Friday, January 28, 2011

KDE SC 4.6 Review

Hot off the oven, the latest from KDE was released just a few days ago. It builds on the greatly improved and reliable KDE 4.4 and 4.5 series and brings a number of exciting new features and enhancements, as well as a cool new look thanks to an impressive new wallpaper and much improved Kwin effects.

Click on image to enlarge.

KDE SC 4.6 brings a completely redesigned activities system. I have to say I have never used it (and honestly don't see it that useful), but it seems this new system is less complicated and more intuitive. On the other hand, one piece that has also got a new design from the ground up, one I do care about, is the Power Management configuration system.

Click on image to enlarge.

Settings have been split into two categories now: Global Settings and Power Profiles. Not that different to what was there before, but perhaps organised in a way that makes life easier for users. The Power Profiles piece is the one that has experienced the most relevant and evident changes, as depicted in the screenshot above. There is nothing revolutionary about the new interface, perhaps not even attractive, but it does what it is supposed to do.

There are many more new features/enhancements, such as improved program start-up notifications and a revamped Oxygen GTK theme that should allow for better integration of GTK apps (it's an improvement, but can't say I am impressed with this one, I still see applications that look out of place). Dolphin gets advanced filtering features and other applications from the KDE realm also get updates, such as Amarok, which is now on version 2.4.

Click on image to enlarge.

Click on image to enlarge.

In my opinion, the most welcome changes come from Look&Feel and Kwin compositing, probably the changes users will notice first. The new wallpaper provides a new vibe to both the KDM theme and the desktop, quite an impact for such a simple change. The widget interface is similar to what we saw in KDE SC 4.5.x, but got a subtle face lift. There are some new widgets added to the mix as well.

Click on image to enlarge.

I wanted to save the best for last: Kwin effects are finally worth using! Ever since I started using KDE, I thought its effects were slow, limited and lacked the smoothness and flexibility of Compiz. As a result, I have alwasy tried to fit compiz in, probably not the best idea. This time around, though, Kwin effects look and behave great... it is the first time I decided to get rid of Compiz! Don't get me wrong, I still miss some Compiz features and effects, specially how you can tweak an effect in any imaginable way. I also miss how Compiz provides a much more flexible interface to create shortcuts you can call effects with, but like I said, this is the first time Kwin stands the comparison.

Click on image to enlarge.


Unfortunately, there are still things I dislike about KDE. First and foremost, I think the project is going down a path that can make it quickly become a thing of the past. Efforts continue to make an already thick client thicker, when the rest of the world is looking elsewhere, to the cloud, to be more specific. Projects like Marble or the work put into revamping Kontact (which incidentally had to be left out of KDE 4.6) seem redundant, tons of hours of work that will probably (and sadly) go unnoticed and used by a very small minority.

The way I see it, KDE is stretching way beyond its reach. Instead of using its growing community of developers to take on new projects, I believe they should concentrate all of those resources into creating a truly amazing desktop manager, one that is ready and relevant for the next ten years. I think it's about time we realize Windows is no longer the reference, because even Microsoft is falling behind in this race.

Computer Science has always been about adjusting to change quickly, and the cloud revolution, empowered by an ever growing market of smartphones and tablets, is only accelerating change. It will take no prisoners. It's time to realize efforts to create an ever more powerful client are essentially wasted time. Computing is no longer about productivity, social interaction and enhanced integration are just as important now. People want to see interfaces that are pretty and work efficiently, and they don't want to have to learn new stuff when what they use is already awesome. Honestly, can Kontact ever beat the integration Google is putting in place with all its applications and services? Can Marble ever hope to be better than Google Earth? Can KDE really develop a mobile device interface that is able to compete with iOS or Android? (Note that we are no longer competing against a proprietary monopoly based on closed technology, for many of Google's projects are open source and community supported)

I think the answer is clear. How long will it take for the project to realize many of its elements are already offside? Time will tell, but it is a shame to see so much potential not fully coming to fruition.


Is KDE SC 4.6 a good release? It is indeed. It still bugs me that certain basic shortcomings, such as a proper icon theming system, are getting no love from developers and keep failing release after release, though. Other than that, there are some undeniable improvements and some exciting new features, so make sure you give it a go.


Kubuntu users can easily upgrade their KDE desktop to SC 4.6.0 using the PPA available. Simply follow the following steps.

1.- Open a terminal window and enter the following commands:

sudo add-apt-repository ppa:kubuntu-ppa/backports

sudo apt-get update

sudo apt-get install kubuntu-desktop

2.- The last step may or may not work (it did not for me), so you may need to run the following command:

sudo apt-get dist-upgrade



  1. An interesting review and I have to admit KDE looks everytime more and more promising but nevertheless after trying it I always go back to Gnome, I think Im too much used to it simplicity.

  2. Well I for one have finally decided to switch to KDE, After trying out KDE 4.5 I was still "siting on the fence" between GNOME and KDE. Well 4.6 did the final push over to the KDE side. And one of the reasons were Activities. I love how you can associate applications with activities and then when you switch between activities the applications automatically close and restore. This alone is pure ingenious to me. The new filter sidebar in Dolphin is also simply awesome. It's also incredible how they manage to improve the looks with every release, even if it already looks gorgeous. It looks so good that even my sister (a graphic designer with MacOS) is starting to have some jealous looks over at my Linux/KDE screen. Anyways, I'm a really happy new KDE user and judging by the fact that GNOME2 is mostly stagnating now-days (GNOME3 Shell is a complete disaster in my experience), and how fast KDE is improving I'm only going to be happier in the future. great job KDE developers!

  3. I cannot wait for 4.6 to arrive on Debian!

    What font are you using in the Power Management screenshot (the bold font on the right half)? It's horrible! It isn't the default, is it??

  4. I enjoyed reading your review. I'm normally a GNOME user, but I've been curious about KDE 4.6 and have been trying it out for a few days on Archlinux. I agree that Kwin has come along well. It doesn't have quite everything that Compiz has, but Kwin is much better integrated with KDE than Compiz is with GNOME. I agree with Steven that the KDE designs did a job, and it is quite visually stunning.

    Still, from a usability standpoint, I find several shortcomings in KDE. I find it to be a little too "click-happy" where it seems that you often have to click through 2-3 extra windows to get something done, such as set a keyboard shortcut. Also, changing the appearance should be integrated into one portal, rather than making the user have to jump around to change the windows borders, the panel, and icons. Lastly, I agree that some of the KDE native applications are lacking. I tried KMail for about 10 minutes, and quickly went back to Thunderbird. Konqueror was once glorious, but it doesn't hold a candle to the modern browsers. Finally, 4.60 is still a little buggy, where I've been experiencing some crashes from time to time.

  5. Thanks all for your comments

    @lefty.crupps: Yup, I am using the default Ubuntu font, size 9. That font you refer to is (I believe) the small setting (Ubuntu font, size 8) which is set to bold by design.

    I agree it looks bad, but I already said Kubuntu fonts don't work that well.

  6. VLC is messed up on KDE 4.6.0 and the devs are pointing fingers at each other!

  7. After a few more days using 4.6, I have to say it feels immature, not solid enough. It's not just that there are some bugs here and there, which is somewhat reasonable for a brand new release, but the whole behavior feels a bit weak, like it is about to break.

    Opening several applications makes things slow down. Internet browsing is behaving very strangely and I sometimes lose my mouse pointer or keyboard cursor for no apparent reason.

    I think 4.6 will eventually settle down and become a great release, but it will take a few stability patches before that happens.

  8. Hi Chema,

    > Projects like Marble or the work put into revamping
    > Kontact (which incidentally had to be left out ofKDE
    > 4.6) seem redundant, tons of hours of work that will
    > probably (and sadly) go unnoticed and used by a very
    > small minority.

    Well, if you consider Marble redundant then there is only one thing you should do: don't use it!

    Judging from your blog it seems that ...

    * you think that the Marble Team works on Marble with the sole intention that we want to add value to KDE.

    This is not the case. Our mission is to create the best and most popular Free Software virtual globe on every platform. We are just providing Marble for KDE (among other platforms) because we think that it's a great desktop with a great technology and platform behind!

    * you have concerns that just because Marble exists the quality of the KDE workspace ("desktop") is magically lowered. This is not the case. Marble doesn't make the KDE Workspace more fat. It's just another application/library that is optional and that we recommend to install. You can install Google Earth instead if it suits your needs better (but that - while based on the same Qt technology) will take more resources).

    * you think that all those people that work on Marble should work on the KDE Workspace itself. This would magically improve the KDE Workspace quality and feature set and make it magically more competitive.

    Again your idea of how things in the OpenSource world work is faulty: Putting more people on a project doesn't make the fruits of the project better (Too many cooks ...). For a great desktop experience you need a small dedicated team of experts on this field which believe in their mission. Of course you could delegate some more people to work on the KDE Workspace. But why should that be the Marble people? Why shouldn't it be you? Or some other people who are not yet part of the KDE developer community?

    * You have concerns that Marble is lacking users and is only used by a minority

    Thank you, but we are quite happy with the amount of our users. We would still like to see more people using it. But there is no reason for us to be unhappy. Our project is only 5 years old and we have evolved in one of the most popular virtual globes. And we ship on every single bigger Linux distribution and we know that there are thousands and thousands of happy Marble users.

    Best Regards,

    Torsten Rahn

  9. > Can Marble ever hope to be better than Google Earth?

    Yes! Better in which discipline? Routing? Showing your house? Being more usable? Being Open Source?

    This question is about as useful as the question "Will Gimp ever get better than Photoshop?", "Will Linux ever get better than Windows?". Some people consider Marble already a better solution than Google Earth - since Marble addresses their use cases better. But to the Marble Team Google Earth is not the reference. Our aim is not to "clone" Google Earth as Free Software. Our aim is to create a great Free Software virtual globe. And the current pace of development is great. We are exploring more use cases that we consider important (Education, Navigation) and the progress is faster than in other projects (Keep in mind that Marble was only started 5 years ago while Google Earth's software exists since more than ten years already).

    > cloud revolution, empowered by an ever growing
    > market of smartphones and tablets, is
    > only accelerating change

    You think the web is a great platform? We agree. That's why we are making use of it in Marble already. I also have plans for a nice small version of Marble using HTML 5 ... which would help us to get Marble available on all platforms.

    Will the only application that runs on future platforms be the browser? I doubt it. Web applications work pretty well for use cases where collaboration is key. But it doesn't work well for others.

    > I think the answer is clear. How long will it
    > take for the project to realize many of its
    > elements are already offside?

    Even if that was the case, what should "the project" do? Should it recommend not to develop applications using KDE libraries?

    > Time will tell, but it is a shame to see so
    > much potential not fully coming to fruition.

    If you have concerns like that I suggest that you join the KDE Workspace developers!

    Best Regards,

  10. @Torsten: First and foremost, thanks for reading my blog and specially for taking the time to write such a long and informative post.

    Unfortunately, I think I either didn't explain myself well or you didn't understand my point. I will not go into a full reply because it would take a long time and you seem so convinced of your ideas that it would be a lot of time wasted.

    Let me just say that lots of people in the Linux world are concerned about the amount of time that goes into reinventing the wheel, into creating something that is already there except for a few changes. All I am saying is that a revolution is about to knock down many pre-established concepts in personal computing and the KDE project should react accordinglym instead of following a path that was the right one 10 years ago.

    The cloud technology evolves at lightning speed and it won't be long before most of what can be achieved in a computer today can be achieved from the browser, only faster, better integrated and way more resilient. In my opinion, the choice is clear, people will always go the easy route, so why not creating a great desktop environment that simply nails the basics down?

    Let me ask you something... What would you say if I told you that my friends and I want to create a project to manufacture spoons, only made of carbon fiber instead of metal? You would probably tell me there are all kinds of manufacturers out there which already make better spoons at a cheaper price, have distributors in place and all legal paperwork down... Why not use the existing spoons and dedicate my time to something more original, useful and with higher adoption chances?

    It's just my opinion, of course, and I know that open source doesn't necessarily have user acceptance as its primary goal, but still...

  11. Ok, so plese give me example of linux DE which is "cloud" oriented. Let se Ubuntu unity no, Gnome shell, no. So...

  12. @Anonymous: I am not comparing anything here. I am commenting on KDE because a brand new release just went live, that's all.

    In any case, GNOME has the basics down. It just works, and in that sense, I think there is room for KDE to improve before they do anything else.

    - Rationalization of the application catalog
    - Polishing some rough edges that still make KDE cumbersome and unintuitive at times.
    - Simplifying some areas that are overcomplicated (color theming comes to mind) and don't bring that much value.
    - Fixing VERY old issues (when, oh when, will we get proper icon theme support?)
    - Moving past Oxygen (it is a great, but getting old quickly)
    - Bringing consistency (way too many old applications don't have modern icons, making the KDE menu look poor)

    I could continue, but let's just say there are many basic things that users would benefit from the most that are not getting any attention. Meanwhile projects that are meant to fail get lots of hours of development. I think that is a mistake. In my opinion, KDE would benefit from getting basic things down already. If they want to expand into other projects after that, fine, so be it, but at least make sure that the basics are rock solid, consistent and looking good.

  13. About KDE cloud initiave:

  14. Greetings,

    Oxygen is getting old? What about Raleigh, the default theme of GNOME2?

    While I understand what you are saying don't forget that KDE is an open source collaberation of volunteers, each working on what satisfies their itch that come together at the end to make KDE. There is no wasted resources because the projects they work on satisfies their need. It is just even more valuable when more people find a good use for it.

    Telling those independent volunteers that they should focus on something else would be like telling you that you are wasting time writing a review on kde because someone else already did it. That would be just plain silly because after all this is your blog, you are scratching your itch and kindly sharing it with the world.

    Well guess what, that is just what the developers of marble, caligri, konqueror, kopete, amarok, plasma, and so on are doing. They are solving their needs while kindly sharing their results with the rest of the world.

    Just like we can post our comments here to add to your blog you can elect to add to kde with what ever talents you have, be it bug reporting, graphic designing, making plasmoids, etc.

    And just like we don't have to read your review if we don't like it, you don't have to use any part of kde or related programs if you don't like them. You see, that is the beauty of open source. There are ways to get involved and improve your favorite project or use a different project altogether.

  15. @Travis: I see what you are saying, but let's be honest, while that sounds beautiful, it's plain and simple not true.

    If that was the case, why would KDE try to stretch and be used on as many devices as possible? Why would it constantly try to improve on stability, performance, energy management... Why try to consolidate menus in a way that's more intuitive? Why spend a single minute, much less put together a project to discuss what KDE should be like in 5 years?

    It's simple, because KDE, just like any project with so much effort behind it (and with donations and sponsors) has user acceptance as one of its main goals. It is true that most developers also do something they like to do, but let's not kid ourselves here, they would not last a minute inside the KDE projects if they didn't adhere to their "rules". Even if they are not closed in an Apple way, they exist.

    On a similar note, I also like putting together this blog and there is a lot of it that I do for my own sake, but if nobody read it, I can assure you I wouldn't do it. I could achieve the same thing using springpad and it would take me a fraction of the time and effort.

    Let's face it, there is something truly gratifying in having someone else using and appreciating your work, and that quickly becomes important, a driver for any project, specially opensource.

    On a different note, I think it's important we grow up and stop bringing that stinky "if you don't like it, don't use it" argument up. If it's opensource, it might as well be open to critics and feedback. There was nothing in what I said in my article that was irrelevant, much less offensive, so why not taking it for what it is? Instead of taking that "take it or leave it" attitude, I think opensource projects would benefit from keeping an open mind and accepting all kids of feedback.

    As for your comment on GNOME, I must repeat I was not comparing KDE with anything else, so not sure why you bring that comparison. In any case, GNOME icon theming DOES work, so you can simply download one of the many icon themes that look awesome and forget about it.

  16. What exactly is the problem with icon themes? I agree it could be better, specifically because many of the themes on can not be downloaded directly from the icon theme preferences dialog. But if I download the icon themes and using the install from file option it works fine. My problem with icon themes in kde is that the oxygen icons are so awsome that everything else looks amaterish by comparison!

  17. Hi,

    > Unfortunately, I think I either didn't explain
    > myself well or you didn't understand my point.

    I understood your point quite well. I agree with the importance to adapt to a changing world - we do this all the time.
    You just seem to think that just "focusing" on a desktop that gets the basics alright will magically get more contributors involved and will magically lead to a better, more competitive product. Unfortunately the world is not that simple -- especially not in the OpenSource world -- there's more to it.

    > Let me just say that lots of people in the Linux
    > world are concerned about the amount of time that
    > goes into reinventing the wheel, into creating
    > something that is already there except for a few
    > changes.

    The KDE people had taken lots of criticism for reinventing the wheel regarding the KHTML browser, since it would be "just Firefox with a few changes". And indeed from a user's point of view it didn't seem to make sense at all to reinvent the wheel since Firefox seemed to be "the" solution already. Nowadays the KHTML derived Webkit is one of the most popular solutions which is the foundation for Apple's Safari, Google's Chrome browser and many many browsers on mobile devices.
    The reason is that "reinventing the wheel" sometimes provides a better foundation for innovations.

    > so why not creating a great desktop environment
    > that simply nails the basics down?

    That's a great idea. But you will not achieve this by telling people that they are wasting their time with what they (passionately) do already.
    You will also not achieve this by building up a fear (oh god the big big cloud is going to get us...).
    The only way how you can get people enthusiastic about contributing in the OpenSource world is to create a nice positive vision. And to create a nice environment that has enough niches so that all active contributors who work on their common goal are also able to scratch their personal itch.

    > Let me ask you something... What would you say
    > if I told you that my friends and I want to
    > create a project to manufacture spoons, only
    > made of carbon fiber instead of metal?

    I'd tell you: Given a good business plan and the right sales channels you might become rich with that idea. If you lack those ingredients after exploring your opportunities then you should move on.

    Best Regards,

  18. @Torsten: I think you keep interpreting my words from the wrong angle.

    As far as I am concerned, the people involved in the KDE project are the only ones entitled to say where it should go. As a user and reviewer, I am just sharing my thoughts about that vision. I am nobody to try to convince KDE members, or help them achieve anything, but I have the right to speak my mind, of course.

    KDE developers are surely more technically skilled than I am, certainly more aware of where technology is going and whether certain solutions will be more relevant than others. Therefore, if they chose to follow a certain path, I am sure they will have their reasons. I simply don't share the same view, time will tell if I was wrong.

    In addition, I must say I don't understand why you say I am building up fear about the cloud. It's not like that at all, I think the cloud, like any other "technology" is good if used the right way. I am just saying something obvious that we have seen many times before: The technology market does not wait. That's nothing to be scared about, it's simply reality, something we have seen 1000 times already. Will the market wait for KDE when it did not wait for Apple, IBM, Sun or Nokia, to name just a few?

    Again, that's nothing to be scared about, it's simply something that should be understood, for those are the rules of the game here.

  19. I'm about to install Mint 10 KDE RC to check out this 4.6 release.

    I've been repeatedly going back and forth between KDE and other alternatives, with KDE being my favorite in terms of functionality and "fullness".

    My only issue is that in every release since we met with the 4.x branch, KDE is so horribly slow. It'll start up fine, you get an hour of two of responsiveness, but sooner or later you find yourself in the quicksand, trying to batch-select files in Dolphing, but finding yourself accidentally moving files out of folders because the interface lag is so horrible.

    I'm hoping the better KWin will help with this, but your follow-up suggested that we're still barking up the same tree. Hope not :(

    About the cloud thing vs. time on Marble and other things, I would agree with you if we were dealing with a corporation with limited resources. As it is, however, Marble and every other app is simple maintained and develop by people because they're engaged with their project of choice and want to use their spare time doing that particular thing.

    I don't think telling the Marble developers to work on the KDE platform instead would yield anything good. In the FOSS world, you can't just re-allocate people to different projects. They'd just go drinking instead ;)

  20. And, as it turns out, it's the same lag hell as with the previous 4.x releases. Guess it really is time to go Gnome. Wish I didn't have to, though :(

  21. @Daniel Smedegaard Buus: I have not experienced problems with speed in Kubuntu, but I have seen very strange issues that I presume are caused by Kwin effects. First off, Kwin effects work OK and show reasonable performance when nothing else is running. As soon as you start an application that is somewhat demanding (say Firefox), effects slow down and don't look as smooth any longer. Aside from that, which is a mere cosmetic problem I could live with, both Chromium and Firefox behave very strangely since I upgraded to KDE SC 4.6.0. Sometimes I seem to lose control of what I am doing, a tab may become unresponsive for no reason and I have to jump to another one and come back to bring it back to life. Sometimes text fields simply do not respond and I seem to lose the cursor for some seconds...

    Anyways, strange issues that are not consistent, so I can't reproduce them and log a bug report, but I very much know they are there. Frustrating, but hopefully something that will go away come KDE SC 4.6.1.

  22. @Chema Martín: Well, it seems the issue isn't KDE specific, but due to a buggy Oxygen theme. The issue seems to have to do with pixmaps not being released properly, which over time will bog down your graphics card. It doesn't apply to all graphics cards, however, but particularly nVidia cards are prone to be suffering from this bug. It also doesn't appear that it's an issue that'll be fixed soon, if ever, so the workaround is simple: Chose any other style! I'm currently running Bespin with much success. It's aeons more responsive and doesn't bring my system to its kness :)

  23. After your commands:
    `sudo add-apt-repository ppa:kubuntu-ppa/backports` and
    `sudo apt-get update`
    You probably should have tried:
    `sudo apt-get upgrade`
    before trying
    `sudo apt-get install kubuntu-desktop`

  24. Lol gnome. I used gnome for more decade. I always lothed KDE, I installed it lil over month ago. You have your feelins and your comments and reviews. And others have theres. Well I have mine. Gnome is a thing of the past. the bloating with new computers is not there no more..that is old news. KDE performs on a cheap 150 buck amd 3000 dual core just as fast as gnome with the alot lesser graphics and shortcoming layout. (gnome 3.0 is no way for me) KDE is not about being click happy or more steps...its structure and organizing is alot more efficient. try setting up file sharing in gnome...and tell me its faster or more logical then with kde 4.6 when u simply right click folder and click share. Even fetchs the samba install if not there. Also try clicking icon on gnome network in top right corner compared to kde in bottom right...KDE will now give you bandwidth usage and more on a graph. Along with your ip and settings. The list goes on. Sorry guys. but unless your old dog. KDE has way moved past Gnome in simplicity and Looks. Thats my solid 2 cents. And after 10 years of enjoying gnome I have to throw in the towel and move kde for that simple fact. As for buggy from some of comments. name brand computers have never been friendly with linux. I run on full clone systems. They have always out performed equal dell, hp and other higher cost name branded computers, and with no bug problems. (also big major fact. Nvidia and ati is majority of market as windows is on systems nowadays. Dont run linux without it. Nvidia has my vote due to fact it smoothes the output of KDE like no other. That and I play wow, rift, eve, call duty 4 and much more on my KDE box. lil tweaking but they all run just as good on my KDE as they did when I was on gnome.

    Cheers. Oldcode.

  25. i am fulltime on KDE now .. after unity came. I originally thought it was slow .. but it was just some settings i had to change to make it faster .. Also i wish it gets a little bit more lighter to start with .. meaning users should have the option to turn on all the akonadi ... etc .. god knows how many such services. But overall i feel KDE is obviously more polished than gnome 2 or even gnome 3.

  26. kde is jewish as fuck eat dicks gnome for life dog