F42 Change Proposal: Fedora Plasma Workstation (System-Wide)

Project Discussion Change Proposals
,
21

I’ve vouched for Fedora KDE ever since F35/F34 as a user and I am one of the many that think it is the best KDE Plasma experience out there, obviously that means my time with fedora up until now has been mostly with Plasma, but even so I feel like a full swap over might be too much? Moving aside the whole “Fedora Workstation is GNOME” identity stuff, I feel like it’d bring us back to imo, the core issue:

Fedora KDE is NOT being promoted enough like it deserves to be because of its spin status, but if this proposal somehow makes it through as-is, no compromise, this would make GNOME (an absolutely excellent experience on Fedora that’s well integrated that I also vouch for in the GNOME context) be the victim of underpromotion / being unfairly represented due to it being a high quality experience in its own right compared to all the other spins (barring Sway imo).

My view as someone who wholeheartedly loves all the projects involved in the discussion, I think Fedora should expand its “identity” instead of warping itself even if it pushes the desktop I prefer. GNOME is a fantastic desktop, and that’s coming from a self proclaimed KDE fanboy, and we shouldn’t put it down because Fedora doesn’t push the KDE spin enough or because it prefers to deal with upcoming features by sticking to upstream implementations as much as possible (ie. Plasma implementing KDE specific wayland protocols while actively working to push said protocol upstream for all desktops), moving aside opinionated choices on GNOME’s end like not implementing the SSD protocol (whole other can of worms not worth discussing here).

The ideal outcome of this proposal in my eyes should be Fedora KDE graduating to a Workstation variant (Workstation KDE, Plasma Workstation, whatever) alongside the pre-existing Workstation GNOME, it is very much deserving of the status once KDE upstream implements everything that’s needed for full parity with Workstation (Initial Boot/Account Setup, the expansion of Plasma-Welcome’s functionality allowing for things like third party repo setup to be tightly integrated).

Either way, I hope for the best here, the Linux Desktop is best when we work together, much love to GNOME and the Workstation group. <3

25 Likes
22

I don’t have an issue with Plasma, in fact I was a die hard KDE user for some time. I think the KDE project is great, but it is out of sync with the Fedora project, which causes some issues for us, whereas Gnome is in sync. This has been pointed out above by others. I have always wanted all DE’s to be offered at first install, but that should be for the “Everything Installer” IMO, not the WAS edition.

5 Likes
23

I don’t think there’s any indication from Fedora, but it would be skewed anyway because many people would just use the default no matter what it is.

In terms of popularity I am aware of two statistics:

  1. Gaming on Linux: currently 40% Plasma, 29% GNOME
  2. Arch Linux: currently 35% Plasma, 22% GNOME
6 Likes
24

While KDE is not bad, I personally prefer GNOME…

4 Likes
25

Arch???
Gaming??

I have to ask, what are the bearing of these stats on the topic at hand?

4 Likes
26

Let me know if you have any better (and neutral) Desktop Environment surveys/statistics.

27

Agreed. That’s a major problem which definitely hurts KDE’s viability to be “a primary offering” currently.

According to this proposal though, the release schedule will be aligning to be closer to Fedora’s by 2025 (coinciding with F42 release), so I’d say once we’re past that hurdle, Fedora will be in a position to give Plasma a bigger spotlight.

My idea of making the two DEs equal doesn’t need to be “now” or “immediately”, but rather “whenever it’s viable”. :wink:

5 Likes
28

I agree with these. In fact, having two equal versions (or a choice in the installer like openSUSE does) is a more realistic proposal anyway. You often see people on e.g. Reddit asking questions like “what’s the best KDE distro?” and people saying Fedora will get little upvotes or even comments saying “but KDE is only a spin, it can’t be good / taken seriously”. When actually Fedora KDE (and Kinoite) is amazing!

5 Likes
29

I’m on gnome and would like to stay on gnome.

mainly because gnome was superior on wayland adoption
and i want my AMD Advantage (R 7 6800HX, RX 6800M) to stay on a stable wayland platform which does not break while i’m doing my (for me) critical sre/devops engineering stuff
(with this fedora already struggles hard at the moment, with the latest fedora 39 kernel my system freezed the last weeks A TON of times without any reasonable reason)

as fedora is aimed on professionals which are relying on stability to get their work done I think gnome > KDE
(sorry, but if i wanna play games, i shutdown fedora and boot into my windows on the other disk)

4 Likes
30

This is an amazing idea, KDE was an absolute game changer for me when I first got into Linux and I didn’t realize until diving down the rabbit hole of Fedora Spins.

The workflow and seemingly strange approach of Gnome’s interface made my initial experiences with Ubuntu back in 12.04 feel like I was playing with a foreign toy and not a ‘computer’ like my old g4 Mac on Tiger or the Windows 7 box it was hopefully replacing.
Not to mention that future attempts at running it on an old dual core intel core-m processor with limited ram just couldn’t handle how heavy out of the box GNOME was at the time.

The app draw felt clunky and slow to load, not having a desktop to put links or applications seemed absurd and having sites tell me to install GNOME extensions to do any form of customization seemed like unnecessary overhead and odd loop to jump through which proved to be a security nightmare.

The over simplification and seeming restrictive way GNOME approaches their users in only providing one seemingly right way of approaching a workflow always smelt like the sort or Apple-ish “your holding it wrong” mentality that I was trying to get away from. There were certain fundamental expectations of how to interact with a Desktop across what felt like all other OS’ that had served me reliably all the way back to my childhood that did not work and it acted as a major speed bump to my adoption of Linux as a daily driver.

Thankfully it lead me to diving down a whole plethora of alternates including XFCE, MATE and then finally settling on KDE.

Sorry for the rant, but there is a good and solid reasoning behind SteamOS picking KDE plasma for putting their whole weight behind, and while I can’t speak for Lord Gaben himself, I would hazard to say it is broadly because for the vast majority of users, it strikes a balance between modern style, functionality and a UX that feels familiar from what they learnt on, i.e Windows or Mac.

We shouldn’t just be doing things because “thats how the rest of workstation distros do it”, Fedora is supposed to push the envelope and help set a positive trend for the future. If we want Linux as a whole to reach a reasonable share of user adoption, I think providing them with a familiar starting point that doesn’t require them to re-learn longstanding assumptions of how to interact with a computer is an important first step.

11 Likes
31

Since we’re talking about Fedora, I would expect Fedora-centric stats instead.

1 Like
32

I agree that that would be a very good idea. I honestly just didn’t think it was in the cards lol

3 Likes
33

Yet you “want” the Windows and Mac experience.

3 Likes
34

Absolutely. The intent is not to push Plasma as a default before 42 or more promoted experience when it already has received concessions as noted by @mxanthropocene related to its release cycle. There have been efforts made by others (for example @ngompa and Nate Graham) on working towards Plasma release cycles being more aligned with that of Kubuntu and Fedora, which would no doubt reduce maintenance and QA burden.

Of course, it is completely understandable that desktop environments and forms of desktop user experiences (traditional vs non-traditional) are a personal preference. This was touched on very little in the change proposal for that very reason, but of course it is naturally a key part that people will gravitate towards.

Fedora has always been at the forefront of innovation in the Linux ecosystem, while the KDE community and Plasma has been at the forefront of innovation when it comes to software architecture (KDE frameworks), standardization (as noted in the Change Proposal), and Wayland support. That is not to diminish all of the amazing work that GNOME has done (seriously!), however this is a proposal by many members of the Fedora KDE SIG, as well as myself (non KDE SIG member), so it will highlight key areas where Plasma has been a key stakeholder.

There is bound to be questions raised on how this would impact Red Hat (keep in mind, it doesn’t have to) or if any additional resources from Fedora or Red Hat could be provided to ensure Plasma would be able to meet the quality expectations raised. I imagine these are points that Fedora leadership would like to discuss.

I just hope we all keep this discussion as focused on the technical merits as possible. I think @mxanthropocene for example provided many good points that need to be taken into consideration (Qt LTS). I look forward to hearing how this will be discussed further and how folks like @mattdm can raise this genuine discussion to places where parties from various working groups and SIGs can address items to either bring this proposal to fruition, or an equally beneficial solution for all parties (e.g. more resources for Plasma, improved marketing in Fedora that is less focused on one specific experience, etc.)

:heart:

13 Likes
36

KDE was loathe to support Wayland in the beginning, while Gnome has from the start.

2 Likes
37

Does the proposal to move Plasma to a 6-month schedule include Gear and Frameworks too? I don’t remember ever hearing about that; the fact Gear is on a different 3-month schedule to Plasma is a major cause of burden, as is a new Frameworks release coming out once a month…

2 Likes
38

and Wayland support

~2 years ago i have choosen gnome over kde, because gnome had proper wayland support, kde did not

1 Like
39

That is not true at all, do not spread misinformation. KDE started work on Wayland support around the same time GNOME did.

40

What? KDE started working on Wayland support in KWin 13 years ago (2011).

Here’s an old blog post that details some of that early history: The History on Wayland Support inside KWin – Martin's Blog

5 Likes
41

Is it a joke?

I have checked, and I’m assured that this isn’t an April Fools prank, despite the initial date in the wiki.

Wait, wait, hold on a minute please!

I’d like to please put this on hold as a Change as a procedural matter. Under the Fedora.next framework where we established the Fedora Editions, the Workstation working group chose GNOME for implementation. The correct initial venue for this is that working group.

Now, practically speaking, since Fedora Workstation has been GNOME-based for … a decade now, I don’t think it’s hard to guess how a decision there would likely come out (even though not everyone in the WG is primarily a GNOME contributor or partisan). As someone just now said to me on Matrix, “I feel like that’d be like asking the employees of Apple to support changing the iPhone’s default OS to Android…”

Nonetheless, I think that is the correct approach to start. In the very early days of the Workstation Working Group, there were some proposals for that group to act as an umbrella overseeing multiple desktop environments. That is potentially something the group could re-consider.

@joshstrobl, @marcdeop, @tdawson, @farchord, @aleasto — could you please reach out and have that conversation? See Fedora Workstation — getting in touch.

Edition Promotion?

If the Workstation working group doesn’t like that, the next step would be to consider the Fedora Council Process for promoting a Fedora deliverable to an edition.

It is possible that that is a viable route, but I have quite a bit of skepticism that there’s an alternate-desktop proposal that fits the first criterion:

  • addresses a distinct, relevant, and broad use-case or user-base that a Fedora Edition is not currently serving

There was (again, about a decade ago) a proposal for a Fedora Scientific Edition, a desktop to appeal in the academic and research space — and which happened to feature KDE as a desktop. I’d have to look up the exact details, but basically we rejected that 1) as too niche (while still overlapping with the Workstation mandate), and 2) because it was very clearly a “get the specific desktop technology in as an Edition somehow” play (which felt inherently antagonistic to the goals of the Fedora.next plan).

So, while that would make procedural sense, I don’t want to waste everyone’s time with that requirement.

So, what then?

Well, KDE is (and has long been) by far the most popular desktop in Fedora after GNOME. The KDE SIG has been doing good things, and in particular has built great bridges with the upstream project. The change in release cycle in particular catches my eye. I also don’t doubt that the upstream project is healthy and capable of delivering a stack we can use to make a very polished Fedora experience worthy of being a top-level offering.

If the Workstation Working Group is not interested in a dual-desktop Workstation approach, I think we can and should take this up at the Fedora Council level and find an approach that works, and which better highlights the work that the KDE SIG and KDE fans in Fedora have put in. That may be a change to our Edition rules, or a particular exception, or something else — I won’t promise a particular outcome.

I don’t think that demoting the existing GNOME-based Workstation is a likely outcome. (I personally would vote against it — it is overwhelmingly popular and successful, and we have good relationships with GNOME that we don’t need to just throw out.) I know that we ended up with this kind of feeling like a zero-sum game, but it doesn’t have to do that way.

So, I do promise that if something else can’t be worked out, the Fedora Council will give serious and fair consideration to all reasonable options. I promise we’ll listen to all stakeholders and allow room for community input and discussion. And, I think that we can come up with a solution that will be an improvement on the current state.

Meanwhile…

I’m not going to shut down this discussion topic (at least not immediately!), but I do ask:

  • proposers, please withdraw the Change Proposal, or
  • FESCo, please decline/defer

in favor of the process outlined above (tl;dr: talk to the Workstation WG, and if that does not come to a satisfying outcome, file a Council ticket for next possibilities).

Thank you!

34 Likes