Everyone who has used desktop Linux has asked, “why isn’t this more popular?”
And everyone who has used desktop Linux knows the answer(s).
See, there are a few game breaking issues that prevent desktop Linux from taking any significant market share. Lets talk about why.
The Steep Learning Curve
It is no secret that Linux is harder to maintain and administer than Windows or macOS. It requires command line utilities and editing config files – you can try and say everything can be done graphically, but you’d be wrong. Inevitably every desktop Linux user will look up how to get help with something, and will be met with esoteric commands for their favorite text based terminal interface.
I’m not saying the command line is a bad thing. It is an awesome aspect of Linux, and one of the most powerful aspects of a *nix system. However, for the average person, it is scary. Point blank, they don’t like it. And 90% of the time when you look up how to do something on Linux, you’re getting instructions for the command line interface.
Not to mention, understanding how your Linux system works is complicated. Not as complicated as Windows, I’ll give you that, but still complicated. Ideas such as “everything is a file” requires a rethinking of how a system works. Most people already know how to administer Windows, because it is mostly easy even if you don’t understand the system fully. On Linux, you really have to understand the system fully in order to properly maintain it and not break it.
So what is the solution? Well, look at the Steam Deck. The Steam Deck has abstracted away all the complication of a Linux system in order to make an appliance. That really is the secret, keep that in mind. But the Steam Deck isn’t complicated to maintain. The Steam Deck isn’t complicated to use. You don’t have to fully understand it to use it and keep it running. This is because of several reasons, besides abstracting away the complication it also is an immutable system. Meaning, it isn’t meant to be tinkered with. Many Linux users would scoff at this – but it is required for the OS to become more popular.
Choice Overload
Which do you want, KDE, GNOME, XFCE, or Cinnamon? How about X or Wayland? Do you want Qt apps, or GTK? All of these things might mean something to me or you, but to the average user, it is gibberish. See, here is the problem. In order to get started with Linux, you have to either go in blind (which is a horrible idea) or spend days researching what all these different choices mean.
So from the end user’s point of view, you have to spend days researching what all these different technologies mean, try them out, and form an opinion on them before you can really start getting work done. That is insane. No one is going to do that. Or rather, no more than 3% of people are going to do that. A rounding error.
What is the solution? Obviously, look at the Steam Deck. The solution is to make appliances, not general purpose machines. See, Ubuntu, Fedora, Elementary, all of these are “general purpose” OS’s. There is a disconnect here. The average user is asking “what is the best solution for what I want to do?” Specifically, “for what I want to do?” We need to make distributions that are appliances, built to a specific use case. Something like Ubuntu Studio, or Nobarra. That is the path to removing the choice overload.
Compatibility
So, we know the end user is asking “what will enable me to do ‘x’”. It isn’t about the OS. It isn’t about the interface, or the terminal. It is about doing something specific. And that happens in applications. The applications are what the user cares about. To 97% of people, the OS is a vehicle that delivers them to an application, and then should get out of the way.
The problem? Those applications they need/want aren’t available on Linux. And I’m not saying there aren’t great alternatives. There are! But here is the thing. The average user wants to collaborate with others. That means using industry standard formats. Libre Office, GIMP, Kdenlive, none of these are 100% compatible with industry standard formats.
People’s coworkers will be doing one thing in Photoshop for example, and the Linux user won’t be able to learn from them or do the same thing because they’re using the (hilariously bad, and embarrassingly named) GIMP.
We need these big Linux organizations to start courting software developers. Quit spending all this money on executives and chairmen, and get the software vendors to give a rat’s ass about your OS.
Easy to Break, Hard to Fix
I have broken dozens of desktop Linux installs. I have fixed far fewer. I’m a cybersecurity engineer. I’ve been using Linux for 12 years. I still can’t keep a desktop install running for more than a year. And often, when they break, it is faster to just reinstall the OS.
As long as it is faster to reinstall the OS rather than fix the issue, people won’t come back to Linux. They’ll just install Windows instead, or like me, buy a Mac.
Maybe it is a skill issue. Hell it IS a skill issue. But the level of skill I’ve acquired over 12 years still isn’t enough to fix the damn system when it breaks. Yes, there are 1% of uber nerds out there who can fix any Linux issue. But even for long time users, it is a daunting task. Besides, after a system breaks, let’s say you fix it. You can never be 100% sure you’ve fixed all the possible issues that you caused.
What is the solution here? Immutable distros. Appliances. Quit making general purpose desktops, and make something tailored to a specific set of tasks and court people who do these things.
The Path of Least Resistance
People use Windows because it came on their computer. People don’t LIKE Windows. People use it because it is what is already there. Asking someone to make a bootable thumb drive, boot their computer off it, go through an install process and wipe their hard drive – insane. 97% of people will never do that.
See, people don’t care about what OS they’re using. People care about investing the minimum amount of time necessary to get done what they want to do. And that means – putting up with Windows or buying a Mac.
How can we solve this? Well, we need more computers that come with Linux out of the box. And they need to be advertised. They need to be marketed. Look at ChromeOS. It is based on Linux. And people love it. It isn’t impossible – but it is an appliance. A Chromebook is an appliance made for browsing the web. It does one thing, and abstracts away all the underlying complication. That is why they are popular. And because they come installed by default, and are advertised.
Conclusion
People don’t care about their OS. They care about what they can do with it. We need to shift the narrative from being about an operating system, and make it about performing common and specialized tasks. What makes Linux better than Windows? It isn’t the graphics, it isn’t the lack of spyware, it isn’t the fact that it is lightweight. It NEEDS to be “it does x better than Windows”. That
Leave a Reply