Great to see the KDE development community being so active. I think it has really been showing in the end product. Over the past 2-3 years KDE has easily become my favorite environment, even over macOS or Windows as of late. For the most part things just work, all the default apps are solid. In particular Dolphin is hands down the best file manager out there. Kudos to the KDE devs.
I always preferred thunar or pcmanfm back when I was into exploring all those Desktop Environments and Window Managers ; some of the xfce/lxde tools are particularly light on additional libraries needed, which makes them great options if running a wm only and wanting a slim system. lxterminal with i3 or dwm for example. Actually preferred that compared to rxvt-unicode. But happy to agree on the larger point: it's been showing in the end product. Particularly the KDE community has done a wonderful job slimming down its codebase.
Tried KDE a few months back for a short time, was surprised that KDE's footprint (in mb of space on disk, not ram) was not that much heavier than LXQT. The desktop experience seemed polished, though for me it has far too much gimmickery (in settings), and at this point I'm quite used to Ubuntu's Gnome. But happy for a healthy ecosystem. When I got into Linux (back in ~2008 probably?), I remember the choice with regards to full DEs was KDE vs Gnome vs XFCE vs LXDE ; there are more today, but I would hate to see one of them gone for good, and wish all of them well. Good to read that KDE seems healthy.
As a long-term (since last century) KDE user, it took me a long time to get used to Dolphin and wean myself off Konqueror.
I still see some weird artefacts / behaviour in Dolphin, notably around the sizing of the quick-access panels on the left, after window resizing, or display switching (effectively causing the former).
The copy/move option was present in Konqueror also, along with the same ctrl/shift modifier option to force copy / move without the pop-up.
FWIW with Microsoft Windows, for as long as I can remember, in Explorer you can right-mouse drag directory entries around, to trigger a similar copy/move pop-up. A little-known feature, but increasingly handy as Microsoft continues its war on its own users and obfuscates actual locations & paths, with the default move-if-on-the-same-drive / copy-if-not becoming harder to intuit due to these location abstractions.
I am not that aware of the various file managers and their specific features these days. I guess it's about use case. I always want primarily something "which doesn't annoy", since the file manager is mostly there "just in case" for me. I live in the terminal, even for most file operations.
Back when I tested them out (years ago, so not relevant today, still): Considering dependencies (if not running KDE), but even then Dolphin always seemed to do something quite not intuitively for me. Can't name examples anymore.
Yeah, totally. For some strange reason I'm envious of people who are mostly in the terminal, while at the same time that wouldn't at all suit my needs (I do graphic design, mapping, web and app development... and I just like using a mouse and clicking stuff!). And I haven't hopped around and tried anything new in a few years. I've lost my desire to break and fix things or try anything new, and now I just want something that works and that I'm used to (which for the last 4 years has been Arch + KDE Plasma).
For the split views, there is Krusader which I have found a little more stable than Dolphin of late.
The last annoys me at times. I'm used to years of clicking and dragging to copy, and holding down shift to move instead. All DEs and even Windows behave this way, I think they should bring it back at least as an option.
Like GP I'm a very happy pcmanfm user most of the time, but I use spacefm sometimes and I know it has split window as an option. Not trying to convince you to leave dolphin, just mentioning in case you like exploring.
- Split window (just use your window manager?)
- Thunar has Tabs, and I'm sure lots of other file managers do too
- That is a nice option, and others should copy it.
Agreed about tabs and split windows. I think they take over the window manager's job, and do it poorly.
About the menu, it is nice, but not unique. Windows' Explorer, for example, shows it when dragging with the right mouse button. Options include copying, moving, zipping and actions from any registered shell handlers.
That said, I think KDE is still the best desktop environment across all the operating systems I've tried.
Thanks. I haven't explored other file explorers in a while, and I figured that some of the features I listed exist elsewhere. But I was typing from my phone and was thus too lazy to type a qualifier.
"Just use your window manager" - and have two separate Dolphin windows? I mean, yeah. But it's very nice to have everything contained in one window. Right now I have five Dolphin tabs, four of which have split windows (I don't NEED this setup, but I like it).
It's a good thing for those "offended" by in-depth configurability that KDE's System Settings now opens to a reduced view called "quick settings". It's the "simple" part of the "Simple by default, powerful when needed" slogan manifesting itself.
I agree. I had been using OpenSUSE for years, then converted to Kubuntu about 2 years ago. No regrets. KDE is currently excellent; quality is high and they are thankfully being very conservative in design and development. This has produced a pleasant, complete and stable DE.
The only real problem KDE has now are the scars of past mistakes that have kept people away and all too often mired in Gnome.
What did bring you to Kubuntu from openSUSE? I'm just curious, as I've been exploring the latter as a possible alternative to Canonical's offering. Compared to Kubuntu it feels more friendly to enterprise administration than polished for personal use.
Everyone that distributes Linux software targets both Debian/Ubuntu and RHEL/Centos/etc. and OpenSUSE is frequently neglected. I have to deal with a great deal of such software. I know this can be worked around in most cases. The added friction is, however, not helpful.
And that's a shame, because OpenSUSE is excellent. If KDE continues to thrive SUSE deserves a lot of credit for keeping it alive.
The author also links to a similar assessment in the Gnome community, which I find really interesting.
With Gnome we can see the interest went down after Gnome 3 was released. Many people like me ran away from it scared by the usability problems and unreliability of the extension system. And Gnome kept the same complained-about interface until today, and the project never picked up the pace it had before.
While KDE4 was also seen as a big failure during its release (many bugs!), the biggest issue was simply bugs: not so many people complaining about the direction of the project, although these people probably existed. The project was able to fix the biggest issues and keep moving on (although KDE is still quite famous for being buggier than some of the alternatives), and now they have a community that can probably be described as "healthier than before", and certainly much better when you compare to what's happening to Gnome.
To give another opinion, I also thought GNOME 3 was not good, until around a few years ago. It's a lot better since then, they really ironed out a lot of the issues so you see the contributions start to raise up again recently. The extension API is also getting a lot more stable, but it's not totally there yet.
One area where I think KDE does a lot better is with QML. At the moment GTK doesn't have anything as nice as that to build apps, and that hurts contributions. There's some neat new declarative UI things like vgtk [0] that are looking promising though.
Interesting write-up. I was a big fan of KDE 3 but never used KDE 4 or 5 in earnest because every time I gave it a serious try, it would start crashing randomly or some critical piece just wouldn't work the way it was supposed to. And that was _after_ disabling a bunch of the "fluff." But a few weeks ago, I decided to give it another try (via the packages in Ubuntu 20.10) and it's been _wonderful_. I can tweak all the knobs I want to tweak, and it has been very solid so far.
I really appreciate software projects that place a higher priority on evolution rather than rm'ing huge swaths of working functionality, designing a worse replacement, and then calling it innovation. KDE did this with 4 and then took a decade to get back to something stable. Let's hope the cycle doesn't start over again anytime soon.
The plans for plasma 6 and kde frameworks 6 are mostly around removing old deprecated features, simplify our dependency tree and porting to qt6. And there is no plans of adding any big features.
Mostly really boring stuff like deprecated API functions with better replacements and such. KDE does not remove features for the heck of it, if it happens it is almost always because of rewrites. There are no rewrites of that nature planned AFAIK.
You can find all the discussion here: https://phabricator.kde.org/tag/kf6/. Don't worry the features are mostly in the API, often because there is now a better available api (e.g. KPluginInfo vs KPluginMetaData, QRegExp vs QRegularExpression, ...).
KDE 4 frequently failed in weird ways for me, but KDE 5 has been stable and much more pleasant than Gnome. It's not only that I learned to avoid the weirdness, it definitely improved significantly.
If only it didn't assume my external screens are part of the same touch surface of my laptop, because that makes touch unusable... maybe KDE 6.
> If you are wondering who in the 1999 cohort made most of the contributions, it’s Laurent Montel and who in the 2010 disappeared in 2018 it’s “Montel Laurent”. Unfortunately I couldn’t find a nice way to merge both.
Just wanted to chime in and say I've had a great experience submitting a small bugfix patch to KDE, even as someone new to C++. The dev docs provided a good intro, the devs were helpful in pointing out relevant sections of code (as comments to my bug report) and constructive feedback was given on my first patch submission, leading to a second attempt that was merged.
In the beginning (95-ish) Qt was license encumbered, there was serious agitation about KDE becoming popular while using it, leading to the very-open GTK based GNOME desktop competition. It was a big kerfuffle at the time.
100% agree, I missed that - at this point in time (and maybe still a bit today?) there was a lot of grift between C and C++ being used for Linux systems things around that time as well (including how big the resulting binaries were, because modems to download!). Thanks for the reminder.
My impression was more than a little regionalism. KDE came from Europe. GNOME came from America. Most distros that I know come from America.
In addition, Richard Stallman made lawyers nervous of KDE by claiming that using GPL libraries with Qt (which at the time was under a proprietary license with source code available via FreeQt, definitely not a FSF-approved license) meant the KDE developers had no legal right to use GPL programs at all, even after KDE and Qt were relicensed under GPL.
Stallman eventually relented and granted permission to use the FSF’s own software, but the KDE project didn’t ask for it and didn’t bother to get forgiveness from other GPL license owners. Whether they need to go through that ceremony was never tested in court. GPLv3 switched to a much more reasonable standard, just provide the source under a compatible license.
By the time the license tempest-in-a-teapot reached a conclusion, GNOME and KDE already had substantial-enough installed bases and corporate sponsors to keep going indefinitely.
GNOME was backed by RedHat and other American developers, which included the likes of IBM, Intel, etc.
The bulk of QT and KDE developers are European; the only major European distribution, SuSE, went with KDE - but apart from that, the project could never attract the same level of investment as GTK-based efforts. It’s kind of a microcosm of the US/EU imbalance in digital tech.
Add to it that C++ (Qt) was never as popular as C (Gtk) in academic circles, which are major contributors of fresh developers to opensource in general.
I don't think the reason is really c++ vs c. But the regional aspect is spot on. Also when SuSE was bought by an American company (Novell), they fired all the KDE devs.
Another aspect is the licensing, GTK was licensed under LGPL way before Qt and RMS casting doubts on the legal aspect of KDE even after Qt was re-licensed under GPL didn't help.
All these played a role in the early 00s, nowadays the problems is mostly that GNOME has many more enterprise features and integration with other RedHat products and it is hard to catch on.
I can't say I'm familiar with the repositories behind KDE, but can somebody point me to say the top 10 - 20 repositories with the most activity, that would be good for determining overall health? I'd like to analyze them and I'll share my findings.
I've studied hundreds of popular open source projects and I've found that "days active" and having lots of developers with mid to high level impact to be extremely good indicators of project health.
vscode and GitLab are probably my favorite repos that best highlights this. If you look at the vscode project
and switch to the impact view and sort by "days active" you can get a sense of what I mean. The more developers above 25%, the better. "First commit" which is also discussed in the post, is also another good indicator. For vscode, you'll find a lot of developers with 5 years of experience.
And something that may seem counter intuitive is, you don't want contributors to have "very high" impact, as this indicates high bus factor risk. For impact, you actually want people to lie on the bottom of zone 2, as that indicates high individual impact and evenly distributed effort. Since GitLab isn't on GitHub, I can't share the insights for it, but that's what GitLab looks like.
Note: Don't install my tool as the docker image has an expired license, which I need to fix.
Always happy to hear good news about KDE. Recently I've installed OpenSuSe Tumbleweed for my wife (previously she was using macos and she is not 'technical') and it's great, even better than I remembered. Stable, customizable, one can make it beautiful with a minor effort.
The biggest issues I still have are not KDE related:
- touchpad experience on Linux just sucks compared to macos. Even started sponsoring the gitclear because of that. I haven't tried the mtrack yet, but lack of the reasonable configuration tool is a bit discouraging.
- there is no modern mail client for Linux. The only "serious" mail client is thunderbird. Unfortunately it hasn't received enough love/attention in the recent past :(
Shout out to the great people in the KDE Community.
I was in high-school and being enamored with the open-source concept ("You mean I can modify anything I want about it?" - hah) I wanted to make my mark in a simple little way. I remember fondly when lurking around in IRC and somehow decided I would change the login screen. Got pointed to the KDE greeter channel and that is how it all started for me. I kept annoying them, did not know how to build, compile, nothing. Thanks to d_ed for answering my annoying questions, that is how it all started for me.
> If you are wondering who in the 1999 cohort made most of the contributions, it’s Laurent Montel and who in the 2010 disappeared in 2018 it’s “Montel Laurent”. Unfortunately I couldn’t find a nice way to merge both
The thing I don't like about KDE is that it is that KDE apps aren't individual components but part of a large framework. Anytime I go to install a KDE app on Arch, it wants to install like 30+ packages and unnecessary apps with it and I just end up finding something more lighter weight that doesn't have any or only a couple dependencies and it usually ends up performing a lot better.
The apps are already individual things that don't depend on each other, except in some cases where KParts are used. A KPart is kind of like an application turned into a library. Usually, this is not something users need to be concerned about. The integrated terminal in the Dolphin file manager is a Konsole KPart. openSUSE (and probably others like Fedora and Ubuntu) actually split Konsole into `konsole` and `konsole-part`, so it's not a problem.
That sounds like an Arch packaging issue, a misunderstanding of what the packages actually do or you're installing a metapackage and not realizing it.
Before we had a big libraries called kdelibs and kdepimlibs and people would complain that they were too big and so nobody outside of KDE would use them.
Now we split everything, and people complain that they are too many libraries. There is no way to make everyone happy. Note that all this packages are very small and all provides features you really want to have.
