KDE Developers in Germany Planning for Plasma 6.0 | Hacker News Reader

KDE Developers in Germany Planning for Plasma 6.0

phoronix.com

71 pointsprofwalkstr21 comments

I've been using Linux long enough that I was around for KDE 3 and KDE 4. I did not enjoy using them, but I love 5. I'm nervous about this upgrade. I can't think of anything I want that would merit a major version upgrade.

Oh, KDE 3.5 was amazing. When KDE 4 came, I packed my stuff and never looked back.

Good to hear KDE 5 is doing fine at least, let us hope for the best.

I get that many people are more excited about rewriting stuff than fixing bugs but oh man is the risk is rarely ever worth it. If you have a good thing going, don't kill your golden goose.

I feel the same. The widgets/cashew stuff was totally insane.

People used to look forward to software updates. How did it come to this?

I get the impression that attitudes are starting to shift back to anticipation, but I suspect that many people are still leery due to the disruptive changes brought by KDE 4, Gnome 3, and Python 3 (to name three projects that immediately come to my mind). Of course, those weren't the only products affected and open source software wasn't the only thing affected. The years around 2010 felt a bit like the early 1990's, with the distinct difference that disruptive change in the early 1990's brought with it quality of life improvements that were more often well received.

Now that software is starting to mature again (at least in my mind, and at least with open source projects), changes are more welcome. Watching updates is more interesting because the bullet points look interesting from a utility standpoint rather than a drama standpoint.

on specific consumer products people tended to look forward and probably still do. On major framework, user interface or OS changes which KDE falls into I think people always have always been at best ambivalent about.

The reason is pretty straight forward. If you break something at that level you tend to break almost everything, and people have spent a lot of time building stuff on top. Shaking that in a big way is always risky. People tend to be more conservative the more you go towards the bottom of the stack.

When I was younger (in my teens, early twenties), I looked forward to every feature upgrade. Now I’m getting anxious at breaking changes and upgrade paths and every new major release of RHEL/Debian makes me a bit sweaty and they feel so quick (what, Debian 12 is coming out this year? I just migrated to 11 last year.) Now I’m an old grumpy 29 year old man yelling at the cloud.

There are a couple of reasons not to worry about this upgrade:

A. 3 -> 4 required very very significant changes. 4 -> 5 and now 5 -> 6 are nowhere near that.

B. KDE really learnt their lesson on this and they are very careful not to create anything similar to that. Thus, not big rewrite/changes like 3 -> 4 are in the works.

This is just to get the whole project going with Qt 6 which isn't that much different from Qt 5.

Edit: Disclaimer, I used to be a very minor contributor to KDE around 3 -> 4 time.

I'm keen to see a fairly minimal change with KDE6 / Plasma 6 - so much is working great in KDE5 / Plasma 5 that it would be a shame to suffer reversions but presumably the odd one or two will crop up.

The only things I can think of I'd like tweaked are:

1. to make shifting pinned icons about more initiative (I must have the wrong mental model of editing the desktop layout as I always end up flailing around with something that seems like it should be trivially easy!)

2. Sort out Baloo: perhaps off by default and ideally to have better handling of exceptionally large compressed files (these usually result in it going into an hours long process of chugging away often with little to show afterwards!)

Throw in better handling of network shares for non-KIO apps and finally fixing that bug where the text cursor randomly decides to jump forward or back on Wayland sometimes and their work is done as far as I'm concerned.

My fantasy wish list:

- Let me globally use Meta as default keyboard shortcut modifier. There's code in QT somewhere that does this but it's restricted to MacOS.

- Get back custom window icons in the task manager. Apparently Wayland doesn't allow this, so it got disabled for X11 too. I used to identify my n terminal windows by icon, but now they all look the same.

- A way to persistently attach shortcuts to select particular windows (again, said terminals). You can manually set shortcuts, but they don't persist. I used to have some scripting based on inspecting window titles, but it was fragile some update broke it.

- In Dolphin, Mac-style column view and ‘quick look’. (On the bright side, Dolphin will now let you disable the focus-follows-mouse behaviour of the preview panel; that just missed 23.04 though.)

What is missing to have Plasma providing a full experience on Linux phones or tablets, e.g. PinePhone/Tab2?

The biggest problem is probably driver integration for machines targeted at android.

Bitch of a problem to have to constantly patch between them, and you're not getting any support from the vendors.

My pinebook pro seems to like plasma. Presumably that’s true of the pinetab, right?

I don't really know to be honest.

But there's this that might be worth reading: https://tuxphones.com/purism-librem-5-vs-pine64-pinephone-li...

> On the other hand, Pine64 opted for a community-based approach: they provide the base hardware and OS to users and developers, and the community develops the needed software to make it run. postmarketOS and (Ubports) Ubuntu Touch developers have already ported their OS to the PinePhone development kits, and KDE Plasma Mobile will be the primary desktop designed for this phone. About $10 for each device will be donated to open source projects.

I don't hate KDE, I love many many things about it.

But have they ever considered moving their software beyond alpha quality before releasing a new major version? A major reason I'm not putting any more time into KDE is the expectation that it'll never be stable.

I know this is an incredibly snarky, cynical and just negative take, but that's how I felt when I was trying to use Plasma Wayland. That it was on a touch device made the experience considerably worse, but a lot of the UI glitches would likely have happened without it.

A friend of mine has a collection of KDE glitches that look like the ones I experienced, but are completely different.

Every KDE update blog post talks about bug fixes, but during the time I had been using KDE I couldn't sense any improvement. It's like they either fixed a minute subset of bugs or kept putting in new ones.

Oh yeah, what I saw of Kirigami ... kind of sucked.

From what I could tell (purely from use!) they have at least 3 different UI toolkits: Qt Widgets (the traditional, solid feeling and looking applications), Qml with Qt controls (looks mostly traditional, but has some weirdness around scrolling, sometimes it's extremely spongy) and Kirigami, which is the worst of them all with the weirdness of Qml-based apps AND a look different from the standard platform theme (kind of like material design but with the proportions being all off). There might be some configurations/frameworks in between these because some applications would have normal touchpad scrolling, but were then also scrollable by just dragging the mouse cursor (touch scrolling was hit and miss in Qt Widgets apps, Dolphin for instance just wouldn't do it, touch-based right clicks are of course another horror).

Ideally I'd be totally behind Kirigami, I like the idea and I like Qml quite a lot too. But the UX was HORRIBLE.

CADT again. Did they fixed the bugs in 5 ? Or are they rushing to be ahead of Gnome GTK 5 ?

When are they finally going to move on to CSDs like GNOME?

Never, I hope. I want my window management to be consistent across windows.

Reply on news.ycombinator.com