As I keep saying, I would much rather see people take the time the fix the problems they have with KDE 4. There isn't anything in principle that should keep KDE 4 from acting like KDE 3.5. Prefer that the panel act like the 3.5 panel? Then someone can just make one. Prefer the task manager behave like the 3.5 one? Then someone can make one. They won't be included into the default versions, but one of the great things about plasma is that everything can be re-made, there can be more than one type of everything. There is no reason that people can't make it so, for instance, the KDE 4 desktop behaves identically to the KDE 3.5 one. The reverse is not true, though.

In the long run it will take a lot less effort for the people who prefer 3.5 to fix the problems they have with 4 than it would to maintain 3.5, since with 4 they get the benefit of all the maintenance and new features implemented by other people rather than having to keep the whole thing running by themselves. Linux is a fast-moving target and keeping all the various parts of KDE working well with it is no easy task. It is better to let the experts do that and let the people with coding skill who like KDE 3.5 do the much smaller task of re-implementing KDE 3.5 features.

I'm an old KDE 3.5 user. I love KDE 3.5. I used KDE 3.x starting back in early 2007, first in Kubuntu (nobody's perfect) and then in Debian; first Etch and then Lenny. I tried KDE 4.0 back in January of 2008 and thought it sucked; it was installed less than 48 hours.

But I love KDE 4.3.x, and I think 4.4.x is even better. I continued using KDE 3.5 right up until January of this year, but I found the transition to be quite painless and actually enjoyable. For the first time ever I'm running Debian testing rather than stable; I'm doing that mostly because I want to run KDE 4.x.

I am not a casual user. A computer is not a toy to me; it's a tool that I depend on to get my work done. I find KDE 4.3 to be completely usable for that purpose, and in many ways superior to KDE 3.5.

As for the plasma widgets, yes, they are very useful. I have a number of them on my desktop and in my panels, and I use them (although I wish I could make those miserable configuration handles stop popping up).

More features and options are never a bad thing.

On the other side of the coin, in what way was 3.5 "horrible"? 3.5 was magnificent.

I think a lot of opinions are still tainted by the early experiences with KDE 4.x... and yes, I do think the roll-out was awkward to say the least. KDE 4.0 was nowhere near ready for prime-time... it was so bad I didn't touch KDE 4.x again for two years. But I was willing to give it another try after awhile, and it would probably be wise for others to do the same.

By the way, if you're using Kubuntu that could be the problem by itself... that's the worst implementation of KDE I've seen.

I'm getting sorta sick of all the baseless Kubuntu bashing. And of course it's all vagueness like "it crashes".


95% of the time, crashes are upstream problems. A combination of luck of the draw multiplied by a lot of users means more bug reports.

Speaking of, Linux Mint uses basically identical KDE packages compared to Kubuntu. So if KDE apps were more stable in Mint, you just got unlucky when you used vanilla Kubuntu. (Plus, you should have reported these crashes to bugs.kde.org so that they can actually get fixed.)

But this is drifting off topic. Kubuntu really doesn't deserve the bashing.

Sorry, Jon, but I stand by my statements (and just for the record I don't like Mint much either). To me stability is the beginning, middle, and end. (K)Ubuntu, Mint, Fedora... they're all a little too eager to stuff new software that hasn't been adequately tested into the repos.

The Mint stuff was mainly directed toward's pazuzuthewise's post, where the mainly identical Linux Mint packages are mystically more stable.

JontheEchinda wrote:I'm getting sorta sick of all the baseless Kubuntu bashing. And of course it's all vagueness like "it crashes".

If you have an alternative explanation for why kubuntu users run into so many bugs that aren't seen in other distributions I would love to hear it.

A larger install base, for instance. Plus, without specifics the accusations are rather vague.

I don't entirely believe that, Jon. I'm not trying to start a flame war here; I just think we're dealing with a basic difference in philosophy. I've run Kubuntu and can tell you first hand it's a lot buggier than Debian. Most of the really popular "desktop" distros seem to feel the need to keep up with the very latest in software technology, which leads to (in my opinion) an unacceptable level of bugginess. Reliability means sticking with proven technology.

It's not really just a comparison with Debian. I really wasn't trying to pick on your post in particular, but with vague Kubuntu bashing in general.

2handband wrote:...although I wish I could make those miserable configuration handles stop popping up...

Click on 'lock widgets' on the desktop mouse menu; should do the trick.

OFF:

Well, one thing I can confirm is that Kubuntu is buggy. After 2 DAYS on usage I experienced such errors, that driven me crazy. For instance: 10 minutes after boot ALL the Icons become jittery. And this is only one out of many. On the other hand with openSUSE I don't have anything annoying, and my desktop works just fine.

ON:

Back to the actual topic: My first Linux experience was with Knoppix 5.2, It had KDE 3.x back than. I hated it. Not because of KDE, rather because of Linux barely supported my hardware (It was in 2006 - 2007 or 2008, I don't really remember). BUT I remember, that Knoppix sometimes ran FASTER on a 2200 Mhz Pentium 4 machine with 1 gig of ram with KDE 3.x, than any KDE 4.x distro runs now on my AMD X2 5600+ machine, with 4 gigs of ram. This is crazy. Although I absolutely love my openSUSE 11.2 with KDE 4, but I sometimes feel nostalgic to how fast KDE 3 was.
This is just my opinion on KDE 3, and how I remember it. Now it is old and unsupported, yeah, and KDE 4 is the future. And more complicated the software is, the more often it crashes. This is how things work. So now I guess people, like me, who loved the speed and stability of KDE 3 have to help the devs, -- or even better -- contribute to KDE 4, to make it better, faster, and more stable.

spoovy wrote:

2handband wrote:...although I wish I could make those miserable configuration handles stop popping up...

Click on 'lock widgets' on the desktop mouse menu; should do the trick.

But I want to move them around to from time to time...why this weird choice:

movable plasmoids w/annoying handles
non movable plasmoid w/o annoying handles

there are way to many popups is kde4, it's just to "alive"

aapgorilla wrote:
But I want to move them around to from time to time...why this weird choice:

movable plasmoids w/annoying handles
non movable plasmoid w/o annoying handles

there are way to many popups is kde4, it's just to "alive"

Define from time to time. Because, you can still lock them. And when you really want to move them you can unlock and lock again.

The question is how, exactly, you would move them. Plasmoids, I believe, are able to define how they react to mouse interaction, so allowing you to move by just dragging them will lead to problems with plasmoids not designed for that sort of interaction. For instance clicking and dragging in the blackboard plasmoid makes you draw, such a plasmoid would become unmovable if you remove the bar. So doing what you suggest will lead to inconistent behavior in plasmoids and will violate rules about plasma's basic functionality that widget developers based their designs on. It is not that easy to just remove something like that, you will need to completely redesign some plasmoids in order for them to be functional. And developers are not going to give you an option that they know could very likely result in a broken desktop because that would just lead to many more complaints and bug reports about "broken" plasmoids that were only broken because developers catered to a fairly niche use-case.

Not necessarily. If moving is done by a "press-move-relase"-sequence, it would be transparent to plasmoids if the container swallows these sequence. Technicly it would be identical to the way it is now, despite no popups show up. Same goes for resizing, you could have the plasmoids borders act like window handles. Benefits: plasmoids would work more like expected.