I would argue the fragmentation is a feature, not a bug. Part of the “free” is to be able to do whatever you want. There’s a price to pay for it but unifying requires discipline and governance. And governance requires authority. And the whole point is we can write whatever the hell we want.
I'd argue that fragmentation is also a symptom of a lack of truly high-quality solutions, where people keep reinventing wheels because they think they'll solve problems in the existing solutions (and end up creating a different set of problems).
Having a group of people come together and agree on some structure (i.e. governance) doesn't take anybody's freedom away. But such a group could hopefully achieve more than all the "lone wolves" or small teams on their own.
I think the big difficulty is that every person has their own taste and preferences when it comes to the desktop UI/UX and doesn't want to contribute to something that is somewhat different from that. It's easier for a company / organization which can hire people. The payment makes up for any small misalignment of ideas.
But the article also talks about compatible interfaces, so all those fragmented projects could be used together in a meaningful way.
In some ways this reads to me as follows: 'why are people wasting time making things they want to use instead of just using this thing that already exists?' when the answer is obviously that they aren't happy with the thing that already exists. Why did Bram Moolenaar create vim instead of just using vi or ed?
Excellent analysis of the problem. Universal config parsing, UI/ux consistency, and text stream support are foundational to a unified Linux. I would like to see a concerted and organized effort to unify coding progress in a similar direction... Sounds a lot like a software development team, that would want money
A lot of comments here about how diversity is great etc etc. Yes, but also a lot of people value consistency. Now that Win10 is gone and people are reluctant to use Win11, it would be a great opportunity to switch. Yet even I am still reluctant to switch entirely (after using Linux as my daily driver some years ago) due to the lack of UI/UX consistency (I have to use Ubuntu at work; have Mint in my jellyfin pc at home, and even between those there are a lot of annoying differences)
The problem is that Gnome/KDE/XFCE and others aren't strictly-speaking Linux-specific, but support other Unix-like operating systems. This makes possible standardization much more comlpex.
I think the fragmentation is good, it allows many talents working in the same area at once and lets users choose what fits best. The problem is that this leaves a lot of the polishing work to the users, and the defaults - which most stick with - are often the most boring, safe choice.
I think what DHH did with Omakub (and Omarchy) was a constructive solution here - use the myriad of options to pick out a set of components and configs that work well. Polish the selections, hide and ignore the capabilities that don't fint in, and document how to use the resulting "bundle" in great detail.
I'd say if you want a unified desktop, you create an organization, fund it and hire all the desktop developers globally to work on your vision for a unified baseline desktop.
I don't have to like Gnome's org, or the fact that KDE uses Qt, or any number of other bits... Not to mention System76 doing their own thing with Cosmic, which has had a rather impressive development pace. They're all doing their thing and running their projects. Much like MS doesn't guide MacOS development.
In the end, people can use what they like... I'd rather not exist in a space ruled by totalitarian/communist/fascist or any other centralized national constraints in place.
Even if it is wasteful, and it is, people are allowed to waste their own time that they put into a hobby or otherwise donate. For those getting paid, or paying for it, cool, for those not, they are free individuals. I've often said, if I didn't have to work for a living, I'd create a better, floss version of Exchange+Outlook. I don't have to join an existing project, or convince everyone to join mine.
Oh no, let's not. Let's not turn the Linux desktop into a locked down Microsoft or Apple shithole.
Because, IMO, it is one of the most beautiful features of the ecosystem. If you get tired of KDE, you can easily switch to GNOME on the same system and still keep all the things you care about. And if you decide that GNOME is not your thing you can easily revert back on the same system. And unless you are running Gentoo (wink, wink) that only takes a cup of coffee on modern hardware.
Better yet, if you get tired of both you can get something like Mango (https://github.com/DreamMaoMao/mangowc) or Hyprland (https://hypr.land/) and hack your own DE together, or if you really have spare time to shed you can pick up something like QuickShell (https://quickshell.org/) and build a DE pretty much from scratch all by yourself. That's how caelestia (https://github.com/caelestia-dots) came to be, which is also prominently showcased on the QuickShell homepage. It's just that beautiful (and useful too).
Speaking of useful, from someone who installed Linux for the first time way back then when CD-ROMs and Windows XP were still things and Slackware was the most popular distribution the UX of all desktop environments has improved a lot.
And that's actually a understatement. I'm currently migrating (for real this time, and for good) and also want to ween my parents off Windows 10 so I casually slapped a KDE Neon live CD on a USB stick in order to demonstrate that the UX in Linux has improved a lot over the years and is nothing to worry about.
After booting up from the stick I was absolutely blown away myself, because of how good it was. Hell, I can imagine if less people get freakish about Linux a lot of people will finally make the switch. The UX IMHO is so damn good compared to Windows 11 or current OS X for beginners to advanced users alike (in case of the latter meaning that if you want to customize your shell to oblivion within the borders of KDE's UX guidelines go right ahead, nobody is stopping you)
TL;DR: Nothing needs to be changed. The only thing what is required is that people learn there's more than macOS Big Sur or Windows 11, and that these alternatives are just as easy to use.
As much as I will never touch the DE ever again, GNOME's reliance on simple text files made it an absolute joy to use with Nix home-manager. I would be 100% on board with using gnome settings infra across the board.
GNOME settings? Where you need to use 3 different applications to change the settings (settings, tweaks, and dconf)? To be honest, it's been a while since I last tried GNOME, but I don't remember it being very convenient.
Which is why Chrome OS, Android, WebOS, and Steam OS have the lions share of consumer attention, not caring that their devices are powered by the Linux kernel.
I'd settle for a consistent way for apps to get the selected DE's text settings and baseline color palette. Even that is inconsistent between DEs and distros.
I don't know who is the writer of this article but it is a huge load of horse shit...
For the rest, regarding the diversity of DE, we can lament the lack of a strong contender for the general public, Ubuntu was about to become that before their ego took over, but otherwise that is the true value of Linux stacks since the beginning that you can have it and tweak it to your taste...
Not really. I have invested a lot of time and effort to finding and configuring my "perfect" desktop environment. I failed. As soon as you start tweaking (i.e. changing the default settings) and install some additional tools to fill some gaps, you discover loads of bugs and inflexibility.
I feel similarly... I spent about a year with Budgie and really liked it... I had it mostly tweaked to my liking. In the end, I felt better just going back with Pop and keeping it close to defaults. I don't really tweak my mac or windows settings that much either.
In the end, I just want a pretty solid foundation to get work done.
There has been a consistent desire to unify the desktop environments, but the fragmentation has largely been because of differing use cases and philosophical perspectives.
This divide is being further driven on the issue of X11 vs Wayland, and now the drama of decentralized libertarianism vs centralized corporatism. The latter manifests itself as a culture war over the code of conduct or woke software. Now it is coalescing into a political line between hyprland & X11 and GNOME & Wayland. (hyprland uses Wayland, but it and X11 have a similar political affinity by loud and divisive proponents.)
The woke have an affinity with centralized corporatism and want to unify freedesktop collaboration under it and ensure that there is identitarian representation, so contributors don't have to worry about petty discrimination and office politics.
The opposition have an affinity with decentralized libertarianism, and they reject identitarian politics as they want their personal freedom to do what they want even if it is not politically correct, believing that this is the best way for their ideas to flourish into better software.
I personally believe there's some merits to either side, and a fine balance has to be made. We can try to get the best parts (the best software) without the bad parts (discrimination and office politics).
I'm a bit mixed on things as well... I like some of the technical solutions that are coming from the more woke organizations, but I really don't like the abuses under their Code of Conduct or war against the non-woke. Such as activists inside Gnome calling to reject funding from Framework because they're also funding Omarchy and Hyperland.
Identitarian politics can either be explicitly enforced or implicitly allowed. The woke chose the former. The opposition can do the latter but they also believe in checking identitarian interests in favor for personal freedom.
There are concerns over this escalating into fascism, but the logical extremification of ideas only muddles the waters and makes communication difficult. A pragmatic and balanced solution gets moved out of reach, and as a result, the corporate watchmen can push lighter opposition to the extreme fringes. Who then watches the watchmen?
I would argue the fragmentation is a feature, not a bug. Part of the “free” is to be able to do whatever you want. There’s a price to pay for it but unifying requires discipline and governance. And governance requires authority. And the whole point is we can write whatever the hell we want.
I'd argue that fragmentation is also a symptom of a lack of truly high-quality solutions, where people keep reinventing wheels because they think they'll solve problems in the existing solutions (and end up creating a different set of problems).
Having a group of people come together and agree on some structure (i.e. governance) doesn't take anybody's freedom away. But such a group could hopefully achieve more than all the "lone wolves" or small teams on their own.
I think the big difficulty is that every person has their own taste and preferences when it comes to the desktop UI/UX and doesn't want to contribute to something that is somewhat different from that. It's easier for a company / organization which can hire people. The payment makes up for any small misalignment of ideas.
But the article also talks about compatible interfaces, so all those fragmented projects could be used together in a meaningful way.
In some ways this reads to me as follows: 'why are people wasting time making things they want to use instead of just using this thing that already exists?' when the answer is obviously that they aren't happy with the thing that already exists. Why did Bram Moolenaar create vim instead of just using vi or ed?
I read it as "why don't people doing their own things do mines instead?"
Excellent analysis of the problem. Universal config parsing, UI/ux consistency, and text stream support are foundational to a unified Linux. I would like to see a concerted and organized effort to unify coding progress in a similar direction... Sounds a lot like a software development team, that would want money
A lot of comments here about how diversity is great etc etc. Yes, but also a lot of people value consistency. Now that Win10 is gone and people are reluctant to use Win11, it would be a great opportunity to switch. Yet even I am still reluctant to switch entirely (after using Linux as my daily driver some years ago) due to the lack of UI/UX consistency (I have to use Ubuntu at work; have Mint in my jellyfin pc at home, and even between those there are a lot of annoying differences)
I'd rather not. I like choice. You can disagree with my choice, but isn't it great that you can disagree?
The problem is that Gnome/KDE/XFCE and others aren't strictly-speaking Linux-specific, but support other Unix-like operating systems. This makes possible standardization much more comlpex.
FreeBSD 15 is planning to ship with an option in the default installer to setup a KDE Plasma desktop!
Why, Windows itself ships with multiple Windows desktops these days.
Nope, everything boils down to Win32 and COM.
The ability to replace Windows shell with something else, like AfterStep clones, is long gone.
Man, I miss those days.
I think the fragmentation is good, it allows many talents working in the same area at once and lets users choose what fits best. The problem is that this leaves a lot of the polishing work to the users, and the defaults - which most stick with - are often the most boring, safe choice.
I think what DHH did with Omakub (and Omarchy) was a constructive solution here - use the myriad of options to pick out a set of components and configs that work well. Polish the selections, hide and ignore the capabilities that don't fint in, and document how to use the resulting "bundle" in great detail.
I'd say if you want a unified desktop, you create an organization, fund it and hire all the desktop developers globally to work on your vision for a unified baseline desktop.
I don't have to like Gnome's org, or the fact that KDE uses Qt, or any number of other bits... Not to mention System76 doing their own thing with Cosmic, which has had a rather impressive development pace. They're all doing their thing and running their projects. Much like MS doesn't guide MacOS development.
In the end, people can use what they like... I'd rather not exist in a space ruled by totalitarian/communist/fascist or any other centralized national constraints in place.
Even if it is wasteful, and it is, people are allowed to waste their own time that they put into a hobby or otherwise donate. For those getting paid, or paying for it, cool, for those not, they are free individuals. I've often said, if I didn't have to work for a living, I'd create a better, floss version of Exchange+Outlook. I don't have to join an existing project, or convince everyone to join mine.
Oh no, let's not. Let's not turn the Linux desktop into a locked down Microsoft or Apple shithole.
Because, IMO, it is one of the most beautiful features of the ecosystem. If you get tired of KDE, you can easily switch to GNOME on the same system and still keep all the things you care about. And if you decide that GNOME is not your thing you can easily revert back on the same system. And unless you are running Gentoo (wink, wink) that only takes a cup of coffee on modern hardware.
Better yet, if you get tired of both you can get something like Mango (https://github.com/DreamMaoMao/mangowc) or Hyprland (https://hypr.land/) and hack your own DE together, or if you really have spare time to shed you can pick up something like QuickShell (https://quickshell.org/) and build a DE pretty much from scratch all by yourself. That's how caelestia (https://github.com/caelestia-dots) came to be, which is also prominently showcased on the QuickShell homepage. It's just that beautiful (and useful too).
Speaking of useful, from someone who installed Linux for the first time way back then when CD-ROMs and Windows XP were still things and Slackware was the most popular distribution the UX of all desktop environments has improved a lot.
And that's actually a understatement. I'm currently migrating (for real this time, and for good) and also want to ween my parents off Windows 10 so I casually slapped a KDE Neon live CD on a USB stick in order to demonstrate that the UX in Linux has improved a lot over the years and is nothing to worry about.
After booting up from the stick I was absolutely blown away myself, because of how good it was. Hell, I can imagine if less people get freakish about Linux a lot of people will finally make the switch. The UX IMHO is so damn good compared to Windows 11 or current OS X for beginners to advanced users alike (in case of the latter meaning that if you want to customize your shell to oblivion within the borders of KDE's UX guidelines go right ahead, nobody is stopping you)
TL;DR: Nothing needs to be changed. The only thing what is required is that people learn there's more than macOS Big Sur or Windows 11, and that these alternatives are just as easy to use.
As much as I will never touch the DE ever again, GNOME's reliance on simple text files made it an absolute joy to use with Nix home-manager. I would be 100% on board with using gnome settings infra across the board.
GNOME settings? Where you need to use 3 different applications to change the settings (settings, tweaks, and dconf)? To be honest, it's been a while since I last tried GNOME, but I don't remember it being very convenient.
And dconf settings are stored in a binary database, aren't they?
Oh, you're completely correct. I got confused by dconf dump and import (which is the text format), the latter home-manager does transparently.
As for the grand parent comment, yeah, I guess the UI situation sucks - but I did everything through home-manager anyway.
Only a single entity can do this, if you actually want it done then fork them all and wrap them with your Linux specific API.
Which is why Chrome OS, Android, WebOS, and Steam OS have the lions share of consumer attention, not caring that their devices are powered by the Linux kernel.
I'd settle for a consistent way for apps to get the selected DE's text settings and baseline color palette. Even that is inconsistent between DEs and distros.
Isn't the beauty of Unix the amount of options in FOSS built specifically for it, at this point? Why would you restrict anything in this space
Hardly, the UNIX wars became the Linux wars.
MacOS is still there...somewhere
Yes, I did not mentioned that all UNIX wars were fully gone, they are only a bit less relevant.
Aix and Solaris are also still there
I don't know who is the writer of this article but it is a huge load of horse shit...
For the rest, regarding the diversity of DE, we can lament the lack of a strong contender for the general public, Ubuntu was about to become that before their ego took over, but otherwise that is the true value of Linux stacks since the beginning that you can have it and tweak it to your taste...
> you can have it and tweak it to your taste...
Not really. I have invested a lot of time and effort to finding and configuring my "perfect" desktop environment. I failed. As soon as you start tweaking (i.e. changing the default settings) and install some additional tools to fill some gaps, you discover loads of bugs and inflexibility.
I feel similarly... I spent about a year with Budgie and really liked it... I had it mostly tweaked to my liking. In the end, I felt better just going back with Pop and keeping it close to defaults. I don't really tweak my mac or windows settings that much either.
In the end, I just want a pretty solid foundation to get work done.
I never said that "tweaking" has to be easy or config(-script)-kiddy level, but it is possible.
Try to have a proper menu bar under OSX in any way, even hoping to code your own menu bar software and you will understand what I mean.
Typical xkcd 927 case.
I don't know them by number and I wanna guess before googling.
The one about having N standards, wanting to create one that unifies them all, and ending up with N+1 standards?
Edit: I guessed right.
There has been a consistent desire to unify the desktop environments, but the fragmentation has largely been because of differing use cases and philosophical perspectives.
This divide is being further driven on the issue of X11 vs Wayland, and now the drama of decentralized libertarianism vs centralized corporatism. The latter manifests itself as a culture war over the code of conduct or woke software. Now it is coalescing into a political line between hyprland & X11 and GNOME & Wayland. (hyprland uses Wayland, but it and X11 have a similar political affinity by loud and divisive proponents.)
The woke have an affinity with centralized corporatism and want to unify freedesktop collaboration under it and ensure that there is identitarian representation, so contributors don't have to worry about petty discrimination and office politics.
The opposition have an affinity with decentralized libertarianism, and they reject identitarian politics as they want their personal freedom to do what they want even if it is not politically correct, believing that this is the best way for their ideas to flourish into better software.
I personally believe there's some merits to either side, and a fine balance has to be made. We can try to get the best parts (the best software) without the bad parts (discrimination and office politics).
I'm a bit mixed on things as well... I like some of the technical solutions that are coming from the more woke organizations, but I really don't like the abuses under their Code of Conduct or war against the non-woke. Such as activists inside Gnome calling to reject funding from Framework because they're also funding Omarchy and Hyperland.
It's a bit silly, but serious all the same IMO.
Identitarian politics can either be explicitly enforced or implicitly allowed. The woke chose the former. The opposition can do the latter but they also believe in checking identitarian interests in favor for personal freedom.
There are concerns over this escalating into fascism, but the logical extremification of ideas only muddles the waters and makes communication difficult. A pragmatic and balanced solution gets moved out of reach, and as a result, the corporate watchmen can push lighter opposition to the extreme fringes. Who then watches the watchmen?
Don't invite corporate watchmen to the party.