40 comments

[ 3.0 ms ] story [ 95.4 ms ] thread
(comment deleted)
I'll hit a completely opposite tone to the article's and say this is actually pretty honest for once.

When the regular person thinks of an OS, they think of the desktop environment provided for the most part. So in my opinion, this is just the right things taking their rightful position.

If anything, certain people should now start fearing the incredible potential these decisions have in terms of ecosystem consolidation. Interesting times ahead.

KDE Plasma desperately needed this. It's a second-class citizen on every distro, except recently on Fedora where they decided to upgrade it from being just a "spin".
What makes you think that? I always had this impression when using kubuntu (more than a decade ago)

On arch or manjaro everything has been nice and smooth instead (I install manjaro for the non computer savvy, arch for me)

They have KDE Neon.
However during the Plasma 6 upgrade debacle earlier this year the developers made it clear they don't indend KDE Neon to be suitable for production usage.

I'd been quite satisfied with KDE Neon for quite a while, but looking to move to another distro due to this.

Not that there's anything wrong with their stance, though I do think it was not communicated well enough.

OpenSUSE is very much "the KDE distro".
[flagged]
Yawn, it's been a fantastic daily driver for 20 years now.
That wouldn't be wise. Everyone who does everything in the cloud (a.k.a. other people's computers), like Ms 365, Teams, Youtube, Tiktok, Instagram, Google Mail, Google Maps, Google Spreadsheets, WhatsApp, etc, ... should think twice. There is no gain for them. It's all outside of their control anyways. They gave away their sovereignty to the point the OS does not matter at all. They should stay where they are and not burn their (and others') time with learning a new OS. Linux doesn't need those people, and those people don't need Linux. Make more Insta Reels instead. Everyone is happier then.
You can use GMail with any client with IMap. Youtube has several frontends. Libreoffice it's far more featureful than Docs and Spreadsheets. Google Maps? Organic Maps works offline and you can always use OSM maps with just GPS support to orient yourself in a disaster.

These people will get burned and want their computers and media back.

Okay, you can use GMail and Youtube with other frontends. Doesn't change much, though. If you are just using cloud services, you are better off staying at whatever OS you use.

> These people will get burned

They decided this way. I can remember when the Facebook plague began. There was no real need to start living _inside_ Facebook, but most people did. They want their stuff back now? That's ridiculous. 15 years of social media didn't make them smarter in any ways. They want to get deeper and deeper, and I'm not going to stop them.

The author seems to think that someone should tell all these volunteers to focus their efforts somewhere, and that the volunteers should listen.

Fortunately there are proprietary desktops that receive substantial income from hardware sales and licensing fees. Unlike volunteer-led efforts, they have leaders paid to set direction and workers paid to implement that direction. One can use such a desktop and OS, or perhaps one can be grateful that these unpaid volunteers provide an alternative at no charge, no matter how uncoordinated the end product is.

>The author seems to think that someone should tell all these volunteers to focus their efforts somewhere

Gnome and KDE are in a pretty workable state because a lion's share of the support comes from Red Hat, Suse and Canonical (and to a lesser extent from Valve or QA people from other distributions like Arch), who are probably not going to fund another distro. In particular a lot of the core dev salaries come from those companies.

It'd be on a shoestring budget and team which in the Linux world usually ends in disaster, in particular if you're talking about building an entire distro.

> Gnome and KDE are in a pretty workable state because a lion's share of the support comes from Red Hat, Suse and Canonical

I think's now Red Hat and Endless. Other than the odd patch, I don't think there's much from either Canonical or SUSE.

It was my impression that code and money for Gnome primarily came from Red Hat, but also Canonical and Suse.

I didn't think it was the case for KDE, and at least in their latest report they seem to be more Bernie Sanders in their income breakdown[0]. As for code contributions I don't know but I'd be pleased if a more knowledgeable reader could give us a clearer picture.

[0] https://ev.kde.org/reports/ev-2023/#welcome-to-kdes-annual-r...

For a long time SuSE and Mandrake were the KDE distributions, and some of the money came from there.

This was 20 years ago, dunno about nowadays.

Which is what I have been doing since 2010 thereabouts.

Got fed up with The Year of Desktop Linux, even though I am probably in the minority of HN readership that actually subscribed to Linux Journal during its whole lifetime, did a couple of Gtkmm articles for publications like The C/C++ Users Journal tool create awareness and such, still subscribe to LWN.

Apple, Google and Microsoft take better care of what I consider a desktop experience without ideologies in the middle.

Fedora Silverblue set to the Testing branch with Flatpak it's almost the official Gnome distro, if you care.
Isn't Red Hat the official Gnome distro since forever?

Ubuntu then resist the changes, try their own thing, to finally have to give up and use whatever Red Hat is selling. Currently they are not giving up with snaps, everything else comes from Red Hat/Gnome land.

The point is, we already have some kind of official Gnome distro. May be the Gnome guys are disagreeing with Red Hat, or, as it appears to be from the article, they want to play with some new technology.

For KDE, we used to have Suse and Mandrake. Mandrake was forced to change the name to Mandriva Linux, and now it has disappeared.

The current KDE based distros have a tiny market share compared to the other distros.

They should focus on why, and I would believe the commercial nature of QT is one important reason.

It doesn't matter if there's an official KDE distro, if people doesn't use it.

> Isn't Red Hat the official Gnome distro since forever?

You mean Fedora? It was considered the "unofficial GNOME OS" for a while, the maintainers of GNOME Shell, Mutter etc were also the maintainers of the packages of Fedora. Fedora Silverblue uses OSTree, which originally got birthed out of very early versions of 'GNOME OS'.

But it was never official, Fedora didn't even ship with Flathub for the longest time and IMHO Fedora has become hostile to casual desktop users.

Adapting Flathub and purging the Fedora channel at Fedora Silverblue it's just a matter of seconds.
> You mean Fedora? It was considered the "unofficial GNOME OS" for a while, the maintainers of GNOME Shell, Mutter etc were also the maintainers of the packages of Fedora.

Yes.

>commercial nature of QT

That "reason" has been obsolete for decades.

It's written "Qt", everything you need for a Linux desktop is LGPL, and it gets regular outside contributions.

I get the discomfort, but if the Qt Company ceased operations tomorrow, Qt would only stagnate... kinda like Gtk already does.

QT may be an important reason, yes, but I don't think it would change much with lets say GTK. Nobody wants to deal with the madness you find in KDE. I think most people like a configurable desktop but with sane and solid defaults. That's why Gnome is preferred. And I don't think it's gonna change anytime soon. KDE has a very opinionated userbase, which is not bad, but it also has its downsides.

I'd like to change a couple things in my Gnome desktop that I can't but I rather have it this way than deal with KDE.

The article says

> > This is entirely in keeping with how the KDE project as a whole works: for instance, if you search the KDE Applications website for "text editor", you'll find three: Kate, KWrite, and Nota. If you search for "file manager", you'll find four; and "web browser", three. Within the desktop, there are multiple start-menu tools, multiple app-switching panel-button bars, and so on. Even the "About" option on the "Help" menu is duplicated: one tells you version information (in one of two version formats, either decimal-number based or date-based), and one tells you about KDE as a whole.

I think they are ALMOST right. Yes, different people use different apps. Yes, it should be possible to download fifteen different web browsers or twenty five different text editors. However, there should be a coherent DEFAULTS story. By default, when you install an operating system, we should have ONE default COMPLETE web browser, one COMPLETE text editor, and so on.

I don't think when you install an operating system, it should install an office productivity software suite such as Microsoft Office or Libre Office by default.

> I don't think when you install an operating system, it should install an office productivity software suite such as Microsoft Office or Libre Office by default.

I don't see why not (apart from the Microsoft example which would be monopoly abuse). If I install a standard Linux desktop distribution (rather than a server oriented one) then I'd be quite happy to have LibreOffice or Gnumeric etc. installed. Office tools are used so frequently that it makes sense to include them along with e.g. photo editing tools.

And even ignoring OneNote and trial version of MS Office in more recent Windows versions, Windows has always shipped with WordPad. Microsoft's vision in the 90s was that the operating system should come with basic productivity tools like a plain text editor (notepad), a rich text editor suitable for writing letters (wordpad), a graphics editor (paint), software for sending and receiving fax and scanning documents (Windows Fax and Scan), along with a media player, a web browser and some games. You could be productive in the office with just a plain Windows install.
I find it weird if a distribution doesn't have it.

At the end of the day, a big selling point for FOSS products is always going to be the free-beer aspect. Bits are free, take as many as you want.

But there's also the appeal of centralized management through the distribution. Even if it's technically an opt-in checkbox during the installer, it's convenient that 95% of your software comes from a central source with a single standard way to get updates. I'd expect this is even more exciting at scale-- if you're managing 5,000 desktops, the FOSS answer is "have a cron job run apt-get update;apt-get upgrade a few times a month" and the Windows answer is "here's an aftermarket tool, and you're probably going to have to script a bunch of individual installers and deal with licensing gimmicks to make them all properly launch."

> there should be a coherent DEFAULTS story.

I see the appeal to someone who is new and shows up with an intention like "idk, just give me the internet lol". That's a perfectly fine and overwhelmingly common way to use a computer. I think even a mainstream distro that's offering a KDE desktop should do that.

For the KDE project itself though it's always put a lot of store in offering lots and lots of choice (maybe the main ideological difference with Gnome). Of course all settings have to have a default, and randomizing or forcing a choice on first use would be a hassle, but you aren't pushed down any particular path. I like this and I'm glad there's a desktop environment with an attitude like this.

I think that there is an attitude being evinced by the popularity of search terms like "best photo editor free 2024”. People do they same on forums, they would like to get the "best" one and ideally wouldn't have to spend the time and energy on finding it. Sometimes there is a natural choice, sometimes there's a choice that seems natural to some and not others, and the common case (imho) is that it's complicated. If you're just doing something once and it doesn't matter, use any tools that works, if you'll use it often and the outcome is important then there's not substitute for trying different things.

You might view it as multi-armed bandit problem with a subjective value function. Whether the value varies much between people is something else you'd have to figure out.

I completely agree about the importance of defaults. I’ve always believed in the phrase: “It’s the defaults, stupid.”

In my experience, 90% of users just stick with whatever defaults they’re given, while the other 10% are the power users who customize things.

This explains why you see poorly designed defaults like the Windows 10 taskbar, where the search bar takes up about 33~50% of the space by default:

https://www.isumsoft.com/images/backup-recovery/install-wind...

Another thing I think Linux distros should focus on—and I know some already do—is renaming default apps to be more intuitive. For example, instead of calling it “Caja” (or is it “Kaja”?), just call it File Manager (Caja) so new users know what it is.

As for including an Office Suite during installation, why not? At the very least, make it a pre-selected option that’s easy to deselect if someone doesn’t want it.

I thought this was already the case with KDE Neon and Fedora.
Philosophically, it always bugged me that distros were so centrally in control of packaging: app developers are not allowed to package them in the way they see fit, which has caused some friction in the past[0]. It seems healthier and more scalable if each developer could package it autonomously without work from the central maintainer group.

KDE Project Banana mentions little about packaging, apart from this:[1]

> Instead of legacy packages we target modern deployment systems such as flatpak and systemd-sysext

I hope they succeed in that respect. Flatpak has the promise of decentralizing distribution. That said, my experience of it so far has been brittle, with things like changing your locale or shell prompt causing your apps to freeze[2]. I hope they make a stable operating system nonetheless, but it is not easy.

[0]: https://news.ycombinator.com/item?id=41407768#41408628

[1]: https://community.kde.org/KDE_Linux

[2]: https://github.com/flatpak/flatpak/issues/5564

> It seems healthier and more scalable if each developer could package it autonomously without work from the central maintainer group.

> Flatpak has the promise of decentralizing distribution. That said, my experience of it so far has been brittle, with things like changing your locale or shell prompt causing your apps to freeze[2]. I hope they make a stable operating system nonetheless, but it is not easy.

That's not a coincidence; you just got rid of the pesky maintainers who made sure that the collective system worked, leaving only a bunch of separate apps that are free to do whatever they want.

On the same note, this ensures that security updates are decentralized, so instead of running 'apt upgrade' and being done, you get to hope that every dev providing you with packages updates their dependencies (which is also harder because flatpak encourages vendoring as much as possible).

I have to say that the best KDE distro always has been Slackware, either -stable or -current. Slackware with slapt-get get had, for a while, a pacman-level package manager allowing you to install all the dependencies in a breeze. It even had repos covering all the slackbuilds from SBo made into packages.

A pity Freenix/Freeslack (libre Slackware) died, because if I had to use some $DBUS based distro instead of Hyperbola, I would choose that for professional usage where you need standard tools for your enterprise related work (Docker and such).

From end user perspective I think the desktop environment that i.e. gnome gives users is good enough from ui/ux perspective. It’s not worse than macos or windows. The thing where Linux distros can improve at is still at stability and making things just work for the end user. Especially around hardware support and os version upgrades and packages upgrades that often just don’t work.
Maybe it is true for Gnome, but KDE is full of critical bugs that would be considered showstoppers for Windows.

KMail doesn't work with Exchange accounts with OAuth/2FA (mine stopped syncing and won't budge).

KRDC messes up input between itself and Plasma (e.g. Win key is sent to both, and then wreaks havoc in remote session).

(comment deleted)