows comes preinstalled almost everywhere). But I argue that this year, the last barrier has been torn down that prevented Linux from gaining these crucial preconditions.

Corporate Support

Linux is an entirely Open Source ecosystem. Nobody has to pay to use it. But there was always company support behind Linux. Most famously, there is a derivative of Fedora that is made and distributed by Red Hat. They sell enterprise licenses and support contracts to companies which have to run high availability servers. But that only affects business customers. RHEL (Red Hat Enterprise Linux) is a server operating system, not something you install on your laptop.¹

This focus on business customers shifted in 2006, when another company, Canonical, decided to throw its support behind a more user-facing operating system. Using Debian as a base, they released Ubuntu 06.04. The promise? Better support for hardware and easier setup due to a company focusing on its improvement in a more structured way than a horde of volunteers could. And it shows: Ubuntu is one of the most famous variants for people wanting to try Linux.

The effects of having a company behind a Linux distribution also means that other distributions will benefit from this. When a piece of hardware is suddenly supported by one distribution, it typically doesn’t take long until all others support it, too, since that support is often baked into the core piece of the operating system — the kernel. And that is the binding factor between all distributions: regardless of how they look, they all share the same kernel. What differentiates the various distributions like Debian, Ubuntu, Fedora, Mint, Manjaro, etc., is merely the look and feel, not the underlying hardware support.

Simplicity

The second important factor, simplicity, is a direct derivative from this. As a company, there is typically a coordinated effort behind working on a system. That means that the team that makes the settings dialog doesn’t follow completely different design goals as the one developing the desktop interface. This means that the entire OS works more consistent.

I can remember having tried out some less popular distros, and using them was atrocious. If you pick and choose your interface components from different teams, you may get exactly the OS that you want. But the price you pay for that is that every puzzle piece follows different UX standards. Some auto-save settings, others may want you to click save, and most default to the de facto standard in the Linux world where you need to press the button labeled “Apply.” But again, since there is no company coordinating the development, different settings apps can easily follow different interface standards.

To add insult to injury, there is not one, but multiple settings apps that you can install and use to tweak your operating system. Same desktop environment, different apps. Because someone may have not liked the default approach to modifying your system settings, and decided to quite literally write a wholly new app for that. That's the opposite of simplicity. To be fair, this is less and less of a problem. Typically, there is only one settings app installed by default, and that one works very well. Installing other settings apps is still a possibility, but what is important is that you don't have to do it.

The setup experience, on the other hand, has always been pleasant and simple. I can remember back when I first tried Ubuntu 06.04, there were already “live USB” which essentially allow you to boot your computer directly into Linux to try it out without installing anything. That is something neither Microsoft nor Apple give you, and it’s amazing. This way, you can just download a bunch of distributions you may be interested in, and try them out one after another without actually installing them. You try them out, and once you're settled, you install only the one you liked the most.

There are other parts of the OS that were always far and wide between. Some work great, others not so much. One of my pet peeves to this day is that switching between a light and dark theme based on the time of day is still cumbersome. But then, Windows does not even allow one to do so at all, while every Linux distribution that I've tried at least has such a setting, even if that doesn't always work reliably.

In short, Linux is intentionally made in such a way that you can LEGO your own distribution if you are so inclined. There's not much you require apart from the actual Linux kernel, so it is possible to puzzle your own distribution together yourself, and especially in the Arch Linux community, this is still a popular method. But the important thing is that this is no longer a necessity. When I tried out Ubuntu 06.04 in 2006, I still had to somewhat puzzle together pieces of software to make my computer work. But nowadays, that's no longer the case. As long as you choose one of the relevant distributions (such as Ubuntu, Mint, or Fedora), they will quite literally work out of the box and have everything they need. The only thing you need to install are your own applications.

An operating system must stay out of the way. Configurability is cool, but it works against the fact that what you use your operating system for the most is to work with it. You don't want to fight with the window manager, and instead focus on that war crime of a grammatical construction in your paper. And nowadays, most Linux distribution do exactly that. They just work, and you don't have to fight them anymore.

Software Support

That leaves one part: software support. And that’s the big elephant in the room. Above, I casually wrote that "The only thing you need to install are your own applications." But here's the catch: that can become an entire rabbit hole that eats away your weekend.

Two decades ago, when I first tried Linux, I had to spend weeks trying to make my operating system detect the hardware I had installed in my computer. That's no longer an issue, as everything is supported out of the box. But what is still a problem is software support. When all you need are web apps (Google Drive, Microsoft 365, Teams, etc.), you're golden because the only app you have to install is a web browser, and that even comes preinstalled. But this changes once you require some specialized software.

This is an issue for almost everyone, because most people typically need at least one program that only works on Windows and/or macOS: Adobe programs, photo and video editing software, CAD software, and, most famously, video games. All of these apps often only work on Windows and macOS, sometimes even only Windows, and won't run on Linux. The reason is that Linux is just not popular enough for companies to warrant the extra effort to develop and test the same program also for Linux. Even Davinci Resolve, the video editor from Blackmagic, is only verified for the Linux distribution Rocky OS, which I haven't even heard of before watching a video by a YouTube creator who did switch to Linux entirely.

And that has been the biggest showstopper for widespread Linux adoption, and that is what I believe has changed in 2026. You see, it's a classical chicken-and-egg problem: companies will only start supporting Linux if a significant part of their user base uses Linux and demand is high. But since their users need that software, they will probably stick to an already supported operating system, so the company has no incentive to support Linux in the first place. And that is one of the primary factors that have prevented widespread Linux adoption in the past.

But how do you solve this? Well, you'll need yet another company that just makes Linux support Windows software.

Enter Valve.

Thirteen years ago, Valve released a first preview version of their own take on Linux, "SteamOS." Their goal? Create a new operating system that is optimized for gaming consoles. Since you can't build much on Windows, and macOS is bound to specific hardware, and Linux was entirely free and customizable by design, the choice for the company was easy. But the biggest part was not to write some software that made the OS look nice on handheld displays. Instead, the problem was how to get all the games they wanted to support on their hardware such as the Steam Deck to actually work on it.

Long story short: For the past 13 years, Steam worked on a piece of Linux software called Proton that takes a piece of software designed to run on Windows and translate all the instructions from Windows to Linux. That's extremely complicated and that's the reason it took them so long, but it paid off. This year, Steam finally released their "Steam Machine," a custom-built computer that runs SteamOS (read: Linux) and that can be placed in your living room to play video games on your TV. And, because Proton works on any Linux operating system, even people who already have a computer benefit from this. Proton works on any Linux version, and as such everyone can play Windows-only video games on Linux.

The Matthew Effect

And that's, I think, the important thing that has just turned the tables for Linux. Valve has essentially solved the chicken-and-egg problem for one large group of computer users by simply side-stepping the issue. Instead of waiting for video game developers to start publishing their games for Linux, they have ensured that Windows-software can just run on Linux without extra steps for both users and developers. That means that, from now on, people who just wish to play video games can just switch to Linux, and it will all work as they expect it to. Which means that Linux adoption for this group will increase significantly over the next few years.

Furthermore, since many people also tend to stream their video game sessions online, and produce videos from it, I expect that, over time, some companies will start to port their software to Linux. This will, in turn, increase the amount of people switching to Linux. Once that market share is big enough, other companies will follow suit, because suddenly there is money to be made with software developed for Linux, and nobody will want to miss out on that opportunity. This will, in turn, push more people towards Linux. And so on.

This is known in the literature as the Matthew Effect, or cumulative advantage. This is what has kept Microsoft afloat for so long: Because the company has strategically placed itself as the only choice for many early on, they had such a market penetration that Windows just received more attention. But now that Valve has solved this issue for one large group of users on Linux, the same will happen here. Once companies start to support the growing Linux gamer community, this will have beneficial effects for non-gamers as well. The more software becomes available natively on Linux, the more groups of people can switch to Linux without risking their livelihood.

Is 2026 the Year of the Linux Desktop?

I argue yes. I know many people have been burned by declaring previous years the “Year of the Linux Desktop” and then having to admit they were wrong. I haven’t yet. So we’ll see how well I fare. To be sure, I can be wrong.

Yet, I am confident. While many people won’t buy a steam machine due to its price tag, the big win for Linux is that it’s free. You don’t need to buy a $1,200 mini computer from Valve to benefit from their work. You can just download SteamOS and install it yourself in the computer you already have. And you don’t even have to use SteamOS. There are alternatives out there.

We are finally in a place where Linux for Desktop is mature enough to become part of a triad of desktop operating systems. In more than 90% of all cases, you will be able to “just install it” and it will just work. No driver hunting, no troubleshooting before you even wrote your first word in LibreOffice. You just set it up, connect to your Wi-Fi and get going. Just like it was with Windows 20 years ago.

No, Linux won’t supplant Windows as the most popular operating system this year. And it probably never will. But what has changed is that there is now a drive behind the increased adoption of Linux that was lacking in previous years.

You need three things to get people to switch to your operating system: ease of use, support from companies, and popularity from users. But all three depend on each other: users won’t use it if it is difficult to use, and companies won’t improve the compatibility of their tools when most of their customers do not use Linux.

What the persistence of thousands of volunteers and Valve have done for 20 years is that they have eliminated the company support part from the equation and solved the ease of use problem. This means that the only remaining factor is user popularity, and because the other two issues are solved, users will start switching to it. Microsoft’s current behavior with awful decisions for Windows is only a further accelerator of this trend.

This is mostly true for gamers, of course, and professionals will have it more difficult. Photographers and videographers will likely be stuck with Mac for the time being thanks to Adobe, and office workers will remain stuck with Microsoft due to large business contracts that tie their companies to Microsoft long-term. But gamers are a big group, and many of them also stream their playing, so there is a chance that having solved the popularity issue for this group of society will likely have ripple effects into other market segments.

And that’s why I think that 2026 will be the year of the Linux desktop. Because solving the usability issues of the past, and eradicating company support from the equation has made it viable to achieve significant market share over time.

There are two developments that I feel support this statement.

First, I see more and more influential people switching to Linux. After two Linux challenges in 2022 and in 2026, the channel Linus Tech Tips seems en route to replacing more and more of their Windows computers with Linux. The company's owner, Linus Sebastian, allegedly is currently using predominantly Linux for his day-to-day work. Last year, PewDiePie also switched entirely to Linux. Recently, Hardware Haven also announced a permanent switch to Linux. And, just before I finished this article, Thinking different released a lengthy video on switching to Linux. All of these various channels serve very different audiences, but they are all highly visible and share one conclusion: There are some rough edges around Linux, but they are not any more difficult to solve than working around Windows's issues.

Second, I have, without really noticing it, replaced Windows with Linux on all my computers. While I always had a Debian server running, I recently also added a Network Attached Storage (NAS) that is running TrueNAS, yet another Linux distribution. And I ditched Windows from my gaming computer and switched to Cachy OS. And my work computer has been running Manjaro for four years. The only thing that keeps me from completely switching to Linux is my presence within the Apple ecosystem. With an iPhone and a Watch, it is just too convenient to stay on a MacBook as my primary device.

But, and that is important, if you're using macOS, you are essentially running Linux. Now, it's not technically Linux, but since both Linux and macOS stem from the same roots, UNIX, they work very similarly. macOS and Linux have the same terminal, the same commands you can run, and a lot of the same file and folder structure. So most code is directly portable between macOS and Linux. The same cannot be said for Windows, which works entirely different.

Finally, while I always played around with Linux, in these past three or four years, switching to Linux meant that I did not have to run a single command on the command line. It just worked. Everything I needed to set was possible from a graphical user interface. And everything has been supported from the get-go. It was as simple as it could get. No problems, no driver issues, nothing. It just worked.

Final Thoughts

It is great to see that there is finally motion again in the operating system space. Until now, people either had to pay a lot for a usable operating system, or were stuck with Windows. In XP and Windows 7 times, that was indeed not an issue. But Vista, 8, and 11 have shown that Microsoft can also produce bad software. And with its current push for AI, it is only getting worse. VS Code, Teams, Office 365, Sharepoint — most Microsoft products are degrading significantly in functionality. They are more bug-ridden, less user-friendly, and have constant issues. If Linux manages to gain the necessary market share before Microsoft reverses course, there will be no turning back.

I, for one, will leave Windows behind happily. 2026 really is the year of the Linux desktop, at least for me.

Onward!