I usually don't open topics about vague and overly general matters. But as someone who uses KDE for years and wishes to see it reach perfection, this is constantly on my mind: Why is KDE permanently buggy in one way or another? Is there any hope that this might ultimately change someday?

Since I know some people will probably claim I'm wrong, I can certainly confirm KDE is constantly buggy. Usually it's indeed little things, and part of them happen during distribution upgrades: A menu entry that's not persisted, little features missing, a component occasionally crashing and starting back up, and others. Still, during the years in which I used KDE4 and the months after I've switched to Framework 5, there hasn't been a moment I can think of when there wasn't some little but obvious issue.

If anyone runs Framework 5 and wants simple examples they can test right away, a few that pop to mind are: Nothing in System Settings -> Online Accounts works, clicking any of the buttons there does nothing. If under Desktop Settings you set your wallpaper type to slideshow, then set the seconds to 0, the field becomes 1 next time you open up the menu... so for example, 15 minutes and 0 seconds turns into 15 minutes and 1 second. An issue that was recently fixed was the password field of the screen locker losing focus if other programs started up in the background, even while you were writing your password. Before that, enabling the Baloo file indexer caused search in Dolphin to not work at all and return an error, which was supposedly fixed now. That said, I'm willing to bet nearly everyone can confirm desktop effects cause artifacts and flickering, or say that plasmashell or krunner have crashed for them at least once during the last few days. Yeah... and icons in the system tray have black boxes behind them now, whereas the device notifier sees my hard drives as removable storage (memory sticks) and offers a button to unmount even the root partition:




Now don't get me wrong: Most of these don't pose an obstacle to using the system, and are at most just annoyances. What mostly upsets me is that KDE has reached and maintained a great level of quality throughout the years... yet all that is ruined by the presence of a bug that's rubbed into the user's face. This is especially a problem if you're a former or present Windows user, looking up to KDE as an open-source alternative of the same grade: KDE offers greater or equal quality and visual fidelity to Windows, often more features than Windows includes by default... yet in Windows you never notice menu entries that don't work, or the desktop periodically crashing because of a widget you added. I worry this might be part of why fewer Linux distros default to KDE, with openSUSE being the only one I know of that currently does.

Usually, when it's a considerable issue, I do report it on the bug tracker. Problem is that many bugs take months or even years before anyone even confirms they're taking a look at them, even some of the simplest! Although I understand that KDE is open-source software and people aren't always paid to handle such issues, that still feels like a lot. The purpose of this thread isn't for me to complain... it's just something I felt I should bring up, and hopefully draw a bit more attention to. Perhaps more automated tests could be implemented with qt5, to catch little things like settings that don't persist in menus or custom plasmoids crashing plasmashell or rendering glitches?

Please report those little bugs as soon as you find them, especially if you are running testing versions in your distro. In fact, the more people who do that, and report glitches *to their distro*, the better. And better yet: volunteer with your distro to triage bugs, and make sure that KDE bugs are reported on bugs.kde.org. Well-reported bugs help all of us, whether the issue is "little" or big.

Of course it helps if you can provide a patch to fix the problem, but even if you can only provide detailed instructions on how to trigger the bug, that helps. And naturally if you want to help triage bugs for KDE, that would help as well. Please see the KDE-quality list for more information, or the wiki: https://community.kde.org/Get_Involved/Quality

I recently installed Kubuntu 15.10 with Plasma 5.5.3 and I have to say that i've been experiencing a lot of little bugs. I really think that the KDE team should focus on bug-fixing, optimization (performance improvements) and stabilization instead of adding new features, but that's only my opinion.

kde-lunarparks wrote:I recently installed Kubuntu 15.10 with Plasma 5.5.3 and I have to say that i've been experiencing a lot of little bugs. I really think that the KDE team should focus on bug-fixing, optimization (performance improvements) and stabilization instead of adding new features, but that's only my opinion.

While running Plasma 5.5.3 I encountered quite a few glitches as well. I'm happy to report that once I got 5.5.4 through the backports PPA [1], all of them disappeared. Your experience might be different, of course. It proved to me that the Plasma team did exactly as you asked, and focused on "bug-fixing, optimization (performance improvements) and stabilization."

1. https://kubuntu.org/news/plasma-5-5-3-a ... ntu-15-10/

Valorie

valoriez wrote:

kde-lunarparks wrote:I recently installed Kubuntu 15.10 with Plasma 5.5.3 and I have to say that i've been experiencing a lot of little bugs. I really think that the KDE team should focus on bug-fixing, optimization (performance improvements) and stabilization instead of adding new features, but that's only my opinion.

While running Plasma 5.5.3 I encountered quite a few glitches as well. I'm happy to report that once I got 5.5.4 through the backports PPA [1], all of them disappeared. Your experience might be different, of course. It proved to me that the Plasma team did exactly as you asked, and focused on "bug-fixing, optimization (performance improvements) and stabilization."

1. https://kubuntu.org/news/plasma-5-5-3-a ... ntu-15-10/

Valorie

That's fantastic! but I am only seeing Plasma 5.5.3 in the Kubuntu Backports PPA instead of 5.5.4 so I can't jump to that version, did you obtain Plasma 5.5.4 via Kubuntu Backports though? maybe I am missing something...

Oops. I forgot that I tested 5.5.4 before it got to backports. Sorry about that. It should land as soon as a few more kinks are worked out.

Valorie

I agree that kde devs should, for at least one cycle, stop developing new features and just focus on fixing bugs.

I couldn't agree more.

KDE is awesome, plenty of useful and handy features that besides surpisingly work without wasting -in general, there are exceptions, of course- too many resources so you can have the latest KDE on a machine 8 years old ¡working smoother and snappier than a 3 y. o. Windows one! I sincerely believe is the best desktop environment even for Windows users. But, as you say, it has always to have some minor bugs that make impossible to me, and I suppose many others too, to recommend my acquaitances to switch from Windows. I just can't show how well it works and a week later to have a call from my acquaitance complaining because something has stopped working, some configuration has vanished or some silly failure has made something that yesterday night worked, don't work today morning.
Accustomed users like us know, usually, how to solve silly "microfailures", and when we don't, we at least know how to find solutions on the net or ask for help; but average PC users don't, and even if they would, they should not have to, because those little glitches should simply not occur. I know no software is perfect, but in Windows don't occur such things, it may be slow, heavy, uneccesarily complicated sometimes, whatever, but you know that the next day things are going to work exacty the same; in Mac don't occur either; in Gnome occurs less (or at least so it was when I used Gnome, I admit it has rained a lot since then); and the most uncomprehensible thing for some of us is that the "fat" things of KDE work well; they're usually silly details what fail: one day is this feature/app/wahtever; another day is that other, the next desktop revision those are fixed but a new one appears; one day, without having installed anything nor touched any config file something stops working or works weirdly... How is it possible that developers do such fabulous work with the big things and then never get rid of small glitches? Fortunately those glitches have usually an easy workaround until they are fixed, or are so small that we can live with them, but if you comment with other KDE users you see that almos all of them can tell the same experiences, and most of us have tried more than half a dozen of distros on several machines too, so they aren't distro-specific problems or hardware failures.

@valoriez, yes, we report bugs (at least I do as you can see if search in BKO), but many times the process you need to follow to triage a bug is so time consuming that you report one instead the 3 or 4 you'd want to, because the time you lose trying and reporting is many often way bigger than the time you lose dealing with that little bug; and you lose even more motivation when you know that no matter if the bug you report is solved because the next upgade there will appear new ones; sometimes I have the impression that what I do serves for nothing, that I can't contibute to make KDE better because a fixed bug means another new bug by other side, a sort of "fatality" o who knows, hehe. For non programmers like me it's very surprising how devs can build such big and complex "structures", and make them work very well, and then constantly forget to put a screw here or a gear there.
Also sometimes bug reports are ignored, plain and simply, or the developers tell you that they will have a look on them, and there dies you bug report. And we do understand that sometimes the developers can't handle all the bug reports they receive, especially the minor bugs reports; after all, if there are really important bugs, those are the ones that need to be prioritized, clear. But then that's perhaps part of the problem too, perhaps there should be a percentage of working hours a month dedicated exclusively to these minor glitches. And what about those companies and institutions that use KDE to earn good money but don't contribute back? Couldn't be done some pressure on them -"ashaming" them, publicizing that they use KDE but don't contribute back or contribute ridiculously, for instance- to convince them to contribute with manpower or donations so more developers could be hired?
Said all that, I'll keep reporting bugs and collaborating with the developers as much as I can, of course, but hey, let's admit that bug reporting isn't a panacea at all.

I don't know, I don't have the solution, but MirceaKitsune is right, KDE doesn't need to improve too much in features; ok, it obviously is quintessential to software getting better features each revision, of course, but I mean that KDE already has great features and works very well, we aren't second class citizens anymore. The prioritary goals have already been reached thanks to the great work of KDE team... except those damn annoying small glitches that year after year, version after version, persist like a plague and keep away many many potential users and donors. (u_u)