tldr: For Linux adoption it would be better for devs to focus on 2 (“main”) distros which are very similar to Windows and macOS and then 2-3 further (“big”) distros which give a bit more room to experiment. All the other distros create confusion and analysis-paralysis for the user who wants to switch or wants to help others to do the switch.
Edit because some people got emotional and I was being imprecise: Disclaimer: I dont want to dictate any foss dev, I understand that “Linux” isnt a company. By “Linux” in this post I only ment the desktop OS for personal and work use.
— (sorry for the long paragraph, i ranted and brain dumped the idea)
I see a problem: Even “stable” distros like Debian and big and “fully developed” DEs like KDE or GNOME arent ready for the majority of the users switching from windows. Missing software compatibility and the need to fall back on the commandline are just some of the problems. The biggest one is the confusion for the average user: They google “install Linux” and then need to do research for at least 30minutes, figuring out which of the popular distros is the right one for them. If decided, then (depending on the distro) they then have to choose the DE.
Its a sinilar problem to the adoption of the Fediverse: You are expected to decide what instance you want to be part of. This makes it also very hard for a linux enthusiasts to convince/help install a distro for a family member, as you dont know their preferenced or how they use their Win/Mac machine. So either you as an expert have to observe and then do research on what distro+DE fits the usecase or the enduser themselves need to distro-hop, which is obviously not happening.
Now you are thinking: But just install Linux Mint and they probably do most of the things in their Browser anyways.
But in my experience the switch of potentially the browser, the mail-client and ontop of that the OS is a pretty tall ask for an average end user. So the whole switching thing becomes a multi year operation where they first switch the software they use to FOSS one. Which is a tall order and it makes it even harder to explain and convince someone. Heck, it already takes multiple days to get my grandma up to speed after the change Win10 -> Win11, because some buttons moved and the context menue looks different.
Now my utopian idea: If there were only a handful of popular distros+DEs, one could map them on a 2D-plane or even on a spectrum of “fixed, you have to adapt” to “flexible, you have to adjust the settings”. Mac users could switch to a distro which is quite fixed (comparable to macOS). This fixed distro should out of the box be close to the mac experience. With windows the same.
Very very rough prototype of the spectrum to visualize my idea. I dont know enough about it but tried anyways:
flexible <Arch + Hyprland>
<Debian + KDE Plasma>
Windows 10
<Linux Mint + Cinnamon>
MacOS
<Debian + GNOME> fixed
If then most of Linux Devs (from Kernel to distro to UI to software) mostly focus on the 4-5 main distros, then they would get more stable and they could be made to behave closer to their proprietary counterparts.
This then could make the switch from Mac/Win so much more easy because:
-
The distro is closer to the old proprietary OS. So the enduser just has to learn other “new” software, the OS doesnt demand a learning curve but just replicates the Win/Mac experience.
-
The decision which distro to use is easier, as there are the main ones which are easy to choose because they are distinct from one another.
Disclaimer: No, i am no expert, I probably dont know enough of the technical side, I just wanted to share the enduser experience. Obviously there will always be countless distros by enthusiasts who tinker with their dozends of dev-friends for their personal-perfect distro. There will always be the people who deliberately do some frankensteined distro, and I am not here to forbid any of this. The confusing diversity of all the options is just not helping the wider public.
I feel like there really are just 2 or 3 main distros for Linux adoption. Every article, forum, discussion, etc… it’s always Mint, followed by either Fedora or Ubuntu. IMO distro is less important for converts than desktop environment.
I think the most important thing for adoption is actually little quality of life stuff.
- The first question during installation should be “are you new to Linux” and if you select yes it doesn’t ask you about file systems or partitions it just installs the goddamn operating system with a default configuration, and casual friendly software.
- Photo and video thumbnails that just work.
- An idiot proof way to get a video player with support for every video codec.
- More GUI based “intermediate” applications. If Grandpa has to figure out samba config files just so he can open up his photos on his laptop he’s going to second guess his decision.
An idiot proof way to get a video player with support for every video codec.
Is this really still a problem? I haven’t used anything but Gentoo (and some Arch) for 18 years, so for me it hasn’t been a problem for a long, long time. By now it should be clear that these patents are unenforceable and distros, especially non-corporate ones should just ignore them.
Kinda. It’s not hard, but it’s also not idiot proof.
On Fedora for example you just need to use RPM Fusion instead of the standard Fedora repos. The problem is that you need to know that you need to use RPM Fusion.
Fedora is a pretty common recommendation to new users (with good reason it’s excellent) but plenty of casual users will run into that problem and decide that videos don’t work right on Linux.
I’ve seen this opinion voiced quite a few times for the last 28 or so years I’ve been a Linux user.
Guess what? It’s free and open source software. People work on what they feel like when they feel like if they feel like. You can’t mandate “let’s just have a couple of distros, think of the public!”. It doesn’t work like that. Yes, life is not perfect.
Furthermore, Linux (as a whole) is not a for-profit project, or a singular organization.
Desktop Linux is far from it’s only purpose, and many of the devs are far more interested in their own use-cases: servers, embedded systems, supercomputers, phones, special purpose OSs. Wikipedia even has a page for the wide range of use beyond desktops and servers. So we can’t simply treat devs as a unified group with a common goal like we can generally do with Microsoft, Apple, Google, Steam, etc. unless you pick a particular distro!
Yes, sorry, that wasnt my intention.
When I talked about “Linux” or “distros” i only ment Desktop OS for personal use. Sorry!
It’s fine.
Every Linux user goes through this, because the freedom means choice, and choice means lots of options.
That’s alright, and I’m also a little bit sorry for nitpicking! I just saw it as an opportunity to illustrate how complex this whole software mess is.
I’m not sure if you’ve come across it yet, but there’s a well-known copypasta posted to satirize the way many Linux users will nitpick terms.
I’ve read that before, didnt realize that this was a joke…
Exactly. I need Debian, Alpine, Manjaro, OpenWRT, MoOde Audio Player, Lakka and SteamOS.
They all serve different use cases. That’s the beauty of it, the utter flexibility to turn it into whatever you need because you can.
100% agree
Agree! I dont want to dictate anyone and I understand that my rant wont change anything.
It was more about the hypothetical optimum “if we one wanted to optimize for user-share of the desktop OS market”, then there should be fewer but better distros.
My feeling on this is that or the “general public desktop” use case we have to defer to corporate supported distros (RedHat, Ubuntu, Suse), because they have to work with hardware vendors that are typically averse to the idea of sharing driver code, and you have to make sure your desktop runs smoothly on your average PC.
I don’t see it happening, honestly.
Tldr: Why can’t Linux be more like Windows! With less options and more corporate control!?
No corporate control, not less options.
There need to be a handful of distros with less settings, being more “fixed”.
I personally like to being able to set everything how I like it.
But it just isnt necessary to have THAT many actively developed distros. If devs could drop 1/4 of the distros and redirect their efforts on bigger projects, those would profit massively.
You’ve given it a lot more thought than the average person pointing this out, so thanks, but ultimately the same stuff that makes Linux and Fediverse good is what prevents most people from adopting them, you can’t really have it both ways, and when you try you just end up with the worst of all worlds like Canonical with Ubuntu and it’s enshittification, GNOME with their extreme vision causing mass user flight to KDE, or .world having what seems to be a printer of dumb motherfuckers in any comment section.
I guess its in the nature of open source. However really need to get on top of the search results of “install linux”. I like the End of 10 campaign, however i also have just noticed that in their install guide they don’t specify where to get Linux exactly just
Download the operating system you want to install. Search for Linux distributions for beginners to get some suggestions.
I’ve personally always been of the opinion that it’s great to have so many different options, and i also never understood why people get so stressed about choosing a distro. You can always switch to something else later if you find something better, but first you need to find out your likes and dislikes about a distribution to begin with. Pewdiepie said in his linux video “just pick one!”, and that’s how i went about it as well. When i saw an LTT video about trying linux instead if windows 11, they recommended PopOS, so that’s what i chose. Ended up trying a bunch of different ones later, and as of right now i’m on Void linux.
I agree with the sentiment because it is a pain to find a distro which you want. But the reason for this is that Linux has given you the luxury to pick and choose what distro and DE you want. When you go to Windows or Mac, people just accept that it is what it is.
That being said, I will blame the Linux community to some extent for promoting “complicated” (like Arch) or too barebones distros (like Debian) to newbies. The shock of moving from Windows to Linux is already a hurdle for most. When you add the need for tinkering and troubleshooting from day one, I can see why people would quit.
We are indirectly focusing on a handful of “distros” as most distros ship with KDE, Gnome or something similar.
True! And for an enthusiast who wants to spend only a few days on finding a distro and setteling into it, like me, its nice to have only three (big) DEs, which you can test and choose in one day and then are set for the further journey.
Now “bundle” a distro to each DE and a newbie would have that experience for the whole distro-finding-experience
the average user isn’t going to install linux; the average user is going to use the linux that came installed on their computer.
It’s like cooking vs going to McDonald’s. Lots of choice and thinking vs. being fed with whatever they put you on a tray.
The confusing diversity of all the options is just not helping the wider public.
Agreed that that’s the case, but don’t quite agree that that’s a problem
It is a problem if the goal is to increase the “personal desktop OS marketshare”.
But diversity is a good thing for itself I agree. I have the feeling that it is a bit sad that it seems that there isnt one or two “main” distros, which one could recommend that tech illiterate family member.
The problem isn’t diversity of distributions. The problems are people who go on describing history of GNU/Linux when a newbie asks them what distribution to start with; and ‘top 10 Linux distributions’ articles which litter the Internet. Just the other day someone shared a link to Distrochooser, a website which gives newbies ten distributions to pick from.
When a newbie asks about Linux, point them at Linux Mint Cinnamon Edition and that’s it. Or at most ask if their primary use-case is playing games in which case recommend Bazzite. That solves the ‘problem’ of distribution proliferation.
See also New to Linux? Stick To These Rules When Picking Distro.
I switched to LMCE when Win 10 becoming unsupported started being announced. I fucking love it. Sure it was different from what I was used to, but it worked, right away. I was able to install software that did what I needed with 0 hassle with very few exceptions and now I’m used to it. I also have no urge to distro hop, but I’m glad that the people who want to can.
which one could recommend that tech illiterate family member.
I’ve given up as their thinking is so fundamentally different, and they refuse to meet even one inch towards the middle :-)
Your problem is that you’re starting from the wrong premise: the primary goal of most people working on Linux is not to make more people switch to it, strange as that may sound, it’s to create an operating system that they personally want to use. Which can mean a lot of different things, depending on the person. So it’s inevitable that there are a lot of different distros, and the only reason there aren’t even more is that most of the one-man shows that don’t attract many users peter out and vanish after a few months or years.
Well put, thanks
What you say could be boiled down to:
If just everybody concentrated on making quality software for Windows, we could have much more quality software for Windows.Your view is common but irrational.
The concept is described by Linus Torvalds as “Scratch your own itch”. The richness and diversity of Linux distros is a strength not a weakness.If you want to make a Barbie themed distro you can. And if you want to, why should anyone try to prevent you?
If development was concentrated around fewer distros, it is far from a sure thing this development would go in the direction you would personally want. You would just have fewer options.My first thought as well :)
No, I dont suggest that there should be another new distro. The existing ones are pretty good at the goal of being userfriendly, at least some of them like Linux Mint
But mint wouldn’t be possible without the work that the Ubuntu community puts into making a stable and polished distribution. And Ubuntu wouldn’t be possible without the Debian community who put in the effort to make the distribution the best for their usage.
Bee hearing that since 25 years ago
