Rant: 2026 will not be the ‘Year of Linux on the Desktop’ – and I’m glad
It’s a running joke in the Linux world: 2016 was the Year of Linux on the Desktop, apparently. Then it was 2017. Then 2018. Every year somebody, somewhere, claims boldly that the time has finally come: we’ll all be buying laptops with Linux pre-installed from PC World. Those of us who use Linux every day look around in bewilderment, wondering why this happy state of affairs stubbornly continues not to come to pass. The alternatives to Linux continue to get worse and worse, and more and more expensive, with ever-increasing hardware requirements but, in the end, it still never seems to be the Year of Linux on the Desktop. I confidently expect that 2026 won’t be, either.
Discussion forums are full of arguments about why Linux isn’t more popular on the desktop, and what should be done about it: there are too many Linux distributions; there are too many integrated desktops; we should all standardize on this, that, or the other thing, because too much choice is confusing to new users. And so on, and on.
What these arguments all seem to miss is that there already is a mass-market Linux for the desktop: it’s called ChromeOS. It’s what all the Chrombooks run, and, despite running Linux, Chromebooks aren’t known for technical elitism – they are true mass-market appliances. I wouldn’t touch ChromeOS with a ten-foot pole, but its mere existence proves that a desktop Linux for non-technical users is not merely possible: it’s here already, and it’s quite successful.
So why aren’t we saying that the Year of Linux on the Desktop has already come and gone?
Linux advocates would likely say that ChromeOS isn’t ‘real’ Linux because it’s specifically made for the mass market. It doesn’t have the power or flexibility of enthusiast Linux distributions like Ubuntu or Fedora. But it isn’t intended to: most computer users don’t need sophistication. So long as the operating system provides a way to run a browser – and ChromeOS is essentially an operating system wrapped around the Chrome browser – that’s good enough for most people, most of the time.
Arguably, the more we make traditional desktop Linux like ChromeOS, the more popular it will be in the consumer market. It isn’t hard to see this kind of thinking behind, for example, the Gnome desktop. Gnome is easy to use, so long as what you want to do aligns with what the Gnome maintainers think you should be doing. Recent Gnome releases don’t look anything like the desktop Linux of the 2000’s: much of the complexity and configurability have been stripped away. Nevertheless, however easy it might be use Gnome, or KDE, or any of the rest, there hasn’t been a mass migration from Windows.
Why? Because there’s still too much, well… Linux in there. Where ChromeOS scores is in not looking like Linux at all. You don’t have to know anything about Linux to use it, because all it does is run a browser.
So, while traditional Linux distributions are failing to make headway on the desktop, ChromeOS is cutting through the waves with the wind abeam. Currently, only Google has the resources and enthusiasm to market such a thing, which is why ChromeOS stands alone as a Linux-based mass-market platform. Well, almost alone: there’s also the Steam Deck, but that’s even less a general-purpose computing platform than a Chromebook is. There’s nothing in principle stopping other businesses offering their own take on Linux, if they think there’s a market for such a thing – and clearly there is, judging from the uptake of Chromebooks.
What Linux advocates are really saying, I think, when they claim that desktop Linux would have more mass-market appeal if we do this thing or that thing or the other thing, is that that their interpretation of Linux would be more popular. ChromeOS and the like are excluded – there seems to be a tacit assumption that we can retain the look and feel of traditional desktop Linux, with all its power and flexibility, while at the same time making it tractable for non-technical users. We want, it seems, a one-click, hands-off installer and a point-and-click user interface, but we also want to be able to do 3D animation and electronics design.
But is that even possible? Can we have the simplicity and ease of use of ChromeOS, with the power, versatility, and broad hardware support of, say, Ubuntu or Fedora?
I would argue that it probably isn’t possible, that every move towards mass-market adoption will rob Linux of its soul. Any changes of that sort will tend to degrade those features that make Linux appealing to the folks who currently use it. Let me explain.
I don’t care that there are twenty thousand Linux distributions, or a hundred different integrated desktops. I don’t care that we have battles over systemd or Wayland. All these things are emblematic of a vigorous, technical community, one that allows for diversity and choice. This makes it fun and rewarding to develop software for Linux; many of its most significant applications are maintained by enthusiasts, not for profit.
I don’t care that Linux doesn’t offer a bug-for-bug replacement for Microsoft Excel, or that some video games won’t run on it. The software I need runs just fine on Linux. I don’t mind that I have to use a command line. In fact, I like the command line – it’s the beating heart of Linux.
I’d venture to suggest that most of us who are everyday Linux users didn’t adopt Linux because it was cheaper than Microsoft Windows. We adopted it because we found it more productive, and were willing to put a bit of effort into learning how to use it effectively. Probably that learning included finding and adopting alternatives to Microsoft Excel and the like. That this saves us money is a fortunate side-effect, not the reason for our choice. We adopted Linux because we get more done, with less pain. And, these days, less intrusion into our privacy.
I don’t want to use a Chromebook. For me, ChromeOS is no more productive than Microsoft Windows, and ChromeOS gives me all the same privacy and advertising concerns; worse concerns perhaps because, you know… Google. So I don’t want traditional desktop Linux to be more like ChromeOS; I don’t even want it to be more like MacOS – I want the Linux world to be just as fractured, opinionated, and argumentative as it is today. I want there to be a dozen different ways to do the same thing. In short, I want Linux to work for me and other technically-minded enthusiasts, as it currently does, and has for the last twenty-odd years. I don’t care who else uses Linux – frankly, I don’t care if nobody else does.
Is this an elitist view? I don’t think so. We don’t pretend that everybody should drive the same car, or eat the same meal. Why should operating systems be any different? I use my computers for programming, maintaining this website, and composing music. What reasons are there for thinking that I need the same operating system as somebody who works with spreadsheets or edits photos all day? Linux won’t suit everybody; that’s not elitism, it’s just pragmatism.
So I don’t really care if there never comes a time when you can buy a Linux-based laptop from PC World. The kind of Linux it will be running is not the kind I want: it will resemble the ill-favoured offspring of ChromeOS and Windows 95.
Don’t get me wrong: I think it would be nice if non-technical buyers could choose a desktop operating system that had the broad application support of Microsoft Windows, but was more robust, less intrusive, more configurable, mostly open-source, and cheaper. I’d also like shorter waiting times for public healthcare, and affordable public transport; but I doubt we’ll be seeing those any time soon, either.
I do understand what motivates Linux advocates to push for better mass-market adoption. We look around and see Windows users repeatedly punched in the face by their ad-ridden, unreliable, privacy-wrecking computers – computers whose service lives have been reduced to a year or two – and think there must be a better way. But for most people, I suspect there isn’t a better way, and trying to make one out of Linux will be like tying a strong swimmer to a drowning man. It won’t end well, is what I’m saying.
2926 will not be the Year of Linux on the Desktop. When that year comes – and I, for one, hope it never does – everything that makes Linux what it is will have sunk under the waves of corporate self-interest, never to be seen again.
So instead of complaining that the Year of Linux on the Desktop never seems to arrive, let’s celebrate instead. Let’s rejoice in the fact that desktop Linux remains what it’s always been – a niche operating system for people who want to avoid corporate snooping, opt out of the Andy and Bill rat-race, and squeeze every drop of computing power out of their systems.
Long may it continue.
