Having a look at the old/new Desktop Environments

2011 seems to have finally been the year of the Linux Desktop. But not in the way people always anticipated it by Linux Desktops finally dominating the desktop market (something which will probably never happen unless Microsoft fails badly), but by releasing many new Desktop Shells.

Both large desktop communities have introduced a new shell. The GNOME community has released GNOME Shell, the KDE community released the touch oriented Plasma Active and Canonical decided to go alone with Unity. All these shells introduce new User Interaction concepts and interestingly 2011 was also the year of new classic desktops.

GNOME 2 got forked as the Mate Desktop, GNOME Shell got forked as Cinnamon to provide a classic user experience. In the Qt/KDE world Razor-Qt got introduced and although being around a little bit longer Trinity (a fork of KDE 3.5) got some attention probably based on the fact of the forks of GNOME.

My personal expectation is that Mate will fade away due to Cinnamon. For Cinnamon I expect that it will stay as there seems to be a demand for a classic desktop experience in the GNOME world (btw Plasma could do all that for you). My hope is that Cinnamon and GNOME Shell will rejoin again as I don’t like forks. GNOME Shell will of course also stay around and I expect it to become quite successful as time passes by and I expect many former Ubuntu users migrating to it when Canonical finally decides to go for Tablets only.

Trinity – Who needs it?

In the KDE world I find the Trinity project most interesting both from a social and from a technical point of view. I am very interested in understanding why developers decide to keep an outdated and orphaned code base alive. What motivates the developers to continue the development given that the original authors moved on? Also what are the targeted user groups? Who needs an outdated system like Trinity? I tried to identify the possible user groups of Trinity and came up with three groups:

  1. Users searching for a “lightweight” desktop
  2. Haters of “KDE 4” technology
  3. Users wanting a “classic” desktop on KDE

The first user group mostly confuses lightweight with old hardware. If we see the specs of Spark I would not call that a high end system, so the reason that users have trouble with KDE Plasma on “lightweight” systems is not because the hardware is “lightweight” but just old. But even KDE 3.5 was far from lightweight. The saying that KDE is bloated is probably as old as KDE itself. So the users looking for Trinity as a lightweight alternative to KDE Plasma just fail to realize that five years ago they were screaming that KDE 3.5 is too heavy. Neither KDE 3.5 nor Plasma tried to run well on old hardware – this is a market very well served by e.g. LXDE. So keeping KDE 3.5 alive for such users seems to be the wrong decision.

The second user group is the most dangerous to the project. My fear is that the project gets dominated by users and developers hating the KDE 4 technology. Emotions – especially hate – are very bad to base decisions on. If better technical solutions are discarded just because it might optionally pull in a “hated” KDE technology, the project will run into problems. Also if the hating user group is too big it brings the danger that they will block any progress of the project by proclaiming that they will leave (hate) the project as well. The result would be stagnation and a backward oriented project. My personal opinion is that Trinity is already dominated by haters – more to that later on.

The last user group is the one demanding a “classic” desktop. Well this seems like a valid use case to me. But if I look at a recent KDE Plasma 4.8 and at a KDE 3.5 I don’t see much difference. With a few clicks you get a desktop which looks and behaves the same way as 3.5. There might be slight differences, but is that worth keeping millions of outdated lines of code alive? Wouldn’t it be better to just write another Plasma shell to mimic the KDesktop/Kicker behavior? And even if you don’t want a Plasma powered classic desktop, there is Razor-Qt which offers a modern approach to the problem instead of building on legacy solutions like Trinity.

Trinity – the desktop of duplication

One of the biggest issues with the development of Trinity is that they took over the KDE 3.5 code base without having a plan what to do with it. There are applications which got dropped in KDE 4 (e.g. Kicker) – for those I can understand that the development is continued. There are applications which use a “hated” technology – e.g. KDEPIM using Akonadi. Unfortunately continuing KDEPIM 3.5 will not solve any of the issues which lead to the hated technology nor will it improve the modern project. Last but not least there are application whose development continued in the same way offering a much better product nowadays – for example KWin, Kate, Okular or all the games and edu applications. For those applications I don’t see any need to continue the development of the KDE 3.5 product: it does not serve any of the identified user groups. Personally I find it very sad to see valuable FLOSS developer power wasted on such products instead of helping the current versions.

This is an unfortunate duplication of work. But it’s not the only one going on in the Trinity project. They migrated to git – again (KDE already did that). They migrated to CMake – again (KDE already did that). They started porting to Qt 4 – again (KDE already did that). They have to migrate away from HAL – again (KDE already did that). It’s a pity to see all this duplication with that limited manpower.

But it’s not only the duplication which is a real issue, but also the lack of knowing what is going on in KDE. For example there is an enterprise branch of KDEPIM 3.5. I just dared to pick out one random commit from KDE’s enterprise branch and checked whether the change is present in Trinity’s version of the changed file. I think you can guess the result. I also checked one random change for kdelibs and that is also not present in “tdelibs”. Monitoring directories for changes is extremely simple in KDE.

This means that Trinity does not only duplicate the work already done by KDE, but also that a user running Trinity gets a worse product than stock KDE 3.5 as offered by some distributions using the enterprise branches.

Trinity – the haters desktop

Some time ago I decided to contact the Trinity development team with the aim to reduce some of the ongoing duplication. Seeing commits to “twin” really hurts and our current version of KWin is just better in any area. We have not just integrated compositing but also fixed many, many window manager related bugs. All those fixes are not present in the “twin” fork.

But there is a more dangerous issue. I had taken the time to look through the commits to “twin” and they are in the categories from “plain wrong” to “might lead to a crash that twin does not start any more”. Developing on a foreign code base without any help is very difficult and nobody has ever contacted the kwin development team before (at least not me personally, the kwin mailing list, the bug tracker, the IRC channel when I was around).

The reactions to my offer to help were mostly (there are a few positive exceptions) negative. And unfortunately it resulted again in duplication of work. The lead developer decided that it is important to allow multiple window managers. This is a great thing to do, but what a pity that the KDE developers thought about the same years ago and found a better place to put it instead of the window decorations KCM.

But I wanted to write about the hatred. One of the reactions to my proposal was:

Personally I would hate to have to install kwin, which relies on a bunch of other KDE4 libraries and automatically installs that akodani garbage scanner stuff, just to use TDE

After pointing out that KWin does not depend on Akonadi, I got:

I meant nepomuk, not akodanai.

I think especially the typos speak for itself. Not knowing what the technology is, what it does or what it is for. But obviously it is a “garbage scanner”. This is not a random developer or user, but the project leader.

Another highlight of a recent thread:

This is another area that we should be able to do better than KDE4

Where I ask myself why would anyone have that aim at all? Why not work together with KDE to have a shared and improved code base? After pointing that out, I was informed, that

This Email was not supposed to go to the public list

We see here unfortunately a pattern and I dare to quote another mail from the same thread:

I don’t have time or the desire to pick through KDE4 at the moment but KDE4 is still less efficient for my workflows then TDE, period. It takes more space on the screen to display less information in a harder-to-digest format for starters. Then there is the whole assumption that people have low-resolution or small screens and one-button mice (in TDE all three mouse buttons can be used to interact with on screen elements–much of that power is just gone in KDE4).

As I said I don’t have time to play with KDE4, and all of my prior attempts to use KDE4 as anything other than a shiny toy failed miserably. Please remember that there is a bit of a frog-in-the-pot syndrome when users are forced to use an inefficient interface for long periods of time; the best way to break this is to go back and use Windows XP or KDE 3.5 for a day or two, then go back to KDE4. If KDE4 is truly better then that will be obvious; if it is not then this fact will also be obvious. (I have done this and KDE4 still looks like a shiny toy).

I recommend to read this paragraph several times. Let it sick in…

… thanks. Now let’s see: someone claims that KDE 4 (probably Plasma) is still less efficient, although he did not test it. This person is in charge of the development of Trinity. How should he be doing any sane decision with such a world view? I don’t get it.

It’s pure hatred against KDE technologies and this dominates the development of Trinity. And this is not a single developer issue, it’s unfortunately a common pattern. I have for that one more mail to point to (only click the link if you have not eaten anything and prepared a bucket next to your seat). The worst are the reactions on such a mail like “100% On the nail ! Well said Luke…” or ” I am stupid enougth to imagine that the leader of the KDE developpment team could receive two salaries: with one coming from MS. I see KDE as something self destroying!”
Nobody called the people to order. Spreading pure hatred against KDE technologies and their developers seems to be perfectly acceptable inside the Trinity project.

Trinity – where will it go?

Given what I had seen over the last months by following the Trinity mailing list and partly the development it is quite clear that Trinity is a project for haters of KDE 4 technology. This is the only focus of the project and this is the reason for the ongoing duplication of work as they are unable to just even look whether KDE has already solved the issue.

Overall it results in pretty severe and dangerous design decisions. The hatred results in things like:

At one time there was a desire to port to Qt4, however months of solid work showed that Qt4 cannot provide the features needed to create a fast, efficient desktop geared towards mouse/keyboard interaction and high on-screen information content.

Well, I’m personally not surprised that you get this result, if you add a wrapper around Qt so that you can use Qt 3 and Qt 4 alongside. I’m also not surprised that this is the result if you have a mindset as illustrated above. If you want Qt 4 to be “worse” than Qt 3 in your project, you are able to achieve that. Don’t trust the benchmark you did not manipulate yourself 😉

So Trinity will keep stuck on Qt 3, which reached EOL on July, 1st 2007. Trinity thinks they can continue to maintain this orphaned code base – alongside the millions of line of code from KDE.

Can this work out? I doubt it. No developer knows the code and there is nobody to ask (and they don’t ask anyway it as I noticed myself). Given my personal experience of offering help I understand each KDE developer who does not want to have anything to do with the Trinity project and who would refuse to help them.

Looking over the mailing list there seems to be only a hand full of active developers, looking at the git repository it seems to be a one-man show. One developer to continue a development previously carried out by hundreds? At the same time facing the legacy issues like the deprecation of HAL, having to take care of packaging for all distributions. Finding rare issues like kernel is now 3.x and no longer 2.6.x? I rather doubt that this can succeed.

I have read quite often that “Qt 3 is now maintained by Trinity”. This is a pure lie. The development team is magnitudes too small to maintain this code base. Anybody running Trinity runs the risk of being exposed to newly discovered security vulnerabilities. It is quite clear that security issues discovered and fixed in enterprise distributions (e.g. RHEL, SLED) would stay open in the Trinity project (c.f. missing commits from enterprise branches) probably without even issuing a CVE.

Is there hope?

So what for those users who want a modern, classic Qt based desktop? Trinity will never be the desktop solution for it. The good news is that there is a sane project providing a classic Qt based desktop: Razor-Qt. It is built up on Qt 4 and has a clear project aim. Instead of just forking everything and the kitchen-sink, razor concentrates on providing a desktop shell. And the code is clean and well written, following modern approaches.

So my recommendation for all Trinity users is to either try again KDE Plasma 4.8 or to give razor a try. Trinity is no project anybody could seriously recommend and a stock KDE 3.5 is most likely a better solution than Trinity.

247 Replies to “Having a look at the old/new Desktop Environments”

  1. I think much of the hate comes from people who feel betrayed or burned by the initial release of KDE 4 (or GNOME 3, for that matter). If those people could be convinced to try the latest versions of KDE and GNOME just one more time, I’m sure many would not go back.

    1. Well, maybe they have old slow computers, then KDE3 might be the better alternative.
      BUT couldn’t they just run CentOS 5? It has KDE3 and it gets updates, doesn’t it?

      1. Yes exactly, that’s something I would recommend as e.g. CentOS 5 also comes with HAL and all the other outdated technologies.

    2. I don’t know why, but my pretty modern HW setup (quad-core AMD CPU, 4Gbs of RAM and decent Nvidia video-card) is veeery slow with KDE4, both on ArchLinux and FreeBSD.

      That is not experience of the initial KDE4.0, definetly. I’ve tried KDE while it was 4.3 and 4.6, and have not seen any 4-branch since 3.99-beta (in late 2007?), which was not so slow on much less powerful HW.

      This is not hate. This is not feeling betrayed. This is plain ‘I’m to lazy to solve this trouble while having KDE3 in ports, which just works’.

      1. what exactly is slow? Desktop effects? In that case have a look at userbase which holds some recommendations for various hardware how to make it faster. Also consider to try 4.8 again – from the feedback we got people are very pleased with the performance improvements.

        1. That was not the effects issue. The thing I absolutely hate is useless compositing and such.

          That was issue of slow reaction on every user action (e.g. run KControlCenter, don’t remember how was it called, open new tab in Kopete, open any website on Konqueror, etc.). Its the case of probably twice as slow as KDE3 does. Which makes no real performance trouble, but is very annoying. Also not the KDE4 trouble, but Qt4 (Psi was slower that Qt3 version, for example).

          Also I’ve ran some KDE4 apps on MS Windows 7 and it was pretty fast. On different HW, though (less powerfull laptop).

          1. Technically, compositing shouldn’t slow things down unless your graphics drivers are buggy (ie falling back to nv or vesa). All that compositing does is render your windows to a texture rather than render directly to the framebuffer. This means less redraws among other things.

            There were some NVidia blob bugs that the KDE guys unearthed (the code paths hadn’t been worked before), but IIRC these have been fixed in later blobs.

            If you want some help,could you please post the output of:

            glxinfo | grep renderer

            1. %glxinfo|grep render
              direct rendering: Yes
              OpenGL renderer string: GeForce GT 240/PCI/SSE2

              %ls /var/db/pkg|grep nvid
              nvidia-driver-285.05.09

              Dont think that the blog comments are good way to solve technical issues. As I feel I have enough free time to test, I’ll write to the mailing list better 🙂

  2. That phenomenon/behaviour is called groupthink. It includes all kinds of self-defense and self-sustaining mechanisms. Thus forget about arguing with them but keep an eye on the people you work with to prevent the same happening within KDE4.

    By “the same” I do not mean that there is hating or similar going on within KDE4. But still there is always the risk of decision making being influenced by groupthink.

    1. This also depends on whether the development team is interested in going to git. Btw trinity did the easy way and lost the history by migrating to git – they just imported the last point of svn history.

          1. A decision was made to stop the history at the 3.5.10 release from which TDE was forked. If someone desires further history they may go back in the KDE SVN for it, otherwise it is simply taking up space and cluttering the GIT tracker.

            In practice we have never had to go back further than KDE 3.5.10 when fixing bugs in TDE.

            1. So just to be absolutely clear… you HAVE lost the commit history in the migration to git. Cool, gotcha.

              1. Nope. We have ***truncated*** part of the history, that is all. There is full history going all the way back to KDE’s official 3.5.10 release, and that is all that is needed going forward.

                There is a technical reason for this in reality. Saving the entire history for a large project spanning decades is not only not useful, it actively degrades the performance of the SCM system in daily usage, both in terms of time to interact and in terms of disk space required to work on an active clone of the repository. GIT is not like SVN in that you must have an entire copy of the repository on your system when working with it, whereas SVN allowed you just to download the current snapshot and modify it.

            2. In practice we have never had to go back further than KDE 3.5.10 when fixing bugs in TDE.

              Interesting, my most important tool to fix bugs is using “git blame” which brings me quite often back into times like KDE 2. But well it’s not my problem that you have no history. In fact it is impossible to understand any code which only got documented through commit messages

  3. The fact that you personally like kde4 and think it is better doesn’t mean it is actually better then kde3 was. Many people (inlcuding me) still like kde3, why should people chose using “always-crashing” plasma over kicker? Personally I’m still using kirocker, tastymeny, polyester, amarok1 which don’t have alternatives in kde4 world. And I like how kicker looks like comparing to all-glamour plasma.
    You ask why this has started? I remember 3.5.9, it was best DE ever and it had all I needed from DE. Then kde devs said that kde3 is obsolete and introduced kde4/plasma. I tried kde4.0beta, plasma was crashing all the time,it was not possible to run it for more then a minute. The I tried kde4.1, 4.2, 4.3, 4.4 and plasma was still crashing all the time. While it was wide known that plasma is buggy, kde devs said “look what plasma can do” and release a fluffy buny theme. Yeah guys, we needed a stable plasma and you developed a shitty theme for plasma to crash even more. And the fact tha a single plugin can crash the whole desktop is very frustrating (I don’t see why plasma is considered better then kicker). And in 4.8 it is still true, a single plugin can still crash you whole DE, while I should say that sutiation improved.
    You say here about kde3/kde4 choices and transitions as if kde4 was the only choice. But many people still think that instead of full rewtite (yeah, duplication of effort) kde3 should have improved, bugs should have been fixed, features should have been added. And instead it took devs four years to rewrite from scratch what was already written and working. Nice waste of time.

    1. Just correcting one thing:

      And in 4.8 it is still true, a single plugin can still crash you whole DE

      No this is not true. The preferred way to contribute 3rd party plugins is using QML which cannot crash the Desktop Shell. Also Plasma’s own plugins are slowly migrated to QML, so this is actually being addressed right now. It had already been solved through JavaScript plugins for quite some years.

      1. > The preferred way to contribute 3rd party plugins is using QML which cannot crash the Desktop Shell.

        It doesn´t answer the question on wether a non-QML plugin (most probably a C++ one) can crash the DE.

    2. I use Debian Testing witth KDE 4, and I have not got a single crash in Plasma for months. It is a fact that third party apps can make your machine weak and unstable, even in the MS world, so maybe it is not a good policy to put on your desktop any plugin you find interesting.

    3. > … amarok1 which don’t have alternatives in kde4 world

      Try Clementine. http://www.clementine-player.org/ 🙂

      > While it was wide known that plasma is buggy, kde devs said “look what plasma can do” and release a fluffy buny theme. Yeah guys, we needed a stable plasma and you developed a shitty theme for plasma to crash even more.

      Seriously, that’s one of your arguments? I probably don’t even need to point out what’s wrong about it…

      1. it does demonstrate nicely what Martin has been writing about though.

        for those who aren’t aware, and so as to counter the mindless slander in the original comment: fluffy bunny was not made by the plasma team; it was indeed meant as humor; we were focusing on stability and performance when it came out.

        but we can see exactly what martin observed here: angry comments made without knowledge. no one wins anything from this and it is very, very sad.

          1. Everybody with half a brain knows that Plasma is more robust than Kicker. Plasma is not perfect (the aforementioned crash with C++ applets being an example) but overall Plasma is better.
            With QML that issue is being resolved. It’s just a matter of time to migrate all applets.

    4. Yes there’s an alternative for Amarok1 – it’s called Clementine.

      Try it, despite the fact it’s coded using Qt4.

  4. I’ve read the whole pointy-click linked thread and the conclusions you base on it are much too harsh. It is also clear there was a communication (language?) gap, as on one occassion your tone appeared superior and on another you actually threatened them with this blog post, which will only make the relations worse.

    With that in mind, it’s hard to see you as consistently constructive. Also, why bother with some trolling users at all — it is clear that that list isn’t just inhabited by developers?

    I agree that code nplication is a *waste* of time, but when has that ever stopped anyone and why wouldn’t it be a legitimate thing to do?

    1. as on one occassion your tone appeared superior

      Yes that is possible that it appeared like that. Keep in mind that I’m not a native speaker.

  5. I think you hit the nail right on the head.

    NO DECISION should be made from a place of hatred. And, it seems to me as if the Open-Source community has a sizeable dose of it, poisoning it’s users against various technologies.

    But, I guess it’s not surprising given Linux’s small desktop share: The vast majority of people who find it, are people who are looking for an alternative to their current setup. And usually that means that something bad has pushed them away from Microsoft or Apple. And no matter what company or person that it’s aimed at, hatred is very contagious.

    I don’t think people really appreciate how small of a step it is from “I hate the evil empire’s proprietary piece of crap,” to “I hate those evil [gnome/kde/distroX,etc] devs for not doing what I want them to do.”

    What makes the whole problem so much worse, is that it’s the people with an axe to grind that are the most vocal, taking on themselves to “save the world from these evils by telling how horrible they are and exposing their true colours”. But, in truth, all their doing is helping to create an atmosphere that drive away a lot of potential users.

    [/rant]

    1. I think you can remove the [/rant] – I quite agree with what you said. A lot of manpower is wasted in a too close world view and we could see this happen over the last year several times.

      What I personally consider as a much bigger problem is the fanboyism shown by some very vocal users.

  6. Let me get my angle of view clear before I start commenting: I was using KDE since the 3.5 times, but switched over to Mac OSX about 1.5 years ago. Since then, I do not use KDE on a daily basis anymore. However, my Gentoo box still enjoys to compile the newest releases and I give them a shot.

    In my opinion, KDE emphasizes the existence of technologies too much, but the end-user experience provided by those comes short. Take TDE’s examples Akonadi and Nepomuk as well as Strigi and KParts. I know there are similar technologies in OSX, but I can’t tell their names. However, I can search using buzzwords, I get results from different applications like emails, bookmarks etc. I also get related content and I can preview results, but I don’t know which parts of the system are responsible. It’s an integrated solution that is “just working”; no configuration from my side in order to get the job done.

    I guess the actual complaint behind the “KDE technology”-point TDE users/devs are constantly insisting on, is that the KDE out-of-the-box experience is not good enough. Thus, the hyped technological advancements are not always worthwhile for end-users and thus tend to bloat the system.

    Here’s an example of your work, Martin (no offense): KWin effects written in Javascript. How would you sell this feature to someone who cannot write code? I personally think I don’t need that at all. I’d rather like to hear about having mastered the 60fps goal instead of such features.

    However, I think that’s only a small price to pay in order to get the latest bug fixes so I don’t see a reason to keep KDE 3.5 alive.

    1. Here’s an example of your work, Martin (no offense): KWin effects written in Javascript. How would you sell this feature to someone who cannot write code? I personally think I don’t need that at all. I’d rather like to hear about having mastered the 60fps goal instead of such features.

      Ok, I sell this as “you can download new custom effects from our App Store”. The technology behind does not matter and no user should know about it. Also no user should know about how many fps the effects run at. So on a technology oriented blog like this one you won’t find user ready comments.

      Mac OS is of course not interested to have others know about their internal architecture, so you don’t find anything about it. That’s one of the differences between a closed and an open world.

      1. I don’t know if the store concept for such small pieces can succeed. How is the content in the Aurorae-“App Store” evolving? That might give some indication.

        The fps stuff is specifically why I visit your tech centric blog. It gives me as an user at least some impressions on how performance advances. And that’s what matters for me at the end of the day. For example I abandoned the use of the dashboard because it took ages to switch over to it.

        After reflecting my post as well as my years of KDE usage, I came up with a better example of a feature/technology/concept that I never incorporated into my workflow because it was complicating things for me: Activities. Especially activities vs. virtual desktops: Why use a photo editing activity if I can have a virtual desktop with GIMP on it as well? Freedom of choice sometimes becomes confusion.

        I don’t want to be purely negative, so here’s an example of something I consider to be the best invention in KDE for years: KPartsPlugin for Firefox. I know it relies on a lot of work in the background (“KDE technology”), but as a user I’m only concerned about this piece I can see. That’s what I’m referring to as “selling features”. Such a demo app should rather be part of the core KDE packages than lying around somewhere at kde-apps.org.

        Regarding the difference between open and closed source software: Don’t forget to mention the price tag that’s actually not justified by the provided support 😀

        1. I’ll give Activities a bash. It’s been difficult to explain basically because it doesn’t really exist anywhere.

          I use Activities as a replacement for Virtual Desktops, as they provide much more power: you can use both to separate out applications and tasks, but with Activities I can stop them to automatically stop the applications running on them, with open documents/applications/etc. saved. When I want to resume that activity, I can start it up again, and it opens up all my documents/applications/etc. automatically so I can resume right off the bat.

          I agree with a couple of your points, though: there are not enough people marketting KDE to end users, and; KDE right now emphasises technology (which I think is great, others hate it), but to a much lesser extent implementation. Which is a shame.

      2. “Mac OS is of course not interested to have others know about their internal architecture {…}”

        Not so.

        Strigi + Nepomuk = Spotlight.
        Akonadi = CoreStorage.
        Plasma = Dashboard+Spaces. Exposé and Spaces were merged to create “Mission Control”.

        KParts? You’re probably right, I don’t think there is another technology that rivals KParts.

        As for Apple not make technology names prominent, how’s Grand Central Dispatch for an example? Or perhaps Bonjour, formerly known as Rendezvous?

    2. “Here’s an example of your work, Martin (no offense): KWin effects written in Javascript. How would you sell this feature to someone who cannot write code?”

      Being able to write extensions in Javascript isn’t directly sellable to users, the results are. Works nicely for e.g. Firefox. I doubt normal Firefox users are interested in that someone can do extensions without native code, but they certainyl appreciate the results

      1. Yes, I agree, it’s the result that matters. And a result is what the user gets. Just have a look at my comment that’s talking about KPartsPlugin. I think that is close to what Firefox achieves with its plugin technology.

        KDE has some pretty decent technology in its backend, but fails to deliver advanced user-experience (=result) out of it. Regarding that topic, let me cite Martin’s article called “Die Bugtracker Lüge” (2011-05-01, “the bug tracker lie”): “[…] hier ist die Realität, dass ich die Wünsche als komplette Müllhalde betrachte. Ändere ich einen Report von Fehler auf Wunsch ist es gleichbedeutend wie “> /dev/null”. Es gibt keine Wünsche der Nutzer, die sich mit meinen Entwicklungszielen überschneiden” (“[…] here, the reality is that I consider wished to be dump. Changing a bug to a wish just means “> /dev/null”. There aren’t any wishes that overlap with my development goals”). That’s the point where TDE wants to be different. Their development goal is the user’s wish for some specific experience. So they totally have the right to exist, only the incarnation isn’t really perfect. It would most certainly be better if they implemented the wishes collected in KDE’s bug tracker, but we have to take it as it is. They think they cannot deliver based on this way of thinking expressed by Martin, so they do not want to cross ways.

        Btw, I personally think that’s also one of the reasons why Ubuntu stopped officially supporting KDE. They want end-user experience.

        1. I think you completely misunderstood my previous blog post. The fact that the user requests don’t fit our development does not say much. It is impossible to implement all feature requests – I could point you to many contradicting requests. Nobody can implement that. Also users report things like “KWin is slow”, which might be addressed with effects written in JavaScript, but nobody would report such a wish. It’s just a different level of thinking about the application.

          1. I’m fine with ignoring wishes like “make KWin faster”. I’m also fine with ignoring wishes contradicting the existing work towards a well established goal. But I don’t get how the user experience can be boosted if user requests are considered “not to fit our development” in the generic way you say it. As a result you get a system with that is not suitable for general purpose usage. If I wanted such a system, I could use QNX on my desktop.

            For me KDE is like a nuclear power plant together with the whole infrastructure around it. A lot of high end technology as well as maintenance is needed, but what you are actually doing is nothing more than heating water that will rotate a propeller. And what you are getting is nothing more than electricity. You might get this result to your user in an easier way, that does not force so many worries.

    3. > In my opinion, KDE emphasizes the existence of technologies too much, but the end-user experience provided by those comes short.
      > It’s an integrated solution that is “just working”; no configuration from my side in order to get the job done.

      I’m having several woah-moments with the recent posts about KDE usability, design and perception by end-users. This is basically what drove me to GNOME 3 when it came out, as I posted on the systemsettings post on Planet. While I immediately liked GNOME 3, KDE (used it till 4.6) always felt like a bunch of sometimes more, sometimes less polished things bound together with duct-tape. Trivial example: it ***blew my mind*** that I could click on the GNOME 3 clock applet in the top bar and it actually displayed upcoming events in a nice way, so I immediately knew what I had to do today or tomorrow!

      KDE has good stuff, no doubt, but *imho* all of it needs to be worked into an overarching, consistent design. GNOME 3 is doing that and it’s tickling my feel-good senses 🙂

  7. I must admit that I find this blog a bit curious. Even if everything you said about Trinity is correct (and I have no doubt that you are correct), who cares? Let them waste away.

    So, what I find curious is what is the motivation behind such a blog? This blog brings much more attention to Trinity than simply ignoring it. I certainly knew about Trinity before, and was about as interested in it as watching paint dry. Now, it suddenly seems more interesting (even if for the wrong reasons). In addition, the title of this blog suggested that it would be a different subject than what I read. I would imagine that this was in an effort to get as many people to read it. (Example: I would not have read a blog entitled “Read why Trinity Sucks”.

    Martin, I hope I am not offending you too much. I respect your work. This simply surprised me.

    1. The motivation for this blog post is, that again and again I see on the Internet recommendations to use Trinity. I want to give out a warning that software of the Trinity project is dangerous to the users.

      1. I personally don’t see any value in that. If somebody wants to recommend something then they should. People can decide for themselves whether to follow a recommendation and if they do whether they thing they have been served by recommendation,

        1. Dont forget that users arent aware about the danger using such outdated software such as trinity. Yet most online mags that recommend this DE fail to point out the problems lying in the backround and they base their recommendation in phrases like “try this feature-full de” without say that is based on weared out unmaintained technologies such as qt3.

          1. How about the dangers of running a desktop environment that values social interaction over stability? Folks, every time a crash occurs, no matter how small or where, there is a possibility that a malicious individual can hijack your session or take your data. This risk exists primarily in Internet connected applications, so if you are using a social media Plasma widget and Plasma crashes, it would be possible for an attacker to take *any* data from *any* widget that Plasma was running, or worse.

            Stop spreading FUD Martin.

            1. I should mention that the chances of this happening are vanishingly small, however a crash usually means that data is being overwritten somewhere in memory. The point is that if someone was quite clever they could figure out how to make the computer execute third-party data as code, and potentially hijack your machine if they provided malicious data and triggered the crash.

              Crashes are not just annoyances, they are also security risks if third party data is involved (such as data from the Internet).

                1. But I gave you a very clear and working attack vector on my very first mail to the Trinity mailing list – and it is still not fixed!

                  1. You mean this message?

                    http://trinity-devel.pearsoncomputing.net/?0::4312

                    No mention of a potential attack vector was made, unless it was buried in one of the external links you provided.

                    And furthermore, the commits you referenced *enhance* the integration of twin with TDE, instead of “breaking things” as you claim. I have not violated ICCCM, NET, or any other applicable standards anywhere. Why don’t you bother other third party window managers about the usage of custom atoms, etc. if you are going to claim that we did something wrong with those enhancements?

                    Furthermore, while I am not an expert, I *am* fully aware of how window managers function, and can modify them to provide new features that I or the TDE users want to see. If kwin prefers to stick with outdated standards instead of extending them where appropriate, you will have a hard time competing against the more integrated solutions offered by Apple and Microsoft.

                    1. why have you actually never asked which commit will cause twin to no longer start? It has nothing to do with the atoms.

                      But that’s just the thing I have been writing about. You have no clue about the applications you develop and you just guess about everything.

                    2. Ah, so finally you gave me some information. Your complaint centers around twin not starting (remember, before I had to *guess* from a handful of *unrelated* commit diffs what exactly you considered broken–a daunting task for any software developer). You are right, that has nothing to do with the new atoms or modified window control logic (which leads me to wonder why you posted those commit diffs to your original message, if they were not valid).

                      Interesting that twin always starts 100% of the time for me and all of the users on Debian, Ubuntu, RHEL, Slackware, Fedora, etc.

                      Honestly I think you are just making things up. I think you looked at the commit diffs, said to yourself “that can’t work”, and went merrily on your way, oblivious to the fact that it *is* working.

                      I won’t even respond to your insult execpt to say that it is completely off base, borderline libel, and that I won’t be coming back here for more.

                    3. Sorry, but you clearly illustrate that you have no idea of the problems that might arise with the change you have done. Of course it will always start on your system. And it will also start on every system you test. But it might fail on your users systems in case they use a 3rd party window decoration. Any kwin developer would have spotted it right away, because we are aware of such issues that might arise when changing the library.

  8. I can see the point in wanting a ‘lightweight’ DE for _old_ hardware. KDE4 is no fun with old hardware (especially with some bad graphics card).

    I can understand that that story mentioned in ‘that linked thread’ pissed off everyone involved. It’s frustrating not to know how to do things in a DE [1] and having to watch people go back to *Windows*. *Vista*. Bummer indeed!

    But the conclusions he draws from that incident are exaggerated and inapt IMO. Had he followed at least your blog, he would have known that KDE devs are forward-looking and keeping up with current technology development, providing for improvements and better DE experience also for the time to come! After all, the uncomfortable _early_ 4.x (mostly 4.0/4.1) times are behind us , let’s deal with the present/future!

    Sure, we all like performant DEs/window managers but as I understand it, this is something KDE devs do keep an eye on anyway! Even KDE’s startup time seems to have improved with 4.8 (at least in most cases, the desktop is usable more quickly than in 4.7).

    So, I’d recommend the Trinity devs to have a look at _current_ versions of KDE and not rely on outdated (bad) experience they may have encountered.

    [1] Only recently did I have that feeling when I tried gnome3 once again (which I do once in a while to see what’s going on in gnome). I could not find the place where I could customize the DE (shortcuts etc.).

    1. Well, I think it depends a lot of the definition of both “usable” and “old hardware”.
      I have KDE 4.4 successfully running on a Pentium III box running at 822 Mhz with 768 Mb of RAM and a Radeon 9000 PRO graphic card. While not blazing fast, it is surely usable for things like programming, viewing photos, browsing the web, and even some basic 3d modelling. I think things will only improve when I will switch to 4.8.
      For Pentiums II and below, while it is true you can get them running with a lightweight desktop, there will be very little you can do after booting: even opening a photo will either take ages or consume all of your RAM.
      So, that’s why I find kde amazing: it is able to provide a comparable experience on computers produced in a ten-years timeframe, giving access to the latest technologies even on ancient hardware.

      1. It’s funny that I remember running a DE (was it fwvm2 or windowmaker?) on a pentium with 16 MB (sic) of RAM and managing to do most of what I needed (even some photo editing with gimp). Times they are a-changin’ 🙂

  9. Although my default desktop is KDE4, I understand the feelings of the KDE4 haters.

    The first big problem with KDE4 is kdepim: We have now reached KDE 4.8, while KDE 5 is being worked on. Finally KMail does not corrupt and delete silently and randomly my e-mails any more. This was a really bad experience I had in KDE 4.6.

    With akonadi, it has become really hard to find out what is going wrong.
    Some akonadi service A has failed? -> Now a button is inactive. You have no clue what’s wrong.
    Akonadi service B did not succeed for whatever reason. -> You get about five warning messages that tell you that something went wrong. You do not know what. You have no idea where to look.

    Probably it would be easier to find out what is going wrong if there would be a place to collect error messages. Wait, there is such a place: .xsession-errors! Oh right, it is useless as it is already flooded with 2.4 MiB of garbage after only two hours of normal desktop usage.

    After several hours of tracing and fixing errors, I can send plain e-mails via SMTP, a task that should be trivial since about 25 years. After several more days of digging in log files and configuration files in ~/.local, ~/.config and ~/.kde4, I even managed to send an encrypted e-mail. The receiver told me that, although he managed to finally decrypt it, it was a mess.

    Another disaster is Konqueror. Development of the default browser that comes with KDE, has basically stalled, as far as I see. KHTML did not get much love in KDE4, and even when using WebKit, the browser is just broken. Actually, Konqueror in KDE3 was far more stable, although I must admit that the new feature to restore a session after a crash helps a lot.

    I remember KHTML being a pioneer in passing acid2 in KDE3. Regarding acid3, KHTML had a good start, but is now the worst of all rendering engines being used on the desktop. In fact, there are only two engines left that do not pass: KHTML and Trident (from Internet Explorer). And Trident does a much better job by now.

    Even with WebKit integration, Konqueror as a web browser has lost nearly 100% of its users.

    Or take a look at more basic things: krandrtray. worked fine for me in KDE3. In KDE4, I avoid it completely in favor of command line xrandr. I even do not dare to start up krandrtray, because it is dangerous.

    kcalc: what was wrong with that? It worked in KDE3. Try this: set your decimal mark to something different from a simple point. Then type in a number with a fractional part. Now just copy and paste this number in kcalc: nan. Really? kcalc does not understand its own number?

    Oh, and kmix is broken in KDE4. Repeatedly updated with shiny, fancy new look and feel. But it just does not work right. Was fine in KDE3.

    And so on: many things that worked in KDE3 just don’t work in KDE4, even four years after the first release of KDE4. And there is no intention visible to fix things faster than breaking them. There were similar problems with KDE3. However, with KDE 3.5.10, big parts of KDE worked fine. Many users who stick with KDE3 probably do this, because they want something that works. That is probably also the reason why they do not want anything from KDE4. Because then they risk getting all the breakage and there is no more point in staying with KDE3.

    1. “while KDE 5 is being worked on”

      just a comment on this one, there is no KDE 5 planned, what you’ve heard/read about is KDE Frameworks 5, which is *only* the libraries getting refactored, nothing changes in the user experience.

    2. I have exact the same opinion, you and as sbndrf above. I mean it’s great to see KDE evolving but as much as I try to use it every new update release, I am always switching back to Gnome & Unity. I am always looking back to KDE 3 times. It was such a great DE, sorry to say that, but that’s why I don’t use Trinity. Martin is right in this point.

      But as sbndrf states. KDE has so many great technologies but when it comes to daily use, it’s not working. Example: Everytime I am try KDEPIM I always ending up configuring it 1-2hrs until it finally works with my 3 mail accounts.
      Same with akonadi. It’s fine that it not crashing anymore, but I can’t use it in daily work. At first I have to wait days until it finishes indexing (at a high CPU load), then the simpliest search fails.

      Even starting programs sucks, with this KDE menu (I know there is the old one), but once I used to the Gnome Shell / Unity (experience). Sorry but I can’t get a nice workflow with this.

      Another example is Okular, which even has saving Annotations. Even Evince has this now. These are things why I can’t use KDE daily.

      And after hours of playing around, I ditch everything in favour of Unity.

      If you need more examples, just look at the top-hated bugs in the tracker. You can cry, if you read the list.

      KDE has many superior programs and to my mind Qt is also the superior technology, but I can’t work with. Btw, thanks for 4.8 improvements, KDE feels much smoother now. 🙂
      But today, I am simply using my favourite KDE programs under Gnome and I am fine, although I would really like to use KDE again.

    3. “Even with WebKit integration, Konqueror as a web browser has lost nearly 100% of its users.”

      Wrong. With the Flash 11.2 betas that eliminate all the remaining crashes with Flash, Konqueror + KWebkitPart has become my main web browser, and I recommend everyone to try that combo. In Chakra, at least, it works beautifully.

  10. There’s something that developers always forget…”You are not the code you write!”. I’m programmer, so I know how difficult is to abstract strong critic from yourself and look it just as strong critic against the technology.
    One more think to remember is that developers who _hate_ forks should not run(and write software) for X.org but for XFree86 😉
    It’s true that many people “hate” KDE4 stuff, and they are “hating” it with a very good reasons, in most occasions. IMHO, instead of whining and pointlessly changing a lot of stuff, KDE4 developers is better start fixing real life bugs and usability issues. For example: removing generic shadows from KWin is not a fix – it’s a bug. And when many such silly(from end users point of view) stuff are pushed in KDE4, cons become much bigger number then pros and _working_ “old” KDE3 looks like the best starting point for user oriented KDE.
    Btw, Unix is much older design then Windows, Mac OS 9 … but is still much better choice for the future(Mac OS X already done the transition) 😉
    old != obsolete && new != good

    Posted from my Arch KDE4.8 Linux desktop.

    1. You know… KWin 3 did not even have shadows. And there is no demand for generic shadows. That is quite clear now after one year of new shadows.

      1. The forked KWin3 has some shadows(I wouldn’t call them working, btw 🙂 ). I remember there were some whining about the new shadow system, I also remember that I dropped KDE because of that.
        Anyhow, KWin4 has a lot of usable(and very sane) features that are simply missing(not very sane implemented) from most(all) window managers, so for sure the Plasma tablet will be quite interesting user experience(if it uses KWin).
        Btw, one more thing of “work duplication” – most Linux desktop users never understand KDE and GNOME for their rework(KWin4, Mutter) of what Compiz already has some years ago 😉

        1. That question was answered years ago. Compiz perhaps did compositing, but the rest of the functionality didn’t come close to what KWin provided at that time. It seems only sane to implement the 10% of missing functionality into an otherwise great product, instead of implementing 90% of the missing functionality into Compiz.

        2. most Linux desktop users never understand KDE and GNOME for their rework(KWin4, Mutter) of what Compiz already has some years ago

          quite true – I myself did not understand that. So why did KWin start reimplementing? Well quite simple: KWin is a much better window manager than Compiz and it was easier to bring to KWin what Compiz can do than the other way around. Just think about the fact that Compiz has still not reached a 1.0, still does not support the non-composited case.

          Mutter follows a completely different approach towards compositing.

        1. it’s an experimental feature not enabled by default. Btw Trinity included a huge (and partially wrong) patch for shadows. I think you told me once it was used in openSUSE as well.

          1. No, It was never used in openSUSE. I will not include this patch because it breaks and duplicates the normal kwin shadow function.

    2. It would be nice if KDE team would made a crush-100-Top-hated bugs release. There it should pick especially usability things or KDE4 regressions. Man, I must be dreaming 😀

    3. “One more think to remember is that developers who _hate_ forks should not run(and write software) for X.org but for XFree86 ;)”

      This case is not comparable though, as Xorg has been forked for license reasons (IIRC the switch from a GPL-compatible one to a non-compatible one).

      1. No, that was just the last straw. It forked because people were finding the core team impossible to work with, distros were maintaining extensive patch sets, etc. The time was ripe for a fork. (In fact, that’s why XFree86 changed its licence – because Xouvert claimed to be a fork, even though it never did anything, so they changed the licence to make sure they got credit … and hence made it GPL-incompatible.)

  11. “I am stupid enougth to imagine that the leader of the KDE developpment team could receive two salaries: with one coming from MS.”

    It might not apply to this specific case, but I don’t think that the general idea is so absurd. It is a fact that Wikipedia gets massively manipulated, it is a fact that openstreetmap gets manipulated (list can be continued) so why should this situation not apply to KDE/GNOME et al.?

    With the fanboism as a big accelerator spreading FUD is an extremely efficient (and cheap) way of disrupting open communities. I think we need to accept that this happens and we need a strategy on how to cope with “social sabotage”.

    The first step would be not to play their game.

    1. I am quite sure that this is not the case for code contributions. KDE has quite good working ways of peer reviewing code before it hits the repository. To harm the project from a coding point of view they would have to buy the complete development team.

        1. @Kevin Krammer
          Sorry, I maybe quoted the wrong part. My point is that it is likely that a lot of the FUD FOSS has to deal with is actually placed by paid PR agents.

          @Martin
          Me neither. Its is way easier to frustrate the code contributes and their creativity. Think how much time you spend dealing (and thinking) on the issues similar to what you described in your blogpost. Time you could have spent to create and enjoy.

    2. The only case of KDE manipulation I’m aware of was Canonical with Kubuntu. However their manipulations were done entirely downstream in Kubuntu (eg. shipping broken package manager front-ends, applying experimental Qt graphic modes by default, breaking translations,…) and upstream KDE / other distributions were never affected.

  12. The whole point of KDE3.5 and Trinity-Users is, that they want a fast, lean-and-mean environment like in WinXP set to Win2K Classic. On the other hand, the KDE4 devs are trying to sell KDE4 as positively maximum featured, maximum dependencies requiring, maximum clicki-bunti solution to everyone, and worse, they try to depict everyone who doesn’t like that maximum-bloated approach as hostile, anti-progressive, ignorant, etc. When WinXP and KDE 3.5 came out, they were certainly not really ressource-saving, but that was many years ago. Now they are fun to use.

    The whole point for a KDE5 should be to make applications less interdependent, you don’t have to run servers and databases for every little piece of information, for instance you can store smaller amounts of emails in text files, if another app needs to know about emails, it can simply open the text file. The KDE4 approach is to generate indices of all the emails, store them in sql databases, start services to administrate emails. This is simply overkill for a few hundred or thousand text messages. The bloated approach should be kept optional, if a user needs all that bloat, he should enable it, all other users should be spared from that bullshit.

    The simple message: KDE 3.5 and Trinity-users want to KEEP DA BLOAT OUT. As far as I understand the Trinity guys, they want to stay with QT3 for their KDE3 packages, on the other hand if they get enough man power, they will try to port KDE4 apps like K3B over to their tqt interface, which will allow to call QT4 (KDE4 Lib) functions without installing the official KDE4-Libs-Bloat.

    1. I think you should seriously read about how email works. If it were as simple as you believe the world would be a better place. That is exactly the thinking I pointed out about not understanding the technology. You obviously have no clue about email (protocols) and why a proper solution to the problems is an architecture like Akonadi.

      But keep living in the world of an unbloated Trinity. May I suggest to just google for “KDE 3 bloat” 😉

      1. Rule of thumb: if somebody complains about decisions like “storing email in a database”, they are obviously not talking about Akonadi even if they think they are.
        The are probably talking about Exchange or some other groupware server that uses a database for content storage.

      2. Talking about akonoau^dAkonadi…

        I was watching the git tree and I found a “Mail Notification” resource that gets disabled by default. I tried to compile it, but it doesn’t do anything.

        Why don’t anybody (well, I don’t have the technical competence to do anything programming-relevant, and I would appreciate some resources to begin studying) does a simple hack for Plasma to fetch mail and alert me when a new mail arrives, without KMail running? That alone would justify Akonadi, and would give us the edge, for essentially nothing (all relevant pieces are in place, even the relevant resource, the only thing missing is the interface hook).

        1. try the Lionmail Plasmoid. That’s what I are using to get e-mail notifications and I love it. It allows me to get folder specific notifications, e.g. about comments on my blog 🙂

  13. If you don’t like Trinity why are you reading their mailing list?

    It’s like you’re a masochist who wants to get himself annoyed. Leave them be and they’ll fizzle out with time.

    1. I subscribed to the mailinglist because I wanted to get them to use KWin 4. I am not really reading the mailing list – currently the folder contains > 200 unread mails. I’m just following a few threads.

      1. So you subscribe to ml of project that is all about NOT using kde4 to preach about kwin4? Do also do it lists dedicated to users of Gnome, MacOSX and Windows?

  14. That is exactly the point. I want to consider an email as simple text file. If you want to extract information like subject etc from emails, you can do this from text files too, no database and servers needed. You’d like to use an email like kind of an OLE object in windows, but that is far more than the average desktop user will ever need.

      1. I’ve seen this attitude a lot of times recently, and I really don’t understand it. So I’m curious. What do you think is so evil about databases? (Very) roughly said, their purpose is to store, organize and retrieve data in an efficient way. If you consider emails, what is a bunch of emails for you? From a technical point of view, would you really find it more sane to store all the raw data in a file, and parse the whole file everytime you do something with it? On every access parse all the headers, sender, retriever, forwards, subject, etc. pp.? Whats so evil about using a mature technology that was basically created for this kind of situation?

        I think the circumstances around KDEPIM are much more complex than what people think. It is not only a matter of introducing or not introducing a database for storing some data in the backend. It is an unfortunate combination of circumstances like lack of manpower / funding, porting work, bugs, re-architecturing, etc. pp.

        1. The “databases are evil” is a result of lack of information and thus misunderstanding of the topic.

          For non-developers and especially non-IT people, databases are associated with storage of large datasets, just see how the word is usually used in movies or on TV.

          So there is no knowledge and thus no understanding that database, in this case relational database, refers to a concept of working with data (not necessarily about storing).

          When the problem at hand is relating (meta) data to other data, the obvious choice for any developer is a relational database engine.

          And as long as this is not artificially pushed as a perceived problem, no user will ever care.
          However, if someone with whatever adverse agenda pushes the view that this is bad technology, people are likely to follow it because their understand of database is not what actually is being used.

          Take for example Firefox. It uses a relational database. Do people complain about that? No,

        2. “What do you think is so evil about databases?”

          Nothing, but the choice MySQL for backend in DE was rather bad. The same mistake in Amarok and now this is “cow” and not music player.

          1. can you explain why this was rather bad? What do you think is bad about using MySQL? On my work system I need to have a running MySQL instance anyway and having Akonadi use this one is a real win.

            1. “I need to have a running MySQL instance anyway”

              May I ask in what purpose? And wheter this purpose is avarage desktop/notebook/netbook usage?

              “can you explain why this was rather bad?”

              Beacuse is overkill for normal needs. I understand that developers bulding environment for their needs.

              1. “I need to have a running MySQL instance anyway”

                May I ask in what purpose? And wheter this purpose is avarage desktop/notebook/netbook usage?

                For my work I need a MySQL instance. Quite normal application using a database backend.

                “can you explain why this was rather bad?”

                Beacuse is overkill for normal needs. I understand that developers bulding environment for their needs.

                Sorry, but this is not an explanation at all. It’s just rephrasing different. Why is using a database overkill? Did you know that Firefox uses a database?

                1. “Did you know that Firefox uses a database?”
                  Yes. I know. Did Firefox using MySQL? Using MySQL is overkill not every database. Please do not put words in my fingers I did not write.

                  “For my work I need a MySQL instance. Quite normal application using a database backend.”

                  Ehh… developers probably never will understand the needs of ordinary users.

                  1. And why exactly is MySQL worse than SQLite? Can you please provide the information I am missing here?

                    1. Have you any proof for this claim? Does a MySQL instance really require more resources than an embedded SQLite? Where is the proof for it?

                    2. no, sorry this is no proof. I am running Kontact Touch on my tablet. Also the linked blog post is about 4.6 and not 4.8.

                    3. Yes, i now. If you read comments, will be able to noticing that akonadi is also involved in this. Rather, it shows how these technologies can destroy all the pleasure of using KDE.

    1. “If you want to extract information like subject etc from emails, you can do this from text files too, no database and servers needed.”

      Since I already wrote about the misconception regarding database, I’d like to address the misconception about the use of a server or service.

      The main purpose of using a service for data access is not to provide extraction of information like subject, it is about avoiding concurrency issues.

      Concurrent access to files read-only works nicely, however once you introduce writing into the equation you reach a point where you potentially get access conflicts (in the worse case two write accesses overwriting each other’s changes).

      In theory two processes accessing the same file(s) could use file locking to solve that, in pratice file locking is not reliably available on all commonly used file systems (and can lead to problems like stale locks).

      One other way that will reliably solve this particular problem is to only have one program access the file and other processes communicating with that authorized one.
      This is called a service based approach. Widely used elsewhere, e.g. GNOME’s Evolution does do that as well (called Evolution Data Server there)

  15. >After pointing out that KWin does not depend on Akonadi, I got:
    >> I meant nepomuk, not akodanai.

    I didn’t use KDE or Trinity and now even Gnome/Unity (after they completely broke Alt+Tab), but if kwin depends on nepomuk, I can undestand people not wanting it.

    1. KWin does not depend on Nepomuk. I wrote “KWin neither requires Akonadi nor Nepomuk”. But KWin can make use of Nepomuk if it is available.

      1. the whole akonadi/strigi/nepomuk&Co business is hated, doesn’t matter if one of them is more guilty than another.

        1. I myself, and a whole lot of other people if you follow recent blogs, love Nepomuk and the concepts behind it. I do not say that everyone loves it, but please don’t state the opposite either.

  16. Firstly, I want to thank all the developers for their work which enables me to do my work on my computer every day. Since this is the blog it is, I want to mention, that kwin on my Intel GM965/GL960 just works great!

    I am running kde since late kde1 times, and I was allways happy with what I got. KDE 3 was great, but in general I agree that those times are over. Not at least because QT3 has long reached EOL.

    But it seems also true to me that there are some unpolished parts of KDE 4 right now. Of most concern is akonadi and friends and kdepim. Let me give an example. On every start of my KDE session, I get two errors telling me, that some backend could not fetch some resources. And sending plain text e-mails is an adventure. I think, there is the possibility of frustration, especially in this area. Perhaps slowing down new feature development and doing a round of stabilization for one point release would help to overcome this. While I am suggesting this, and would be happy to see this happen, but please keep in mind that I am not a developer, and it is well possible that the development wheel spins in another way than I think is does.

  17. I understand that it is annoying to see misinformation being spread without anyone pointing out that what looks like fact is not, but I would recommend to do that on channels where it matters.

    E.g. misinformation like “then i heard that the KDE team wanted to adopt the entirety of QT4 into libkde and maintain it as part of libkde” on a KDE communication channel or general user support/information channel should be followed with a clarification that this is wonrg and probably with a question to where this faulty information came from.

    For example I would be interested to know where such a rumor would be heard. Probably where gems like “the KDE developers who had been funded to the tune of $EUR 10 million” come from 🙂

    Of course not all rumor sources can be pin pointed but IMHO countering false information makes more sense on channels where recipients are actually interested in information not just for gossip

    1. People are confusing NEPOMUK with KDE. KDE /uses/ NEPOMUK, and therefore KDE *created* NEPOMUK, and received the reported 11.5M euros for it. (See wp:NEPOMUK framework)

      Of course, KDE /didn’t/ make NEPOMUK and /didn’t/ get the 11.5M euros.

      1. I believe the confusion is rooted in different meanings of the name “Nepomuk”.

        A) First, there was “NEPOMUK”, a big international research project in which several universities, companies and FOSS communities collaborated and which received a couple of million Euro of funding from the EU. (Website: http://nepomuk.semanticdesktop.org).
        It was not a concrete software project, it had more general goals of investigating the possibilities of semantic technologies on the desktop and web, try to come up with technological foundations and a unified set of ontologies that could help with and standardize such technologies, and do lots of fancy academic research into technical and social aspects surrounding such technologies.
        It also produced a couple of concrete software implementations, which were however mostly intended (by the researchers) as “proof of concept” examples, not yet for actual usage.
        Afaik, this NEPOMUK research project has come to its scheduled end and is no longer active.

        B) Then there is “Nepomuk-KDE”, the software now known among KDE users and blog readers as “Nepomuk”. (Website: http://nepomuk.kde.org).
        It originally started out as a small sub-project of the NEPOMUK research project, as an example of a concrete implemantation of some of the investigated semantic desktop technologies.
        But KDE continued to develop it on its own afterwards (and still continues to do so), and now ships it as part of the KDE software collection.

  18. The only hater here seems to be … you. You have the right to your opinions, so feel free to suggest that KDE4 is the holy grail but don’t ask somebody else to stop doing what they are doing.

    1. Yeah that is the easy conclusion to get. But why should I hate Trinity? Does it take us away any users? No! Is it any competition to us? No! Am I personally insulted by the fact that they forked off my old code? No, because I did not contribute to KDE 3.

      Also I think you did not read my post carefully. Nowhere I ask anybody to stop doing what they are doing.

  19. I agree on most of your points. It’s sad to see so much energy “wasted” trying to reinvent the wheel. It’s also sad to see so much hate for insignifiant reasons. Instead of fighting each other it would be really more productive to work together.
    What I don’t understand is why aren’t they (The Trinity people) trying to contribute to KDE4 and fix the flaws they pretend to find in KDE4 rather than doing it on deprecated software.
    But well, haters are everywhere and on every subject. It’s just impossible to avoid them. The best solution is just to ignore them and let them be, they’ll soon or late realise that they are wrong or not, so there’s no point arguing with them.

    I just can’t understand why people say “KDE 4.3+ is slow and bloated”. KDE4 is the most versatile DE I know, and it’s possible to build from an ultra light KDE4 configuration to a super bloaty and eye-candy one. I use KDE4 from the version 4.2 and it’s always been faster and lighter than when I used GNOME2 or KDE 3.4/3.5, and way more versatile. So people saying that just never took some minutes to tune their DE or used the wrong distro I believe…
    I however, I must agree that some aspects of KDE4 were/are problematic. Especially some daemons that had the bad habit to eat my whole CPU (hint: nepomuk, plasma-desktop…) but it tends to be much rarer noadays and no desktop environment is exempt of bugs anyways. But still it’s not a reason to throw so much hate to the developers that are trying to make things better.

    Anyways I want again to thank you all the KDE developers, you guys rock, for providing us an amazing bleeding-edge and versatile desktop environment. Don’t bother with haters, just don’t play their game as other said.

    1. What I don’t understand is why aren’t they (The Trinity people) trying to contribute to KDE4 and fix the flaws they pretend to find in KDE4 rather than doing it on deprecated software.

      I quite agree and actually I have asked in multiple threads to upstream the changes. I also several times pointed out that KDE developers already did the same and that they should at least look at it before reinventing the wheel. It’s a really sad story.

      1. You do realize of course that Aaron (assuming you are referring to Aaron Seigo) is neither involved in Nepomuk nor Akonadi development.

        However, being a pretty smart guy, he understands how features of low level technology can be used to provide features on a higher level, i.e. how to transform features for developers into features for users.

        1. He is pretty vocal supporter. When your users do not want something (as in ‘required’)… maybe you should actually listen to them once in a while.

          The whole situation is a mess. It did not work at all for a few years (Unicode was broken too, but who cares about non-English audience, right?), now it’s still misbehaves and gives exactly nothing to some part of the userbase while consuming memory & CPU.

          The same thing with activities. Some users want just a simple desktop, they don’t need no stinking revolutions in user experience™, databases and what’s not. KDE was always about options and user’s choice, now it is developers feeling high and mighty all over the place. That’s why I mentioned Aaron, he dismisses that as an issue and he, in our (users) eyes, represents those developers.

          1. I don’t understand what the huge problem is. We aren’t forcing Nepomuk down anyone’s throat. There is an option to disable it completely, and most distros actually do disable Nepomuk.

            Some of us work on Nepomuk cause we care about it. We’re not paid. What right do you have to tell us that we shouldn’t be working on something that we want to?

            Ditto with activities. If you don’t want to use Activies, don’t. I think most, if not all, of KDE believes in customization and giving the user a choice.

            Can you elaborate on KDE doesn’t provide a “simple desktop” in your opinion? If you don’t like the fancy kwin effect, and Plasma widgets, Nepomuk and PIM, you can turn all of them off.

          2. > When your users do not want something (as in ‘required’)… maybe you should actually listen to them once in a while.

            The correct solution here is not to get rid of nepomuk/akonadi/etc., but to fix it all until it is working as intended, without users needing to configure anything or intervene with processes eating all their ram or running amok. Then these services can be rightfully marked “required”, developers can reliably build on them and users can have their technology cake and eat it, too. Only if by that time it turns out these services were a bad idea, removal/replacement should be discussed.

            Users usually have no idea what they want even if they can tell you exactly what they’re looking for, they just know what makes them feel good. If N/A/etc. can make them feel good, all is well and nobody would think about removing them. Win7 ships with lots of interdependant background services, is anybody whining about how they’re wasting system resources?

      2. I think it takes a coward to come to this discussion and personally attack someone (even slander) with no proof and no examples, all the while posting as “Anonymous”. Ah, welcome to the Internets…

          1. Just a word of warning: I have no problems to moderate away comments written by anonymous posters. I think James comment was right in pointing out that anonymous posting is bad style.

            1. Anonymous posting is bad style? What? Pro-censorship in opensource community. Now that’s unusual. Do as you wish.

              I’m just a random KDE user, I’m not a developer, and my name will not give you anything. You can close your ears from your userbase all you want. That’s your choice. Don’t blame others when KDE will fail though.

              1. This is my personal blog and insulting comments are in general moderated. I think everybody understands that I would not allow to have people come into my house and throw insults at me or others. The same is true for my blog, so I moderate.

  20. I’m going to break this into a few parts:

    ————————————————–

    1. The kwin/nepomuk debacle: This error seems to have been due to the following sequence:
    a. kwin depends on kdelibs
    b. Some distros have kdelibs set up to depend on nepomuk, although it isn’t an absolute requirement.
    c. Because–sorry, Martin–you clearly had an agenda, no one was willing to take your word for the fact that there was no direct relationship between nepomuk and kwin.

    It was a simple misunderstanding caused by the packaging policies commonly found in binary distros combined with the lack of time to do extensive checking, nothing more.

    ———————————————-

    2. KDE4 haters: Like it or not, the KDE Project made a severe public relations error by releasing KDE4 while it was still in, at best, an alpha state. Most people who tried it in that state and were disappointed are not going to try it again. It’s rather like a job interview: a bad first impression means that you don’t get the job, and bombarding your prospective employer with additional copies of your resume and demands for interviews is not going to help–it’s just going to annoy them.

    Some people did give KDE4 a second chance, but the KDE Project just didn’t get things fixed quickly enough to give them a good experience the second time, either. These people should not be blamed for not being willing to give you yet another chance.

    You simply need to accept that KDE blew it with respect to those users and you are not going to get them back. None of them are ever going to like KDE4, and they *will* be dismissive of it, because *in their experience* it deserves to be dismissed.

    Some of them are going to go beyond dismissing KDE4 and will be rude to members of your project because they feel you were rude to them by ignoring their needs and concerns. It doesn’t matter whether that’s realistic or what the KDE Project intended or even whether KDE4 is the greatest creation of humankind since fire and the wheel–it’s these people’s *perception* of being slighted that needs to be addressed before they will stop.

    —————————————

    3. The “evil” of strigi, nepomuk, akonadi, databases, etc.

    That’s right, I *don’t* actually think these things are evil, despite being a supporter of the Trinity Project. What I *do* think is that they’re solutions to problems that many users (myself included) just don’t have. If they were as transparent as simple text files and consumed no resources, that wouldn’t be so much of a problem, but that isn’t the case. Not needed + opaque/difficult to troubleshoot + consumes resources (not necessarily very many resources, but some) = not desirable.

    In the specific case of databases, being able to manipulate data outside the context of its original application, using simple command-line tools devoted to working with ordinary text, is a Good Thing from the standpoint of most Linux users. Stuff it in a database and you make that more difficult (or lose the ability altogether). Continue a few steps further in the same direction, and you end up with the Windows Registry, which is not something I think anyone here wants. Is structuring data useful? Yes, for certain types of large datasets. In some cases, though, transparency is more important, even if it isn’t as efficient. I think KDE4 has swung too far in the opaque-but-efficient direction for some people’s tastes.

    Personally, KDE3/Trinity is really more desktop than I need, and I use only a small subset of its facilities. The extra facilities provided by KDE4 are useless clutter for me, so I don’t want to have to deal with them–not even just to switch them off, in cases where that’s possible. XFCE4, on the other hand, doesn’t provide some facilities that I *do* use, so while Trinity isn’t “just right” for me, it’s the “best fit” from the desktop environments currently available, while KDE4 is several sizes too large. 😉

    ———————————————–

    4. Trinity doesn’t presently have enough developers to handle such a large codebase: I don’t think there’s anyone involved with the Trinity Project that isn’t aware of this, but honestly, what do you expect us to do about it? “Abandon the project” is not an acceptable solution–that would be like me telling you to drop KDE4 and come to work on Trinity (in short, just plain rude). Most open-source software projects are started (or, in this case, restarted) by just one person. It takes time to accrete developers. In the meanwhile, Trinity can only make do with what it has.

    Would it be productive to triage the codebase and dump some of the less useful/less wanted applications into a bin labeled, “we’re getting rid of this the moment maintaining it becomes too much trouble, and no new features will be added”? Maybe, but I’m not sold enough on the idea to fight for it.

    Would it be productive to bring in some code from KDE4? Maybe. I’m cautiously in favour of the idea *if* the problem of dragging along several megabytes of unneeded dependencies can be resolved. That means splitting kdelibs, which was something you said KDE was working on, and that Trinity just doesn’t have the manpower to contribute to.

    There’s a second problem that also comes down to lack of manpower: Trinity can’t just import willy-nilly from the KDE codebase. Someone has to check potential patches first to make certain that they’re a good fit for Trinity–producing a second KDE4 with a different name benefits no one. That checking has to be done by experienced developers.

    ————————————————

    Overall, your complaints seem to boil down to, “these people aren’t doing things the way I would do them, they won’t listen to me, and they dislike the project I contribute to!” Thing is, the bad blood between the KDE Project and the remaining KDE3 users predates Trinity. That it’s become concentrated at the Trinity Project is unfortunate but unsurprising. And articles like this *widen* that pre-existing breach. You’ll have to pardon me if I don’t think that’s useful.

    1. I don’t think there’s anyone involved with the Trinity Project that isn’t aware of this, but honestly, what do you expect us to do about it?

      What about concentrating on that parts that make sense for Trinity? E.g. no longer developing an own window manager? Porting kicker/kdesktop to Qt 4 and depend on kdelibs? Dropping everything that has nothing to do with the goal of providing the KDE 3.5 desktop?

    2. “What I *do* think is that they’re solutions to problems that many users (myself included) just don’t have.”

      True, however quite a lot of users of KDE use more than one program simultaniously that access the addressbook, e.g. kaddressbook, kmail, kopete, konversation.
      Those users were negatively effected by getting data overwritten when file locking did not work properly.

      An access services makes that reliable.

      “In the specific case of databases, being able to manipulate data outside the context of its original application, using simple command-line tools devoted to working with ordinary text, is a Good Thing from the standpoint of most Linux users. Stuff it in a database and you make that more difficult (or lose the ability altogether). ”

      Absolutely. The question is where does KDE do that? Nepomuk does not do that, it aggregates data from multiple sources that of course continue to exist.
      Akonadi does not do that either.

      Are you referring to Firefox which stores its history data in a database?

    3. 2. KDE4 haters: Like it or not, the KDE Project made a severe public relations error by releasing KDE4 while it was still in, at best, an alpha state. Most people who tried it in that state and were disappointed are not going to try it again.”

      But this was not only the fault of the KDE Project i think, yes they should have made it clearer that it is not a release targetted to end users. But it was also the fault of several popular Linux distributions switching to KDE 4 as their default KDE version too early.

      1. Who was wrong about the fact that the whole semantic desktop did not work for a few years? And it is still not quite there. Maybe someone should actually finish his technology before throwing it at users.

  21. Hi,

    I’m a Trinity user and I belong to the first and third group. I have never tried LXDE, so maybe I should. The reason I used KDE 3.5 instead may be more related to ignorance than anything sane. (Just never took the time to try LXDE or smth else.)

    As for the classic desktop, KDE 4.8 seems to cover things quite well. I’ve installed Ubuntu Precise recently with KDE 4.8 and tweaked it to be mostly like KDE 3.5 and it works quite good. There were some problems with the window manager though (some graphical issues). I installed Trinity beside it for now (there are also some bugs, but they are a bit less bothering). I hope to be able to solve the issues I have in KDE though. Is there any place on IRC where the devs hang out that I can get info? Is it ok I just use the Precise packages or should I run another version?

    Thanks in advance!

  22. While I agree with everything posted here, I’m not sure it was really constructive. This seems like the kind of thing which is only going to further rile up people on both sides. Perhaps just ignoring Trinity would have been the better option.

  23. I used what was becoming Trinity (at the time I last used it, it was really just a KDE 3.5.x PPA) until KDE 4.x became stable on my system . . . which, as a TwinView-using guy on an odder NVIDIA card, wasn’t until about 4.5. That is a long time after distros stopped shipping KDE 3.5 if they were keeping modern on anything else, and honestly, I still feel a bit of resentment over it. I understand that long before then KDE started working just fine again for the vast majority of people, but it was completely unusable for me (plasma-desktop would crash constantly, KWin would start using 99% CPU if it ran for more than a few hours, rendering was sporadically slow and error-filled, etc etc).

    Now, these days KDE is back to being brilliant, and better than ever before. I love the flexibility that Plasma provides, and how by its nature it’s arguably trivial to fit into other form factors and UI/UXes, and anyone who’s interested can contribute in small little ways (I myself added a Pooch Cafe plugin for the comics widget, since I love that strip). But it’s true nonetheless that things behave a bit different from how they did in the KDE 3.5 days, and it’s even more true that the transition was harsh, and for some far harsher than others (which makes it hard for many core devs and supporters to entirely sympathize, even if they understand it intellectually).

    But more importantly, I don’t think we in the Free Software world have the right to really say to someone that they can’t do something, unless it involves withholding code. I don’t just mean in the legal sense (although obviously that tends to be true), but more in the spirit we live and work under. We can certainly say, as you’ve said here, that it is unlikely to succeed, that there are these potential security and longevity issues, etc etc. But remember, to a large degree this whole world started with something that was “just a hobby, won’t be big and professional like gnu”. If someone has a passion for something, they get to do it. If you don’t care for it yourself, ignore it; not only is that more respectable than badmouthing a project (yes, even if the project badmouths yours first) but, as pointed out by others in the comments here, that’s probably your best play anyways.

    From my own perspective, though, even if Trinity has outlived its usefulness, that project got my computer through a time when it was impossible for me to actually run KDE4, and whatever actions or invective the project has generated since then, this project (mainly the one lead dev) got me through the transition to KDE4 with a usable KDE environment the entire time, and I’ll always have respect for it. Other than that, I think you just have to let people scratch their own itches.

  24. The KDE4 developers should look at some of the responses to my hypothetical question here:
    http://trinity-users.pearsoncomputing.net/?0::2681

    Just click the “Next” link to see what TDE users have to say…none so far would choose KDE4 over the myriad of other options available.

    Which begs the question: If TDE users would not choose KDE4 over the alternatives available, why is a prominent KDE4 developer trying to close down the TDE project? It certainly can’t be to boost the KDE4 usage statistics…

    1. Because a better alternative would be to run an enterprise-level OS such as CentOS 5, which not only provides the traditional KDE 3.5 desktop but a whole wad of bug-fixes that don’t exist in Trinity?

      1. Wait what.

        Enterprise is not for a regular consumer. Also, people using Enterprise desktops do not provide the same functionality as newer desktops do. In fact I am very happily running linux 3.2 on my Archlinux box, with the latest everything, and running Trinity. Can I do that with enterprise? No.

        Secondly many enterprise options are for large scale deployment and cost a LOT of money. It’s just not feasible.

        Third, What about all the bug fixes that exist in Trinity. What about the 3+ years of development and time? What about integration features provided by 3.5.13 that aren’t there in 3.5.10.

        Also, old patches laying around have been imported to the TDE environment. KDE:KDE3 for example have many patches and we are looking to import some of those fixes.

        1. First of all:
          >Can I do that with enterprise?
          Yes, you CAN do that with enterprise. Quite easily in fact. The fact that you don’t even know this is quite forboding.

          A person running KDE 3.5 won’t get the same functionality that a person running KDE 4.7/4.8 will. I simply assumed that, if you were looking for the same functionality as a desktop running KDE 4.7/4.8, you wouldn’t be using Trinity.

          If you’re going to run an ancient DE, don’t be surprised when it depends on ancient software only available in enterprise (or just plain old) distributions. See: HAL, Arts, Linux 2.6.*, potentially sysvinit, etc. At least in enterprise distributions, this software is still being supported and fixed. More importantly, at least an enterprise distribution running KDE 3.5 will WORK fully.

          CentOS is completely free. Won’t cost a penny. Installs like any other desktop distribution.

          I am somehow more confident that Red Hat are more capable at fixing bugs and vulnerabilities than a single part-time developer.

          Supporting a Qt 3-based KDE 3 just isn’t feasible without industry support, and if you want industry support you might as well go with a project that ACTUALLY HAS industry support. Like, for example, an enterprise desktop.

          1. Not sure what Calvin was trying to say with regards to Enterprise desktops; TDE *is* an Enterprise desktop.

            Calvin is not part of the steering for the TDE project as a whole; his primary job is to assist with social media, the website, bug management, etc. and he may simply be unaware of these distinctions.

          2. Maybe I am not 100% clear on Enterprise deaskop. My thought was to say that “I would like to use a modern distribution AND I’d like to use Trinity/KDE3 whatever name you give it.”

            In time those distributions will reach EOL and will no longer support KDE3. What then? I use Trinity so that I can have a product now, but also in the future. Using an enterprise desktop from 2008 won’t do me very much good down the road.

            And Timothy is correct, I am not speaking in any sort of official position. These thoughts are entirely mine and not of the Trinity project itself.

  25. Hi!

    I am a KDE3 maintainer in openSUSE and in no way associated with Trinity project. As you possibly know, modern openSUSE ships with normal KDE3 which gets gradually updated.

    I would say that many of the difficulties you noted are not that disasterous. For example, HAL has been dropped from openSUSE long ago (there is even no such package in the repository) and this event did not lead to any catastrophe for KDE3. Qt3 is still required by LSB so it will stay there for many years independently of whether or not we ship KDE3.

    Conversely your idea of using kwin from KDE4 under KDE3 is not that attractive in my impression. There are many disadvantages in using Qt4-based window manager under KDE3 (no kcontrol integration, no shared colors, greater memory and disk consumption due to use both Qt3 and Qt4). Even if KDE3’s kwin would not be available, I still would most likely recommend using Compiz-8.6 rather than kwin from KDE4 because it has an option for using KDE3 window decorator, and written in GTK which is anyway installed on any system. There is also a fusion-icon-qt3 applet that allows to control Compiz from KDE3.

    That said I would very much appreciate your help in maintaining KDE3 in openSUSE, especially if you could point me to useful patches that still not included in openSUSE’s KDE3 and express your opinion on the patches that I include.

  26. My recommendation is for them to make a trinity plasma shell, either fork plasma-desktop or create a new plasma shell much like plasma-netbook. They would benifit from all the new features and plasmoids while being able to tweak the shell to look more / act more like KDE 3 and it wouldn’t be a full fork since there would be a shared codebase with plasma / all of the other applets. Besides the shell they would need to design a plasma-theme that looks like KDE3 theme and update the crystal theme, optionally they could add a dialog that can configure the defaults to be more like KDE3(Konqueror default file manager, crystal theme and KDE3 plasma theme / kwin theme). Qt3 is old and without industry or a large community support it is not a feasable goal for the long term.

    1. trinity with plasma? are you kidding?

      That is exactly, what convinced KDE3 and Trinity users don’t want at all. Maybe the whole proposal was sarcasm, but my experience with KDE4 devs is, that they really think that way. In their opinion, Trinity urgently needs a bkonadi-server, strikxi, depomuk etc.

      1. And for having a desktop shell you don’t need: strigi, akonadi or Nepomuk. That’s exactly the thing this blog post is about. The trinity developers have no clue about the technologies and base their decisions on “we don’t want the hated technology” without even checking the facts.

          1. Yes of course. Nepomuk altogether is a completely optional build dependency. On my CI system compiling kde-workspace I don’t have anything of Nepomuk installed as I don’t need it to test the building.

            1. Thanks. Strange that there is no distribution, where Nepomuk is optional. I even have not distribution where Nepomuk is off by default.

              Don’t get me wrong I don’t hate Nepomuk. I simply I do not see the solution to my problems in Nepomuk.

                  1. “there is a low-fat package which disables Nepomuk”

                    And you must install this package after you install kubuntu? Then Nepomuk is not off by default. 😉

                    Kubuntu is propably the censored with KDE4. Please do not recommend censored.

                    This comment got moderated to remove insults

                    1. where did I recommend Kubuntu? Btw if you want to continue to be allowed to post comments here you should not insult. I do not tolerate that anybody calls a distribution names!

        1. I really should get a life …
          I’m not a Trinity developer, some of my arguments may be a bit unprecise and too emotional. I’ve got an degree in computer science, but I’ve become quite skeptical to so-called progress, I’m not considering facebook, web x.0 and all this stuff as desirable.

          Yet, Martin, your point is, that the Trinity project would be much saner, if it adopted more traits of KDE4, first a small inconspicuous kwin4, then a little nepomuk here, a little plasma there (I understand, that these things are completely unrelated, you don’t have to tell me again), in German this would be called “Salamitaktik”. But that is exactly what the Trinity and KDE3 guys absolutely don’t want, they don’t want to be assimilated. Nobody needs two KDE4 projects.

          If you wanted to e.g. bring KMail4 into Trinity, you would have to make it (in standard configuration) completely independent from all these services like akonadi&Co and mysql (again, I understand, that KDE3 isn’t free of services in general, but nothing like akonadi there), but your stance would probably be, why not allow a little akonadi into Trinity, just a little bit, it won’t hurt, don’t be that intolerant, akonadi’s got a right to live. If the Trinity project ever accommodated to such intents, it would become like KDE4 after a few ‘merges’.

          1. I seem here a very severe and common misconception: considering everything from KDE “SC” as one thing. I am mostly talking about the desktop shell which has nothing to do with KDEPIM and by that Akonadi. The Trinity developers and many others make the mistake of considering KDEPIM as part of the desktop. This is not the case!

            The result is the irrational demand of nothing should be from “KDE 4” and that Trinity tries to continue everything. The decision whether to replace e.g. twin with KWin, kdesktop with Plasma and KMail with KMail 2 have to be considered independently. If the Trinity developers decide that e.g. they don’t like Akonadi but continue KMail 1 that is fine. But it has nothing to do with any decision on whether kdesktop should be ported to Plasma.

            This is one of the real big issues with Trinity: they never got to the state to define what they want to achieve (e.g. keeping this piece of software or that) and go through the complete package and drop all the things they don’t care about.

            Given for example the situation around kdepim, given the complex code base and the high demand of development effort (just look at the bug statistics for how many bugs kmail 1 has) there could be only one conclusion: don’t continue kdepim 1. So what about the users who don’t want Akonadi based KMail? Well there is still evolution, thunderbird, gmail, etc. etc. Those users not wanting Akonadi based kdepim might just have failed to realize so far that kdepim is not the right product for their usage pattern. That there is better software out there to fit their needs.

            1. I’m not officially speaking for Trinity project. Afaict, the Trinity guys first want to make sure, that the KDE3 base packages ‘survive’ and that they get a minimum of bug fixes, so that KDE3 will still be distributed with recent linux distributions, a few opensuse volunteers are doing the same for opensuse. The KDE3 apps aren’t getting any new features (if not maintained by their devs), and very few bug fixes at the moment. The short term plan is just to keep a complete KDE3 DE alive, there’s no discussion about which packages to throw away, no ‘thinning out’ planned, the whole idea would be rejected, in the worst case the users just will have to live with buggy KDE3 apps. The issues in these situations with e.g. KMail3 mostly arent’t about missing “Lotus Notes” functionality and missing fancy facebook integration, it’s just about being able to read one’s emails without crashes, if it works, people will stay with KMail3.

              The whole situation is certainly dissatisfying for Trinity/KDE3 users, if you e.g. would like to use K3B from KDE4, because K3B from KDE3 has a lot of open bugs and was hardly maintained even before KDE3 went out of support, you would have to install all the dependencies from KDE4libs again.

              1. You are unfortunately misinformed about what Trinity actually does. As I wrote in the blog post they are adding features to e.g. their KWin fork, which are all incorrect and dangerous resulting in a worse product than the actual kwin from KDE 3.5.

                1. If their changes are objectivley wrong and resulting in a worse product, then file bugs or send Tim a message.

                  If you just mean, that these changes will make it harder to integrate KDE4 stuff into Trinity, well …

                  1. In my first mail to the Trinity project I referenced actual wrong commits. Nobody ever asked what is wrong in the commits or anything about it. I told them several times if they have questions about KWin they should ask – it has never happened for the code.

                    1. This is because you have not stated what you think is wrong with those commits. Anyone can claim that something is “wrong” without providing at least a basic explainaination of why they think it is wrong; ususally these people are ignored as they normally don’t understand why the commit was made in the first place. How would you respond if I pointed to the kwin compositing commit and just stated “you did that wrong”? I’d probably be ignored. 🙂

                    2. Well there is a difference. Our commits are all peer reviewed by developers knowing the KWin source code. This makes it unlikely that a severe issue hits the source base. Than there is the thing that I am the maintainer of the application, that is I am one of the persons knowing the source code best. Luckily the twin tree is hardly different to the kwin tree, so I do know the source code.

                      One of my tasks as a maintainer is to peer review changes. I do that every day. Believe it or not, you become very good in just reading a diff and seeing what’s wrong with it. Now if I as the application maintainer tell you the commit is wrong, it is most likely true. Remember that I know that source code better than you, that I know what side effects a commit may trigger. Also think about that I only develop on KWin and not on the complete Qt and KDE source base. I would myself consider as an expert of the KWin source code. Can you say the same about twin? If yes, I will do a twin quiz with you 🙂

                    3. OK, but you still haven’t told me what I supposedly broke in twin. 🙂

                      If you would like to move this discussion to the trinity-devel mailing list, under a title such as “twin repairs”, would be more than happy to discuss the twin internals and the reasoning behind the changed code. If you can clearly state in terms of established WM standards what has been done wrong I will correct it ASAP. Technical debate is a good thing, unsubstantiated finger pointing is not.

                    4. Have you ever thought about that maybe your TQt stuff may introduce very severe issues? It’s just a hint where to look to.

            2. Seems to me you can’t even build kdelibs without akonadi, which in turn has nepomuk in dependencies. At least that was the case last time I looked, round 4.4 or 4.5. Am I wrong?

              1. You are quite wrong. Akonadi has never been a dependency of kdelibs. Nepomuk has never been a dependency of Akonadi. Nepomuk has always been only an optional build dependency of kdelibs. As I mentioned before: on my CI system neither Nepomuk nor Akonadi are available to build kde-workspace (and of course kdelibs which is a dependency of workspace).

      2. Plasma is mature and highly flexable, you can get plasma to work / look like Win 7, MacOSX and Gnome 2 / Cinnemon and even Unity one can fork plasma desktop remove the cashew and replace it with a configuration dialoge. With a bit of tweaks it will work and act much like KDE3 and it could be included in with KDE so all one would need to do is change the plasma shell to plasma-trinity. The project could focus on making a modern desktop with a classic look and feel without any huge amount of effort, you won’t need to maintian Qt, Kdelibs or plasma and you get acess to the configuability and power of plasma as well as a modern systray, dbus-menu, notifications and all the random plasmoids that you could imbed in the pannel. On launch it could disable strigi, nepomuk and akonadi. You could dissable krunner but you would loose a lot. Kwin4 is awesome it can (with compositing effects) almost be a functional shell initself, all that is lacking is a clock and a systray.(providing it is paired with krunner). KDE3 was awesome but plasma has made the whole desktop framework flexable that it with can be respun to a netbook interface and a tablet / phone interface without a fork. Why couldn’t you get an awesome KDE3 interface using plasma? And nepomuk is starting to get awesome, with the new music and TV kio that updates your media files with data from the net making searches by actors and other cool things possible. It is getting huge performance fixes and interesting work on reconmendations with plasma-active. But that is a side issue. Trinity should make a new plasma desktop shell and remerge with KDE proper. KDE4 apps have gotten even better, Kate, Konsole even the games there is little reasons I would want to go back to KDE3 at this point. A new plasma KDE3 inspired shell would be awesome and easier to develop and maintain in the long run. To me it seems like a no brainer.

        PS Knetwalk FTW.

        1. +1 – great summary of how I would love to see Trinity evolve (and would have hoped for that to happen years ago before forking).

  27. Dear Martin… while, at least after reading the comments, I see where you are coming from and what you want to achieve with this posting I must point out that you are “off target”.

    I run the latest KDE4 version myself and intend to do so. For my family I use the version the hardware works with (that depends on the underlying distro obviously) and leave it at that. The kids wants more games than anything, my wife wants working email (Thunderbird, hate to admit but it is easier for a daily use than KMail) and social media web + photo and music.

    Professionally I DO use trinity on some older desktops and I WILL recommend it. I am fully aware it is a one-man show (keep up the good work) and the dangers in that but it is a fully functional easy-on-the-hardware desktop environment.

    It is so to speak his problem if he does not want to talk to you or if you miscommunicate. Leave it at that, you just come out as a “hater” in this case yourself. He is not going to migrate to KDE4 and personally I see no problem with that.

    Very respectfully, as I admire the work you do in KDE,
    Sinclair

  28. You know why I abhor KDE4? That’s because you as developers decided what’s best for us, your end users, and you literally shove technologies which I do not want to run or see running.

    You know why Windows 7 is almost universally praised by hardcore Windows Vista haters? Because Windows 7 was made responsive, fast and crash free and in Windows 7 I can disable data indexer, I can disable almost each and every “innovation”, except the reworked start menu (which is actually a lot easier to use than the one in Windows XP, since it has a built in search functionality) and some Explorer inconsistencies (like the missing “Up” button which returned in Windows 8 – yes, Microsoft “usability” designers should be fired).

    When I use KDE4 I have to bear with things which are mandatory to run, yet they are are so heavily intertwined you cannot disable them without making KDE4 unusable. Yes, I’m talking about akonadi, nepomuk and strigi.

    I do NOT want a semantic DE. I do not want to run a MySQL instance on my desktop all the time, even if it’s not required by any running applications. I don’t want file tags, file search or anything else, since all my files are perfectly organized.

    I want a DE which loads fast, uses minimum resources and allows sane and comprehensible out-of-the-box customization. E.g. KDE4’s “kicker” is still a fugly unfinished equivalent of its KDE3 counterpart. I still don’t know how to make KDE4 kicker fully transparent without using semi-broken themes, in fact I don’t want to use any themes, just tell me where the hell this option is? There isn’t one? WTF?? Almost 5 years in development and I should mess with some themes in order to get something that was in plain sight in KDE3?

    Why doesn’t KDE4 runner allow to compute expressions without entering “=” at the end of them? Why?! Hundreds of people ask for this feature, yet there aren’t even sane reasons or reimplementing it.

    I want KSensors and KNetStats for KDE4. Where are they? No, don’t tell me about plasmoids which don’t monitor voltages and fan speeds. Don’t tell me about KNemo, I just like KNetStats more.

    Where’s the long promised influx of new applications in KDE4 after you changed KDElibs license to LGPL? I fail to see them. kde-apps.com is so dead. Independent developers were a lot more active in … KDE3 days than now. The writing on the wall is that even developers shun shiny bloated KDE4.

    I remember in KDE3 days you actually cared about optimizations and code clean-up. Nowadays the only application in KDE4 that receives such attention is KWin. All other applications get slower and more and more bloated every release.

    I hope I will jump back on the KDE5 bandwagon once it gets released. But I don’t hold my breath. KDE4 developers are now obsessed with tablets and JavaScript everywhere.

    1. I’m sorry but most of your “facts” are just incorrect. You should recheck each of them before you expect an answer.

      1. Would you care to elaborate?

        ““facts” are just incorrect” sounds too meagre and shallow.

          1. Even if everything said in the initial comment was wrong, don’t you think the “I don’t have the time or any interest”-attitude is just causing the misinformation of users? You, Martin, are actually the only developer I know of* that is providing nice insights for users. I really appreciate that, but I think you fail at end-user PR like the rest of the KDE team. This is not the first time you try to end a discussion with statements recommending “If you don’t get it, go away and never come back.”

            Gosh, it took me ’til KDE SC 4.8 to get everything compiled without Akonadi, Nepomuk, Strigi or any SQL backend. And it’s not that I’m a complete idiot or haven’t tried before. Yet, I still cannot get rid of Plasmoids I don’t need, specific wallpapers or technologies like the Aurorae theme engine, kactivities, kwallet or libplasmaclock. Meanwhile other compulsory dependencies are introduced like PolKit. I really hope that KDE’s package modularization I read about is helping with this problem.

            * ok, there’s Aaron Seigo’ blog, but for me his posts are rather written in the “I tell you a bed-time story”-style and thus tldr. VizZzion’s blog is not updated regularly enough and Thomas McGuire’s one is simply dead although he as the KDEPIM guy should put a lot of effort into spreading information.

            1. fair enough. But keep in mind that I’m not a PR guy but a developer. I don’t always have the time to explain everything in detail and that is really difficult for someone using to have a very deep understanding of the technical aspects he is writing about.

              Sure it’s not best style to just block the questions, but I have to. My blog is not the right place to discuss such questions. I wrote that the facts are wrong. For most of them google will give the right answer, for everything else there is forum.kde.org which is much better suited to resolve misinformation than comments in my blog.

              Concerning your wish to remove technologies: not all technologies can simply be removed. E.g. Aurorae is bundled with KWin because we think it is an additional advantage to the user. Well it’s not used by default and just installs one library < 200 kB. So I don't see an issue with it. Of course you can modify the CMakeLists.txt to not install it. Same of course with many other of your listed technologies. You have to understand that we cannot make every user's dream of how KDE should be become true. We have to have a sane default. And that's what we do. We provide sane defaults and additionally distributions can customize everything. If a user needs more he can change the source.

        1. Well I guess the easiest to point out wrong “fact” is
          “…after you changed KDElibs license to LGPL…”

          KDE libs have always been licenced under LGPL or BSD style licences.

          One weird but probably accurate point is “I do not want to run a MySQL instance on my desktop all the time, even if it’s not required by any running applications.”

          The question however is, why to you run a MySQL instance if you don’t want to and it is not used by anything.
          And looks like a strange thing to place in a comment about why you do not like software from the KDE 4 series, since it obviously has nothing to do with that

          1. I assume the MySQL stuff comes from the wrong assumption that everything requires Akonadi and because of that MySQL is started for Akonadi. Which is to a certain degree even the case as the digital clock setting uses Akonadi to query for events which starts MySQL. But this is a usage and so Plasma uses it. If a user does not want it, it’s just one checkbox away and Plasma won’t start Akonadi and by that also no MySQL instance.

            1. There’s patches and work arounds to get around it out there as well. I believe my first attempt succeeded in 4.6. Any informed package can find them. I wrote an article on it on my block. MySQL takes 50+ Meg of Mem. Not cool, it’s not that I don’t like databases just not ones that take 50+ Meg of Ram. I guess you can call me uneducated, or just assume I think desktop environments should be efficient and MySQL for desktop use is not a efficient choice (really option).

              1. “Not cool, it’s not that I don’t like databases just not ones that take 50+ Meg of Ram.”

                All software components have the advantages and disadvantages. SQLite’s disadvantage is that is less good with concurrent access (threads accessing the database). MySQL/embedded had, at least back then when a default backend had to be chosen, no support for transactions yet (InnoDB backend not availble).

                In any case it would be interesting to see how one measures the memory need for a component that is not in its own process. Which kind of memory analyzing technology are you using to measure e.g. QSLite’s memory usage when operating on an equivalent dataset MySQL was working on when it used 50+ MB?

          2. In Windows I see lots of programs running their own Jet DB instances (e.g. Windows Media Player, Windows Live Mail), and even different releases of them (with their own bugs), and nobody complains…

          3. Always?

            From Wikipedia: On January 14, 2009, Qt version 4.5 added another option, the LGPL,[82] which should make Qt even more attractive for non-GPL open source projects and for closed applications.[83]

            Until this day you couldn’t legally create anything using kdelibs without releasing the sources of your application (since Qt was using dual GPL/proprietary license before this date).

            1. I was just pointing out that your sentence about kdelibs licence change made no sense due to kdelibs not having changed its licence.

              Your new statement is not accurate either as most Trolltech customers have not released their code even before Nokia added LGPL as another licence option.

    2. “Why doesn’t KDE4 runner allow to compute expressions without entering “=” at the end of them? Why?! Hundreds of people ask for this feature, yet there aren’t even sane reasons or reimplementing it.”

      You mean that thing you get when pressing ALT+F2? Evaluates expressions when you prepend an =. No suffix = needed (at least since 4.6, which was the first time I discovered that functionality, IIRC).

      1. Nay, it doesn’t work in Arch either, so you are … wrong or lying:

        2+2 (Press Enter) – nothing happens (I’m now running KDE 4.8.0 in Arch)

        In KDE3 it evaluated an expression you entered.

        1. He’s neither, you got to read closer:
          “Evaluates expressions when you prepend an =. No suffix = needed”

          ie. he says you can use a prefix operator, no claim of no operator present, more like a misunderstanding 😛

          The entire wish in that bug report is btw. flawed because the pre/suffix serves as disambiguation (what is probably even more required due to the expanded functionality of krunner, see Aaron’s comment #1) – “5+3” is a valid command, “=5+3” is not (must be escaped to “\=5+3”, at least in BASH) so w/o that pre/suffix krunner would shadow executables (or trigger unwanted executables if such binary really exists, leaving aside this obviously extreme case of a command name, bc is pretty powerful)

          If you really, really, reaaaallly would worry about the 2nd level “=” key, you’d just map it to caps lock (the obvious wildcard key on every board)

          Or you go for a distinct calculator like speedcrunch.

          1. Fact N1: It worked in KDE3 beautifully.

            Fact N2: It doesn’t work in KDE4.

            Fact N3: Now you try to make up arguments why it cannot/shouldn’t/not allowed to work in KDE4 even though technically it can work without any problems (even Aaron said it’s perfectly possible if I use “” as a separator).

            Now tell me in what ways KDE4 is getting more user friendly by making simple things complicated and imposing limitations which didn’t exist back in KDE3 days.

            People amuse me.

            1. Fact #0: You wrongfully claimed “Tom” to be wrong or lying – i stepped in to correct that.

              Fact #1-#2: KRunner and the runner of KDE3 are not of the same scope. Aside the online filtering you’ve also random plugin extensions to krunner.

              Fact #3: I pointed a collision inherent to that wish. The fact that it does not have to happen does not render it invalid. (Nukes don’t have to explode, yet they’re dangerous)
              So please don’t imply i’d “make up arguments”

              Disclaimer: I’m not involved in krunner development or anything, I don’t even use it.

  29. I just want to note on the claim that Trinity is a one-man project. While it is true that Pearson is the primary and mostly alone contributor to the Trinity repository, at the same time there are multiple KDE3 maintenance projects virtually in every distribution. All of them make their patches. There are patches circulating from Chakra (arch), from Trinity, from openSUSE, from kde-sunset (gentoo), from Mandrive International Backports, from Alt Linux, from the KDE vanilla branch, from KDE enterprise branch, from Red Hat etc.

    The share of Trinity project and Timothy Pearson in particular in creating all those patches is quite minor. The main defining feature of Trinity is the use of tqtinterface layer instead of calling qt3 directly.

    That said the majority of user-visible changes in Trinity and other KDE3 projects are the patched developed by diverse groups, contributed by many people, far from being a one man show.

    1. I’m a Chakra user, and I’ve never seen a modern patchset for KDE3. The KDEmod patch set you may be talking about is as outdated as KDE 3.5.10 itself.

      1. There is possibly no single project that incorporates all post-3.5.10 patches. This is a consequence of the closure of the KDE3 branch in KDE SVN. Instead of contributing into one centralized place tens of contributors and maintainers started to contribute to their own projects. That said the number of KDE3 contributors still remains high but they are scattered across projects and it is quite complicated task to find that or another patch sometimes.

    2. Sorry, no. Chakra is a full KDE4 experience which originated from the *same team* who was developing KDEMod 3 and 4, which btw existed and finished way before Trinity was made. Get your facts straight.

      1. I am talking about Chakra-kde3. Also a lot of patches originates with Pardus Corporate team. KDE3 developmen did not stop. It just got complicated and decentralyzed.

  30. I think this is one of the Trinity TWin patches (originating from the Chakra project) that are disputed: http://paste.kde.org/408152/

    It breaks the KDE3’s own KWin windows shadow functionality and introduces a new one. The control dialog for setting up the normal shadow functionality does not get hidden, but just ceases to function (this is one of the problems with this patch).

    Other problems are related to lesser features of the new functionaliry compared to the original one. For example, in the original shadows you can set whether to show the shadow when moving the window. The newly introduced shadows just disappear when you are moving a window. No possibility to display them in motion.

    openSUSE’s KDE3 does not include such patches.

  31. About a classic desktop experience, I can understand that. In my case, I even filed a bug requesting that the KDE1 windeco be ported to Kwin 4, but it was not possible (lack of manpower, which I understand). The customizability is there, but there are much less alternative iconsets, styles and Kwin native decorations than ever, and with that, you can’t get a classic-looking desktop. However, even in that case I’m still more than happy with KDE 4.

    Solution: I think KDE 4 needs a new kde-look drive 🙂

    1. concerning win deco: it is now possible to design decorations without writing code. So it should be possible to design a KDE 1 styled decoration fairly easy.

        1. “To a certain degree” – it had an IceWM engine.
          KDE4 had (has) the dekorator engine, it’s just not been in the “SC”, but every distro shipped it. There’s also later been smaragd (beryl/emerald theme interpreter) and Aurorae (despite Martin wrote it twice or thrice 😉

  32. Maybe, but I’m stuck. I’m neither a coder nor a designer 🙁 However, the lack of variety in themes should be a matter of slight concern, IMHO.

    1. To me it is no concern, but a very good sign. It means that most users are very well suited with our default offerings and that makes me quite happy

  33. Martin, don’t waste your time and energy with Timothy Pearson and the Trinity project.
    I’ve read interviews with him where he simply lies about KDE and Plasma. Eg. there was an interview during the 4.2 timeframe where he claimed that Plasma Desktop is incapable of adding icons to the desktop.

    1. Thanks for the advice, but it also shows how important my blog post is to point out the problems with the Trinity development efforts.

  34. Martin,
    First of all, congratulations for you work and it’s great to have this kind of feedback with a developer of a team like KDE.
    About your article and the thread discussed here… Open source is freedom of choice!
    Timoty has the right and the freedom to patch KDE 3.5 and if he feels happy with that, well…. great! it’s just his choice and the choice of many users.
    I see you had good reasons and good intentions to help them (Trinity project) and I understand you are worried about the few human resources wasted in many projects but it should be great to recruit new people/devs instead (I’m not a developer, I would help if I could!)
    I returned to KDE 4 series when 4.7 (Chakra) version was released and I saw there was a huge improvement. Also QT/KDE ecosystem is great.

    Greetings

    Agustín

    ps: sorry if I write in a weird English. As you, this is not my native language

      1. After having been involved in a very long discussion on the trinity mailing list yesterday, I don’t think that he lies consciously. To phrase it positive I would say he lives in his own small dream world with a very restricted world view. If something does not match the world view, the world is turned upside down often enough that it fits the world view again. His world view involves Qt 4 being bad, KDE 4 being bad and if something is in fact better than in Qt3/KDE3 the view of the world is changed. Actually I feel pity for him.

        1. If you are going to claim libel then you may want to go after all the other bloggers, reporters, etc. that didn’t like your precious KDE 4.x either. Oh, and calling someone delusional is usually libel as well.

          Grow up and learn to live in a diversified world.

          1. I did not claim libel and I did not call you delusional. Both things I cannot read in my last comment. I don’t want to talk about what I recognized while reading your threads as my English is for that not good enough and I think it touches a very private issue of yourself.

            It has btw nothing to do with that you don’t like KDE 4.x (which does not exists) but with your general view of the world. Just read the threads again and watch explicitly for the things Aaron stated. View everything in a broader context, try to read it from someone not involved.

            Actually I have thought about writing another blog post with the results I have learned. But I am very unsure as that feels to me like I would beat someone who is already lying on the floor. On the other hand I consider it is as quite important to have realized that your hatred against KDE has nothing to do with KDE that the root of the problem is much deeper. It is a very sad story.

  35. I don’t hate KDE4 on some kind of emotional level. I have found comments from the KDE4 team about legitimate criticisms to be disdainful, though, which doesn’t exactly make me feel confident that the the developers have anything like the user’s interests in mind. Instead, they seem to have latched onto some kind of ideal of how an interface should be and are demanding that everyone adapt to it.

    I’ve actually used KDE4 quite a lot, and in the end, it simply does a poor job of meeting my needs. It gets in my way, in is not intuitive, and it makes me chagne my established work patterns without giving any real benefit in exchange for doing so.

    Yes, there are ways to mitigate many (but far from all) of the worst aspects of it — but even these ways are difficult to unearth and are not obvious. Why did I have to do a google search in order to figure out how to get shortcuts & documents back onto my desktop where I need them? Why can’t I get rid of that cashew? I have no need for it whatsoever (since activities have proved useless for me), and I keep tripping over it. And so on…

    The KDE team has made it very clear the direction they want to take the UI, and that they aren’t interested in addressing the needs of people they believe are too stupid or lazy to conform to their way. I have not left KDE, KDE has left me.

    I do not hate the new KDE. Instead, I am saddened by it.

    1. I find such comments sad. First of all it has nothing to do with what I blogged about – no matter how bad some people thing KDE (4) is, it is no legimation for the behavior and problems with the Trinity development I pointed out. Second of all, it makes me sad that people talk about a general “KDE 4” – which just does not exist. My interpretation is that you dislike some points of the Plasma Workspaces. And that’s another point I criticised: most disappointed users would benefit from a port of kdesktop/kicker.

      Last but not least I am sad, because you have an expectation that the software has to be tailored to your needs. If it does not do what you need by default, you are not even willing to read our documentation on how to use it. That is really sad.

      1. Actually, I have read the documentation as well as lots of searching the web trying to figure out how to make KDE 4 work in a way that is useful to me. But I shouldn’t have to do any of that. A DE should be intuitive enough to be essentially invisible. it should be possible to figure out how to accomplish things just by poking around for a few minutes. With KDE 4, it is not.

        What makes me sad is that KDE 4 has left me behind. It does not meet my needs (or at least, I cannot figure out how to make it meet my needs, even after reading the docs). It is clumsy (for me). This isn’t so much an issue of the defaults not being what I want — I’ve been very active in the Linux world from before KDE even existed, and have been using KDE since 2. I’m very capable and willing to adjust things to my liking.

        My problem is that I cannot do this with KDE 4, at least not to the extent that I prefer. KDE 4 gets in my way and I can’t figure out how to fix that.

        That’s what makes me sad.

        It may very well be that this change to KDE gets it a larger audience, and if so then kudos! I’m glad that it works for so many people. I just wish that I could enjoy it as well.

      2. Oh, I forgot one important piece of my last reply: the connection to your post is that you listed three groups of people who are attracted to Trinity. I was just pointing out that there are people who don’t fit into those groups.

        1. given your comments you clearly fall in the category of those users wanting a “classic” desktop. You are unhappy with the Plasma shell and would like that nothing changed at all. That’s just fine, but it is exactly one of the three categories I named.

  36. I am running OpenSuse 11.4 with KDE 3.5.10. The last version
    of KDE4 that I tried was 4.6. It had several problems:

    1. The names of the virtual-desktops (I use 20 of them) are
    not readable in the buttons of the virtual-desktop pager
    applet, because each button has miniature versions of the
    window-outlines of all the windows in that desktop. These
    outlines obscure the name-text and are not removable. Also the
    name is written in a very thin font which is hard to see.

    Activities does not have a small 1-click pager, so it’s not
    a useful substitute for virtual-desktops.

    2. KDE 4.6 insists on the Phonon audio system, which does not
    work with my ASUS M2N-E motherboard audio chip(s). The System
    Settings 1.0 program assumes Phonon, and its GUI for setting
    audio options doesn’t even work — I can’t move the sliders
    or buttons or whatever they are.

    3. There’s no way to manually retract the panels.

    KDE 3.5.10 doesn’t have these problems, and is fast and does
    everything I need.

    1. The desktop names are easily visible in the pager buttons.

    2. The ALSA sound system works great with my motherboard
    audio chips.

    3. There’s a little button at the end of each panel that I
    can click on to retract the panel.

    I also prefer the simpler look and feel of KDE3. I never
    use compositing. The “cashew”, and the methods for adjusting
    the panels and their contents, are annoying to me. I think
    that KDE3/Trinity follows important Unix principles —
    as simple as it can be and still do the job well, and keeps
    its configuration and other data in human-readable text format.

    However, I am using the KDE 4.6 versions of konsole,
    konversation, okular, kmix, klipper, and System Monitor,
    since they are improved over the KDE3 versions. I use KDE
    System Settings to set the gamma of my display.

    I’m very appreciative of the efforts of the Trinity people,
    since they give me assurance that KDE3 functionality will
    continue to be available in the future.

    When I asked about the first problem nearly a year ago in the
    KDE IRC channel I was told that I had eye problems that were
    preventing me from reading the desktop names in the pager
    buttons, and that nobody else had this problem. KDE4
    enthusiasts elsewhere told me that I should get a newer
    motherboard that would work with Phonon, and that the KDE4
    developers were intentionally making virtual-desktops difficult
    to use because they wanted everyone to use activities instead,
    yet he could not tell me how to switch between 20 activities as
    easily as I can switch between 20 desktops in KDE3.

    My question: Have the three problems I’ve described above been
    fixed in KDE4.8?

    1. I’m sorry but my blog is not a bugtracker nor a user support forum. It is not the right place to ask the questions. But let’s have a look at it:
      1. I don’t use the pager – no idea
      2. Phonon uses Alsa underneath – your description does not make any sense to me
      3. I don’t get it – I can resize the panel if that is what you mean.

      I would suggest to ask questions at forum.kde.org. I am pretty sure that people can help you there without telling you to buy a new mainboard. That’s just a stupid answer.

      And an important thing to know: Trinity will not be able to provide KDE 3 functionality in the future. That is exactly what my post is about.

      1. I don’t understand why you even try to respond to them. Let them think what they think. You have a very big user-base that just Loves KWin, you don’t need to convince some 3-10 haters that KDE 4 actually can do all the things they need.

          1. which does not say much given that the KDE3 repo has been around for quite some time and the KDE 4.8 repo just for a month.

              1. yes sure, but see that the KDE 3 repo has been around longer, so the chances that users included it are higher than for the 4.8 repo, where it is also quite likely that most users are very happy with 4.7 and don’t include the repo. At work only two of my group use the repo, everyone else uses the standard repositories.

                1. The KDE:Release:47 is also in the list. KDE:Release:47 and KDE:Release:48 combined do not exceed KDE3 repository. And KDE3 users of course also can install all they need from standard repositories, KDE:KDE3 is needed only for those who wants updates faster than from official repo.

            1. These numbers are the base for the OpenSUSE Build Service scheduler priorities which decides which packages are more popular so to build them first. They are updated on a regular basis (I believe once a week).

  37. I didn’t read all through, but it seems to me that only because there is something new the older one is unnessessary or fals or bad.
    And “some clicks” to have a KDE4 behave like KDE3 is a lie , to name it.
    I recomment the author to give that argument to a 65 year old women (AKA mother), which has already a hard time learning computers in general and insist that KDE4 and KDE3 aren’t much diffrent

  38. Martin, stop arguing with those from Trinity, you know that they will respond like little kids to anything you say. Just ignore, it’s really easy, just let them think what they thing, it’s important that you thing the other way and there are like hunderds of thousands of persons who agree with you compared to this haters which are supported by ~3-10 people. Which we know for sure that they are just mind sick, or hundreds of thousands of people are 🙂

  39. I’m coming to this debate very late, but I thought I’d share my reactions. I don’t use Trinity, but I completely understand the desire to resurrect the old KDE. It worked, and it worked really well. Whether it’s the bugs (and I stumble over more KDE 4 bugs than I’d meet in KDE 3 on a day-to-day basis), or just that everything is moved around, there does seem to be a large group of people who don’t like the new KDE as much as the old KDE.

    This shouldn’t be taken personally — I’m guessing many of the same people take credit for both versions — but KDE 4 might benefit from spending more time figuring out why it has so many haters. I’m not one of them, but I am in the ranks of people who feel that progress hasn’t made them happier. And I’d love to think that KDE 4 made me enjoy using my computer more than KDE 3 did.

    1. The number of haters is rather low and the usergroup for Trinity is also rather low. Just look around, look at the various recent awards where KDE Plasma is considered to be the best desktop environment. If you look at those competitions you also see that the Trinity user group is less than 1 % of the Linux user base.

  40. “Emotions – especially hate – are very bad to base decisions on”

    Emotions are likewise a poor basis from which to launch an analysis of the entire situation. If a product garners a large group if dissatisfied followers, the problem is the product, not the followers. It’s really very simple. Failing to get the message simply makes the developers look inept.

    You seem to have given in to the temptation of presumimg that the KDE folks know “what’s best” for KDE users. Perhaps if more attention was directed at improving what had obviously created a highly devoted following instead of attempting to create the next big “thing”, the division never would have happened. Apparently, the lust to be recognized as the next Steve Jobs with a new “innovative” GUI proved to be more than some in the KDE movement were able to resist. No one “hates” KDE technology, they are simply dissatisfied with what KDE did with it.

    KDE4 is an amalgamation of disparate agendas lacking any useful direction or practicality. While it certainly looks good, it simply doesn’t serve any real practical purpose. KDE4 has lost that “just works” element that linux users pride theselves so much in.

    But, lets not focus on that, I’m sure if we can just squelch all the “haters”, it will start working great. 😉

    1. I think you clearly missed an important piece of information: Trinity got started when KDE had sorted out most issues with 4.x. So no, your analysis doesn’t hold. Also thinking that on my side emotions have been involved is quite wrong. Heck what should I care about it – I joined KDE development with the 4.0 release, I have never contributed to 3.x.

  41. I’ve stuck with KDE 4.x since kubuntu switched to it. Unfortunately, 4.8 still doesn’t work.correctly and I’m switching to trinity. The two biggest bugs are background image for konsole (somewhat finally fixed in 4.8 but still lacking shading and transparency that was in 3.5) and now panel submenus are broken (They open up dolphin instead of behaving like a menu and they worked correctly until 4.7).

    Trinity is fulfilling the need to have a working desktop environment while transitioning the code to more modern technology. All this talk of hate, sounds more like someone is projecting his own hate of forks onto the Trinity developers. All the hand wringing around KDE4/Trinity makes we think the KDE4 folks know they’re producing a substandard product.

    Forks are a good thing. That’s the whole point of open source. If you don’t like the direction a project is going, make a fork that works the way you want. It maximizes everyone’s happiness. We all get what we want. People who like KDE 4 get KDE 4 while people who want something that works get Trinity. Get over it and move on.

    1. I’m sorry that you are lacking features of Konsole this is an application not developed by the Plasma developers and I’m also sorry that there is a regression in Plasma 4.8.

      But I have to inform you that Trinity is not “tranitioning the code to more modern technology”. That is what KDE already did in 2008 and is currently doing again by embracing QML.

      I want to also point out that I have not contributed a single line of code to KDE 3.5, so I am quite certain that I am not “projecting my own hate of forks onto the Trinity developers”. That’s just ridiculous, as I pointed out in my mail the hate is from Trinity developers onto KDE.

      Oh and forks are not a good thing but a very bad one. Instead of working together let’s fork, so that nobody has enough resources to deliver a great product and so that we are fighting each other instead of combining our strength to have a competitive product. Yeah great idea those forks.

  42. You said it a few times: KDE3 is mostly for people with old computers, e.g. PIII with around 1 GHz and 256 to 512MB RAM. KDE4 does not work well on such old machines.
    But LXDE is no alternative. I assume that you did not use it yourself up to now. What about the LXDE development in the recent months and the last year? So you compared outdated software with outdated software.
    The most important point is that Trinity should not change the KDE3 code of software they do not understand, KDE3-KWin comes to mind. This is very dangerous. In this regard the best tip is indeed your advice above, i.e. to use KDE 3.5.10 with any version of openSUSE from 11.1 to 12.1 or to use RHEL5/SL5/CentOS5 with still supported KDE 3.5.4. This means that KDE3 is actively supported by Red Hat until 2017 with reference to security and a few bugfix updates. This should be enough, old software versions cannot be supported eternally.

    1. I have to admit that I have not used LXDE, but LXDE has a focus on “lightweight” while KDE never had. That’s the difference I wanted to point out.

Comments are closed.