KDE SC 4.7.0 Officially Released

[BenderRodriguez]

There are always two sides on the stick

Some would complain that they introduce new stuff instead of stabilizing it, the others would complaing that they don't introduce new stuff (but you never hear them complain that it is stable ).

I think KDE has come to the point where it is mature and quite stable. The good thing is that they are not going to make some instabilities when moving to Qt 5 and ultimately to KDE 5 so there will be no 4.0 experiences

[Alejandro Nova]

My experience.

1. I made that post in Martin Gr??lin blog. That's true: my GeForce 6150 is running just great, and I tried to stick to stock Fedora packages, so, no OpenGL ES for me. Maybe if I compile the OpenGL ES KWin I can get even better performance.

2. About this:

Nepomuk/Strigi - i'm curious - when will be usable? Here's simple test and big fail (again).

There are memory leaks in nepomukstorage and a lot of regressions, partly because of Strigi. You NEED an updated Strigi, and you won't get it with any distro (no, 0.7.2 won't cut it). Vishesh Handa of the Nepomuk team is working hard to fix those leaks, with the backtraces I provided. Let's hope to have a better Nepomuk in KDE 4.7.1.

[Awesomeness]

I only upgraded to kde 4.7 yesterday but compared to previous upgrades there seems to be a lack of noticeable refinements to how you actuality interact with the gui.

Is this a sign they've got it close to how they ultimately want it to work or a lack of new ideas? Maybe I'm missing something or perhaps I'm simply expecting too much from every release? A new version of kde 4 does typically deliver noticeable improvements.

It's a sign of the developers working on other stuff. Many core devs are currently working on further modularizing the Frameworks, even replacing technology that is based on functionality which is going to be deprecated in Qt 5.0 (eg. they're writing QML bindings), and Plasma Workspace devs are currently mostly concentrating on Plasma Active which in the long run will also be the future for a purely QML-powered Plasma Desktop.

KDE devs are still working like crazy but with so much going on behind the scenes on boring technical stuff, it's hard to show off fancy GUI results.

[V!NCENT]

Totaly unrelated, but I whish I would know how to develop for KDE, because I would do so many things differently.

I like KDE more than anything, but someone needs to fill the gap between total usability [Gnome] and total technology and feature [KDE]...

I dreamed up an interface and functionality today... Maybe somthing for the future...

[Luke_Wolf]

It's a sign of the developers working on other stuff. Many core devs are currently working on further modularizing the Frameworks, even replacing technology that is based on functionality which is going to be deprecated in Qt 5.0 (eg. they're writing QML bindings), and Plasma Workspace devs are currently mostly concentrating on Plasma Active which in the long run will also be the future for a purely QML-powered Plasma Desktop.

KDE devs are still working like crazy but with so much going on behind the scenes on boring technical stuff, it's hard to show off fancy GUI results.

Of course these efforts allow KDE to expand out onto other devices using multiple specialized UX's rather than one generic one, which then assuming these efforts take off means many more developers, when I look at the whole of whats going on, I think there's a KDE everywhere strategy in play here.which ultimately could work as a catalyst for Linux domination on the desktop. Think about it, if you can move the core of usage into KDE programs, then the user doesn't care about the backend do they?

Totaly unrelated, but I whish I would know how to develop for KDE, because I would do so many things differently.
I like KDE more than anything, but someone needs to fill the gap between total usability [Gnome] and total technology and feature [KDE]...
I dreamed up an interface and functionality today... Maybe somthing for the future...

How does gnome (assuming you're speaking of 3) have absolute usability? I thought usability = use ability, the ability of an average user to sit down and use it without prompting. I'll grant you that gnome 2's interface is downright obvious how to use, but Gnome 3 like OS X and Unity is just backwards. KDE can be argued to be less usable in the default paradigm than gnome 2 however that is only valid if and only if the person has never dealt with the paradigm that windows introduced, it could also be argued that lancelot as opposed to the kickoff applications launcher would be a better usability default. That being said I think that I could grab an average user off the streets set them in front of KDE and they'd know what to do, they might not utilize some of the more advanced functionality but they'd fulfill their purposes easily enough.

[devius]

There are memory leaks in nepomukstorage and a lot of regressions, partly because of Strigi. You NEED an updated Strigi, and you won't get it with any distro (no, 0.7.2 won't cut it). Vishesh Handa of the Nepomuk team is working hard to fix those leaks, with the backtraces I provided. Let's hope to have a better Nepomuk in KDE 4.7.1.

Well, that video illustrates my own experience with KDE since version 4.3. Search simply does not work with or without Nepomuk/Strigi, so I suppose that "a better Nepomuk in KDE 4.7.1" is one that finally works. In 4.3 nothing was ever returned when I did a search. In 4.5 some results started appearing, but I could be in a folder looking at a file, typed that file name in the search bar and dolphin couldn't find it... holy crap, how can they have failed so badly in implementing a search engine?

[pejakm]

I have a solution for all of you guys: simply don't use strigi, nepomuk and akonadi.

[devius]

I have a solution for all of you guys: simply don't use strigi, nepomuk and akonadi.

Thanks but I already came to that conclusion 3 months ago and ditched KDE in favor of Ubuntu 11.04. Not just because of that, but obviously that was also one of nails in the coffin.

[RealNC]

Is it normal that moving a window around (any window, it's not application-specific) results in the kwin process eating 60%+ CPU? Because that's what happens here. Actual effects (like minimizing/maximizing or "present windows") make kwin eat almost 100%, and that's when FPS go down the drain and the animations feel sluggish.

If there's load elsewhere (like watching an HD video or emerging [Gentoo user here] something) then it turns into a real crap-fest. Since everyone else has so great results with this, using the same driver even, I wonder what the hell is wrong on my end.

[devius]

Since everyone else has so great results with this, using the same driver even, I wonder what the hell is wrong on my end.

When/if you find out let me know because KWin also ran like crap on my HD4200. I always thought KWin just wasn't optimized enough and on more powerful GPUs nobody would notice the bad performance. BTW, I don't know if you tried with fglrx, but if you didn't let me just save you the trouble and let you know that in my case it made no difference whatsoever.