One thing that I think this argument sorely misses is an honest discussion of how e.g. bad (or perhaps unliked) decisions get made that have big impacts here.
For example, count me in with those folks who think the "new" GNOME sucks. Now, maybe you disagree and that's fine -- but so often those discussions start and end with "Well it's open source and so because you're not making anything better you can't even talk."
No. Some big players put their thumb on the scale and had a vision and a direction for GNOME and what role it would or should play; someone thought it was a good idea to try to out Steve Jobs Steve Jobs.
THOSE moves need more discussion and transparency in order to REALLY talk about "the Linux Desktop."
> While this could absolutely happen, the way that Linux as a whole has been developing over the years isn't always conducive to making the world's Windows and macOS users convert en masse.
Its the way Windows is developing that is driving this change. GNOME might be hardly usable but Microsoft managed to top that.
Edit: I retract the last sentence. I'm currently trying GNOME and its less usable than Windows.
Linux desktop has already arrived for me. All the apps and utilities I need is there, all installable via apt. It's not a Linux problem anymore when the hardware manufacturers won't support it.
GNOME is nice, KDE is nice, and we have other options for people that don't like the two previous one. The issue we have now is walled garden, when some proprietary software won't support standards and even their own file format.
> The problem is that these are "wins" because they bring Linux closer to Windows or macOS.
I disagree that this is an issue. The main advantage of Linux for me is that I have choice (including using various desktop environments that the author is annoyed by; I used GNOME for years and eventually had too many problems with it so I switched to KDE), and those choices are not controlled by one entity which, in the case of Apple and Microsoft, view me only as a customer to extract money from.
It's 2025. The "year of the linux desktop" has been a meme for years. No one says it in earnest. No one is having init or DE wars. And while there is plenty of healthy discussion about flatpak and other alt forms of software distribution, this is exactly the kind of innovation and experimentation that leads to the usability improvements the author wants to see. Linux is doing just fine, and I'm glad there are multiple options to accomplish similar tasks.
I HIGHLY agree with this article and IMHO all the comments bashing Gnome just don't really get it. For all of it's faults, I still roll with Ubuntu/Gnome every day because it just freaking works, gets out of my way, gets the job done, and doesn't require a weekend of tweaking software to get everything working just right.
I don't care at all for the SystemD/whatever else flame wars. Sure if you work on these systems you probably care deeply about the differences but please realize that most of your end users do not give a shit. The same goes for the various packaging systems, I prefer to still use DEB's when I can but at the end of the day it really just comes down to how easily can I get the apps setup on my computer to get my work done, myself and most other users also don't really care.
What I care about are things like: why is multi monitor support still half assed? why does full screening my chrome window crash my monitor? why is it that half my installed apps don't conform to my theme? why is it that when I switch on X window manager instead of Wayland my wallpaper goes away?
But it seems that the folks that actually work on Linux don't care about these issues because when I ask why VSCode crashes my monitor, all I get is answers telling me to use vim or emacs, or when I complain why the themes look all janky all I get is: "well this wouldn't be an issue if you used <insert obscure window manager that requires a week's worth of configuration to get running and a steep learning curve>.
The vibe I get around these issues is that it's below most Linux developers as they are too busy arguing about some flag in the kernel or whether to use systemd or not. But those same people bitch and moan why "Year of the Linux Desktop" hasn't come yet. Figure out that these issues are not below you, they are the issues that people care about. Fix those issues and I'm positive that adoption will go up.
This article is full of nonsense. The Linux desktop push isn't failing because it has experiences and apps that are similar to Windows and macOS. Being able to run Windows apps on Linux is a benefit, not a failure. As for religious wars over init systems, desktop environments and package managers, competition is making the options stronger, not weaker. Competition is a reason why package management on Linux is far better than equivalents on Windows and macOS.
The main reason for Linux not taking off on the desktop is because most users don't care about what OS they run, they just want a computer that works. If the PC they buy comes with Windows out of the box, they're going to stick with that. Until you get manufacturers shipping PCs with Linux as the default OS, you're mainly going to see desktop Linux as an enthusiast-only option. It's no accident that one of the devices helping to spread Linux (the Steam Deck) comes with Linux as the default option.
This article is kind of old hat - it's basically been true that Linux is fine as a desktop OS for Grandma since some point circa 2010 +/- a few years. The big requirement is that Grandma just uses web browsers and other basic software from the OSS ecosystem, hardware was relatively compatible to begin with, and somebody does the OS upgrades for her every 3 years.
The real issue is that these kinds of "grandma" users maybe just don't use computers anymore. And the folks that do are joined at the hip to proprietary software like Photoshop or CAD programs or whatever else they care a lot about and don't want to relearn, and also make enough money that the costs are invisible. Or they're business computers and not using what's familiar (Windows) is a support cost.
From this perspective, gaming and specific hobbyists are basically the only feasible audiences for the Linux desktop unless people are very much pressured by software costs, or annoyed by proprietary software (DRM, lockdowns, upgrades, etc.) enough to switch their major activity to an open source option. In which case they awkward situation of "software works better on Linux, but won't try Linux until confirmed they like the not-totally-integrated-and-nice-on-windows-or-mac software running not on Linux."
I do think there ought to be more of a business case for Linux as a business OS as you should get reduced hardware and software and support costs, but there aren't actually a lot of people with the right experience and expertise to run a business off Linux as a desktop OS to begin with and so those savings can't be realized effectively.
That said, as computers get more locked down, I think there will be a bigger drive for power users who influence friends and family to switch.
Any case, my house has had year of the Linux Desktop ongoing since circa 2006.
It first complains that the Linux wins, such as running more games, etc are the wrong wins because they make Linux more like Windows.
And yet, later, it says the reason Windows for ARM failed is the apps users wanted not running for it, and what users really want is their stuff to just work. But that’s literally the stuff the author called the wrong kind of win at the beginning.
Further, the author complains about the multiple DEs, init systems, etc and considers this fragmentation to be the cause of LOTD’s failure. And yet, getting rid of this would actually make Linux like Windows/Mac unlike adding compatibility for more games.
Finally, I think this is substantially wrong as well. The variety in Linux is what made it possible for Valve, for example, to put together the steam deck. They were able to choose the combination of different options in different areas they worked best for the Steam Deck’s use case. Further, Linux’s tremendous success on the server is also likely due to fragmentation. The fragmentation meant that several different companies could survive and flourish, such as SUSE, Ubuntu, RH, etc and each one of them could contribute different improvements that either helped the entire ecosystem or initially provided an advantage on their ecosystem which the competitors would need to come up with an answer for.
This desire for “year of Linux on the desktop” is always attributed to “Linux lovers,” and other nebulous fan voices. This article uses the word “win,” but as far as I can see the prize they are looking for is… a lot of additional non-technical desktop users? Why? What’s the goal here that actually benefits Linux or broader open source development?
I love Linux, and I use it daily for software development and gaming, but I am often exasperated by the Linux community. There's a strange combination of newbie-friendliness combined with gaslighting; if you're completely new, you'll get tons of enthusiastic help, but the moment you experience genuine frustration, you're castigated or told that your preferences are inherently wrong. I think that many hardcore Linux users don't get that some people will never grok the command line, or do not intrinsically enjoy tinkering with their OS. Luckily, I do, but I am hesitant to recommend a Linux box to just anyone.
Most of the uses for a casual computer user have been replaced by smartphones and tablets.
Then you have walled gardens like the Apple ecosystem where interoperability is superior among Apple products which cross sell each other. If you got an iPhone, now get the Apple Watch and a Macbook and all integrate well.
Then you have games, where consoles will give you a decent experience for less money.
Then you have professional users where the most common use case is office documents. This remains contentious but there are more alternatives now like web apps, MS Office clones like Libreoffice, Softmaker Office/FreeOffice, WPS Office.
Then you have specific desktop apps for specific OSes and there you are tied to an OS. This is one of the few legitimate uses for Windows I can think of.
I feel like I could have read this article back in 2004. The main benefit is that you get to choose. The other two big operating systems don't really allow much choice.
> Meanwhile, the Linux community spends enormous energy on debates that rarely affect mainstream adoption. Consider the “init wars,” where systemd sparked endless flame wars (and memes) about the proper way to boot a Linux system.
This is almost in anything. We had play ground arguments over whether SEGA or Nintendo were better. Then Playstation vs N64 vs Saturn. There was Amiga vs Atari. BSD vs Linux. Vim vs Emacs. Ford vs Chevy.
Developing GUI applications for Linux is also a huge pain. Developers not familiar with Linux need to learn so many things...
What is X11, Wayland, GNOME, KDE, d-bus, application ids, portals, etc.
Then once you have a working application, users request having it distributed as .tar.gz, snap, flatpak, you name it. Then dependencies are missing on some Linux distribution or there are random bugs with Nvidia graphics cards.
Compare that to developing for Windows, where most things "just work"
The year of Linux on desktop is not come=ing, because the year of anything on desktop is not coming, and hasn't been for maybe a decade. Every new mass-market thing runs either in the browser, or, more rarely, specifically on mobile phones. Or maybe it's a game, so it completely eclipses whatever platform experience. If not that, it's entrenched ancient desktop software, like Excel (turning 40 in a few weeks), which is also its own world.
The "desktop" itself, the underlying OS, is irrelevant to most users who are not hardcore pros, like, well, software developers.
I need to have Outlook, Zoom and zScaler working seamlessly. Not with wine or whatever but right there bam it works. Otherwise I cannot use the OS in my company.
I can install them if this is easy.
Now, I've been using Linux since 1994, wrote a small bit of the kernel at that time (you may have ran it, it was for a well known NIC) and I use it daily on my systems. Via ssh or remote dev in vscode.
I have no idea what the graphical interface is today and how to configure it. I could learn if this is easy.
One thing I know is that the sound did not work on my Thinkpad last time Ininstalled Ubuntu and the second screen would not wake up (the third did). Surely googling and chatgpting would help but in Windows 11 it just works.
(This is a dupe so I'll copy my comment on the old, unloved post from 5 days ago...)
Linux is a fish, stop trying to make it a bird. The fact that a significant portion of the Linux-user population thinks/believes/hopes/expects that it will someday be a bird won't make it so, or do anything to unblock the technical, legal, and organizational roadblocks.
If you want a FOSS desktop OS that can win the "right battles", here's what you do:
1. Come up with a name and a logo. Trademark them. Make a basic set of rules that people have to adhere to if they want to use your logo. Obviously, get a lawyer to look it over to ensure it's ironclad.
2. Fork FreeBSD (or any other open-source-but-not-copyleft-licensed kernel)
3. Pick a GUI layer. GTK, Qt, WxWidgets, doesn't really matter as long as you keep the API stable so programs written in 2028 will run in 2038 (good luck doing that on Linux).
4. Create a driver API interface so someone can write a Realtek Wifi driver once and it'll never need recompiled or updated for a newer kernel. The driver file will work in 2028 and 2038 (of course, excepting the case where there's a new CPU architecture, or a security vulnerability).
5. Stabilize the application-level API as well. That means, probably pick a version of glibc and stick with it forever. Patch vulnerabilities, but maintain backward compatibility as much as possible. Application binaries should work forever.
If your instinctive reaction to these bullet points is to think "who's gonna do all that" then yeah, I agree with you. It's not going to happen.
26 comments
[ 4.2 ms ] story [ 45.4 ms ] threadFor example, count me in with those folks who think the "new" GNOME sucks. Now, maybe you disagree and that's fine -- but so often those discussions start and end with "Well it's open source and so because you're not making anything better you can't even talk."
No. Some big players put their thumb on the scale and had a vision and a direction for GNOME and what role it would or should play; someone thought it was a good idea to try to out Steve Jobs Steve Jobs.
THOSE moves need more discussion and transparency in order to REALLY talk about "the Linux Desktop."
Its the way Windows is developing that is driving this change. GNOME might be hardly usable but Microsoft managed to top that.
Edit: I retract the last sentence. I'm currently trying GNOME and its less usable than Windows.
GNOME is nice, KDE is nice, and we have other options for people that don't like the two previous one. The issue we have now is walled garden, when some proprietary software won't support standards and even their own file format.
I disagree that this is an issue. The main advantage of Linux for me is that I have choice (including using various desktop environments that the author is annoyed by; I used GNOME for years and eventually had too many problems with it so I switched to KDE), and those choices are not controlled by one entity which, in the case of Apple and Microsoft, view me only as a customer to extract money from.
Time for a Cathedral and the Bazaar refresher?
https://web.archive.org/web/20250307173133/https://www.catb....
Where is the translation layer that lets me seamlessly run x64 apps on Linux on Arm?
-no ads
-no tracking
-no vendor lock in
-no preinstalled or unremovable crapware
That's enough for me. Yes, it's not perfect, but you're simply allowed to say no.
I don't care at all for the SystemD/whatever else flame wars. Sure if you work on these systems you probably care deeply about the differences but please realize that most of your end users do not give a shit. The same goes for the various packaging systems, I prefer to still use DEB's when I can but at the end of the day it really just comes down to how easily can I get the apps setup on my computer to get my work done, myself and most other users also don't really care.
What I care about are things like: why is multi monitor support still half assed? why does full screening my chrome window crash my monitor? why is it that half my installed apps don't conform to my theme? why is it that when I switch on X window manager instead of Wayland my wallpaper goes away?
But it seems that the folks that actually work on Linux don't care about these issues because when I ask why VSCode crashes my monitor, all I get is answers telling me to use vim or emacs, or when I complain why the themes look all janky all I get is: "well this wouldn't be an issue if you used <insert obscure window manager that requires a week's worth of configuration to get running and a steep learning curve>.
The vibe I get around these issues is that it's below most Linux developers as they are too busy arguing about some flag in the kernel or whether to use systemd or not. But those same people bitch and moan why "Year of the Linux Desktop" hasn't come yet. Figure out that these issues are not below you, they are the issues that people care about. Fix those issues and I'm positive that adoption will go up.
Sorry for the rant.
The main reason for Linux not taking off on the desktop is because most users don't care about what OS they run, they just want a computer that works. If the PC they buy comes with Windows out of the box, they're going to stick with that. Until you get manufacturers shipping PCs with Linux as the default OS, you're mainly going to see desktop Linux as an enthusiast-only option. It's no accident that one of the devices helping to spread Linux (the Steam Deck) comes with Linux as the default option.
They usually do not need to know - they just see a software centre which is app store like.
> they care that their favorite apps will work
That depends on app developers.
> that updates won't break anything (which Windows does all the time)
Already done
> and that they don't have to learn a list of text commands to make basic changes to their computers.
Already done.
She's 58 and a book keeper.
She even went so far and got some windows apps running with wine. All just with the help of a forum posts she found via a web search engine.
The real issue is that these kinds of "grandma" users maybe just don't use computers anymore. And the folks that do are joined at the hip to proprietary software like Photoshop or CAD programs or whatever else they care a lot about and don't want to relearn, and also make enough money that the costs are invisible. Or they're business computers and not using what's familiar (Windows) is a support cost.
From this perspective, gaming and specific hobbyists are basically the only feasible audiences for the Linux desktop unless people are very much pressured by software costs, or annoyed by proprietary software (DRM, lockdowns, upgrades, etc.) enough to switch their major activity to an open source option. In which case they awkward situation of "software works better on Linux, but won't try Linux until confirmed they like the not-totally-integrated-and-nice-on-windows-or-mac software running not on Linux."
I do think there ought to be more of a business case for Linux as a business OS as you should get reduced hardware and software and support costs, but there aren't actually a lot of people with the right experience and expertise to run a business off Linux as a desktop OS to begin with and so those savings can't be realized effectively.
That said, as computers get more locked down, I think there will be a bigger drive for power users who influence friends and family to switch.
Any case, my house has had year of the Linux Desktop ongoing since circa 2006.
But MacOS always gets it right: in any browser, the websites look “juicy” for the lack of a better word, and pleasant to look at.
Why can’t Linux fix this and render closer to MacOS?
Hopefully, without all the other “value added” stuff.
It first complains that the Linux wins, such as running more games, etc are the wrong wins because they make Linux more like Windows.
And yet, later, it says the reason Windows for ARM failed is the apps users wanted not running for it, and what users really want is their stuff to just work. But that’s literally the stuff the author called the wrong kind of win at the beginning.
Further, the author complains about the multiple DEs, init systems, etc and considers this fragmentation to be the cause of LOTD’s failure. And yet, getting rid of this would actually make Linux like Windows/Mac unlike adding compatibility for more games.
Finally, I think this is substantially wrong as well. The variety in Linux is what made it possible for Valve, for example, to put together the steam deck. They were able to choose the combination of different options in different areas they worked best for the Steam Deck’s use case. Further, Linux’s tremendous success on the server is also likely due to fragmentation. The fragmentation meant that several different companies could survive and flourish, such as SUSE, Ubuntu, RH, etc and each one of them could contribute different improvements that either helped the entire ecosystem or initially provided an advantage on their ecosystem which the competitors would need to come up with an answer for.
Then you have walled gardens like the Apple ecosystem where interoperability is superior among Apple products which cross sell each other. If you got an iPhone, now get the Apple Watch and a Macbook and all integrate well.
Then you have games, where consoles will give you a decent experience for less money.
Then you have professional users where the most common use case is office documents. This remains contentious but there are more alternatives now like web apps, MS Office clones like Libreoffice, Softmaker Office/FreeOffice, WPS Office.
Then you have specific desktop apps for specific OSes and there you are tied to an OS. This is one of the few legitimate uses for Windows I can think of.
Otherwise Linux is king.
> Meanwhile, the Linux community spends enormous energy on debates that rarely affect mainstream adoption. Consider the “init wars,” where systemd sparked endless flame wars (and memes) about the proper way to boot a Linux system.
This is almost in anything. We had play ground arguments over whether SEGA or Nintendo were better. Then Playstation vs N64 vs Saturn. There was Amiga vs Atari. BSD vs Linux. Vim vs Emacs. Ford vs Chevy.
What is X11, Wayland, GNOME, KDE, d-bus, application ids, portals, etc.
Then once you have a working application, users request having it distributed as .tar.gz, snap, flatpak, you name it. Then dependencies are missing on some Linux distribution or there are random bugs with Nvidia graphics cards.
Compare that to developing for Windows, where most things "just work"
The "desktop" itself, the underlying OS, is irrelevant to most users who are not hardcore pros, like, well, software developers.
I can install them if this is easy.
Now, I've been using Linux since 1994, wrote a small bit of the kernel at that time (you may have ran it, it was for a well known NIC) and I use it daily on my systems. Via ssh or remote dev in vscode.
I have no idea what the graphical interface is today and how to configure it. I could learn if this is easy.
One thing I know is that the sound did not work on my Thinkpad last time Ininstalled Ubuntu and the second screen would not wake up (the third did). Surely googling and chatgpting would help but in Windows 11 it just works.
Linux is a fish, stop trying to make it a bird. The fact that a significant portion of the Linux-user population thinks/believes/hopes/expects that it will someday be a bird won't make it so, or do anything to unblock the technical, legal, and organizational roadblocks. If you want a FOSS desktop OS that can win the "right battles", here's what you do:
1. Come up with a name and a logo. Trademark them. Make a basic set of rules that people have to adhere to if they want to use your logo. Obviously, get a lawyer to look it over to ensure it's ironclad.
2. Fork FreeBSD (or any other open-source-but-not-copyleft-licensed kernel)
3. Pick a GUI layer. GTK, Qt, WxWidgets, doesn't really matter as long as you keep the API stable so programs written in 2028 will run in 2038 (good luck doing that on Linux).
4. Create a driver API interface so someone can write a Realtek Wifi driver once and it'll never need recompiled or updated for a newer kernel. The driver file will work in 2028 and 2038 (of course, excepting the case where there's a new CPU architecture, or a security vulnerability).
5. Stabilize the application-level API as well. That means, probably pick a version of glibc and stick with it forever. Patch vulnerabilities, but maintain backward compatibility as much as possible. Application binaries should work forever.
If your instinctive reaction to these bullet points is to think "who's gonna do all that" then yeah, I agree with you. It's not going to happen.