Quo Vadis, Kubuntu?
October 31st, 2007Events of the past week, before and after Gutsy’s release, have led me to think deeply about Kubuntu. No, I haven’t been thinking about what new features to implement (not that I could anyway), what bugs to fix (I’ll try, of course) or what Ubuntu features we need to catch up with. No, I was thinking much deeper, less technical, more abstract, more long term. Not that those goals are unimportant. But they will only be important depending on who/what we are and what we have planned.
Some disclaimers. Firstly, my views are my own and do not represent the official view of Canonical, Ubuntu, or Kubuntu. If they reflect the views and thoughts of other developers and members of the community, that would be a coincidence. Secondly I have been with the community for almost two years now, but my area of contribution has largely been focused on (but not limited to) the community aspects of Kubuntu. And lastly, my technical know-how of the various process involved during the whole development cycle is a bit limited. So I’m not 100% sure if my views and suggestions will be technically possible and feasible.
What happened?
Ubuntu, no doubt, is a great distro. It’s popularity has helped katapult Kubuntu into the limelight as well. Kubuntu definitely has merits of its own, but Ubuntu’s popularity has helped a lot in marketing Kubuntu. But that popularity is a two-edged sword. Because of that popularity, Kubuntu has also been put under a microscope, compared and contrasted with Ubuntu. But while Ubuntu has grown by leaps and bounds, Kubuntu has somewhat been behind. This has been the case in the past, but probably never so obvious until Gutsy. During Gutsy, I’ve seen people compare the two, feature by feature. They noticed the discrepancy, and they ask why.
Canonical, Ubuntu, and Kubuntu
The answer is simple but true. We lack people. We always could use more hands. Kubuntu is, in principle and in practice, a community-driven distribution. This brings into light Kubuntu’s relationship with Canonical and Ubuntu, as well as a big difference between Kubuntu and Ubuntu. Kubuntu has quite a unique standing with Canonical, a bit different from other Ubuntu derivatives like Xubuntu or Edubuntu. Kubuntu is “officially supported”. We also have a Shipit option. And most of all, we are very fortunate to have Jonathan Riddell, the big man behind Kubuntu, to be employed by Canonical. But at the same time, unlike Ubuntu, Kubuntu is an almost 100% community-driven distribution. In this we are a lot similar with Xubuntu, Edubuntu, and other Ubuntu-derivatives. This is one big difference Kubuntu has with Ubuntu. But does it matter? Or am I just sour-graping that Kubuntu doesn’t get some preferential treatment? Given some “promises” made a few years back about Kubuntu and KDE, maybe I should be. But that’s beside the point. I love community distributions/projects. It means that the community is in control. No ulterior motives, no hidden agendas. But when a community project has to contend, be compared, and in a way, compete, with an older, more supported sibling, then the comparison falls short of being fair. When the community distro’s identity and goals are defined by how much we have to catch up to the features of a more developed base distro, then there’s probably something amiss.
What is Kubuntu?
In my most humble opinion, it all boils down to two questions: what is Kubuntu and what it wants to be. The common conception about Kubuntu right now and the goals we bothers me a bit. Based on what I’ve observed, the common conception is that Kubuntu is just Ubuntu with KDE (and without GNOME). In some ways, this is correct. Kubuntu does make use of Ubuntu as its base, with KDE as the desktop interface. There would be no problem with this definition if it were true. But is it? Is Kubuntu simply just Ubuntu with KDE slapped on top? If it were true, then the Kubuntu team is really nothing more than just a subset of the wider Ubuntu developer community, just like the KDE teams of Debian or Fedora. And the Kubuntu team is responsible for maintaining all KDE packages in the Ubuntu repository after all. But then, what would Kubuntu’s official support from Canonical mean? And what does it mean for our users? Well, based on user feedback in Gutsy, it means that users expect the same amount and kind features that are in Ubuntu to be in Kubuntu as well. They begin to expect that, since Kubuntu is just Ubuntu with KDE, it will have the same features as Ubuntu, but with a KDE face (which sort of also poses a problem for KDE a bit). And perhaps they expect that, since Kubuntu is officially supported, it should have all these things too. And boy, were they very disappointed. This (mis)conception, these expectations, are now also reflected in where Kubuntu is heading. Right now, we have a lot of catching up to do with Ubuntu. But why does it have to? Why have we started to define ourselves, define our features, define our roadmap, based on what Ubuntu has? If Kubuntu just uses Ubuntu for its base system, why do we need to catch up with their non-base features as well?
My very personal answer
Again, I’m just one out of a number of more experienced developers, contributors, and users, so my own definition may not reflect the official definition, if there is one. But still I will give my unsolicited opinion, in the hopes of probably starting some reflective thinking and some discussions.
The first step to a definition of Kubuntu is to accept that we are not Ubuntu. We will never be like Ubuntu, and we might probably never be able to stand up to Ubuntu. Why? Simply because we are a community project. Ubuntu isn’t, at least not in the same way. If we were to set out and match Ubuntu feature by feature, we would forever be catching up with them. The amount of developers we have alone makes it almost impossible to work on the same features simultaneously, side by side, with Ubuntu.
Kubuntu is not Ubuntu. We have to make this clear, to ourselves, and to our community. A large part of user expectation is largely based on what we have been doing in the past: trying to catch up with Ubuntu. And users have now grown to expect that. But why do we need to define ourselves and our goals based on Ubuntu? Can Kubuntu not stand on its own merits?
But does this mean we ditch those features and go on our merry way? Of course not! Most of those features are good, and some are necessary for an enjoyable user experience. But to acquire those features for the sake of matching up to Ubuntu, or to force a feature or app even if it’s not that ready yet just to catch up, that becomes a problem.
Does this mean we “divorce” Kubuntu from Ubuntu? Hardly. We would still be part of the Ubuntu project, use the infrastructure and support that it affords. But I’m sure we are quite free to define our own roadmap, our own identity, that is not Ubuntu.
Once we have accepted that fact, we can start to define Kubuntu as it really is. And then we can decide where we are headed, what we really want Kubuntu to be. This is where I may go wrong. But for me, Kubuntu is more than just “Ubuntu with KDE”. Kubuntu is not a KDE-fied Ubuntu, nor an Ubuntufied KDE (either way hurts KDE in the process as well). I think the About Kubuntu pages here and here probably comes closest to what I’m thinking. Perhaps I’ll blog about my own Kubuntu definition… Maybe…
The three-fold road
Anyway, I see three possible paths for Kubuntu right now.
1. The easiest option. There is actually no problem. I’m just overreacting and being melodramatic. We can just go on with the way things are. Indeed, if I have been totally wrong, I can live with that and accept the status quo. I personally don’t believe that there’s is no problem, though.
2. The ugliest option. We actually become simply a KDE team within Ubuntu, a team that will be called, “Kubuntu”. Kubuntu will become a sort of Ubuntu KDE Spin like Fedora’s. We will have less pressures. We can tie ourselves more closely to Ubuntu’s development, use Ubuntu/GNOME utilities when it matters, or just come up with KDE frontends to those utilities a few releases after.
3. The hardest option. If I have been correct and there is a problem, now’s the time to redefine ourselves, create our own roadmap, our own destiny, if possible away from Ubuntu. Again, we don’t have to stop being part of the Ubuntu project. The whole community will probably have separation issues. But if we survive through it, I believe we will have a more solid and firm distribution.
Well, I don’t know what will be the fallout from this post. I’ve been having second thoughts about posting something like this. I just hope that, IF (and that’s a big if) there really are problems like this, a fruitful discussion could be started.
Just some small updates, while I'm trying to get off my ass and migrate the site completely to Textpattern.
October 31st, 2007 at 4:44 pm
I always considered Kubuntu to be the *same* distribution as Ubuntu, featuring another Desktop Environment. That is, equivalent to your “choice 2″.
This looks like the obvious definition to me. Debian, Mandriva, Fedora and the like support different DEs, but they kept an unified distribution. That choice seems wiser to me, and your blog post confirm my opinion.
Why would you need another distribution around ? What would be “independent Kubuntu” goal, achievement, motivation ? Do you have something new to give to the community ?
At the moment i just see a devoted KDE team of a maybe-too-popular distribution. Nothing that would justify another workforce-division. Gnu/Linux distributions are working right now on some delicate issues (downstream/upstream collaboration, standards, package management, etc.), but i don’t see any problem that a new “own roadmap and destiny” would ease to solve.
If you consider forking, it’s necessary to define your roadmap *first*. Without any motivation apart from “we’re not able to catch up”, i hardly see how such a fork could be successful.
October 31st, 2007 at 5:04 pm
What about option 4: Canonical could redouble its efforts and make Kubuntu as much of a priority as Ubuntu is.
I think that they’ll have to put more effort into it when KDE4 comes out, since there’s a lot of positive buzz around it, even for the casual Linux user. I can’t count the number of times that a commenter on Digg says “I’m switching to KDE when KDE4 comes out.” How many of those users, who currently use Ubuntu, will switch over to Kubuntu? I’m thinking lots.
October 31st, 2007 at 5:05 pm
I totally agree with you! I think that actually Kubuntu is only a bad copy of Ubuntu… Think about Strigi (really ugly if compared to Tracker).
KDE4 is an opportunity to redefine the role of Kubuntu, creating its own roadmap and its own release schedule (why does Kubuntu follow the Gnome release schedule??).
October 31st, 2007 at 5:14 pm
+1 big Problem
* pushing gtk/gnome techs into a kde env/world..
* fixed release schedules
* broken kde builds
* mispatched / bad customized kde builds!
* political decisions in technical issues!
..
..
community$ rm -Rf kubuntu/*riddell*
October 31st, 2007 at 5:17 pm
There’s nothing wrong with Kubuntu being Ubuntu with KDE on top. After all, that’s why the derivative was created in the first place.
It seems to me you forget the magnitude of having KDE instead of GNOME. Having a highly configurable desktop, tight integration between applications, desktop-wide protocol support (KIOs, obviously) is what’s missing when you use Ubuntu.
The desktop experience is what sets Kubuntu apart from Ubuntu. Granted, in some areas its not up to par but that will change. When KDE 4 or 4.1 arrives, the difference from Ubuntu will become even more significant.
In other areas, it simply makes sense to participate in the Ubuntu community. Not just technology areas, but community building, marketing / getting mindshare etc.
Why change that now? Over time, Kubuntu will improve, the (KDE) difference from Ubuntu will become more clear and there is a change we will benefit more from the wider Ubuntu community. It would be great if we could convince, say, graphics designers that are involved with Ubuntu to also do their stuff for Kubuntu.
But I emphasize with the identity problem. Naming the distro Kubuntu was a boon at first, but how to grow beyond being Ubuntus little brother?
My take on it is we wait for the bigger improvements that will come with KDE4.x and Koffice. Eventually we will earn the right to more mindshare.
October 31st, 2007 at 5:24 pm
I agree with Eric’s option 4: Canonical to give KDE the attention it deserves. Here are areas where Kubuntu should catch up:
1. Bulletproof X - someone needs to implement failsafe server in KDM
2. A KDEfied version of Synaptic - Kynaptic looked so promising. Adept is more cluttered, has a less visually appealing interface and does not support automatic removal of unneeded dependencies when the package that requires them is uninstalled
3. Better Compiz fusion support would be nice but is not a biggie. After all, with KDE 4 we will eventually get a great window manager that offers compositing effects out of the box. This is one example where Kubuntu need not copy its GNOME sibling.
I can understand your concerns about Kubuntu being always the bridesmaid and never the bride, but there is hope that with the emergence of a strong KDE 4 series Canonical will sit up and take more notice. Fingers crossed
October 31st, 2007 at 5:30 pm
Has Mark Shuttleworth indicated that if kde 4 is impressive enough they would drop gnome? If that were to happen, it would be a good thing for ubuntu, kubuntu, and kde. Probably not a good thing for gnome. A good thing for linux? That’s a matter of opinion.
October 31st, 2007 at 5:43 pm
No probably not. There’s been whispering in blogs etc. that this might happen, but it’s pure speculation and could launch a major flamewar with those preferring GNOME. What I really meant was that if Canonical would see very high uptake in Kubuntu after KDE 4.0/4.1/etc, it would hopefully devote even more resources to Kubuntu, e.g. it would pay more Kubuntu developers or delay implementing features in Ubuntu until a similar solution for Kubuntu was available.
October 31st, 2007 at 5:50 pm
Actually, I want to make a few more comments. BTW, Jucato, you’ve always been very helpful to me when I had problems on #kubuntu.
Firstly, I meant to say “Hasn’t Mark Shuttleworth…”
Anyway, I think that we should supply only bug fixes and basic improvements to the kde 3 series and start focusing on getting kde 4 in shape. I think that, for hardy, we can supply a kde 3 version and an optional kde 4 desktop that can be added after installation. Is this too much for the kubuntu project to handle? After all, making kde 4 ready for kubuntu will probably be a year of work, so we have to start now if we want to have it by default in hardy+1, am I right?
By hardy+1 we should have a functional kde 4.1 environment that will completely replace kde 3. Freeing up open office with koffice will give us a lot of room on the liveCD too.
Also, you indicated that we should have a different release schedule than Ubuntu. I’m reluctant to make such a broad mutiny from our older brother, he hasn’t treated us that badly. Furthermore, the kde project is moving to 6 month release schedule*, and they cited being in line with kubuntu as one of their reasons.
*I’d personally prefer a debian like system with conservative updates trickling down over time through a 3-tier release system (stable, testing, unstable)– we should have hardcoded bugs stuck in the OS for 6 whole months, but that’s just my opinion.
October 31st, 2007 at 5:54 pm
I’m a Kubuntu user since the very first release. I’d like Kubuntu to go the “third” route and find its own identity. For example look at the website kubuntu.org. It’s ugly and unprofessional.
Kubuntu needs some serious marketers and artists but the devs seems very shy about promoting Kubuntu as a brand of its own.
October 31st, 2007 at 5:56 pm
I tried Kubuntu 7.10 and wasn’t happy. I can’t compare it with Ubuntu as I haven’t tried that since 6.06 because I use KDE as wm (and plenty of Gnome apps on top of that).
I see 2 options:
-1. better support from Canonical, Fedora with KDE is very much OK so it’s possible (again: don’t know how Fedora + Gnome is),
-2. spend your time on a KDE centric distro. I’m using Pardus atm and it’s behind *buntu in features/ packages/ etc. but far better in desktop behavior.
Kubuntu did nothing really wrong but also wasn’t really right for me. Surely Shuttleworth can afford a KDE team to give Kubuntu the love it deserves?
October 31st, 2007 at 6:00 pm
Personally, I like KDE better than Gnome. If it were not for Kubuntu, I probably wouldn’t use Ubuntu and would still be with Fedora (or swearing about all the compiles under a Gentoo variant like Sabayon).
I don’t know all the particulars of the release schedules, but it seems to me that it makes sense for Ubuntu to be “Ubuntu Core with Gnome” and Kubuntu to be “Ubuntu Core with KDE”.
Am I missing a fundamental issue that makes that inherently impossible or implausible? From what I’m reading here, the definition of Ubuntu seems to be “if it doesn’t run Gnome, it isn’t really Ubuntu”. That sounds too close to MSFT Windows to me (easy, folks… easy) — which has defined its operating system based upon its window manager. I think Ubuntu is above that.
Perhaps the evolution of KDE will be something to bring new life to “Ubuntu Core with KDE”.
October 31st, 2007 at 6:06 pm
> Has Mark Shuttleworth indicated that if kde 4 is
> impressive enough they would drop gnome? If that were to
> happen, it would be a good thing for ubuntu, kubuntu, and kde.
How could that be ? Is ubuntu’s gnome support hurting someone ?
Ubuntu is involved in the Gnome community, and I think that’s a good think. What’s hurting is that they are not as supportive to KDE. But how could you wish they “drop” Gnome ? Should the Gnome community then say “i wish they dropped KDE, that would be a good thing for us” ? Nonsense.
The lesson of that story should be “diversity matters”. Not “KDE is better than Gnome” or whatever.
Asking Canonical more support looks like a good idea (although you must be careful here : nothing is due, and Ubuntu supporting Gnome does not mean they *have to* do the same for KDE), but asking Canonical to “drop” another project and “use KDE instead” looks like a hurtingly bad one.
October 31st, 2007 at 6:53 pm
Whew! So many replies already?
1. First of all, option #2 up there is actually a sort of “slap in the face” option. Something I personally don’t want to happen, but could happen on a worst case scenario.
2. Regarding Canonical giving more support to Kubuntu and KDE (as once “promised”), see the logs of “Ask Mark” in the Ubuntu Open Week: https://wiki.ubuntu.com/MeetingLogs/openweekgutsy/AskMark
It’s pretty much a closed topic (for now maybe).
3. Kubuntu would go nowhere without Jonathan Riddell, who is our first and foremost KDE developer in the team. (He’s not a GNOME developer).
4. No where did I indicate a separation of release schedules. In fact, I like the predictable 6-month schedule and sometimes wished KDE did too. However, there are certainly some instances where I wished we could have held off releasing for a week or so. But in any case, I think the fixed schedule is in fact, an advantage.
October 31st, 2007 at 7:09 pm
What I find amazing is how *good* Kubuntu still manages to be despite next to no support from Canonical. If that’s not a testament to the power of the KDE infrastructure, I don’t know what is…
Still makes me a little sad when I think of where the Linux desktop could just be if the workforce was split more fairly between Ubuntu and Kubuntu though. :/
October 31st, 2007 at 7:58 pm
Jucato, good post and it has triggered nice discussion in comments.
I think that resources are not main problem of Kubuntu. Even though it would help substantially to have more full-time developers (particularly now with kde4), it would not help with general positioning of Kubuntu vs. Ubuntu. What we are lacking is clear message in the communication that we see going out during regular announcements. They focus on Ubuntu with no separation what is GNOME desktop specific implementation and what is not. Why? Because Ubuntu is considered to equal GNOME.
On the other hand, I think most users expect that Kubuntu is Ubuntu with KDE, very similar to OpenSUSE or Fedora KDE spin. I think this is to some extend expected: we are sharing a bit more than just ubuntu in name.
If we want to keep Kubuntu “brand”, than we should to some extend behave as one of the “spins” in Ubuntu family. This probably includes regular release cycles, sharing common infrastructure, develop most of disrto apps in DE-agnostic way and to some extend also matching Ubuntu in features that we think make sense for our users. But I doubt that we will ever be able to implement all the Ubuntu features (catchup), as our developer community is in fact much smaller.
This is why I think that proper positioning of Kubuntu is key. And it is not just about kubuntu.org web site (we need volunteer here!), but also about messages give by Ubuntu - they should be more clear on what to expect from Ubuntu/GNOME and what from Kubuntu.
For future, I think the most important thing is how we will bring kde4 to our users and this is what I personally want to focus in next months.
October 31st, 2007 at 8:04 pm
*buntus suck. Try a real distro like OpenSUSE or Fedora.
October 31st, 2007 at 8:05 pm
I don’t think Canonicals will give more attention to KDE. Why would they do that? It is not a secret that Canonicals are trying to make their Linux profitable. They managed to make their Linux distribution the most popular one. (But I think that in some technical points Ubuntu is still behind the other Linux distros like OpenSuse and Mandriva. And this popularity is only result of marketing and a lot of fuss. But that’s the different story.) So if they still can’t make profit I don’t believe that they will hire more developers. No KDE developers for sure. That is why I am absolutely positive that there is no reason to hope for Canonicals.
So what to do? My personal point of view is that for now you should accept the status quo and wait. Let me explain why. Gnome is evolving very slowly. Look at the latest release. They call a great and big feature that they have united three different utils into single one. I mean their new look and feel configuration utility. They’ve just took three different single panel applications and made one new application with 3 tabs which are equal to the old separate applications. And this is called a big and great feature. Look and Gnome 2 years ago and now. There are no really big changes. KDE can evolve much faster. The reason why it can lack some features gnome currently have is that KDE developer are focused on the KDE 4 already for a two years. That is why you don’t see great improvement in KDE during last 2 years. But after KDE 4.0 will be stable it will bring a huge amount of new and great features which will allow KDE to evolve and become better much more faster than Gnome can do.
The technical background of Gnome (gtk, glib, etc) is very limited compared to QT which is technical background of KDE. After table release of KDE 4.0 KDE will be able to evolve much more faster than Gnome. And I am absolutely positive that in a two years time from now Gnome will be almost the same as it is now. Gnome just can’t evolve real fast. There would be no new major features in Gnome for years. There will be Gnome 2.xx.xx.xx.xx.xx with no major changes.
That is why I think that you should focus on some strategic directions for now. Stop working on KDE 3.x at all and devote all your efforts to KDE 4. Look what is currently missing in KDE 4, choose the most important things and start work on this. A good start could be porting of Guidance system settings modules to the QT4/KDE4. Maybe few more new system settings modules.
Like a good boxer you need to wait. Let the opponent to be more active in first few round let him to gain more points for these few first rounds. But you know that you can do better than him later and that hi can’t do more in future. Let him show the best he can and than beat him. Because you know that in later rounds you will be in much better shape and you can do much more than your opponent.
Sorry for such comparison. It’s a bit ugly but just I wanted to explain my point of view.
October 31st, 2007 at 8:43 pm
[...] dann habe ich noch einen Blogeintrag zu Kubuntu gelesen:http://jucato.org/blog/quo-vadis-kubuntu/ So weit waren wir schon einmal. Die Diskussion war vor dem 6.06 Release. Damals ist Mark [...]
October 31st, 2007 at 8:47 pm
Me being just and interrested user, I also had thought about Kubuntu every once in a while, and especially why it’s mostly being ignored, why it doesn’t get the reputation it deserves. And in my eyes there is another problem in public perception: the name.
From outside it seems that there is the Ubuntu universe with “Ubuntu” as the main and general use distribution, plus some additional specialized distributions keeping the Ubuntu name and just adding a prefix. While this is basically true, say for Edubuntu, it’s not for Kubuntu. Kubuntu has actually the same target audience as the main Ubuntu, the general users who need a desktop for their everyday work. But the name just doesn’t reflect that. Especially for users from the outside of the free software world, which we all should be interested to attract. It sounds like a specialized derivate from the “real” Ubuntu.
These users may hear, besides RedHat/Fedora or Novell/OpenSuSE, about Ubuntu, not Edubuntu, Xubuntu and also not Kubuntu, just about Ubuntu. If all goes well they also notice that there are other derivates. They might have a look, but in the end they lack the knowledge to decide about which to take and go the safe way by selecting the main Ubuntu. Even more so, as they might learn that for instance Edubuntu is for eductional purposes, and they don’t need that, but what the heck is KDE?? Why is it different/better than main Ubuntu?? Probably something specialized. So better go the safe way and take what everyone takes… This was nicely illustrated in a recent article in a German maintream news magazin, where the author wrote a journal about his journey into the linux world. And he did exactly that, heared about Ubuntu, looked around, and in the end, you guess it, he installed Ubuntu, and not X-,Y-,Zubuntu.
So IMHO there are 3 options:
1. Keep everything as is, stay forever in the shadow ot Ubuntu. (Am I too negative here??)
2. Integrate with main Ubuntu, at the same level as Gnome. So that you have one general use Ubuntu with either KDE or Gnome. Problem is, the user with no clue has to decide on install whether he wants KDE or Gnome…Bad. Also it yields the comparison problems described in the original aritcle. And it contrains the development of Kubuntu. Play catch up.
3. Change the name to something more unique, independent, more on the level of the name “Ubuntu”. Which suggests to the outside world, that it is also a main distribution, not just a specialized addon.
October 31st, 2007 at 9:03 pm
Me being just and interrested user, I also had thought about Kubuntu every once in a while, and especially why it’s mostly being ignored, why it doesn’t get the reputation it deserves. And in my eyes there is another problem in public perception: the name.
From outside it seems that there is the Ubuntu universe with “Ubuntu” as the main and general use distribution, plus some additional specialized distributions keeping the Ubuntu name and just adding a prefix. While this is basically true, say for Edubuntu, it’s not for Kubuntu. Kubuntu has actually the same target audience as the main Ubuntu, the general users who need a desktop for their everyday work. But the name just doesn’t reflect that. Especially for users from the outside of the free software world, which we all should be interested to attract. It sounds like a specialized derivate from the “real” Ubuntu.
These users may hear, besides RedHat/Fedora or Novell/OpenSuSE, about Ubuntu, not Edubuntu, Xubuntu and also not Kubuntu, just about Ubuntu. If all goes well they also notice that there are other derivates. They might have a look, but in the end they lack the knowledge to decide about which to take and go the safe way by selecting the main Ubuntu. Even more so, as they might learn that for instance Edubuntu is for eductional purposes, and they don’t need that, but what the heck is KDE?? Why is it different/better than main Ubuntu?? Probably something specialized. So better go the safe way and take what everyone takes… This was nicely illustrated in a recent article in a German maintream news magazin, where the author wrote a journal about his journey into the linux world. And he did exactly that, heared about Ubuntu, looked around, and in the end, you guess it, he installed Ubuntu, and not X-,Y-,Zubuntu.
So IMHO there are 3 options:
1. Keep everything as is, stay forever in the shadow ot Ubuntu. (Am I too negative here??)
2. Integrate with main Ubuntu, at the same level as Gnome. So that you have one general use Ubuntu with either KDE or Gnome. Problem is, the user with no clue has to decide on install whether he wants KDE or Gnome…Bad. Also it yields the comparison problems described in the original aritcle. And it contrains the development of Kubuntu. Play catch up.
3. Change the name to something more unique, independent, more on the level of the name “Ubuntu”. Which suggests to the outside world, that it is also a main distribution, not just a specialized addon.
October 31st, 2007 at 9:34 pm
Michael,
YES. Excellent post, thanks.
However, it’s going to be super tricky to implement a name change that would both 1) remain attached to the Ubuntu name, upon which all the marketing machinery is centered, and 2) avoid the negative imagery of being one of the [x]Ubuntu obscure spinoffs.
I was thinking of something along the lines of “Ubuntu [some_adjective]“, where [some_adjective] should bring an idea of advanced, improved, bolder, neater, more modern, sexier.
Then it occured to me that KDE already has some nifty marketing going about some of its new technologies, and that might make for a nice tie-in with an Ubuntu rebranding.
Hence: what about “Ubuntu Plasma”? “Ubuntu Oxygen”? “Ubuntu Appeal”? You get the drift.
October 31st, 2007 at 9:45 pm
Michael,
YES. Excellent post, thanks.
However, it’s going to be super tricky to implement a name change that would both 1) remain attached to the Ubuntu name, upon which all the marketing machinery is centered, and 2) avoid the negative imagery of being one of the [x]Ubuntu obscure spinoffs.
I was thinking of something along the lines of “Ubuntu [some_adjective]“, where [some_adjective] should bring an idea of advanced, improved, bolder, neater, more modern, sexier.
Then it occured to me that KDE already has some nifty marketing going about some of its new technologies, and that might make for a nice tie-in with an Ubuntu rebranding.
Hence: what about “Ubuntu Plasma”? “Ubuntu Oxygen”? “Ubuntu Appeal”? You get the drift.
KDE 4 is a wonderful opportunity to completely bring Ubuntu on par with the modern visual sexiness that is at the core of the Leopard (and Vista I guess) marketing, I think!
October 31st, 2007 at 10:39 pm
I’d agree. Kubuntu works, but it somehow feels a bit…disjointed. In support, particularly: we share the bug tracking, forums, etc. of Ubuntu, and the vast majority of instructions are aimed at GNOME users. Many of the headline features only make it to Kubuntu quite a bit later.
I like KDE, but I’m seriously thinking of moving to another distribution once KDE4 is out. I think a KDE-focussed distro can probably do a better job of making KDE work.
October 31st, 2007 at 11:13 pm
“”1. The easiest option. There is actually no problem. I’m just overreacting and being melodramatic.”"
That is it! Got it on the first try, you are smart.
I go with that answer. I love the Gnome UI on a small laptop screen and the option-filled KDE UI on a big desktop screen. Both work well in kUbuntu — this is not a big deal.
Ubuntu has bulletproof X and Kubuntu has K3b (and better apps for most tasks).
Chill out.
November 1st, 2007 at 12:12 am
As a GNOME user i’ll ask so what if Kubuntu is going down the drain?
If there were only one primary DE out there for Linux, like GNOME, then that DE would become so powerful because
1. new devs and the majority of community people will work on it bringing less bugs and more features
2. there would be less feature duplication from other DEs
3. a standard desktop for linux, no more copy paste errors…
Then the Linux community will be able to beat Windows faster.
The small community support of ubuntu to kubuntu only means either:
a. people are not that interested anymore on KDE
b. kubuntu is not getting enough popularity in the media
So when KDE dies, users win. So too when any other DE dies, besides it doesn’t matter what DE is out there. What matters is you get things done, and I’m so tired of dealing with compatibility issues with both KDE and GNOME, if I were to make a roadmap for both DEs, I’ll opt for better integration. Option 2 for the win.
Hoping I won’t get kicked here because of this blog post.
- Mark GNOME user
November 1st, 2007 at 1:12 am
MarxChan: I fear a consolidation into one “DE” (aka “Window Manager”). That’s exactly what exists in Mac OS and Windows. Those Operating Systems have become defined by their windows managers. The window manager should never do that.
November 1st, 2007 at 2:06 am
> If there were only one primary DE out there for
> Linux, like GNOME, then that DE would become so powerful
If there were only one window manager.
If there were only one web broser.
If there were only one operating system.
If there were only one search engine (oops, that one is true).
You know, sometimes diversity is a win.
November 1st, 2007 at 2:21 am
I would personally drop the name Kubuntu and give it equal status with Ubuntu, which would get rid of the confusion of whether it is a separate OS from Ubuntu, which the name suggests.
Personally, I find that the version of KDE shipped with Gutsy is a back-track on the old (Feisty) version; it is less responsive, the problem which put me off the KDEs shipped with Edgy and Dapper. Things like context menus taking time to open in Konqueror, which is what had me stop using it. OTOH, GNOME in Gutsy is buggy as well - just try changing the default desktop fonts, the preferences window crashes. I am using XFce at the moment.
If possible, they should copy the Ubuntu colour scheme onto KDE (even if copying the theme itself would be too difficult) and stop calling it Kubuntu, and then work on responsiveness which has always been a bugbear of mine with Kubuntu. They could copy the version shipped with OpenSUSE 10.3 which is one of the best KDEs I’ve ever used (although they should leave out the bug which you find when opening up a link in a new window in Konqueror, that it opens it up with as many tabs as the old window had, with the page in the last tab).
I don’t think splitting from Ubuntu is an option; the whole point is that you have access to the Ubuntu software repositories.
November 1st, 2007 at 2:28 am
I’ve used Kubuntu since Breezy or Dapper and while I prefer KDE over GNOME, Kubuntu doesn’t feel very polished. I don’t mind not getting the little extras, but even basic KDE stuff doesn’t work properly (network settings is a nightmare, system-settings is inferior to kcontrol, enabling/disabling user services doesn’t even seem to exist in the GUI).
I tried openSuSE and Fedora to see how they treated KDE and it was good, but not as great overall. Having a Debian base is certainly the way to go. Contrary to some of the overzealous comments; Kubuntu seems to be the best at Just Working, having tons of software without having to add 25 repositories, and getting out of your way so you can get some real work done.
My only complaint is KDE needs to be cleaned up, so I’m very excited about KDE 4.
November 1st, 2007 at 5:26 am
kh: kcontrol is there on Kubuntu. You can invoke it by doing “Run command” or by editing the K menu so that System settings points to kcontrol. I’ve done this myself.
November 1st, 2007 at 5:36 am
kh: you can also add it to the K-Menu. I always do. I have a laptop that I’m trying to keep “pure Kubuntu”, but I keep making settings changes by “alt+f2 | kcontrol”.
Of course, on another machine, I’ve said to … umm … with it, and added the KDE repos and installed KDE from KDE.org.
November 1st, 2007 at 6:11 am
I’ve always been a little miffed that Canonical didn’t hire more people to work on Kubuntu, since KDE at Qt are clearly the superior technologies.
But now that they have practically perfected GNOME in terms of ease of use (BulletProofX, restricted drivers manager, automatic backups, the list goes on!) I think it would wise of them to begin developing Kubuntu as their main product. If they don’t, I see a big opportunity for some KDE4-centric distro to emerge as the “next Ubuntu”, simply because GNOME is progressing slowly, while KDE is poised to make a quantum leap. And I don’t mean to get into a KDE vs. GNOME war - I’m just saying that KDE4 has attracted a lot of attention, and the GNOME developers have expressed no particular plans for a GNOME 3 to compete.
November 1st, 2007 at 6:27 am
Hi Jucato.
I’ve just read this post with some interest; I’m an Ubuntu user, and, as someone who has had a historical preference for KDE over GNOME, I’ve always had an expectation that I would change over to Kubuntu one day. After KDE 4.0, I suppose. This idea has been pre-supposed on the fact that Kubuntu is simply Ubuntu with the GNOME parts replaced by KDE equivalents, or option 2 in your terminology.
Frankly, this is what I read in the ‘About Kubuntu’ page you linked to. You clearly consider Kubuntu to be much more than this, but neither you nor your linked pages describe the higher goals Kubuntu aspires to.
What do you feel would be lost if Kubuntu is restricted to your option 2 scenario?
November 1st, 2007 at 10:34 am
@S: Yes, it is completely amazing at how far Kubuntu has come without that much support from Canonical. It reflects the power of KDE and the amazing people that we have. That’s why I’m confident that we can pull this through.
@Nathan Dbb: I always knew I had a few screws loose.
@MarxChan: Sorry, but this isn’t really a KDE vs. GNOME issue. But since you brought it up, let me try to answer:
1. new devs and the majority of community people will work on it - This is too idealistic. The reality is that there will be people who will leave. It is a presumption that joining two projects together will automatically merge people as well. That’s not always the case.
2. there would be less feature duplication from other DEs - It is the duplication that allows innovation and ingenuity to surface. Being able to see and evaluate your options is also a great thing.
3. a standard desktop for linux, no more copy paste errors - But is that really the answer? The proper answer would be to develop a common ground/specification for this, like http://www.freedesktop.org/wiki/Specifications/clipboard-manager-spec
Then the Linux community will be able to beat Windows faster. - It could be the other way around.
One single DE, one single point of failure. If the DE goes, it takes down everything with it.
a. people are not that interested anymore on KDE - I don’t know where you get that idea. Because first, while we lack developers, we have an abundance of users. And second, while Kubuntu lacks developers, all othe KDE-based distros are flourishing. This doesn’t hold water.
b. kubuntu is not getting enough popularity in the media - This is indeed a problem. But not always, since there are in fact a number Kubuntu deployments, like in France.
So when KDE dies, users win. - Nope. Not always. KDE users and developers will lose. Linux gets hurt, too.
Anyway, moving on…
@kh: One of the best things I’ve observed about Kubuntu is that, at least for me, everything just works. And other users seem to feel that way, too. Of course, we can’t really get 100% on that, and there will always be some problems.
As for System Settings, better start getting used to it, becuase it will be the default in KDE 4.0. And for user services, there’s the serviceconfig in System Settings -> Advanced tab -> System Services.
@Matt: I would probably consider the autonomy and freedom to do what we can/want to make Kubuntu a special KDE experience for users. We probably would also lose that sense of pride of being a derivative distribution and simply become “just a team”. I’m not absolutely certain with how things work at Debian and Fedora, but this it the way I’ve seen it, based on my trials of the two. They both have KDE teams that are part of the larger development team. They are responsible for maintaining the KDE packages of the distribution. They also seem to make some default settings when KDE is installed and used rather than GNOME. Other than that, I’m not sure what else they can do. Kubuntu is almost in a similar vein to what openSUSE has I think.
November 1st, 2007 at 2:48 pm
KDE rocks. Nothing against GNOME, xfce, etc. but Kubuntu just *feels* right, because it’s an overall desktop experience. I can katapult to a folder and open it in konqueror, and then launch a music file in amarok.. or just katapult to the music file.
The point is that Kubuntu (to me) creates a very personable experience, and allows the user to do more things with it’s features than I could ever understand. One day I’ll figure out how to get all the programs synced up… after I get off my butt.
If Kubuntu and Ubuntu do drift apart, for different goals/viewpoints, I dont think it would be the end of the world. *buntu will still run under the hood, and that’s what matters the most. Roadmapping to outline a vision of 1)update amarok , 2) katapult rocks… is not the same as deliberately cutting ties with Gnome/Ubuntu.
I do wish that Canonical would help more with the Kubuntu development, but it is always nice to have more hands, as you said. More work for me. That being said, Ubuntu has always been about the community, and (at least on the dev side) that is the underlying point.
As for your options, I feel it’s a mix between one and three. There is great communication and love between the *buntus, but that doesn’t mean they all have the same goals. The classic case of GNOME/KDE is trite when XFCE is introduced, and only matters because we actually have a choice. It’s true that I’ll rib on gnome users and xfce users, but that doesn’t mean I won’t help them with issues, or contribute their projects when I can.
Summary; blah, blah blah. I like Kubuntu and working on it/with it. Ubuntu is cool too, but with different goals. Let’s identify where the similarity/differences are and agree to focus on what features we love to use.
November 1st, 2007 at 7:07 pm
Interesting discussion. I would like to give my 2 cents from the point of an end user. When we are talking about desktop environments, what matters for such a user is just the desktop. He may not even know what bulletproof X is, he may not even be aware there are more features in a distribution than in another because he most certainly just mechanically powers up his computer, logs in the desktop and starts doing his stuff.
My point is that as long as we are talking about *desktop* experience, the vast majority of users notice only what is directly in front of their eyes. In that sense, kubuntu already has the edge over ubuntu simply by using KDE and this will become much more apparent when a stable and supported enough (themes, useful applets, integrated applications that work satisfactory - see khtml “problems” or lack of voice support for kopete …) version of KDE4 enters the scene. I believe this is supported by the overall percentage of kde users. Keeping up with other feature additions is a good thing but it’s not decisive if you have the edge in the most important area.
Saying this, I really am not sure why any company would choose to support a DE like gnome which seems to be in a dead end over a technology with visible future like kde …
November 1st, 2007 at 8:48 pm
I’ve been a Ubuntu user since Warty and a Kubuntu user only since Feisty. What can I say, KDE just feels right, GNOME doesn’t.
I personally think that the right decision would be to talk to Canonical and get the same treatment as Ubuntu. I recently read that one of the objectives in the roadmap for 8.04 was that Kubuntu had to catch up with all the features that Ubuntu offers. I hope that means that Kubuntu will get as much attention as Ubuntu gets.
November 1st, 2007 at 11:59 pm
Yeah, adding kcontrol to the the menu is always one of the first things I do on a new install. But if you’re going to reinvent the wheel, it should be better than the previous version. I remember reading about kcontrol not making it to KDE4, so I imagine system-settings will be much better by then.
As for user services, I’m not at a KDE desktop at the moment, but I’m fairly certain it doesn’t have what I’m looking for.
For instance, in GNOME, you have Sessions for user startup programs, Services for user services. In KDE, you have one for system services, one that looks like low-level KDE stuff, and maybe something else, but not enough for _user_ services/programs.
Most default startup programs (Klipper, Knetwork-manager) can be killed from the tray and it will ask not to start again. Not so for Strigi, I could find no way to kill it and the daemon from the GUI (with the intention of having it persist a new session). I remember looking around the startkde and other startup files and couldn’t find it (although I was sort of in class at the time).
I would just like to see easier, more thorough, less redundant system administration tools in the GUI. I spend a lot of time in the terminal, but what I love about KDE is that you can do most things from a terminal or the GUI, or use just keyboard shortcuts or just the mouse. Love having the option to work in a variety of ways.
Thanks for the responses.
November 3rd, 2007 at 6:37 pm
Thank you everyone for all the feedback. Sorry if I wasn’t able to reply ASAP. Real life got in the way, as always.
Anyway, I’m planning to make a follow-up post, summarizing on the topic. Probably I’ll get it done tomorrow.
@kh: I think I understand what you’re looking for. But you mixed “system services” and “user autostart/startup” programs programs. System services on Linux are different.
I think KDE’s Session Management is actually very powerful, even without a Session tool like GNOME’s. By default, KDE automatically saves the session at logout and restores it when you login. You can also set it to let you manually save the session, which lets you setup only the apps you need to be autostarted, without having to list or add them one by one, like in GNOME’s Session tool. Unfortunately, KDE’s session manager is highly unknown or underestimated, which is why users keep on asking if there’s a tool to autostart apps on login.
The good news is, there is such a GUI tool, which you can find in kde-apps or in Kubuntu’s repositories. The better news is, I’ve been told by the author that this tool is now integrated into KDM in KDE 4.0. So all we have to do now is wait.
November 5th, 2007 at 6:49 am
Just wondering what you have on your “native” kubuntu dist that I do not on my ubuntu+kde one?
November 5th, 2007 at 8:03 am
I think that we should have separate developers for both gnome and kde, and all apps should be designed to work well under both environments. Doesn’t Linus use KDE too? I’ve been really disappointed with Dolphin at this point though.
November 5th, 2007 at 8:22 am
@kubuntu=ubuntu+kde:
Depending on how you installed KDE on top of Ubuntu, you’ll be missing on the default settings, artwork, application selection, and KDE tools from Kubuntu.
@Joshua K:
The developers are actually separate, and the Kubuntu developers are really very few. Lately though we a few new faces hanging around, so maybe it’s a start of something new (where have I heard that before?).
November 5th, 2007 at 8:29 pm
I’ve always loved Ubuntu for being a simplified, more usable (to me) and standardized Debian distribution. But although I tried GNOME time and again, I always came back to KDE, it’s just more to my taste: To me, KDE simply feels and for me it works better than GNOME! So only when Kubuntu came out did I switch to the *Ubuntu family.
I’ve been quite happy with Kubuntu since then. Only with Guty I noticed a rift, that’s why I’m now running Ubuntu 7.10 on some systems, and keep Kubuntu 7.04 on others. As you said, with Gutsy it became noticeable for the first time that Kubuntu tried to catch up, and this time it failed.
Well, as to my personal opinion, I never liked the separation between Ubuntu, Kubuntu, Edubuntu, etc. I’d rather have a single, complete system and choose the major software components. I always regarded Ubuntu that way, most of the time talking about Ubuntu when I meant Kubuntu or Xubuntu, because to me they were just different frontends to a single distribution, only split so the live CD could be kept smaller. I realize that might not be how Canonical or the various teams regarded it, but I believe this makes the most sense. I don’t want to switch distributions if I want to use a different GUI or software (e. g. Edubuntu’s terminal services), I just want to apt-get some meta-packages and be done with it!
I’m also using both GNOME and KDE applications. I think trying to rewrite everything is a waste of time and energy that would be better spent in improving the already existing and working software. That’s why I use Ubuntu’s upgrade-manager and Synaptic instead of Adept, they are more refinded and simpler to use. And I always use Kubuntu’s Konsole instead of Ubuntu’s Terminal. And Amarok for music, Firefox for web, and so on. Ubuntu (all of it) could be so much better if there was one base with the best apps of all worlds/desktops!
So what I’d like to see, is some sort of unification of all sub-distributions, focussing the energy on what each does best, and making the desktop choice irrelevant (let people use whatever they want, integrate it so it looks and works the same, and let users customize if they prefer it differently).
Does anyone agree?
November 5th, 2007 at 10:42 pm
While both GNOME and KDE have their weaknesses and merits, I personally do prefer KDE. I have tried Ubuntu, Kubuntu, Fedora KDE and Mandriva. I stopped using Ubuntu, because I do not like GNOME as much as KDE. I stopped using Kubuntu because it did not look and feel as polished as other ´native´ KDE distributions and because some basic features did not work (ie. a notification applet for available updates). The Fedora KDE release, on the other hand, looks to me to be the exact same quality release as the GNOME version with only a different desktop environment. Today I prefer Mandriva, a native KDE disto with great looks and user friendlinesses. Based on these experiences, I believe it would be the best to make Kubuntu an exact Ubuntu clone with KDE desktop. If KDE 4 becomes the mega hit that I expect it to be, maybe in time Ubuntu will become the Kubuntu clone with GNOME DE ;-).
November 6th, 2007 at 3:42 am
I agree, Kubuntu needs it’s own identity…actually, I like Kubuntu over Ubuntu, but it has long languished in polish and fit and finish that Ubuntu has…
On the other hand, the Kubuntu team has not exactly been forthcoming with it’s users as to what they want to do, how members of the kubuntu community can get involved, or even what you expect of Kubuntu users.
The first step in getting a Kubuntu identity has to be to engage your users in the development, I think.
November 6th, 2007 at 7:48 am
I have been using Kubuntu since it’s first release and it has always pretty much worked - if a little rough around the edges.
I have always installed the ubuntu-desktop also just to check out gnomes progress although use kde most of the time - I just can’t work in gnome (if only they chose kde at the main desktop…..)
It has never had the polish of other KDE distros - i.e opensuse, mandrive - even my gentoo installation looks nicer than the default kubuntu install.
However it is the most installed linux desktop I know of - i work in a webhosting company - no one uses gnome…
The best solution would be for Opensuse and Kubuntu to join forces (losing any Novel influence ) stop bothering with gnome and lose any Novel packages and replacing opensuse’s package manager with kubuntu’s. - Just an idea ……….
November 8th, 2007 at 2:54 am
I used Kubuntu in the Feisty era, but since september 2007 I switched to the more stable desktop/laptop debian system, Mepis 7 (beta!!).
I liked to think Kubuntu like a Ubuntu+KDE, but it was lesser than this. I was disappointed!
If I will use an *buntu system, I will use Ubuntu of course. All great and innovative software produced in the Ubuntu family is there.
Now I have a Debian+KDE+Mepis_configuration&utilities that is just that I need.
I like and I join the Warren purposes.
If a kde distro must be in the ubuntu family, that can be Mepis. Canonical can support Warren!
November 8th, 2007 at 6:04 pm
[...] just want to give some sort of closure to my previous blog post about trying to (re)define Kubuntu’s identity and goals, just to get rid of this chip on my [...]
November 8th, 2007 at 8:29 pm
For me, it’s definitely option 4
I’d wanted to add it even before I started reading the comments… What I would like to see, is Kubuntu as one of two official “faces” of Ubuntu. So that on the main Ubuntu website we would see “This is ubuntu, the best Linux distribution. It has two versions, Ubuntu, the Gnome version, and Kubuntu, the KDE version”. And in every official release announcement, we would see “We’ve just released Ubuntu x.xx. You can download the Gnome version here, and the KDE version (Kubuntu) here.”… I know, sounds impossible, but I can always dream… 
Or, if option 4 is not possible, then option 1. I’ve been using Kubuntu for about a year and a half now, and frankly, it was always sufficient for me. Maybe it’s because I’m a power user, it’s not a problem for me to install any package that I need, or find information on Google about what package I need to install or what repositories to add to get something… I used to use Debian before, so Kubuntu was rather a step ahead for me in terms of user-friendliness. (K)ubuntu has a huge repository of packages just like Debian, which can be as easily installed, and it looks much better than Debian, and is more up-to-date, and that’s all I wanted. I don’t really care if Kubuntu has this or that utility that Ubuntu has…
That doesn’t mean I wouldn’t like to stop seeing reviews where people say “yeah, Kubuntu is nice, but it’s worse than Ubuntu, it doesn’t have as big priority in Canonical as Ubuntu”. Or than I wouldn’t like to see Kubuntu much closer to Ubuntu on Distrowatch top list. That’s why I wrote that option 4 would be the best. But if that’s not possible… then I don’t think it’s necessary to change anything radically.
I don’t like option 2. Especially the “GNOME utilities when it matters” part… No hideous GTK in Kubuntu, please :> (well, except Firefox)
And I don’t like option 3 either. Of course you know the situation better than us users, because you’re an insider, but I think any attempt to gain more independence from Ubuntu would lead to marginalising Kubuntu even more. With Ubuntu, you’re in a way a part of the most popular Linux distribution. Without Ubuntu, you would risk becoming just another apt- and KDE-based Linux distribution…
And I agree with those that say that when KDE4 comes, things will change for the better. Let’s hope they’re right
—
To MarxChan:
You may be right that having one DE instead of two would be better for Linux. So why not concentrate on KDE, which is clearly better?
See, the number of users that use KDE and users that use GNOME are more or less equal. Maybe it’s not 50-50, maybe it’s 45-55 or 55-45, but it’s definitely not GNOME-80, KDE-20, or “people are not that interested anymore on KDE”. We just have two equally preferred desktop environments and dropping any of them would result in protests and outrage of about 50% of Linux users. So the only option is to admit that both DEs have to stay and try to respect each other…
—
To Matt Smith:
> If possible, they should copy the Ubuntu colour scheme onto KDE
What?? No, please don’t do it!
KDE is blue, GNOME is brown, and it should stay that way… Brown colors just don’t suit KDE and make it more GNOME-y, and I suspect that it’s the same with blue GNOME. Just keep it the way it is, I admit that Ubuntu’s GNOME is one of the prettiest GNOMEs I have seen, because of nice widget styles (Human!), well-chosen colors and icons not looking as if they were made in 1980s. And blue KDE in Gutsy is much better than Feisty’s purple one - I especially love the wallpaper 
—
And one more thing, regarding the System Settings - I generally like the idea (especially since I saw that the control panel in MacOSX looks the same :D). What I don’t like is the fact that some of panels are just missing from System Settings. E.g. I can’t find “file associations” panel in it, though it’s available in KControl (and there are more of such panels).
November 8th, 2007 at 8:54 pm
[...] just want to give some sort of closure to my previous blog post about trying to (re)define Kubuntu’s identity and goals, just to get rid of this chip on my [...]