On Distributions, Kubuntu, and KDE
The open source operating system experience exists in pieces, scattered across a world of projects and technologies. Distributions exist because they attempt to create a unified experience from the bits and pieces of open source functionality out in that world, while establishing themselves as a vendor their users can trust. Users might try a few distributions to see which one works for them and settle on one which provides the experience they are looking for. Users build a relationship with their distribution from which they can establish a trust in the packages the distribution ships and the defaults they configure.
So when a user decides they no longer agree with what their distribution ships, and exercise an option to download untrusted packages — risking the failure of future software upgrades and possibly severe security issues — with whom does the fault lie?
Is it reasonable for a distribution to expect a user to visit their website on Release Day to be informed about changes and cautions in the new release? Is it reasonable for a distribution to be alarmed and concerned when rogue packaging interferes with updates, upgrades, and security? Is it reasonable for a distribution to encourage their users to resist or desist using such packages? Is it reasonable for a distribution to encourage users who are not interested in the offerings of the latest release to fall back to the still-supported version? Is it reasonable for the distribution to support their users the best they can while maintaining their own project goals?
I think these are all reasonable expectations. These expectations allow the distribution to to its job: provide a cohesive product which they can reasonably support while protecting its brand and reputation. If users are going to lay the responsibilities of security, stability, and trust on their distribution, then I don’t think it is unreasonable for the distribution to take actions to protect itself and their users.
But what do you do when your users try to reinvent your distribution? There seems to be some confusion as to why the Kubuntu distribution exists and who it exists for.
Kubuntu is a KDE distribution.
Although some would like to say that KDE competes with GNOME (and therefore Kubuntu competes with Ubuntu), I disagree. KDE has completely different market, user experience, and technology goals compared to GNOME. KDE is changing the way people interact and use their Desktop and developing new technologies. KDE also focuses on usable functional design rather than universal usability. One of KDE’s primary user groups is made up of Early Adopters. These are users who are interested in new technology, trying new things, and like to stay on top of what is fresh and new. They don’t necessarily want KDE while it is still breathing and bleeding, but the fresher the kill the better.
I find it ironic how I’m discovering so many “KDE users” who are resistant to change.
Kubuntu attempts to balance itself between its user community and KDE. As an Ubuntu project, we are subject to Ubuntu policies and inherit a very user-driven community. Ubuntu’s strength is in its user community, but so are its weaknesses. There is a necessary balance to everything, and sometimes it is hard to make user goals and project goals even on the scale.
Nearly a year ago last January, KDE 4 was more than fresh — it was still bleeding. 4.0 was clearly a “developers” release, but the excitement surrounding the new and pretty and shiny still drew some users in — at a cost. Kubuntu supported its user community by shipping the stable and reliable 3.5.9 in Hardy instead of adopting the still raw KDE4 technology. We tried to appease traditional KDE users (the early adopters) by also releasing an unsupported KDE4 Remix distribution. Everyone was happy and the world was a good place.
When Intrepid rolled around, it was 10 months since KDE 4 was first released and time for Kubuntu to realigned itself with its goals and focus on KDE. We shipped 4.1.2. This has become a lose-lose situation.
At the time when decisions were made, there were still some bugs we were concerned about in 4.1.2. In an attempt to support our users, we temporarily disabled some not-quite-working functionality until it is fixed in 4.1.3. This pissed off the traditional KDE users.
In the other hand were our traditional 3.5 users. By refocusing our attention from them to KDE and shipping 4.1.2, we pissed them off as well. As a result, instead of sticking with what was known and supported (Hardy), some users installed Intrepid and then resorted to third-party 3.5 packaging which has now damaged their systems.
What’s a distribution to do? There are a large sect of 3.5 users who are not ready or willing to make the move to KDE 4. Continuing to support them stagnates the Kubuntu distribution. Limiting support forces them to resort to other (sometimes damaging or dangerous) sources. It is times like these where you must take a step back and remember what your purpose and goals are.
Kubuntu is a KDE distribution.
As a KDE distribution, it is our responsibility to ship what KDE is developing, not a stagnant branch of what it was. KDE 4 is the future of KDE. It is the purpose and goal of Kubuntu to support, package, and distribute KDE and make it easy to install and fun to use. If some Kubuntu users can’t get over that, then maybe Kubuntu isn’t for them.
I’m sorry to tell but kubuntu 8.10 is one of the worst distribution I tried since some years. So many things are not working and worst some that I personally fill a bug report on launchpad and were the problem was only from the packaging as krita2 cannot open any jpeg file because of kubuntu packaging or worst nepomuk ad strigi are not compiled with sesame2 backend so they are slow, buggy, cpu hungry and space eater… Bluetooth is useless without the gnome version of ubuntu (ok this was not kubuntu fault). Honestly I’m testing kde4 since the first version published and I like it, yes it’s not as polished than kde3 but it’s ok now I’m using it every day (but for kdepim because kmail4 is useless with imap+gmail and the installation didn’t convert my kmail3 data to kmail4…). I’m waiting to see what we will have in jaunty but that will be my last try with kubuntu if the problems above are not solved. It’s not a menace it’s just that I’m tired to fill bugs for nothing and to have buggy desktop because of strange packaging.
This very much relates to the issues I have been facing. I have been using KDE 3.5 for a while, and I like it very much. I was excited about the Intrepid release, but was very disappointed that it couldn’t do multi-row taskbars and auto-hide panels, two things which are very important to me (I value my screen real estate). Sure, Plasma looks very cool, but I need my basic everyday functionality.
Finally I had to decide not to downgrade from 3.5 to 4.1, thereby missing out on unrelated goodies like usb-creator. So basically, “if you want the latest kernel, latest X, new goodies like usb-creator, you have to sacrifice established basic features like auto-hide panels”.
If there are a large sect of KDE 3.5 users who are not ready or willing to move to KDE 4, surely there’s a reason for this?
This very much relates to the issues I have been facing. I have been using KDE 3.5 for a while, and I like it very much. I was excited about the Intrepid release, but was very disappointed that it couldn’t do multi-row taskbars and auto-hide panels, two things which are very important to me (I value my screen real estate). Sure, Plasma looks very cool, but I need my basic everyday functionality..
Finally I had to decide not to downgrade from 3.5 to 4.1, thereby missing out on unrelated goodies like usb-creator. So basically, “if you want the latest kernel, latest X, new goodies like usb-creator, you have to sacrifice established basic features like auto-hide panels”.
If there are a large sect of KDE 3.5 users who are not ready or willing to move to KDE 4, surely there’s a reason for this?
I think part of Kubuntu’s problems stem from it’s name. ‘Ubuntu with KDE’ would be much better if you ask me.
One of the goals of a distribution is to decouple upstream from the users. If upstream delivers products that are not ready to be deployed by common users (in this case KDE 4.0 but also KDE 4.1.2) they should not be shipped.
Or has the goal of Kubuntu shifted to be a bleeding edge distribution like Fedora? In this case it would not really fit into the Ubuntu philosophy of “Linux for human beings”, wouldn’t it?
Hi Celeste,
What makes you say that a primary KDE user group is early adopters? I’ve never had that impression.
After two years of subpar package manager I had enough.
Fix/redo/replace package/network/power manager, font properties configuration tool (good example: GNOME in Ubuntu Intrepid) and … put Firefox on CD. This whole widget toolkit preaching is pretty stupid endeavor, there is no point in pure KDE distribution nowadays.
It is a big and incomprehensible fallacy that a distribution would be defined in terms of its desktop environment.
I have never understood Ubuntu’s way of defining its derivatives in terms of different desktop envirnments.
It is a completely unwarranted assumption to say that kde users are all early adopters.
I have been following kde since the beginning and deploying it professionally since kde3 and most kde users are actually rather conservative and want a powerful but easy to use desktop. In all of my tests, a common theme for users refusing gnome was gnome’s placement of screen dialogs and taskbars did not match their expectations.
People expect functional technology. Somewhere along the lines, many distributions have forgotten the old motto of ship it when it’s ready.
As an early adopter, I am happy to tell you that for all of its bells and whistles, KDE 4 has not been ready for general use until now. KDE 4.2 or 4.3 will likely change that.
If I were to substitute today’s kde 3.5.10 desktop for kde4.1, most of my users would see it as a huge downgrade. They don’t care about the innards of the technology or long-term roadmaps and goals.
Thankfully, I am following SVN and KDE 4 is a lot closer to being ready for mass deployment, but it ain’t there yet.
Wonderful article. I’m experiencing the same problems in the German ubuntuusers Forum as a KDE supporter. We (the German uu team) warned the KDE 3.5 Users about KDE 4 and told them to stick with Hardy if they don’t want KDE 4.
Nevertheless there are at least three new threads opened a day where users complain that they were “enforced” to use KDE 4. That KDE is shit, that Kubuntu is shit. It’s really annoying and depressive. I already thought about dropping the support for some time as I just don’t like reading all these bad comments about KDE and Kubuntu.
Personally I think that a user who is able to do the upgrade (as the button is not shown in Adept) knows about the situation and thinks that KDE 4 suits him.
KDE, GNOME, and Cocoa all are competing for the same users. User users. They’re all 2d GUIs with usage paradigms and HIG, and they’re in the same market.
In my opinion, gnome is the present, and KDE4 is the future. I think KDE3 users should have been advised not to upgrade from 8.04.
That said, I tried the KDE4 LiveCD, and it failed in several different ways before crashing entirely, filling my screen with unintelligible characters and forcing me to do a hard reboot.. I look forward to Kubuntu 9.04, KOffice 2, Amarok 2, and whatever Kdenlive is at when the time comes.
Good post. I’m relatively new to linux and Ubuntu, and I’ve used GNOME as default. I’ve thought about KDE though, the screenshots look nice. But for my purposes it seems likely that I’ll stick with what I’ve got. To be honest I don’t really understand why there are different desktop environments, but this post helps put it in perspective. Thanks!
summary:
kubuntu distributes a product. people provide feedback on that product. One distribution developer disagrees with the feedback suggesting, “this product is not for you”
“Kubuntu is a KDE distribution.” Sure. Shipping with KDE 3 when version 4 is not ready yet doesn’t contradict that, though. You say “KDE 4 is the future of KDE.” I think we all agree on that. The only question is when to jump on the bandwagon.
I don’t know how many Kubuntu users are early adopters but even if it were a large percentage that doesn’t necessarily mean they want to apply that habit to their desktop environment. The desktop is the starting point for most actions and I think it should be stable, reliable and complete. I for one am using Kubuntu on my desktop PC because I want to have exactly that there. My playground is my laptop which has Debian testing/unstable installed but even there I’m still on KDE 3 and that probably won’t change until KDE 4.2.1 comes around.
Well said, Celeste.
Your story of the 8.10 release reminds me of what someone said that is so true:
“Try to please everyone, and you will end up pleasing no one.”
Those who are in love with their KDE 3.5, and upgraded to Kubuntu 8.10 without checking out KDE 4 first, were a little bit negligent, and deserve their disappointment, IMHO. KDE 3.5 is available on many Linux distributions — there’s nothing special about it for Kubuntu. But KDE 4 is the future of KDE, and it is to Canonical’s credit that it was a pioneer, offering KDE 4 early in its development to the Kubuntu user base. No one was “forced” — as Martin says, 8.04 users can stay with it for two more years will full support. I appreciate the efforts of the Kubuntu developer team to bring KDE 4 out for early adoption — thanks!
@dibl
Well, here does another problem rears its ugly head: The lack of seperation between platform and applications on Linux. Staying with KDE 3.5 pretty much means that you have to stick also with the rest of a possibly outdated userspace which in no way relates to KDE4’s current problems. At this point you can either hope that some developers provide binary versions for your favourite application x or your are out of luck (unless you consider exotic occupations like compiling).
In both OSX and Windows users can stick with a stable platform and still enjoy the benefits of updated applications until a newer platform matures. But since the repository model is considered a sacred cow by many users and there’s little hope that distributions could agree on a common stable baseline this problems will remain for an unforeseeable future.
Well said. I threw Kubuntu 8.10 on my Aspire One. While I’m not entirely thrilled with it (seems to be less feature-complete compared to the openSUSE and Archlinux 4.1.x packages I’ve used), I still think you guys made the right choice. 8.04 had KDE 3.5. It’s an LTS release. And it get “.x” releases in the meantime. Those unhappy with the current state of KDE4 can stick with a supported version while the Kubuntu devs can work on getting KDE4 ready for prime time. It’s only “lose-lose” because, well, some of your users not being able to make the right choices for themselves.
Let me preface this by saying I use Kubuntu all the time and am very grateful for the effort put into providing Kubuntu. This article is well written and I’m happy to see discussion. I disagree with much of the premise but please take my comments in this context.
“It is the purpose and goal of Kubuntu to support, package, and distribute KDE and make it easy to install and fun to use. If some Kubuntu users can’t get over that, then maybe Kubuntu isn’t for them.”
Easy, fun and if you don’t like it then **** off seems like a very contradictory point of view. The fact is that KDE 4 still has a lot of work to do on basic stability and useability without which the bells and whistles don’t mean a lot. Corrupted icon rendering, lack of KDE’s traditional comprehensive configurability in many areas, the list goes on. Completeness means a lot in a desktop environment and KDE 3.5 has all the bases covered in a way KDE 4 simply does not (yet).
I love Kubuntu and use it as my distro of choice, but I think not providing an official KDE 3.5 package in Intrepid was a huge mistake. Kubuntu is not a fully independant KDE distro, it is a KDE-centric Ubuntu as is clear from `lsb_release -a`, the release name, release schedule and so on. Now, if I want to move to Intrepid for all the implications it has for packages common to “pure” Ubuntu and Kubuntu then my hand is forced in regard to KDE (major) version. This is what people mean when they feel they’ve been forced into KDE 4; if you are a KDE user who’s not ready to accept KDE 4 in it’s current state, you don’t get to move forward with the rest of the Ubuntu family.
I don’t think it’s reasonable to even split individual users in terms of “Early Adopters” or not. I use Kubuntu at work and at home. At home I’m apt (no pun intended) to want to fiddle with all the new features and can some spend some time investigating issues, tweaking configs, etc but at work reliability and feature-completeness is vastly more important. I moved my laptop to Intrepid, but until either KDE 3.5 becomes a supported option or KDE 4 matures more I can’t justify upgrading my other machines.
There was a choice, people were happy. There is a mandate, people are unhappy.
Well, I am not a Kubuntu user, but I totally can understand the annoyed users.
Some here said, they are not forced to upgrade to Kubuntu 8.10 with KDE 4. But sure in the real meaning of “being foreced” they are not.
But practicaly they are: A Linux system is a lot more than just a DE. A lot of other parts of the software stack and a lot of enduser applications have evolved in the meantime. With 8.04 you are stuck with older versions or have to use some (propably harmful) 3rd party packages.
So when you (the one for who KDE 4 is not ready yet because of features you really need) only have two options:
1st - Update to 8.10 and get KDE 4.1 (which may not be ready for you) and the rest of the system updated
2nd - Stay with 8.04 to have a working DE (KDE 3.5) but an also otherwise outdated system
and well
3nd be angry about those bad options and move on to some other distrbution. And well, moving do a new distrbution means learning new things, having a day of work and so on. Is it so hard to understand that users are unhappy about their options?
If Kubuntu is really so user-driven as you wrote, and there is really a lot of demand for KDE 3.5 packages, why not help those who created the inoffical KDE 3 packages to fix them and get the maker of those packages more deeper involed in Kubuntu and then ship also KDE 3 officaly.
It is technicaly no problem to do delvier both. Well it is more work Kbuntu may not have the man power for that. But as there are already somehow working packages it should be possible to help those who created those the bring them to the quality standards Kubuntu needs.
Or in other words: If their is demand for KDE3 in a user-driven distro it should be possible to find users in the userbase to get KDE 3 packages back. But instead they (or at least Celeste) say you are not belonging to our target group anymore, Kubuntu isn’t for you anymore.
Btw: As a not Kubuntu user: I always though Ubuntu/Kubuntu aims not for early adaptors but more for the general “mass” (as much you can say mass in Linux on Desktops), an easy to use distro which just works, also for people without much Linux or even Computer experince at all. Well I thought that is what Ubuntu/Kubuntu is. And I bet a lot of your userbase though the same.
And now they only option if they want an otherwise up to date system (yes, never kernels support more hardare for example) is to use a DE which is declared for early adaptors only by Upstream. A distro targeting the “average user” should not do that, in my opinion.
Oh I forget one important point:
The graphic cards vendor with the biggest market share for private endusers (nvidia) has yet only a few days released a beta version of its driver which actually make KDE 4 really usefull.
With older drivers and a Nvidia GPU you are lost, with or without desktop effects you are getting a very slow bad experince form KDE 4.
So actually Kubuntu though it would be a good idea to make a DE the default one which is not yet supported by more than 40% of the users hardware.
Kubuntu does more harm to KDE than good.
Kubuntu ships broken KDE releases since years. The release version on Kubuntu 8.10 crashes constantly (mostly Kontact and KNotify) on my work PC. The beta 5 version of openSUSE 11.1 on my private laptop OTOH has a much better KDE experience. Another constant flaw of Kubuntu releases are its broken translations.
Kubuntu 7.10 and 8.04 looked like this: http://www.flickr.com/photos/19616885@N00/2535182834/
Now Kubuntu 8.10 looks like this: http://www.flickr.com/photos/19616885@N00/2991047111/
OTOH openSUSE’s KDE is almost flawless. The translations work, it doesn’t crash, and the tray is not broken: http://www.flickr.com/photos/19616885@N00/2991042741/
I can’t recommend Kubuntu to anyone. On the contrary. I advise anybody to NOT use Kubuntu — it just causes frustrations.
@blahjake
Easy, fun and if you don’t like it then **** off seems like a very contradictory point of view.
I agree. If the objective is to be fun to use, then they should be shipping the software that actually meets that description. And if KDE4 is not yet mature enough to qualify, then shipping it is somewhat at odds with that claimed objective.
I am one former kubuntu user, I switched to another distro faced with the fact that kubuntu would not support kde3.5. I did that because I rely on quite a few kde apps that are not stable in their kde 4 versions, most important KOffice and digikam.
I am sad that I had to make that choice, because I do want to use kde 4.1 or above, and I really like using kubuntu. I also need konqueror from kde 3, since there are things that does not work properly in the kde 4 file managers yet. I will probably be back when my key applications becomes available in stable kde 4 versions, though, as none of the two alternatives I tried feels very attractive after having used kubuntu for a long time.
Oh, why don’t I just stick to hardy then? because of missing up to date application versions. Most importantly, the outdated digikam/kipi packages on hardy contains a crash bug in one of my most used plugins. I even consider resinstalling hardy, I can compile digikam in a current version… Though one of my goals when I switched to a binary distribution was not to have to compile stable software versions.
I think Kubuntu has a hard time because its the gap between what it’s supposed to do (Ubuntu, just with KDE instead of GNOME) and what it’s able to do (ship a pretty vanilla KDE and hope that it doesn’t break on Ubuntu’s low-level foundations), that gap is just too big. I still use Kubuntu because it’s a genuinely community-supported distro and because it’s KDE-centric - OpenSUSE ships the better KDE distro but comes with the Microsoft/Novell issue, and Fedora struggles with Red Hat’s GNOME-centic development methodology like Kubuntu does with Canonical’s. I use Kubuntu because I believe that it shouldn’t take a large company’s support to ship a usable KDE distribution, but sometimes I doubt it.
Part of the high expectations come from the “official stamp” put onto Kubuntu by Canonical - like, “if it’s officially supported, it should deliver what it promises”. Maybe if Kubuntu emphasized the fact that it’s a community-maintained add-on to the more properly maintained GNOME distribution instead of a full-fledged alternative, it would probably lower user expectations to a realistic level. On the other hand, the question is whether such an acknowledgment would rather be beneficial (by gaining greater acceptance) or backfiring (by losing even more manpower).
In any case, if Kubuntu goes on like it has in the past, there will always be lots of disappointed users. Multi-monitor doesn’t work, bluetooth doesn’t work, printing is a PITA, and with every release some new major issue seems to creep up.
As sorry as I am, I have no hopes that the Kubuntu team will be able to fix the necessary stuff properly. All my hopes rest on upstream KDE - if upstream does it right then Kubuntu will (hopefully) work nicely, and if something doesn’t work in upstream then Kubuntu will fail too. But that’s also the nice aspect of an (unwillingly) vanilla distribution - when something works then I know that it works for all (or at least, most) KDE users, which is kinda satisfying.
Still unsure what I should do with my friends that are not yet ready for KDE… might make for a double-upgrade from Hardy to Jaunty. (Which would be nice if it worked smoothly when the time comes.)
That’s why I run Debian. When KDE 4.x is in the Debian testing or stable branch, you can bet it will work.
I still don’t know whether Kubuntu 8.04 is LTS or not. Some say yes:
https://wiki.kubuntu.org/
some say no:
https://lists.ubuntu.com/archives/kubuntu-devel/2007-December/002066.html
If it’s not LTS, there will be no official upgrade path from 8.04 other than to 8.10. Support for non-LTS 8.04 will end in October 2009, but the next LTS will be 10.04. In that case you shouldn’t recommend skipping the only supported upgrade path.
Short follow-up remark:
Yes, I really think that Kubuntu should position itself as cutting-edge close-to-upstream-KDE distribution, because that’s what’s actually possible with the current manpower. Trying to be all things to all people (even if you just focus on KDE 4) is not going to work with the current amount of resources.
Or maybe Canonical really does manage to fill in the missing pieces to make stuff work sufficiently (although it’s unrealistic that it will ever experience the same amount of polish than the “real” Ubuntu). Who knows, one can dream after all.
I think you’re probably right KDE is not for most of us ordinary users at the moment and I’ve been pleasantly surprised to see how far Gnome has come on and switched to that.
Having used KDE 3 for a couple of years I’ve tried to be an early adopter but too many times with KDE 4 I’ve spent a lot of effort only to discover that having an option available to change something doesn’t mean that that option has actually been implemented (I’m thinking saving sessions and hotkeys / keyboard shortcuts off the top of my head).
I just don’t have the time and as much as I appreciate the underlying technology I need something that works now not at version 4.1.6 or 4.3.6 etc. I just need to get my work done and I can’t convince other people that Linux and free software is a viable alternative to Windows if I’m forever saying that things will be ready soon and it’s something else will be great when it’s finished.
Though temporarily losing features when making huge strides forward may be a part of advancing computer software the more popular Linux becomes the more people there will be who don’t understand why upgrading to a new version leaves them with things not working that used to; print screen being one example (apologies if I’m out of date with this but I gave up a couple of versions ago).
As a couple of folks have said if Kubuntu is meant to be a bleeding edge development distribution then please say so and avoid the confusion. But to cut people off from other new developments in the underlying systems of Ubuntu just because they are a bit conservative with the desktop they choose to run (3.5…) does seem contradictory to the stated aims of this particularly distribution.
The whole “KDE 4 with Kubuntu” thing is just another example of people deciding that exciting, shiny stuff is more important than the stated objective of giving people a decent user experience. Why do people like myself want to run KDE 3.5, not some “edgy” stuff like KDE 4? Because we’re not experimenters who, if our desktop goes away and we end up at the login prompt, treat that as a fun challenge and will hack all hours in one of the least fun languages in order to return the desktop to a functional state; we’re people who want to keep all our mail, our pictures, our bookmarks, our files…
My solution to Kubuntu’s inappropriate strategy: stick to a supported release which has had the benefit of being tested, patched, fixed; ignore the half-finished, rushed-for-an-arbitrary-deadline latest release with all the toys everyone wanted to play with.
I wrote what I think about this post and this situation in form of a post of my own:
http://djdarkmanx.blogspot.com/2008/11/kde4-desktop-that-some-distributions.html
In summary I don’t agree with this post because it misses the point on every topic.
I also think Kubuntu 8.10 is a lose-lose release.
The KDE 4 desktop is incredible broken and unusable if you are used to the comforts of KDE 3. There are basically NO appealing applets for the panels (just some stupid clocks and twitter plasmoids for the desktop) but overall it’s just missing too much to be on-par. And that’s me - a power user.
For now I am trying Gnome as my Desktop (with 8.10) and use some KDE 4 applications.
But to be honest - I am pessimistic that the KDE 4.2 desktop will be on par with KDE 3.5 and will bring me back.
Please keep in mind what “Ubuntu” means and what the philosophy of Ubuntu is. The distribution should not be for early adaptors but for “Human beings”.
I think Kubuntu definitely has lost this tag. My girlfried uses a KDE 3 desktop and I would never dare to switch it to KDE 4 as of now. She is also not an early adaptor.
And I wonder why you did not just kept the Kubuntu 8.04 infrastructure to allow easy parallel installation of KDE 3 and KDE 4 - it just worked! This could be kept until 4.2(.1) or 4.3(.1) is out and this way the “Human beings” would not be thrown out of your target group.
Well in the end I’ll probably look for another distribution for me and my girlfriend in the next time.
Sorry to rant but you should realize that this is a way to give feedback and I think it’s better than just to be quiet and ignore all the flaws.
A human beign.
First of all : When an distribution works for you : Don’t change until you buy a new computer ;-) It saves you time and misery.
When you eventually want to change : do a try-out first ! If you see you cannot cope with the imperfectness until they are fixed : fall back to the old one.
If you can cope with the imperfectness of the new version : well think twice if it offers you the advancement you need against the possible hick-ups and if your bussiness is crucial for you :stay a little longer (until the next version) with the old before you change.
If everyone follows the above concept we don’t have any worries.
The problem is those users who are complaining didn’t follow it. Sorry for them : they seeked the trouble themselves. Always listen to the more experienced persons if you aren’t capable to judge yourselve.
Registered Linux user (old#1126, new #22996) linuxpioneer since 1990
Not sure what to make of the blog post, but then again I don’t really understand people’s resistance to the current KDE 4.x stuff either.
I am part of the early adopter crowd, at home it’s primarily Debian unstable with some bit from experimental, Ubuntu as a secondary distribution. The machine I do stuff on during the downtime at work is Ubuntu. In both Ubuntu installations I have been pulling from the Jaunty archives for the last 2-3 weeks.
Individual cases where there is instability, features in KDE 4 apps that aren’t supported yet that require them to stick with the KDE 3 version, I understand the resistance, the users needs are the users needs, but the amount of resistance seems out of proportion with that.
I’ve been primarily a Gnome user since the early licensing issues before Qt was relicensed. This is primarily because of the more loose knit development that, as a user, gives the appearance to me that outside projects have an easier time working with them dbus, hal, compiz, etc…
When I first started reading about this plasma stuff I was a bit skeptical and apprehensive, but at the same time reading about Aaron’s experience watching how people work and how that affected his vision for the desktop, it was the first time in quite a while I’ve been excited about the desktop.
4.0 was rough, but it laid my apprehensions to rest, as I see the stability improve, aesthetic improvements, see the pieces falling into place to match the vision, I continue to be excited. I can still use kmix as my mixer in Gnome, use a different window manager in KDE, etc… so I’m happy, and these KDE 4 releases are the first KDE release where I am content to run as is without changing window managers, running a Gnome top panel and KDE bottom panel, etc…
While I am perfectly content with KDE 4 the attitude expressed by others would seem to indicate that in the same way the KDE 4 desktop was previously completely installable in parallel to the KDE 3 desktop as blah-blah-kde4 packages for the early adopters it would be good if KDE 3 could be completely available and installable as blah-blah-kde3 packages for the stragglers in the upcoming release.
Later, Seeker
@Antal Istvan Miklos - Very good post. I agree in almost all points (I just had not the time to try 4.2 SVN)
@Patrick Op de Beeck - Well I installed a new system and used Kubuntu 8.10 in the assumption it would be usable and any good. But I was wrong. And if this was some flaky niche distribution (as it would be a distribution from packagers for them self) I would accept that they make some experiments.
But this is (K)Ubuntu self-claimed “Linux for the masses” and “Linux for human beings” so there are certain expectations. For one, that the USER is in the center of deciding what to do - and that would be providing KDE3 AND KDE4.
openSUSE does it the right way.
I have a theory :)
First off, I don’t think KDE should be mainly for early adopters but *also*. At the moment it’s for early adopters by necessity (we would like to have the time to fix all the problems…), not by choice.
Celeste, you do usability and you like KDE. Maybe saying that KDE is for early adopters is a way for you to resolve a cognitive dissonance a la “I like usability and also KDE”…
Put another way, I think you are arguing based on the practice today, not on the theory of what we’d like to do.
Now everyone should be sufficiently confused :P
I chose installing Kubuntu 8.10 especially to try out a more recent version of KDE 4. I’m a long time KDE 3.5 user and I run SUSE since, well, ever. Kubuntu being a KDE centric distribution I thought it was the natural choice. I did it after trying KDE 4.1.x packages through the OpenSUSE build service because I thought that a spelled-out KDE distribution would provide the best possible experience.
I’m still undecided about this. There are aspects about Kubuntu that I like, but it was also a sobering experience. While I applaud the Kubuntu team for having the guts to shipping KDE 4 as the primary desktop, there are a number of annoying bugs which I don’t know quite know whom to attribute them to. It’s not a lack of features - I’m enjoying the new concepts of KDE4 quite a lot, especially the Plasma desktop, folder views, compositing effects, etc. I agree that there are not too many exciting Plasmoids yet, but the concepts are so very promising.
But how can a distribution ship with broken Bluetooth support? Or printing - you have to manually add the printers because it will not automatically pick up CUPS servers.
And so I still don’t know whether I have the best possible KDE experience. I have no reference, so I just don’t know if something is supposed to work and it’s just the distributor who shipped a broken package. Who would be able to answer this question?
I believe that some level of certification is necessary to avoid confusion. The KDE project needs to work closer with packagers and distributors and define a level of functionality below which something must not be called KDE.
i´d linke to add some points that are not (yet) being considered right now:
i totally understand that there are people (read: users) being frustrated with the current situation.
and most of us will agree that the current state of kubuntu ibex is not polished/stable/whatever you wanna call it.
the kde project made a real big step in releasing kde 4 in january 2008. since this big step is a complete rewrite of the desktop (in contrary to the greater number of smaller updates in gnome) they had (and still have) to decide which way kde4 can be improved and made available to users.
i think there are 2 ways of doing that (with some compromise ways in the middle):
they could have released 4.0 and 4.1 as a release candidates and rely on the people which are confident testing those releases. and by now we would have something like a 6th or 7th release candidate and strive for a final version for….june/july 2009.
the good side of this approach is that there wouldn´t be any frustrated users. but the downsides would be that the progresses wouldn´t have sped up (as they have), and that there´d be the possibility that kde 4 could be outdated even before being released.
the other way is to release what is there even if its not perfect and get a much bigger amount of actual users, increasing the frequency of bug reports…and fixes, getting some attention in the media (and the industry), and finally getting hit with critizism why the product is not really finished.
so both ways have good and bad sides. and i don´t think there are many ways in between.
this had to be decided for the kde project itself as well as for kubuntu (whether or not update to kde 4)
don´t get me wrong i don´t say that these decisions were right or wrong.
and if you ask me, i think that there´s very little use in saying something like: “kde 4 is complete crap, they should´ve waited, there should be a kde3 version of ibex…”.
we can´t change the past, but we CAN change what´s going on right now.
so if you can´t live/work with a system that´s less than 99% stable then you could help making it better (which i think is way better than flaming) or it would propably be best to watch out for other distributions (or flavours of ubuntu) that fit your needs and watch back if the time´s right.
i decided that (at least for now) i can work with kubuntu ibex being a bit buggy and unstable (even as a software dev.) and that i am willing to help improving it.
just for the figures: as far as i know there a 2 full time developers working on ubuntu and only 1 working on kubuntu, which makes comparing those 2 difficult at least…
thanks to every single person that helped making (k)ubuntu what they are today!!!
can´t stress that too much
witti
kde4 is definitely not the future..more like a poorly implemented copycat of the windows vista interface. It’s sad to see the kubuntu users left with such a unusable desktop. There are many informative posts here, hopefully someone listens. Tuxo, good point it doesn’t seem to fit the ubuntu philosophy of linux for human beings.
To cut a long essay short, Intrepid has been a minefield for me, nvidia, -rt and m-audio issues mostly, but I kinda expected this after edgy. The KDE4 side of it has been awesome however, and its the only reason I’ve migrated my desktop. Everything else is staying LTS for the moment.
Maybe the problem is KDE4 is so seductively pretty… But I’ll put money down that the same users that are clinging to KDE3 will be the loudest objectors once KDE5 comes around.
Well if people want KDE 3.5.9 they can stick with the previous release and update the kernel or any new packages they want. It might be Ok to include KDE 3.5 until the end of the year but there is no point in sticking to it anymore. KDE4 is where the KDE developement is going and it is a perfect stance to start focusing in KDE4 and not in KDE3, by the way the latter has already given all it has. KDE4 developement started with new technologies to adapt the environment to modern times or technologies, as of now I am using version 4.1.3 and I love it, I also understand the people who still do not feel like jumping to the new version but there is not point in forcing everyone else to not make the jump. I also use Windows and still have XP, which is 8 years old, I dont have the need and dont even consider switching to Vista, like most of the people who uses Windows. As I said before that OS is 8 years old. How come people have the urge to upgrade from an OS released this very year (8.04) and want developers to stop working to integrate the new technology and contribute to a faster improvement but instead want them to work on something to become obsolote in the developement point of view
Right now when people complain that Kubuntu 8.10 is not usable, they are told “8.04 is supported, stick to that”. But what will happen when 9.04 approaches? What happens to people who want to upgrade to 9.04? At that time, they will probably be told “you’re on 8.04, upgrading to 9.04 is not supported”.
I’d certainly like to be an early adopter, but does that require having to give up basic things like auto-hide panels and bluetooth?
Celeste, is a taskbar that insists on having only one row regardless of how tall it is and how many windows are open supposed to be usable?
I totally do *not* agree that KDE is for early adopters.
However, in my opinion Kubuntu 8.10 is the right thing. Because non-LTS versions of Ubuntu *is* for early adopters. Since Kubuntu equipped so many cutting-edge packages and kernels why object the the newest version of KDE4?
As a 3-year ubuntu user, I know ubuntu actually is a very unstable version, don’t expect a flawless ubuntu out-of-box when it releases, My 8.04 is not working well until all upgrades up to October, that’s 6 months after it releases.
So for all ubuntu or kubuntu users: don’t expect any of your ubuntu works well if you install it at the same month it release, instead wait for five or six months and most issues solved and you’ve got a good linux distribution.
I am a happy user of kubuntu now, because I install 8.04 when 8.10 releases, and I will install 8.10 when 9.04 releases, never trying the newest version of kubuntu and I see kubuntu a decent distribution, I mean it.
Well Celeste, you are a primary usability voice for Kubuntu and judging from the comments above and our early adopter user experiences, Kubuntu 8.10 was not ready for release. And yes, KDE 4 is great now and becoming better.
What the Kubuntu team have totally missed is that there is a basic set of functionality which is the minimum for a successful release. The ability to connect to a network with any interface, add printers, see other computers on the network, share folders, load proprietary drivers, boot successfully every time. Seamlessly and easy for standard users. These are all broken in Kubuntu 8.10 and all work in Ubuntu. You can’t even ping other computers due to the existing setup. Knetworkmanager has been broken for years, etc. The list is too long.These are all very basic and what really creates a bad user experience if it does not work.
Kubuntu has to decouple from Ubuntu for releases. Kubuntu has to change its decision making model for what has to work in any release. That is the only way it can regain credibility with users who has been supporting the distribution for a long time. Making philosophical excuses don’t cut it.
The first question that pops up in my mind is: how much did the KDE team have to loose if they released KDE4 next year? Was it really so urgent to do this in 2008?
This rushed release reminds me of Microsoft’s bad move with Vista. Btw: KDE4 smell like Vista experience.
Although I use Debian, I usually test all kinds of distributions. And my clients get new stuff too. But this time, I just couldn’t give them the new stuff with the updated/upgraded softwares.
It is quite annoying that KDE has advanced photo processing software, office suite, and Kopete still cannot receive files from Yahoo Messenger users. I guess, everyone who has internet, uses instant messaging. It became a vital thing. And strangely it is not on top of priorities list.
As some users say, it is a forced transition due to software deprecation in KDE3.
I am also not convinced that Kubuntu development is user-driven enough. People who work with FL Studio music production software since the beginnings can tell you what user driven evolution means.
Dear KDE team! I appreciate your efforts. Without you, we would still not know what does it mean the experience of using an advanced and free desktop environment. Please try to learn from the users’ reactions you are facing in this period and try not to repeat this mistake.
Free software’s quality may not be the victim of some sort of greed for success or exaggerated competing instinct. In this world everybody began to run, to hurry to release his product without caring enough about clients/users.
In this period free operating systems have a very big chance to recruit new users because of Microsoft’s politics and mistakes. But with a problematic desktop experience these people are kept back and put in stand-by. And I tell you, if competition releases a better product than before, people won’t care that much of free but problematic softwares. We all tend to make a compromise and pay for quality products.
I am an early adopter KDE user who uses Kubuntu. Your article is very interesting. However, regarding Intrepid - I’ve now upgraded 4 boxes and every one of them breaks under when using KDE 4 (causes X to crash). Running Gnome on Intrepid is fine. So for now I’m having to slum it with Gnome. I really think Kubuntu needs a lot more testing and polish, and it seems to be getting worse with time. On the up side, when there is a new KDE release, the Kubuntu backports are second to none in release time.
I’m a Gnome user, and have been trying out KDE 4.1 on Intrepid this week. I agree with everything you said, especially about early adopters and the support of alder, stagnant releases. My Kubuntu 8.10 experience has actually been rather pleasant, and in my opinion, the Kubuntu team has been making the right decisions and been doing a good job.
KDE just happens to be in a bit of a difficult state at the moment. You want to make an omelet? You have to break some eggs :)
“At the time when decisions were made, there were still some bugs we were concerned about in 4.1.2. In an attempt to support our users, we temporarily disabled some not-quite-working functionality until it is fixed in 4.1.3. ”
I assume you mean the zoom in/out here? I agree it’s not perfect, but it feels rather strange to read someone claiming that KDE is mostly used by early adopters (and as much as I try, I can’t understand where you got that idea from) and at the same time removing a feature said early-adopters would be interested in trying out.
However, that’s only one of the problems with Kubuntu Intrepid. It’s general bugginess is worse. The new adept is so slow is is next to useless, the crash-handler crashes, plasmoids that don’t stay where you put them, and the list goes on.
I’ve used Kubuntu since Feisty, but it is impossible for me to continue to do so. Now I’m using OpenSUSE 11.1, which is still considered beta. I’ve even updated it to a pre-beta version of KDE 4.2, and it is STILL more stable than Kubuntu.
KDE4 is the *FUTURE*, but is nowhere near the functionality KDE3 provided.
As long as KDE4 does not match feature-for-feature KDE3, it is vastly inferior.
Very very well put, Celeste - couldn’t have said it better!
The sad point is, that the KDE Team seams not to concentrate on the basics in usability (Taskbar issues i.e.), which once made the KDE 3.x tree such a smasher. But this is something to blame on KDE, not on Kubuntu. The good side of Kubuntu shipping KDE 4.x is (beside being on the technological edge), that a huge group of people get a view into KDE4 for the first time - and of course they will mention what they miss / do not like /expect. That’s a chance for the KDE team to listen closely and make KDE 4.2 brilliant - hopefully.
I can’t judge over the quality of packages - as I haven’t the time to look into other distros at the mo. But if the package quality is in fact such of a problem as some comments state - then that should be a phat hint to Canonical to think over the developer situation (which seems to be like anyone for gnome / no one for KDE). So again thanks for your time to pointing things out so clearly and so detailed :) Regards, Herr Irrtum
So you can’t please both user groups with the same decision, those that want KDE3 and those that want KDE4 (surprise !). And those that want to stick with KDE3 nevertheless want to dist-upgrade to get all the other updates (re-surprise !).
But why oh why do you have to put *either* KDE3 *or* KDE4 in your distro ? You had planed to support both anyway (in different kubuntu versions). And they can be installed alongside each other quite easily. Please everybody.
When reading such stories I’m glad to stick with gentoo. So many problems disappear when using a streaming-release, compile-locally distribution.
@Pan, Shi Zhu
Why wait for months? Just use another distro. It’s embarrassing for (K)Ubuntu that its final release version is less stable than a beta version of openSUSE.
Where on earth have you got the impression that most KDE users are early adopters? I’ve been using KDE since, hmm, 1999 or 2000 or so, and my reasons for doing so has always mainly been: stable, well thought-out UI, and the Konqueror file manager. I switched to (K)ubuntu after using Gentoo for a year and starting to long for a stable distro that is easy to keep up-to-date.
I’ve just started to get my girlfriend used to Linux and KDE 3.5 (in Hardy) instead of Windows, and I myself mostly use KDE 3.5, even though I’ve installed 4.1 and use it once in a while. But I will quite frankly not upgrade to Intrepid precisely because the desicion to not include KDE 3.5. Unless 4.2 is a *huge* step up in usability from 4.1 (which still frankly feels like a pre beta version — it’s slow, unstable and lacks a lot of basic functionality) I can’t see myself switching to KDE 4.x until at least the 9.04 release of Kubuntu. I may even decide to switch to another distro so as to get access to the latest versions of apps and the kernel but be able to stick with KDE 3.5.
@DanielW
“1st - Update to 8.10 and get KDE 4.1 (which may not be ready for you) and the rest of the system updated
2nd - Stay with 8.04 to have a working DE (KDE 3.5) but an also otherwise outdated system”
Oh please, stop saying such lies. When 8.10 is out it doesn’t mean that 8.04 is automatically so much outdated. 8.04 is still supported with security updates.
Besides, 8.04 will be supported _longer_ than 8.10! 8.04’s support will end in April 2011 while 8.10’s support will end in April 2010. In May 2010 8.10 will be outdated and without support, while 8.04 will still be supported.
@Herr Irrtum
Please if you don’t know what KDE is, don’t talk about it, talk about other Desktop Enviorments that you use, and know what they are about, I’ve seen more usability improvements for KDE4 in the last few months, than I’ve seen for GNOME in the last few years.
But if you want to prove your point you can try to convince me, which is impossible, because I actualy read what the developers had to say, opposed to most people who rant about KDE usabilty.
KDE 4.2 will be the initial version for crying out loud, next time RTFM and ask the KDE developers and KDE users before making up things.
@Andy
“I really think Kubuntu needs a lot more testing and polish, and it seems to be getting worse with time”
That’s what I’m trying to tell, Kubuntu needs a lot more than “wget kde.org/the-latest-rls.tar.gz && tar xvf the-latest-rls.tar.gz && cd the-latest-rls && mkdir build && cmake .. && make && sudo make install” they might as well cancel the whole kubuntu project and provide a shell script for that.
@Bocskai Csaba
“This rushed release reminds me of Microsoft’s bad move with Vista.”
Ask the developers before making stupid statements like that.
“KDE4 smell like Vista experience.”
Can you back this up or you just having some random taughts?
“It is quite annoying that KDE has advanced photo processing software, office suite, and Kopete still cannot receive files from Yahoo Messenger users. I guess, everyone who has internet, uses instant messaging. It became a vital thing. And strangely it is not on top of priorities list.”
Yahoo! Messenger is a proprietory protocol, tell them to open it up, or use skype, and as I tested, neighter does pidgin work right.
“As some users say, it is a forced transition due to software deprecation in KDE3.”
And can’t you run KDE4 software in a KDE3 desktop? where’s the problem?
“I am also not convinced that Kubuntu development is user-driven enough.”
Do you know how hard it was to make this big jump into KDE4 for the sake of better user expiriance?
You clearly don’t know what you are talking about, next time, do a little research.
@jo
“kde4 is definitely not the future..more like a poorly implemented copycat of the windows vista interface. It’s sad to see the kubuntu users left with such a unusable desktop. There are many informative posts here, hopefully someone listens. Tuxo, good point it doesn’t seem to fit the ubuntu philosophy of linux for human beings.”
Again, the KDE4 looks like vista thing that everybody brings up but noone backs up. If not KDE4 is the future than what is? The current GNOME desktop, don’t you think it will be a little boring in a few years?
The conclusion is this:
Everybody who speaks about KDE4 should do a little research before making up nonsense, that they can’t back up. I think that the KDE Developers, would be more than willing to answere questions related to KDE4.
One of the main problem is that (ohhh know here it goes again) KDE is a second class citizen in the ubuntu world, Mark Shuttleworth couldn’t agree with the KDE devs on syncing the release cycle, with a little logic, you can add up, where is this going to. If the Kubuntu team can’t cooperate with the KDE release team, then Ibex is not the first garbage the Kubuntu team is gonna deliver, Ubuntu has a release cycle and KDE has a release cycle too, the swarm of open bugs in Ibex not related to KDE, wich were present in the Beta testing, and remained there after release prove, that maybe the almighty 6 month release cycle is not that perfect. Consider this before telling everyone that the KDE releases are agresive.
@Antal:
“KDE4 smell like Vista experience.”
Can you back this up or you just having some random taughts?
Just look at all the controversial reactions from users.
So some people just “love” Vista and others just say that is was released to soon and does not show the 8 years that passed since XP.
Same for KDE4 (as of now!), where there is a big controversy in the user base. Some users just love it where other think that maybe 4.2 or 4.3 is the first “proper” release (like the Vista SP1).
Vista provides much more “bling” compared to XP and some usability improvements, while UAC is annyoing for most users. KDE4 also has increased “bling” and FOR THE USER besides that not much more (as of 4.1) but more of a decrease in functionality.
Vista brings DX10 and tries to enforce .NET, where KDE4 brings Qt4.
Now of course Vista is mostly a NT4, where as KDE4 is a (nearly complete) rewrite. So from a developer perspective things changed. Using Qt4 gives a much better basis and core developers as well as application developers benefit from it in the long run. But on Windows you are now encouraged to use .NET and WPF which also make many developers happy.
Yes, I am polarizing here. And I know that KDE is FREE with all the nice implications about it. But this is not what the discussion is about…..
@Peter
I agree with your point, Linus Torvalds also said, that the users don’t like changes, and he was right.
But a second thing is, users demand is increasing, users want widgets, users who are not gamers want desktop effects(please no rant about kwin, ask me before), users want instant messaging, users want desktop search, users want a more shiny gui. Many can dissagree at my point, but just think about it, would you like to go back to KDE2? or in the Windows world would you like to go back to Windows 95? Look at what OSX does, Microsoft can’t fall back while Apple is winning, and KDE4 can’t stick to the good old KDE3 way forever, if KDE would have remained with the old way, it could have no chance against OS X and Vista, on the battleground of features. For some this might come as a suprise but, GNOME will also make a big leap, when GNOME3 comes out.
Most of the KDE3 options are already back in KDE4(SVN) and it’s getting to the point where it will surpass what KDE3 is capable of, just look at what plasma can do(besides crashing randomly once a day).
@Antal:
My post was not against KDE4, but just to emphazise that KDE4 is *not* ready for the average “human being” at the moment, even less in the Kubuntu 8.10 incarnation.
And yes, “bling” is important for the desktop of the future. Average users like it and when it’s non-obstrusive maybe even power users start to like it. (Like the window search in the kwin4 “Expose”).
I think the basic functionality like desktop search, IM, Mail, Browser etc. are out of question and where available under KDE3 too.
My personal (and most prominent) problem with KDE4 is actually Plasma (as you mention it). I can’t see any use in widgets other than a one time two minute smile. All the important information where in my KDE3 kicker panels, like Weather, CPU/Memory Graphs, Search, Calculator, Color Picker, ….. .
I will try KDE4.2 soon but of what I heard and seen it does not look like that much has improved for the PANELS. The main focus to me seem to be generic widgets like huge twitter apps, huge weather panels, big clocks, big comic or photo display things, well acutally only stuff that can be placed on the desktop as a widget and nothing that can be placed on the panel without looking ridiculous. (Look at http://techbase.kde.org/Projects/Plasma/Plasmoids )
I do not say that we will not end up with a usable desktop in the future. But for now I personally don’t see how this will happen for MY use-cases - at least not until 4.2 or more likely 4.3. All this “reinvent the Desktop” does not seem convincing. For now there is nothing new. At least nothing that we have seen with Dashboard for years. There is just a broken panel compared to KDE3 with a spicy SuperKaramba (and some may claim that there is a better architecture with plasma).
And yes, KDE4 *is* the future. But it will only be a centrum of gravity against OSX and Windows (and Gnome), if the user is put in the first place. Even if it is open source and most of the work is volunteer. Either it should be FOR ALL users or for the developers ONLY (and some early-adaptors who accept everything). Most of the time people say negative things there are just justifications and very seldom we can see collaboration or acceptance on either side [Plasma again is a very good example for this].
Sorry, I could not resist to write this :-(.
I hope the point where KDE4 will supass KDE3 will come soon. And there are alot of very capable developers working in the direction. So despite my rant, real kudos to all of them.