KWinFT 5.24 Released - Continues To Advance Its Wayland Support, Expand On Wlroots

[mdedetrich]

romangg I can't say I disagree with any of the examples you gave. I, myself, am under the impression KDE is barely held together and the devs don;t have good grasp on the code base. But it's not always about what you say, it's also about how you say it
I should know, as I grow older, I'm more inclined to be blunt when a little more finesse would be required.

I don't think its far to characterize it this far. KDE is a big branch and romangg comments are more specific to KWin and the compositing side of KDE rather than KDE as a whole which is massive. There are many cases of KDE software which has been very innovative and done lots of refactorings (i.e. Krita recently redid their rendering engine for performance).

It does seem that its KWin specifically thats having these tensions which is not too surprising considering that KDE is reliant on KWin to do anything (it is afterall the main compositing engine). Such software commonly permeates an ultra conservative attitude out of fear of breaking things and considering its age and how Linux is not that mature when it comes to dealing with graphics/compositing/X11 there are a huge amount of workarounds/hacks and without having lots of hardware to test regressions there is a lot of fear of not breaking things.

In these kinds of cases I can see why a fork is necessary for such large refactoring especially if you want to revisit things from first principles however I would tone down the language, it isn't fair to brush all of KDE software with the same brush.

[holunder]

It's not like I've given up on KDE, but the organization needs some fundamental changes in the way its development is organized internally and how openly it interacts with other projects in the Linux ecosystem. Blatantly copying their innovations without even saying thanks for it is it not.

The Overview effect is not a Gnome innovation, lol. The Overview effect is a copy of macOS’ Mission Control UI+UX. But Gnome 3 has experimented at the same time with this kind of single-workspace-view-but-still-multi-desktops-reachable approach but has ultimately settled to a close UX to macOS.

And in my opinion, it’s not a great UI for power-users. This works great on a light notebook use but fails on a power-user’s desktop PC with lots of windows and workspaces with workflow meaning. Not everything Apple comes up is gold, especially their catering to pro-users is, software-wise, horrible.
Would be really sad if this would kill the Grid and Present Windows effects (where you can also show windows of all workspaces, not just your current, or windows of the current application).

[Sho_]

Well.... can you say, that like k_ver said, he is friendly in person ?

Roman and I fairly regularly engage in private exchanges, which are civil and enjoyable. Sometimes we agree on things, sometimes we disagree. In fact, we often have a private chat open while we debate things in public on reddit or here. To an outside observer it may appear as an epic feud where there isn't one.

Content disagreements aside, I don't like the way he expresses things in blog posts and believe it does net harm. He knows it. It is what it is.

[pereiraalex]

Roman and I fairly regularly engage in private exchanges, which are civil and enjoyable. Sometimes we agree on things, sometimes we disagree. In fact, we often have a private chat open while we debate things in public on reddit or here. To an outside observer it may appear as an epic feud where there isn't one.

Content disagreements aside, I don't like the way he expresses things in blog posts and believe it does net harm. He knows it. It is what it is.

I am happy it is that way. Really happy! Its good that, although there are disagreements, civil and enjoyable talks still happen. Please keep it up that way.
I hope that deep down, Roman values the work of the kwin team in iteratively improving kwin and keeping its users happy.
I hope that deep down, in the future years, kwin team might look at kwinft and take "fruits" of its work.

I have to agree that "it does net harm". But on my take ( and I guess my culture ), its sad that it is that way. Different takes about things and disagreement should be healthy, not harmful. So, for example, I didn't feel any "less kde'ish" when I read that blog post, and that specific comment.

Thank you for your work, and for your feedback! Its was good to hear that!

[tildearrow]

tildearrow Do you think that Valve might be interested in investing in your project, since they will be utilizing KDE for SteamOS 3.0 and the Steam Deck?

Personally I don't think so. I'd rather see them adopt KWinFT.
I am about to contact Roman to initiate talks regarding my X11 unredirection code.

[theriddick]

Tried kwinft 5.23 and my experience was... black screen and then login screen. I hope it's gonna be ok in 5.24. Manjaro btw.

Same, every-time I've tried it. Doesn't work and doesn't seem like drop in replacement works. Shrug.

[Alexmitter]

tildearrow Do you think that Valve might be interested in investing in your project, since they will be utilizing KDE for SteamOS 3.0 and the Steam Deck?

Kwin and the whole KDE stack only runs in "desktop mode". For gaming, valve uses a own wlroots based compositor called gamescope, Kwin would be way too slow, unstable, unreliable to serve as the base of a gaming system.

[Alexmitter]

Well.... can you say, that like k_ver said, he is friendly in person ?

Because I mostly believe this is culture differences. At least, in my European Country culture... if that comment is said with "honest intentions from friendly person", it shouldn't be considered an attack, really.

Then again, if its a snarky comment, made by a snarky person, its completely different.



Anyway.... as a very old kde user, and really, as someone who wishes the whole KDE team all the best, I think some "growing up" is needed.

SO WHAT if "its not KDE's innovative days" ? even since kde 2 and 3, all the complaints from users/trolls (alexmitter, where are you? oh right, this isn't a "greedy qt post") is that kde devs are always doing new features and not fixing bugs and polishing. WELL NOW KDE IS FIXING BUGS AND POLISHING! no team can be always 100% full innovative mode ON all the time.

And romangg ... dude, I seriously also hope you all the best, but you can't (well, you can, but should not really) say: "Hey kwin users, can you wait 5 years while I rewrite kwin almost from scratch and change everything? just go to sleep for 5 years or use another desktop/system, okay??"

Anyway.... bottom line... hope everyone all the best

I wish KDE the best, and the best is a future without Qt. KDE is better then Qt, and its better then the people behind Qt. KwinFT is a step in the right direction, using a community effort (wlroots) instead of the Qt Wayland Compositor code.

[stingray454]

As someone who hasn't been following this development at all basically (ie not up to speed on what has been said / done in the past) - I understand that KwinFT is a big rewrite / restructure, and that the Kwin team might be cautious to implement large changes and possibly break things, but is there any hope that it would officially replace Kwin in the future once it reaches a stable state and continue as the official version, or will it forever stay a forked drop-in replacement (maybe due to disagreements on how it "should" work)?

[R41N3R]

As much as Iike the development of KwinFT (but I'm not a user yet), I do wonder why it isn't focusing on a pure Wayland future just like gamescope does. Are there reasons to develop the X11 branch as well? I would expect that Kwin will always remain the main X11 compositor in this legacy scenario because of compatibility.