Desktop Linux Keeps Winning the Wrong Battles
howtogeek.com
howtogeek.com
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.
It is a massive moral failure though. It shows that after two decades of work, the Linux community has been unable to build a simple sane functional stable development environment better than Win32.
Even Apple’s famously fast deprecation is like rock by comparison.
It literally means Windows will always exist - as the preferred IDE and Reference Spec for the Linux desktop. It also means all evolution of Linux will be ironically constrained by Win32 compatibility requirements.
Surely WSL is not a moral failure for Microsoft.
Competition can definitely improve things, but it's not universally positive. In particular, endless competition in parts of the operating system makes it hard to build anything on top of them. E.g. if you want to distribute an application for Linux, do you build a Flatpak, or a Snap? Or take a more traditionalist approach and make RPMs, DEBs, etc.? You either pick your favourite and leave out a large fraction of Linux users who disagree, or you have to do more than one of these. This is definitely a drag on the ecosystem.
I agree that most users don't care about the OS, though.
On pretty much every other operating system out there, I as the application author have control over the release cadence of my software. On some platforms, I can simply release an installer and be done. On others, I send the app through some form of certification process, and when it's complete it appears on the App Store.
Linux is rather unique in that it has tasked the distro maintainers to be tastemakers and packagers of the software that appears in their repositories. They control what version is used, what libraries it's linked against and their specific versions, how often it's updated, which additional features are added or removed before distribution, et cetera.
From the outside, this is complete and utter insanity. In the old days, you had to either static link the universe and pray or create a third-party repository per specific distro. Thank goodness these days we have numerous ways to cut distros out of the loop - Docker, Flatpak, AppImage and of course Steam if you're a gamer.
On top of that, some of the biggest Linux distros will only release application updates with their own OS releases, on a cadence of months or years. Software developers expect to be able to deliver updates in hours to days (at least in the consumer space - highly regulated things like banking are different).
There are good reasons why RedHat and Canonical, the companies behind some of the biggest Linux distros, are pushing app distribution systems (Flatpak & Snap) which aim to be cross-distro and have upstream developers directly involved in packaging. There are absolutely downsides to this approach as well, but it would be nice if we could stop pretending that traditional distro packaging is a marvellous system.
.appimage
RedHats and Canonicals paying enterprise customers are whats keeping the Linux ecosystem alive. No one else brings the required manpower to the table.
This is a narrow view about how innovation happens in Linux and related software. Yes, Linux-focused companies are driving many of the changes, but there is plenty of exploration of ideas that happens outside of those companies.
Like, innovative and fresh stuff is cool, but at the end of the day you need to keep your business running and not breaking down.
I even had a look at source code for makepkgs in Arch and they are literally the same commands in the script that the book has you manually type in.
The packaging critique comes up over the years but it is a bit of an overblown.
Building packages for different distributions isn't super difficult. I've build Arch Packages using the ABS, DEBS and RPMS and they are all conceptually the same. Having a quick skim of the Flatpak instructions it doesn't look that different either.
If you don't want to bother with all of that. You can just have a script that drops the installation either in /opt or ~./.local/. I am pretty sure Jetbrains Toolbox does that, but I would need to check my other machine to confirm and I don't have access currently.
The OS does make some difference, but I think that if all the same applications were available, a lot of people could switch without much difficulty. In some ways going from Windows to Linux might be easier than Windows to Mac, because many Linux distros are happy enough to conform to UI conventions that are familiar from Windows.
It's entirely possible for a large enough brand to ship a Linux based desktop OS to mass adoption. It has already been done once with ChromeOS.
Linux will never be the name users remember, and it's not meant to be.
I would venture to guess that kids know the difference between Mac and Windows and probably prefers one over the other.
Never underestimate the identity association in enthusiast communities.
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.
There will always be more “Linuxy” out in the weeds desktops for people who want them. Most people who want that built their own setup anyway, making whatever the big DEs do more or less moot.
The biggest battle desktop Linux is losing is the one where a minority of devs are dictating their preferred compute paradigm to a majority of users that don’t agree it’s a good solution.
I can “fix” Gnome in about 2m with extensions, but that doesn’t help when a new user loads it up for the first time and is hit with the unintuitive ideology of some nerds.
But your use case is why GNOME have extensions. To alter the defaults and add stuff that they don't care about, but you do. In macOS, you have to basically reverse engineer and use private APIs.
Absolutely. A commercial product can succeed while maintaining an auteur's vision, so long as that vision largely aligns with users' needs. In contrast, open-source projects are often not viewed through a "product" lens, to their detriment.
When this happens in open source, we get clunky and idiosyncratic (though sometimes lovable) software like GNOME and GIMP. When it happens in the commercial world, we get projects like Megalopolis.
For example, how am i supposed to discover how to maximize a window?
I don't even understand why they thought it would be a good idea. On mobile there is at least the justification that it helps keep the valuable screen estate for "important things" (although anecdotally I've seen how much more my mother and grandmother both struggle with button-less iPhones because they can't remember the gestures). But on the desktop, even on laptops, it is a very weak argument.
Gnome made several decisions toat make it super complex for normal users. Windows have a taskbar. Mac have a dock. KDE have a taskbar.
And I really like Gnome design language, UI. But the UX is awful.
I'm not sure that means anything. Every time I have to help my kids or wife with a Windows problem, I'm perpetually plagued by how weird it is.
The only people who find Windows easy or obvious are already Windows users. And yes, the same can be said of Linux environments.
-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.
Every distro except arch is full of preinstalled crapware. Some like OpenSuse even have preinstalled crapware bundles where stuff you uninstall comes back after update
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.
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.
I do understand why people bash Gnome and their developers. The hate is deserved.
[1] https://discourse.gnome.org/t/feature-request-show-when-apps...
This is my first Linux desktop success after several attempts over the past two decades.
I still remember my first attempt (probably late-90's) - it took me 3 hours to figure out how to get an mp3 to play.
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.
The "desktop" itself, the underlying OS, is irrelevant to most users who are not hardcore pros, like, well, software developers.
> 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.
Time for a Cathedral and the Bazaar refresher?
https://web.archive.org/web/20250307173133/https://www.catb....
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.
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.
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.
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.
I think one of the reasons is that if you think it's better and see how other users are struggling with all the bs these companies are throwing at them, it's only natural that you would suggest what works for you.
There is also another reason, though, and this is more on a platform level, i.e. think globally, act locally. If more users are on Linux then companies will start to target it as a platform that requires first class support. So, then, even if payable, we would also get native Linux apps for enterprise. This would be a win.
As it stands, Steam was the big push but we need more companies to adopt this approach. Don't get me wrong, even wine was amazing, but Proton is such an easier experience that makes it all a breeze.
So, why? In the long run, it will be a better experience for all of us, Linux native users.
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.
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.
Where is the translation layer that lets me seamlessly run x64 apps on Linux on Arm?
apt install qemu binfmt-support qemu-user-static
sudo dpkg --add-architecture amd64Windows is an operating system made by Microsoft for PCs, macOS is an operating system made by Apple for Macs but a product call Linux Desktop does not exist.
There is Ubuntu, REHL, Debian, Fedora, Arch and many other operating systems based on the Linux kernel - which is a just a kernel - that are built for different purposes, none of them targeting mainstream desktop usage.
In the history of operating systems based on Linux there were a few products targeting mainstream desktop users like Lindows and Ubuntu, but the only really successful, at least for some time, was Ubuntu.
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"
This is a weird complaint, and I'm saying this as a mostly-Windows developer. Windows has an insane number of technologies, many of them replacements for older replacements for older replacements ... etc going 30+ years back. Do you know how registration-free COM works, for example? Probably not, yet it is there, and if you go to the Windows section of MSDN and just start reading through the docs, you'll have to read through that too.
You can reasonably argue that it's not a thing that you actually need to know to write an app for Windows, and it's true. But similarly you don't actually need to know what d-bus is, never mind how it works, to write an app for Linux. If you use a reasonably high-level framework such as Qt, it takes care of all that for you. Hell, you can even go for .NET Core + Avalonia to get something very similar to WPF on Windows.
And then there's Electron, which lets you pretend that everything is just a browser...
Nah, writing the apps for Linux isn't difficult at all. OTOH you're right in that distribution is a pain, and end-user support across numerous wildly different distros can also be a pain (especially when they start packaging a patched version of your code in their package repos!).
Yeah that part is just bonkers. Take Discord, for example, probably the most popular "app" that people install on Linux these days. There is a Snap, a Flatpak, a DEB. I can't imagine how much work it is to maintain all of those. Screen sharing is broken in random versions. People say to install the DEB (if you're lucky enough to be on a Debian-based system), but oh, make sure you install via a terminal command with "--fix-broken" to install all dependencies (wouldn't the app store GUI do that? maybe not). Apparently the Flatpak is better, but don't forget to install something called "flatseal" so you can upload files (Flatpaks are sandboxed by default). I imagine Discord tech support gets a lot of angry emails from Linux users who didn't get the "flatseal" memo and think that Discord just sucks at developing software.
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.
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."
Shuttleworth did so much for desktop Linux by mailing us all free install CDs, but then users everywhere began to expect Linux to be free as in beer.
But until users are willing to pay for desktop Linux again, like some of us may remember, enterprise will always win out.
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.
I see this issue as well. A CLI setup with Emacs/VIM doing C/C++ development is very stable, because that's how the majority of linux devs interact with Linux.
What puts a bad taste in my mouth is when you mention issues outside of that setup, the usual response isn't "oh this is an issue we need to fix", it's "well your setup sucks, stop using VSCode/Gnome/Chrome/etc"
If you want to know why VSCode is buggy, you will probably have to get into contact with Microsoft, I guess. I know this looks like some kind of smarmy sarcastic response, but it really is the truth; the rest of us really don’t know why they put bugs in there.
I've gone to the extreme of writing my own OS because I got fed up with how corporate Linux has gotten
It can't just be a means to an end; that's what happened to Android. You make it an expectation that everything on a given box be FLOSS.
It shipped with WordPerfect Office, which was still neck and neck with Microsoft Office at the time, including WordPerfect, Quattro, Corel Presentations, Corel Draw, all there.
It had migration tools to move Internet Explorer, Netscape, Outlook, mIRC, and ICQ settings, and Windows registry settings to Corel Linux.
It shipped with a Wine-based Windows application compatibility layer out of the box.
It lasted less than a year.
For desktop Linux to be successful, it needs to be the cool different OS -- not the boring Windows that the accountants use, and not the MacOS that the art snobs and marketing hipsters use, but something appealing on its own merits.
And, arguably, that's about where it is now: 70% boring, 15% commercial hipster, 15% indie.