Interview: Aaron Seigo
There are a number of people who felt bitten by KDE 4.0, and who decided to stick with KDE 3.x, or maybe move to an alternative. Since KDE 4.2 is supposed to be ready for everyone – developers, enthusiasts, and normal users alike – what would you say are the defining differences between 4.0/4.1, and 4.2? Because of what improvements should the aforementioned users try KDE 4.2?
I actually sort of answer that as part of question 2:
What feature or improvement in KDE 4.2 are you personally most excited about?
There are two feature tracks in 4.2 that I find exciting: the feature parity track that was focussed on filling in the missing feature of the earlier KDE 4 releases, and the forward looking track which took the unique parts of KDE 4 to the next level.
Both the desktop shell and file management have reached feature parity with the 3.5 releases with KDE 4.2, which many users will appreciate. Things like detailed file previews on mouse over in Dolphin and Konqueror, panel autohiding, printing to PDF and Postscript or good integration of Ark for dealing with compresesed file archives. Up until 4.2 there was something of a compromise some users were making between the new features in 4.2 and the features they knew and loved from 3.5. Those compromises are pretty much a thing of the past at this point.
Which in turn should make appreciating all the new capabilities and polish a lot easier. Akonadi is hit a really good point in 4.2 and applications can start seriously using it. I’m already using it for some of my PIM data, and it seems pretty solid in day to day usage.
KWin’s window effects have gotten very smooth and even more useful in 4.2, such that showing KWin next to MacOS’s window manager is something we can do with a straight face. It’s not just in-your-face-bling, but also very useful and slick (read: often refined and subtle) effects.
Also, not specific to KDE, but nVidia just released a new x.org driver that addresses all the known performance issues with KDE 4.2 so one of the biggest sources of discomfort for users running a full KDE desktop session has also been eliminated. That can only be a good thing.
This isn’t specifically related to KDE 4.2, but a common question being asked on OSNews and other websites has to do with themeing. Is there any specific reason why Plasma and the rest of KDE4 use seperate themes? Was it a conscious decision to not “couple” Plasma themes to ordinary KDE themes?
The theming in applications tends to be limited to icons and things like push buttons, menus, etc. while in the desktop shell the theming needs to cover a much wider array of us cases, from taskbar buttons to system tray areas to notifications. These are things that don’t have great analogs in most applications. So we had to sit back and ask ourself how we wanted these things to look, as it wasn’t a pre-made given as it tends to be when writing a regular application.
While Plasma does share icons with applications, we decided to make the desktop shell components somewhat distinct from applications so that there is a distinction between “things that are part of the environment” and “things that run inside that environment”. By making it possible to create a visual separation, we’re able to see how/if that distinction is helpful to people. It’s a way of putting more emphasis on applications as being items of clear focus.
Of course, it’s possible to make Plasma elements that look exactly like the traditional widget themes and the current defaults are meant to compliment the Oxygen art seen throughout KDE. So if it turns out that we do want everything to look homogeneous again, that’s quite possible.
Looking ahead – what will be the area of focus for KDE 4.3?
Akonadi usage is one area that will be worked on in 4.3. We already have some work on accessing your mail, contacts and calendaring in appropriate areas of the 4.3 desktop shell being worked on with nice progress. So I think we’ll see more and more integration of PIM data outside of the traditional PIM apps, as well as a lot of new work on those traditional apps such as KMail in Kontact.
Nepomuk is also finding its way into more and more places. It’s set to show up in the file dialog in 4.3 and more applications are picking it up as part of the workflow as well. So I expect to see more context and searching goodness in 4.3.
A wildcard entry for 4.3 right now is integration of community and social services with applications and the desktop. The Plasma will be focussing a lot of our 4.3 energy on exploring how we can open up doorways onto and for the community through our software. Community is really the most unique strength of the Free software desktop compared to proprietary offerings, but to date we haven’t really taken advantage of that. We are looking to fix that by using the proposed Community Web API on Freedesktop.org and making the desktop a more social place for those who like to use their computer for such things. This isn’t going to be yet another “aggregate your web content on your desktop!” project, though; while an interesting use case, it’s not one that seems to be very compelling for many users and we feel that there is a lot more potential there. There’s a reason the social websites are popular with people, and it’s not because they agregate each other’s content. 😉
While we usually have a few “big picture” things we work on in each release, there’s also the wave of general improvements across the board. We’ve been consistently seeing 350-450 commits a day for the 4.2 release time period and that doesn’t seem to be slowing down any. So we can expected to also see continued progress on the feature already there, from mapping and geolocation with Marble to grouping/filtering of desktop notifications to improvements in image management… And so on.
KDE 4.2 is a really nice release and the directionality we see for the next releases promises to keep that pace.
Where to get KDE 4.2
KDE 4.2 will make its way to your distribution of choice soon enough, but if you can’t wait, you can always build it from source yourself.
Big Thanks to all the developers and contributors who made KDE4.2 happen and a great release for the 4.x series.
Please can you not keep going on about 4.0?, it’s not like you spend good money on it and HAD to try it or use it. The 4.2 release shows how far the 4 series has come, I personally love the 4 series and contribute because of it.
Many people didn’t seen it for it’s potential(instead banged on about where’s my KDE3.x features), I did see it’s potential.
Edited 2009-01-27 19:27 UTC
No. I don’t think we should ever forget it. KDE 4.0 was a colossal disaster and it’s a great model for open source projects to learn how NOT to do things. Everyone now knows when you release something and expect the users to do alpha testing on real stuff, they will revolt. You gotta get your stuff up to snuff, at least mostly functional, or risk angering a large chunk of your userbase.
Not that I don’t respect the KDE guys or what they are doing, and I am not upset about it, I’m just saying, we shouldn’t just move on. I don’t think users want to have to learn to wait until x.2 until something is usable.
We all learned a lot from this experience, and thankfully, it looks like the KDE team is actually going to realize much of their vision and win back many of their fans. It’s a story with a happy ending.
On a related note, Aaron Siego, for all his bytes, is a great face for KDE for the tech folks. He’s interested and he’s substance. People respect that.
If KDE 4.0 had not been released back than, we’d hardly be at a 4.1’s level by now. To develop and enhance the software you actually need users to use it, spread the word, encourage people to give feedback, find bugs, and point the most missing parts. You gain contributors that way too.
By hiding your project and labeling it WIP you’ll hardly get the fraction of help and interest. It’s not like KDE could have waited till now and release 4.2, it’d be in a much rawer state.
By seeing what we got today, I’d say it was worth it. It did no harm, but instead produced a great DE in reasonable time window. Besides, it’s not like KDE 3.5 disappeared after KDE4.0’s release.
Counterpoint: it did lots of harm. Lots of people ditched KDE, lots didn’t use 4.0, and the team, I think, still suffers from some users who feel burned.
Furthermore, it’s erroneous to say that if they hadn’t released, they wouldn’t be up to 4.2. First of all, it wouldn’t matter the version number if it was actually complete and worked. Secondly, why would you assume the work would go slower? Because there are fewer people using it? So you are, in fact, advocating the release of incomplete, buggy, untested software to users? Just checking.
If I realize an empty zip file and call it an OS today, I can’t after 4 years or work go back and later say “had I not released that zip, I wouldn’t be here today.”
KDE 3 wasn’t gone. It never was. It’s still here now. If you didn’t like 4.0 (or 4.1), you could have stuck with 3.5.x or used GNOME, or XFCE, or some other environment, so basically your point is moot.
Sure, you tried it and didn’t like it. That’s why there’s choice. Hell, you can even run several DEs side by side.
Stop complaining, or help out to make it better. Or fork it and make it better. It’s not like you have to pay for it or something.
I don’t think it’s as clear cut as you make it out to be. x.0 from the beginning of computers has signified that the developers think the release is ready for general consumption. KDE made a 4.0 release and expected everyone would read the blog posts about how it should still be considered a development version. No one did apparently. I don’t think you can fault the users in this instance. KDE could have released 4.0 as beta4.0 or something else like that and continued on with beta4.1, but they didn’t.
The whole situation probably *has* hurt the project a bit too. People hate it regressions probably more than anything else. I know a few different people who stopped using ubuntu completely (including myself) because of the whole premature pulse audio fiasco.
Considering the fact that they had to be following development to know and to install kde 4.0, it isn’t much to expect them to follow official statements by the devs or even read the release notes.
Consider this situation:
If you buy a product and you only look at the model name and number and have certain expectations of the product. The instructions come with several warnings, you don’t read the instructions. You then try to use the product and are disappointed. Whose fault is it? your’s, or the makers?
Instead of listening to the demagogues, you could expect people to maybe pay a little attention to what they are doing.
Nobody forced them to use 4.0
Edited 2009-01-27 21:49 UTC
This dead horse may be beaten to pulp as well but you conveniently forget to mention that the general announcement on kde.org, the one which users will most likely read, didn’t mention the sad state of 4.0 with one word. Zilch. Nada. No mentioning of 4.0 being a development release or only being suitable for very adventurous users who are aware of the development leading to the release, no word about the desktop having less features than Packman and major applications still missing.
There’s no way out of this 4.0 disaster. The KDE Ministry of Propaganda (a.k.a. KProp) failed on so many levels when communicating to the general users that it’s not even funny anymore and can solely be used as a bad example on how to *not* handle public relations. Fortunately, although they never openly admitted their failure the announcements for the next versions were a little more carefully phrased so internally some kind of learning process must have happened.
Edited 2009-01-27 23:50 UTC
Kernel 2.6.0 anyone? And a lot more .0 releases.
The problem are not .0 releases that have a lot of problems, but rather that “normal” users get these by their distributions — why that is the case would need some further investigation.
I also do not like PulseAudio and think it should rather be an option than the default right now as I have personally no gains though something that worked for a few years makes problems here.
You could say 4.0 falls in the same category and that might be true (I did not use 4.0 myself I sticked to 3.5.X) the difference is that with 4.2 KDE 4 is for me perfectly useable while that is not the case for PulseAudio that is “forced” upon me and as it appears its use cases are hardly of interest for me (sound over network, per app sound …).
So to conclude this: Distributors have a great responsibility. They are the direct link between FOSS projects and its users. So they should choose wisely. For that there has to be good communication between distributors and developers.
Edited 2009-01-28 11:18 UTC
KDE follows traditional UNIX version numbering and always has. A major new version numbering (eg. from 3.x to 4.x) denotes binary incompatibility with the previous version and .0.0 denotes that its the first of such a release. It is not true that x.0 has always signified that the developers think the release is ready for general consumption. This kind of thinking is more a product of proprietary software from the Windows world. x.0 signifies that libraries etc are stable and ready for developers to work with. And KDE has always followed the “release early, release often” rule.
RED HERRING! Man, that annoys me so much! It’s the absolute lamest argument in the book. It amounts to this: “It doesn’t matter if we release shit, because it’s free, so shut up, you can’t say anything bad about it unless you coded it yourself.”
So weak. It doesn’t matter if it’s free. It doesn’t matter if I forked it. Those arguments are red herrings and don’t address the actual real issue, which is that KDE4 was not up to snuff. True or not, your “you didn’t pay for it” makes no difference.
It does matter. KDE is a high quality desktop environment that you get for FREE. And, like the others said, you were never forced to use it.
Submit bugreports, documentation, translations, code, or shut the hell up.
The problem is that the OSS world does not say “we release things for free so you get what you pay for”, it states that “it is the best way to create software”.
Blaming users because they complain, because they compare it to Vista/XP/Windows 7/Mac OS X (really very enjoyable and polished experience) and find those offers to be more polished… replying to these users “but you have to pay for it” is like saying “to develop polished software you need to charge your users”.
I do not think people working on KDE4, the Linux kernel, etc… really love defending the quality of their work/their releases with the “hey, we work for free (meant this way: “we do not force you to pay to use our product”)/”you get what you pay for”.
Complaining about users who use the product without contributing code when someone highlights problems with your release is also kind of a dead end.
Sure Vista had its problems (it did not give me much to complain about even in RC status), it is expensive (the Mac tax is even higher being locked to their HW) and all, but I have had a polished out-of-the-box experience without much work on my end and I do value that… why shouldn’t I?
But free software has a different development method. It is the “cathedral and the bazaar” method, the “release early, release often” method. Early releases are buggy and feature incomplete, and more adventurous members of the community run it and test it to find those flaws and get them fixed. Many eyes make all bugs shallow. The more people who participate here, the quicker the flaws are found and fixed, and everyone benefits via subsequent, more stable releases. It is the free software way, and it always has been.
Why should it be changed because some Johnny-come-latelys, probably refugees from the world of Windows-think, are used to something different, something more like what you describe?
The Bazaar method makes better software faster, it has been proven over and over. If some people don’t want to participate in that process, all they need to do is wait for later releases and enjoy the benefits.
I can’t see the point in their complaining about the Bazaar method and its “signature” characteristic of a decidedly immature product at a x.0 release point.
I think the point of “no charge” has gone way over your head. Woooooosh.
The fact that there is no charge for, and warrant with the free software is an ENABLER of the Bazaar method of development. This is what makes it tick. You can’t have an immature and feature incomplete early release and expect to get people to participate in the refinement and further development of the project, and at the same time charge them money for it. Likewise, you can’t wait for it to become mature and thoroughly tested before release while the only people who have ever seen the code or experienced running it are a small development team. That is going to take ages.
What you need to do to make the Bazaar method of development work properly is release early, release often, to as many people who you can find as willing to participate in the early testing phases, with source code available to all for public scrutiny. That is what works. This is where the co-operation and the collaboration comes in to it. Everyone wins.
Edited 2009-01-28 12:17 UTC
Bingo.
Don’t believe the hype. Any hype.
No, but no one wants to work for free on something they don’t like doing or have little interest in.
How about that I tell you to clean my house tomorrow, for free, out of the kindness of your heart, even though you don’t know me personally?
Since we’re talking personal, subjective experiences. I had a polished out-of-the-box experience without much work on my end with Kubuntu. Can’t say I had that with Vista.
I did not say I believed it… just saying that if you claim something you become responsible for that claim.
I know, that can and does happen… so I have little problems if an OS maker tells me “look, we charge for this OS because we are paying developers to make sure sure you get the product you want”… as long as they back it up with their OS release.
You are missing my point, I am more pragmatic than religious discussing about development philosophies.
Closed Source, models which do sell software, and mixes between the two do have their advantage.
Both the OSS and the closed source/non-free software development models can fail and both can succeed at delivering usable innovation.
I see the results they deliver and I see the method in which they deliver it, but with my user hat on the result delivered and its qualities are what I look at when deciding if I will be using a product instead of another one.
Different users different opinions
. Personally I have been using KDE4 since it was introduced in Fedora 9 and with Fedora 10 it has gotten much better that I will admit (especially the networking and audio management), but performance (yes, nVIDIA user) and image quality of the kwin composite-effects enabled desktop still lags behind what Vista and especially Windows 7 (although it is only a Beta) provide.
I am not trying to place blame on KDE, Xorg, Fedora, etc… singularly… just stating what I observe, that as a whole my laptop is quicker/responsive as well as more pleasant to look at during every day’s use with Windows 7 than under Fedora 10 with KDE 4.1.3 (latest nVIDIA drivers installed)… and on KDE4 I disabled some effects that Aero provides (like blurred transparencies).
Opening Firefox with some tabs, IE (do need it for work), Eclipse, some file explorer windows, desktop folder copy and paste, and OpenOffice Calc and working like I always do (which does mean switching between windows, minimizing one window, maximizing another) leaves my Fedora+KDE4 set-up gasping for air more frequently than my Windows 7 set-up (formerly Vista SP1). Just minimizing and maximizing Eclipse and or Firefox shows a clear difference in the maturity of the two solutions right now.
Is it KDE’s fault? Firefox’s fault? nVIDIA’s fault? Xorg’s fault?
That matters, but it also matters that right now one solution provides a better user experience for me than the other.
I do not play politics with my OS choice, I try various solutions, I learn how to do my work in them, and then choose the one that “holds me back” the least so to speak
.
The price of the OS is one of the factors that can influence the decisions I take as far as default OS for me, but it is not the only one.
So many words to say that when a Linux distribution gives me a completely better experience than what Windows can provide me at that time the Windows partition stops being the default-to-boot-into one in GRUB
.
Edited 2009-01-28 13:18 UTC
if nobody complains or points out the flaws (subjective as they might be), the devs would have a hard time improving the software.
They will come back though as long as KDE offers what the wandering user is looking for in a desktop. I think we will find over time that there will be a new generation of converts form other DEs who like the new style. It is all give and take.
Yes. That is the way that free software is developed. Release early, release often. It is the ONLY known way to develop something as complex as a complete new desktop system in a couple of years.
Research about it if you have the time. Google for “The Cathedral and the Bazaar”.
Both Windows and Mac OSX took 10 years or more to go from scratch to a similar level of function as KDE4 has undergone in just two.
Edited 2009-01-27 22:01 UTC
I guess we missed the part where you have to give half-baked versions of the software an X.0 version designation. Your ongoing “release early, release often” mantra doesn’t apply to that particular feat of poor release planning. As google ninja noted a while back, poorly applied, “release early, release often” just translates to “it’s alright to release poorly tested and unready crap, as long as you do it a lot”.
So you didn’t google for it as I suggested.
http://en.wikipedia.org/wiki/The_Cathedral_and_the_Bazaar
Early releases of free software, developed in the “Bazaar” model, are intended for “public testing, scrutiny, and experimentation, the more rapidly all forms of bugs will be discovered”.
Period. Initial releases to the public of free software will contain bugs. The idea is … to identify them and flush them out. Early.
Edited 2009-01-28 00:30 UTC
You’re tiptoeing around the issue, lemur2, as usual. This isn’t about whether or not early releases of Free software will contain bugs.
This is about violating the generally accepted – by both Free and non-Free software developers – idea that a 1.0 release indicates a release which the developers believe to be free of showstopping bugs, ready for general consumption, but may obviously still contain bugs and issues that will be addressed in 1.0.1 and 1.1 releases.
KDE 4.0 violated this generally accepted idea – i fact, it ass-raped it, injected it with heroine, and made it its bitch. Yes, that’s how bad KDE 4.0 was.
And that was simply a mistake. Anyone with more than two functioning braincells realises this.
Edited 2009-01-28 00:37 UTC
I beg to differ.
leos says it beautifully:
Edited 2009-01-28 00:42 UTC
GNOME 2.0 was not NEARLY, not NEARLY as bad as KDE 4.0. Go read some of the old reviews, and you’ll see.
You seem to misunderstand what everyone is saying. This isn’t about missing features, but about stability. The KDE guys never intended KDE 4.0 to be feature complete, so that’s not a problem. It was the total lack of stability in any shape or form that made KDE 4.0 so utterly abysmal.
GNOME 2.0 lacked lots of features as well, but at least it was relatively stable. It worked, had its crashes, but nothing that brought the desktop down or killed x.
… not necessarily a problem with KDE4, but more likely KDE4 uncovering previously unexercised areas of the graphics layers. It is, after all, X that crashed here.
Right. Play the blame game. KDE 4.0 was so wonderful that it made *everything else* crash. Brilliant! You should go into politics. (If you are not already.)
You are apparently not a developer.
If something KDE does in application level code makes X crash, then that is the fault of X (or a lower level). X should never crash if an application behaves badly.
Just like if you are running a program and it does something completely bonkers, you would expect that program to crash but leave your OS running. This is why earlier versions of Windows (the 9x series) had such a bad reputation. At that time it was fairly easy to have a program bring down the whole OS, and that was recognized by everyone to be a bad thing.
Also, Nvidia devs have stated that the issues that KDE4 uncovered in their drivers (slowdowns, artifacts) are entirely their responsibility and shouldn’t be worked around in KDE or X code. Sure enough, they worked hard and fixed their drivers so that now I have no more artifacts or performance problems in KDE with Nvidia drivers.
So if the actual developers admit that it was their problem not KDE’s, then who are you to argue with them?
well. In this scenario he is probably right. Kde shouldn’t be able to crash X, this is an X issue. But the de did crash (and recover), but interestingly apps seldom did.
Edited 2009-01-28 01:10 UTC
Oh come on. KDE4 uses a new system for rendering the desktop, using the hardware accelereated features of the GPU if available. X crashed. X crashes plenty of times when one tries to run newly released (hardware accelerated) games, for example. Is it the game, or is it X? Sometimes it is one, sometimes the other.
Investigate the crash and fix it, wherever it lies. This is free software development in all its glory.
GNOME 2.0 had no audio (where 1.4 did), and a lot of GNOME applications wouldn’t run anymore because they were not GTK2-ready … but GNOME was sort of OK was it? … but every single crash that occurs with KDE4 is the fault of KDE4?
Why the double standard?
I’m wondering can you actually prove that those stability issues where the result of KDE4.0?
I tell you, plasma devs are still fighting some horrid xorg bugs that STILL havent’ been fix for a year now, argb bugs. Just look at how KDE4 showed up the nvidia driver, nvidia practically bent over backwards to fix the situation. Plasma devs where blamed for memory leaks which where tracked down to guess what?, you guess it the nvidia driver.
Before you blame stability issues on KDE4.0 or for that matter KDE4 series, just remember that it exposes bugs in older tech, nothing other than KDE4 can be the problem can it?
Oh, come on… You can’t be serious! GNOME 2.0 was a TRAIN WRECK and nobody, even its biggest fans, could deny it with a straight face back at the time. Heck, if you read some of OSNews reviews about it, you will see that Eugenia had nothing but harsh words about it! Yes, she saw the potential but she couldn’t avoid to report the sad state that the thing was back then.
And note that long time GNOME users were the ones complaining the most as there were LOTS of regressions if you compare GNOME 2.0 to 1.4. Lots of GNOME users left the boat around that time.
As time passed by, GNOME users became used to be told that they don’t really need features on their DE but it became a lot better ever since to the point that they began to add back features slowly and now it is barely acceptable. So the bottom line is that GNOME passed through exactly the same thing but people seem to forget it so easily…
I can’t help but notice that some people here are starting to realize that their vitriol towards KDE 4 was somewhat misdirected – even yourself when you said that you sent a congratulation e-mail to Aaron – but there are still those that seem to think that KDE 4.0 is somehow getting a “free pass” while turning their blind eye to GNOME 2.0 and OS X 10.0. Please don’t do that. You’re just embarrassing yourself.
Oh, come on… You can’t be serious! GNOME 2.0 was a TRAIN WRECK and nobody, even its biggest fans, could deny it with a straight face back at the time. Heck, if you read some of OSNews reviews about it, you will see that Eugenia had nothing but harsh words about it! Yes, she saw the potential but she couldn’t avoid to report the sad state that the thing was back then.
Actually, it was GNOME 2.0 that made me a GNOME user
I had been KDE user before that, GNOME 2.0 just seemed smoother and in my use it actually was pretty stable.
Of course it lacked things..but nothing that I couldn’t live without, really. The parts that were there did work just fine, though, whereas when I tried KDE 4.0 the parts that were there didn’t work, or screwed up after a moment.
I’m not here to start arguing though, just pointing out that not everything you say is true.
There are two issues here, although everyone seems to be confusing them. A 1.0 release is supposed to be pretty good. There are no previous expectations, it is the first major release of a product and in the cause of open source projects, has usually seen many 0.x releases to polish it.
A x.0 release, where x > 1, is different. In this case there already is a (x-1).y set of releases, and they are presumably quite stable. Then some major change was made to justify a new major version number. This is bound to introduce regressions and if we look at history, this has been the case for just about every major software project (I mentioned some in my previous post).
Now KDE 4.0 was a pretty bad example of this, but it was hardly unique. Every big re-imagining of projects starts with one pretty crappy release (OSX, Gnome2, Vista, etc). It would have been better for KDE if the quality of 4.0 was more like that of 4.1, but let’s not pretend they are the only project that suffers from this.
No. I read it back in 1997 when it was in vogue and ESR was still a respected FOSS advocate.
I see you are continuing your tradition of avoiding the issue by responding to my on-target criticisms with off-target points which don’t relate. I wonder why you consistently do that on this topic?
Edited 2009-01-28 00:55 UTC
What a load of crap. KDE isn’t an operating system. It’s a Desktop Environment. It relies on thousands of other projects to remotely get to a modern state. It is constantly broken and guess what? The same projects that run on Linux run on OS X.
Fortunately, OS X which is constantly updated throughout the release cycle of it’s .x existence doesn’t retort with, “suck it up. The updates are free. Release often and deal with it because that’s the way of all Software.”
… all free software. OSX is not free software. You are expected to pay for it … therefore you have an expectation that it has at least a certain amount of finish.
This is not the way that free software is delivered. It is released to the public for trial early, in an unfinished state. It is worked on. After a few releases, it becomes progressively more stable.
GNOME 2.0, for example, was not finished. It was buggy, many things that used to work were now broken or missing, but nevertheless, it was released in that state.
http://en.wikipedia.org/wiki/GNOME
Now we are at GMOME release 2.24. Twenty four releases later. Twenty four releases in seven years … an average of over three per year. That is “releasing often”, is it not? Now it is stable. A little crusty, perhaps, but quite stable.
http://en.wikipedia.org/wiki/GNOME#GNOME_3.0
GNOME 3.0 might have a fair chance of more stability on release than most x.0 releases of free software, because “Although the desktop will undergo a major revision, changes planned so far are mostly incremental”. We shall see. But GNOME 2.0, which was a major change from GNOME 1.x, was quite similar, in terms of stability and finish, as KDE 4.0.
Edited 2009-01-28 02:28 UTC
Actually, afaik GNOME follows the old numbering convention that odd numbers signify development versions and even numbers release versions. Therefore GNOME released 13 versions since and including 2.0 and uses a rather strict 6 months release interval.
OK, that sounds quite likely to be the case, because a 6 month release cycle makes a lot more sense than 3 months. A release every 6 months is still very much a case of “release often”.
empty zip? Wow, did you even look at KDE4? The fantastic APIs and Services, the new possibilities with Qt4, features like Solid or Akonadi?
Yes, I’m sorry that end users didn’t enjoy it as much as developers did. Then again, at a seminar for professionals, it would be strange to argue that the beginners didn’t enjoy it.
KDE 4 gave me as a developer a big boost to play with the code and make KDE better. That, it achieved very well, and we wouldn’t be at KDE 4.2 without the increase in development that KDE 4.0 brought with it.
So congratulations to a great KDE 4.2 release, and to all releases that made KDE 4.2 happen.
What do you mean by “lots of people”? Have any numbers/articles to support your opinion? And how do you measure project harm anyway, after only the couple of months?
Besides, the “ditching” argument is plain stupid. If people are leaving KDE3 because they disliked 4.0, although their KDE3 still works perfectly, that says a lot about the people. Consider this: if KDE4.0 wouldn’t have been released until today, we’d all be using KDE3 so far. How is showing to the people one year earlier a bad thing?
Well, this is a place where we definitely disagree. I stand behind my words, if KDE 4.0 hadn’t been released back than in a state in which it was, it wouldn’t today been close to the quality it actually has now. As much as it sounds unpleasant, Free Software projects, especially big ones, needs users to both test it, spread the word, and in the end – contribute. Because userbase is where contributors come from. Release early, release often, thats how FLOSS works. Unless they have millions on a bank account, that is
If you want a solid proof of work done in one year between 4.0 and 4.2 take a look at Techbase and Userbase:
http://techbase.kde.org/
http://userbase.kde.org/
Stuff like this just doesn’t write by itself. Community is active and growing. That’s what counts.
It’s all a matter of balance. Finding your community is a way to ensure your that project lives on. The earlier the better.
Actually, I suspect it says more about the “impressions” that some people are trying to create about KDE 4. Someone seems to be trying to spread a negative meme about it, methinks.
http://en.wikipedia.org/wiki/Meme_(Internet)#Advertising_and_marketing
Oh, for f–k’s sake. This thread started when I said something *good* about KDE. It was hijacked by a group of people who decided that it’s unfair to say anything negative about KDE4. I’m just not bowing to the pressure and changing my mind. That hardly amounts to an agenda against the entire project.
http://linux.slashdot.org/article.pl?sid=09/01/24/1842218
Haha
To support your argument for lots of people switching you cite an article talking about one person? Great reference.
Anyway, I almost agree with you, I think the state of KDE 4.0 should have been advertised better (like on the main release announcement), but there isn’t any evidence that it caused a significant number of people to abandon KDE. Just look at development activity after KDE 4.0 was released. It really increased a lot, including user contributions to documentation and such. Not exactly a sign of many people leaving.
That being the case, it seems a little counter-productive to have users taking away the impression that they were effectively tricked into doing alpha/beta-testing.
To quote from the beginning of your post, “Have any numbers/articles to support your opinion?” That question / challenge applies just as much to your statements as it does to Adam’s.
Is there any substantial reason to conclude that KDE 4.x wouldn’t have advanced as much if KDE 4.0 had been labeled as a “Developer Preview” instead of a point-oh release?
Look at SkyOS; the big complaint is that it is in perpetual beta, that no one can try it out etc.
KDE 4.0 was released with lots of caveats, that it was only for interested enthusiasts, that it was a not quite ready for the masses etc. It’s really most unfair of you to criticise it on that basis; its deficiencies were plainly advertised.
Call me crazy, but I’d like to think that there’s a reasonable middle ground between “perpetual beta” – and “beta-quality software labeled as a finished release.”
And I suspect that simply labeling it as a Developer Preview would have more effective than all of those caveats combined.
More unfair than criticizing users for expecting that software carrying a “x.0” version number to be release-quality & relatively complete?
The average user does not install KDE, he/she install a distribution. The average user don’t care what version KDE is, they care about the version of their distribution.
The average (typical) user doesn’t run Linux in the first place.
I think you have a misconception of FOSS here.
In FOSS there are no closed and then open beta tests like you find in the proprietary world here there is constant testing by the users and developers. Users can report bugs, make suggestions and even create patches — often not easily possible in propritary projects.
The price for this?
Sometimes buggy software with only few features, but only if your distributor ships it. If you install it yourself and ignore the warnings than that is your fault as you as user ignoring the set paths by your distribution have the responsiblity for your actions.
Another different part is if none uses your software gives feedback etc. you do not have the same motivation and “pressure”. Sometimes you have to release something and then you’ll have the motivation to add and fix a lot for the next release.
You can easily look at the statistics to see my point. The last weeks before a release there are more commits than normally, the same happens if you get closer to the different freezes. People want to get their stuff and the fixes in and work pretty hard for that. Without these “deadlines” (=release, time before release) you would have no reason to work that hard in your freetime.
A lot of people seems to forget that KDE is a project that relies heavily on volunteers. 3.5 was released in 2005, if they had waited until now to release 4.0 it would have been more than 3 years of silence from the KDE team.
What happened after 4.0 was released was that
a)Users got involved. They didn’t just tried a beta version for 5 min. and said, “well is a beta version, I will check it back later”. They actually filled bug reports, and wrote lengthy blog entries about the missing features that they more desperately wanted.
b)People with coding skills started to scratch their own itches. Instead of just looking at the beta version for 5 min and said, “well is a beta version, they still have time to close this bugs”, they said “What a mess of bugs! it looks like I will have to close this bug myself” and then they helped.
c)Developers started to work on the framework. Again once 4.0 was released instead of keep waiting some developers saw the potential and started to use the new framework. Remember the framework was supposed to be ready by the time 4.0 was released, but it hadn’t been seriously tired until them. A lot of bugs, missing parts and the like were found after people started to actually use this frameworks.
d)Potential developers noticed KDE 4. This is an important point. While the KDE team was building the pillars of KDE 4, other projects were making good progress on the user visible part and were attracting all the attention. After 4.0 was released lots of people saw the potential and decided to join, instead of deciding to help in another projects. The number of svn accounts went up after 4.0 was released which helped to accelerate the developing process.
So when people say that 4.2 is better now because of 4.0 this is why they say it. Even so it wasn’t an easy choice and is debatable if this advantages are enough to compensate the damage to the public image of the project. I think they made the right call but it was definitely a controversial decision.
I understand that, same reasons outlined by Seigo himself, but it does not speak too well about either the development process of KDE, or the way it was presented to the users and developers alike during development, or the community itself.
Basically Alpha, Beta, and RC phases were not enough to bring enough attention from users which would find, report, and complain about bugs as well as from developers which would hear complains of users, try their application on the .0 release, make fixes, etc…
Seigo says he was worried about having the project stay in eternal Beta and thus having user and developers ignore it.
You yourself seem to say that not nearly enough users and developers alike did not pay much attention to the project when it was just a Beta because… it was just a beta! And this is like the big pink elephant in the room nobody is talking about.
Edited 2009-01-29 18:04 UTC
We did the release with the future in mind and the 99% of computer users who don’t use KDE yet. This did hurt some of our current users (those who used distributions forcing 4.0 on them as sole KDE desktop). Not much we can do. If you think a bit about it, you might understand why KDE 4.0 released today wouldn’t be as good as KDE 4.2 is right now.
First reason: developers. We gained a huge number of developers since 4.0 – over 300. Most of them would not have joined KDE development without 4.0.
Second reason: applications. Many applications only started porting when 4.0 was released. It’s bad enough we don’t have K3B today. It would be far worse if we had to wait another year.
Third reason: third parties. Imagine NVIDIA starting to fix their drivers right now. It took them a year to get them reasonably working with KDE 4 so if they started now you wouldn’t be able to use KDE for another year.
Fourth reason: KDE is not one. The KDE software suite is huge. Big chunks of that were ready when 4.0 was released. Think Dolphin, the whole of KDE edu and most of KDE games. Not releasing those would have meant less developers, less testing and less motivation.
There are more reasons (for example motivation in the community – it’s no fun to work on stuff which won’t see the light for another year) but these are the most important. I would appreciate it if a reply would either refute them all or say “sorry, I didn’t think of those”.
I did think of those. I appreciate them all, and I appreciate that the decision was a hard one. Personally, I think it’s naive to suggest that you can release something as x.0, which historically, has suggested that software X is a relatively bug free major release, and then expect that most people have read your explanation that it isn’t on a blog somewhere.
Nonetheless, despite the heated debate here, I’m very happy to see KDE as a whole moving forward and stabilizing on 4.x. However, I can’t simply say “yeah, that makes sense,” or essentially, I’m suggesting that every project can just dispense with the concept of beta and just release their unfinished software on the world, users be damned, with the justification that without doing a release, no one would have come around. As a user, that’s not my problem. We saw the EXACT same thing with Vista – people weren’t ready, and no one is letting them off the hook. BUt with KDE, valid criticism is hushed with snide remarks like “it’s free, so either code it yourself or shut up.”
But I do understand why.
Except for the fact that they didn’t expect users to “USE” kde 4.0 they expected developers and advanced users to “TEST” kde 4.0.
Unfortunately people were misinformed, either by themselves or through the negative echo chamber that no amount of “this is not for users” statements could fix.
You have to wonder that if they didn’t release kde 4.0 and got decent feedback (along with all the other crap), how kde 4.2 would turn out.
I do. But I’m not sure any conclusion justifies the path as was.
Although I think the problem here was assuming that they could somehow redefine what a .0 major release meant an expect anyone would take notice.
They can call it whatever they wish. The reason they went with 4.0 is because kdelibs was ready for public consumption. Many that develop for kde (as a platform and not as a DE) are not necessarily tuned into dev cycles. So calling it a developer release would not get their attention as they would assume the platform is still shifting under their feet.
What I can’t see is how people managed to install it without knowing it was in the state it was in. AFAIK fedora was the only one that force fed kde 4.0 to it’s users.
Simply looking at screenshots with links to the repos, told me that the DE was still cooking.
Either way, it’s in the past. Whether the same thing happens with gnome 3 or kde 6, we will have to see…
“I do. But I’m not sure any conclusion justifies the path as was.”
So the path taken is more important than the destination? Or that the path taken is actually the fastest in the long run?
Edited 2009-01-28 00:33 UTC
Precisely the point.
This is exactly the process by which free software is developed.
Release early. Interested people try it. Get feedback. Gather information about bugs. Measure performance (and lack thereof) in real-world usage. Define areas that require improvement. etc, etc. This process is what “community development” is all about.
One cannot use this model for software that users are expected to pay for. The users would rightly reject your product, deeming it not worth the money.
Free software is an entirely different paradigm. Apparently, some people haven’t grokked this fact yet.
Edited 2009-01-27 22:10 UTC
“Release early, release often” does not mean deceiving users by using a version number that is known and understood to mean “final” and “complete”, and then use it on an obviously unfinished, crappy, and buggy product.
“Release early, release often” means that instead of dumping a 1.0 release on your users out of the blue, you feed them 0.x releases until you hit final.
Please, lemur2, you might be the local Free software evangelist around here, but that doesn’t mean you can lecture us on what “release early, release often” means.
Edited 2009-01-27 22:35 UTC
You can go to a 0.x numbering system for a new project, but KDE4 was not a new project.
I’m not trying to redefine “release early, release often” … it says exactly what it means all by itself.
A, version numbers are completely arbitrary and has no meaning other than “some progress as been made”. For example, not everyone agrees with the “major number jump means big progress” scheme (like OpenBSD).
B, .0 has never meant “final” and “complete”. If anything, it quite often mean “sure, it sorta works but there are lots of bugs and problems still”.
Version numbers mean squat. Point-oh releases are never stable (at least I’ve never seen that). KDE 2.0 was buggy, KDE 3.0 was also buggy as hell. KDE tracks Qt version numbers with their major version numbers. KDE 2.x ran on Qt 2.x, KDE 3 on Qt 3.x and KDE 4 uses Qt 4.x. And they release a .0 when the API is finalized. What so strange about that? I think it’s very sensible, and not misleading at all.
What group of users is more likely to offer decent feedback? Developers and early-adopters who have no problem using software that’s clearly labeled as beta / pre-release? Or less technical users who assume that “x.0” means “suitable for public-consumption” and become upset when the software doesn’t meet their expectations?
Every other FOSS project undergoing a major upgrade has released a x.0 version (where x is > 1) early, and with bugs included, and without a full set of features working, and no-one gets upset. This is the FOSS way. FOSS is indeed “suitable for public consumption” (at least for the part of the public that is willing to participate in the development process) for these early releases, because that is the way it is done, and that is everybody’s expectation.
GNOME 2.0 did exactly the same. It too was released to the public early, with bugs, and without full features, for the purpose of public testing, with a view to working out the bugs and getting it polished.
http://www.osnews.com/story/1280
Well actually, it was originally supposed to be released a year earlier than it actually was, so you can imagine what it would have been like if that had happened.
The point is this: this is par for the course for a FOSS project. This is the way it is done. FOSS projects don’t attain the “polish” of commercial releases (that you are expected to pay money for) until at least a couple of releases past the x.0 point. The “payment” that you make, for FOSS projects, is that the public is expected to help in the initial development.
You can always use the previous versions if you want instead to get actual work done. No-one FORCES you to participate, you are not REQUIRED to install a KDE 4.0 or a GNOME 2.0 or any other early release.
So if GNOME 2.0 did this, and FOSS projects in general likewise do it, my question is why is there suddenly all this unwarranted brouhaha over KDE 4.0 doing exactly the same?
Why are people allegedly “getting upset” … but ONLY for KDE4 and not for other FOSS applications developed in exactly the same way?
It sure seems like a beat-up to me.
http://en.wikipedia.org/wiki/Internet_meme
Edited 2009-01-28 23:13 UTC
Every? For example…?
I use quite a few open source applications, but I’ve never bought into the “obligation to volunteer” mentality. The notion is absurd just from an English semantics point of view.
It sounds like you’re saying that no one is allowed to express a critical opinion of anything – unless they’re forced to use it and there are no available alternatives.
Funny that you should mention GNOME 2 again, because 30 seconds of searching turned up the “GNOME 2: A Year Later” article published on this site ( http://www.osnews.com/story/3721/GNOME_2_A_Year_Later ).
A quote from its first paragraph:
Whaddya know.
That would be a reference to criticism of KDE 4.0? Or a reference to the “KDE is being unfairly persecuted” meme?
Are you kidding? That’s the only way open source development can work. You make releases and people start using it en masse when it is good enough.
Try telling that to proprietary software developers, but as said above, that is the way open source development works. The reason why the source code is open is so that code is released and tested early and users can do alpha and beta testing.
I didn’t think most people would need to have the concept pointed out. Obviously I was wrong. I think we should just get past this now.
Edited 2009-01-27 23:17 UTC
Sigh. You’re missing the point Adam was saying: when a x.0 release is made it is usually considered a feature-complete product and useable by the intended populace. That’s how I and all the people I know are atleast used to think. And that’s what caused so many people to get angry or feel disappointed.
The KDE devs did indeed try to redefine the well-known concept of introducing a x.0 release. They could have called it KDE 4.0b1 and POOF, everyone would have known it was a work-in-progress and not feature-complete yet.
Oh, and just as a sidepoint here: KDE4 was hyped all over the place so of course people learned about it even if they weren’t actual enthusiasts or well-learned professionals. And then there were distros that made it easy for people to install KDE4.0 and try it. I too installed KDE4.0 on my Mandriva simply with the default package manager. (It was a complete disaster, but that’s not the point here)
Sigh………….. It’s incredible how many people simple don’t understand how this has worked since the year dot.
No, not in the open source world it isn’t. A release is made when the developers decide that they have a feature complete release by their metrics, not yours or anyone else’s, and they feel they have reached a milestone. People then jump on board and use it when it is ready. That’s the way open source development and usage has always worked.
I love when people assume that being an arrogant ass makes them seem more right. Instead, it makes you look like… well… an arrogant ass.
I disagree with you entirely. I think open source projects thrive and do well when they earn their users’ trust. I guess you think that being an open source, free-as-in-beer project means they shouldn’t act professional or do any testing.
Tagging the 4.0 release “4.0” was a mistake, plain and simple, and many people will not soon forget that KDE .0 releases are not actual releases, but rather, just version-inflated betas. If KDE has learned their lesson, great. If not, then plenty of people will treat their software like it’s not production ready, which is a shame.
Sorry, but this is your mistake. It is your mistake about free software in general, not KDE4 in particular.
All initial releases of new code in free software are unstable. Every free software project is like this.
If you want stable free software, then get this:
http://www.debian.org/releases/stable/
See where it says “stable” in the URL?
Anything newer than this is not (yet) stable.
Like this, for example:
http://www.debian.org/releases/testing/
See where it says “testing” in the URL? That means it is still in the community testing phase.
Welcome to the world of free software and community development.
Edited 2009-01-27 23:56 UTC
Ugh. You can argue til you’re blue in the face, but you’re wasting your time, because your point spits in the face of facts.
These are facts:
– Historically, calling something 1.0 (or 2.0 or 4.0) meant “this release has been tested, we believe the bugs have been ironed out, and this is ready for use.”
– KDE redefined their 4.0 release as “needing testing.” They admitted as much. They made the x.0 release into a public beta (arguably alpha) test.
You are now arguing that every project does that, but shame on you. Every project that properly tests their software and respects the convention should take offense. It’s amazing how so many programs manage to pull this off, but KDE gets a free pass.
Don’t bother responding, the point has been made.
No. Not for free software. You seem to have the concepts for commercial software world fixed immutably in your mind.
Did you ever try to run the Linux 1.0 kernel, for example?
No one in their right mind would try to claim that that was ready for users expecting stable software.
In free software, a 1.0 release means “ready for community testing and development”. It does not mean “ready for sale”.
Really?! So that’s why Mozilla was at 0.x for years, they didn’t want “the public” to test it.
I wonder why all these people keep releasing “beta” versions of stuff, when they just need to tag it x.0 for users to test it.
Now I know. Thanks for clearing that up!
Did you ever try to run the Linux 1.0 kernel, for example?
No one in their right mind would try to claim that that was ready for users expecting stable software.
Actually, I hear 1.0 was rather stable and it had reached the intended features, hence the 1.0 release. Until that it was 0.x, like most open-source products.
The intended audience was also very different from the intended audience of KDE; Linux 1.0 was obviously for enthusiasts and developers and not in any way or form for the general populace.
As said, a x.0 releases are usually mostly stable, mostly bug-free, have reached the planned features and generally suit the needs of the intended audience.
Not at all. Not a bit of it. Not even when it comes to desktops as oppsed to kernels:
http://www.osnews.com/story/1280
“This is 2x boneheaded”.
“Gnome 2 does not come without its problems. I do not have sounds on my Gnome 2. I think that Gnome 2 assumes that you have Gnome 1.4 installed, and while I do have the libraries needed to run older GTK+ apps, I do not have the full installation, therefore, it can’t find the actual .wav files. The Sound panel argues that I should install the gnome-audio package, but a look at the 2.0 directory on many FTP servers, did not reveal such a package at all. On another note, Galeon stopped working from the moment I installed Gnome 2.0. And where are Pango’s multi-language settings to select another language for my keyboard? And where are the system tools for networking, or maybe a universal media player? Also, why the included system monitor can’t pick up both my CPUs? Oh, and I lost almost 2 paragraphs from this review, by typing it on the GTK+ 2.x updated Gedit text editor (which is also the default Gnome text editor). When I just place my cursor on the text, and then move my mouse away in order to type there, the program seems to think that I still have my button pressed and it keeps selecting my text. It happens so often that I was not always careful, and there go two full paragraphs. I changed my mouse settings from its Gnome panel, but it did not help the specific situation. Please note that the Kedit or Kwrite editors do not have the problem. Plus, scrolling this very document with Gedit is shamelessly slow when compared even to itself (or GXEdit) of the Gnome 1.x series.”
Edited 2009-01-28 00:52 UTC
That is the stupidest argument I’ve heard in a long, LONG time, Lemur. By comparison, the Islamic terrorists sound sane.
Give it up. Your arguments based on semantic substitution are childish, boring, and wrong.
No, the rest of the FOSS community does NOT redefine x.0 to mean “Alpha” or “beta”… thats what x.0a and x.0by are fore. Plain old X.0 means that the alpha and beta development stages are complete.
The KDE team SPECIFICALLY argued that 4.0 was ready. The users on the openSUSE list argued for WEEKS on end, saying that 4.0 was nowhere close to usable as anything other than demo-ware.
Despite this, the KDE devs working at Suse INSISTED that 4.0 be deliberately, and deceivingly foisted on users (especially new Linux users who didn’t know any better) as the standard software, with 3.x shuffled into the background, and deliberately made difficult to install by all but the most knowledgeable.
The was not an independant act by Novell… it was due to the ACTIONS of KDE personell working within SUSE.
Now, Lemur, we’re SICK AND TIRED of these same old lies which we’ve been hearing from the KDE 4 people for more than a full year now.
STOP IT, and quit acting like the real life incarnation of Eddie Haskell.
You have a very short memory. Gnome 2.0 was awful and lacked many really basic features. Same with Apache 2.0, the first 5-10 releases of Linux kernel 2.6, Vista, Windows XP pre-SP1, OS X 10.0/1.
All of those were new releases of previously mature and stable, complex software projects. All of them were horrible. I would say all of them were worth it in the end after some fixing. If you think KDE 4.0 was unique you really haven’t been paying attention.
Hahah.. After a year of shitstorm about KDE 4.0 (yes it’s been a year) you think KDE is getting a free pass? Not saying it wasn’t somewhat justified, just as the anger at all those other projects I mentioned when their major revisions landed.
We are used to 1.0 releases like GAIM (now Pidgin) where it really was pretty stable and featureful a year or two earlier.
Or, beta releases like Gmail. Basically, 1.0 says it’s finally stable. KDE 4.0 should have been KDE 3.9.
I tried 4.0.0 and within seconds had a bug, when my entire Firefox window vanished when I wasn’t mousing over the toolbar (bug already filed when I checked). Panels vanished, settings changed at random, things were impossible to see… You can file/add to bugs, and I think I’m still signed up for some of them, but it was bad enough I didn’t know where to begin.
I kept checking back at various point releases and it was still aggravating to use even when they’d fixed the ‘programs disappearing’ bug.
Then 4.1 came out and I’ve been using it as my primary desktop ever since. It’s still got problems, such as buttons and boxes partially disappearing in Firefox, and the background filling with garbage when I’m running multiple apps… but I can USE it.
Oh really. I can remember when RH .0 releases were generally shunned because everyone knew they were incomplete and buggy. That goes for a lot of software really, both commercial and OSS, “Stay away from .0 releases”.
Also, not everyone agrees with that .0 idea. Some projects (like OpenBSD) gives no special meaning to “major” (2.0, 3.0) version changes nor any special meaning to .0 releases.
The fact that you assign special meaning to versions and fail to follow the explicit advice of the developers says more about you than it says about them.
You seem to forget that the KDE devs crowed unrelentingly that 4.0 was READY not just for applications developers, but for users in general — even while whole swaths of basic window-manager functionality were missing, and even simple things like konsole (shell console) were crashing without provocation.
The problem with the KDE 4 project is that, when it comes to PR, they’ve become nothing more than a band of liars unworthy of even an ounce of trust.
They have done more damage to the FOSS movement than ANY other group… right at the time when millions are getting fed up with Microsoft enough to ditch windows…. we had Fedora and openSuSE — at the behest and lies of the KDE 4 devs… pushing the whole KDE 4 mess on new users (people TRYING OUT LINUX FOR THE FIRST TIME) without the slightest warning that 4.0 was not merely a work “in development”, but in rather EARLY development, and nowhere close enough to stable for even casual use, but only for alpha- or beta-testing.
Every single KDE dev needs to be hit across the face and the back of the head with a clue bat so that they NEVER forget the colossal level of stupidity and dishonesty which they perpetrated on the newbies in the Linux community. If I didn’t know better, I would think that they were trying to sabotage the movement to get new users into the Linux community.
Using Debian Sid daily for the past 8 years your attitude about “see the word stable” is cocka.
Stable in Debian means, `outdated but solid.’
Testing means mostly outdated and mostly solid.’
Sid means mostly current and almost mostly solid.
Experimental means, current, improperly QA’d and more often than not broken and non-solid.
But heh! Just roll back and sacrifice being current for solid.
KDE 4.2 is on my Debian Sid box. I know systems engineering and userspace development engineering.
The Linux Kernel is solid.
The Linux Userspace is still f’d up most of the time and it tends to relate to the fact that the QA aspect and lack of standard upon which all distros start from [LSB] is still broken with a lack of consensus.
But heh! This is the year of the Linux Desktop.
People find the fact they can fork fifty billion times as a cool feature. Consumers don’t give a s***. They just want something to work as billed, whether they pay for it or use it and thus help spread the FSF mantra by not using Windows.
I’d throw in OS X but heh, they do give back and thanks to projects like WebKit make it possible for the once forked KHTML/KJS and then some come back to the source, fixed and advanced.
But like all package managers, Debian Experimental is not immune to not making sure the entire KDE 4.2 release is built.
I’m specifically speaking for the lack of KDE SDK 4.2 with Kate. It’s currently broken and the hooks to plasma are hosed with it.
Bravo!
Either test the crap in it’s entirety before releasing or don’t release it, even if it’s just in Experimental.
No, stable means stable. Stable doesn’t mean “Debian Etch” … Etch is just the current stable release.
Lenny is, I believe, about to be “promoted” to stable. Perhaps by June.
http://blogs.dailynews.com/click/2008/10/debian-lenny-stable-not-in…
That means that, after all of the testing of Lenny (when it was Debian testing), the developers now consider that most of the bugs are outed, and it is now worthy of the label “stable”.
Sid is the current “unstable” release.
http://www.debian.org/releases/unstable/
Expect it to be unstable.
When Lenny is promoted to “stable”, Etch will be retired (it then will be called “old stable”). Sid will become “testing”.
Debian “experimental” is the cutting edge.
http://packages.debian.org/experimental/
Debian do not release experimental software, but they let you install it and try it. Experimental does not have a “Toy story” name. The next “unstable” will be created (copied) from Sid when Sid becomes “testing”. I don’t know the name it will be given.
Edited 2009-01-28 02:59 UTC
Sid will become “testing”.
Incorrect. Sid is Sid. Unstable is always called Sid (the villain from the first Toy Story)
When Lenny is promoted to Stable, a new Testing will be cloned from Sid and will get a new name, in this case Squeeze.
One can be forgiven for thinking this is rather complex
Ahhhhh, OK. Got it now.
I was wondering why that break-off distribution from Knoppix called itself “Siddux” if Sid was eventually going to become a stable no-new-features-to-be-added release.
Your explanation answers this beautifully. Sid is always Sid. New versions of “testing” are “sons of Sid”.
My guess is that Lenny will be released before the end of February. But your guess is as good as mine.
Great. It doesn’t make this wrong however.
Your prerogative. You’re still wrong for the reasons given.
Open source projects do testing via release early and release often and getting the code to testers and users as early as possible. When it starts working well then word of mouth spreads and more people start using it. This is the way it has always worked. It has never worked via version numbers.
Who are you to decide that? The KDE people felt they had achieved their goals of feature completeness and completeness of the libraries in particular. Quite frankly, it’s not for you to decide, and it isn’t for any piece of open source software you use.
Open source software becomes production ready (the fact you use that term shows how little you actually know) when enough people have tested it, a critical mass of people have started using it because it is good enough and distros feel confident shipping it. It doesn’t and never has happened when a project reaches version .0. The fact that distros have screwed up over KDE 4 and especially PulseAudio to the detriment of their users is neither here nor there really.
Seriously, what planet have you people been living on?
Edited 2009-01-28 13:41 UTC
As far as pushing out another RC/developer release vs declaring it stable and getting it out of the door with the 4.0 status, I remember (although I cannot find the exact quote at the moment) that there was the concern of declaring it “stable, 4.0 worthy and not an RC/dev release” to get more people to test their software and drivers on it as opposed to attract less heavy-duty testing with another RC release.
Edit: found it…
http://aseigo.blogspot.com/2008/01/talking-bluntly.html
Edited 2009-01-28 14:00 UTC
If I was so wrong, why are there thousands of articles across the internet talking about KDE 4.0 as beta quality software? List as many reasons as you want, you cannot refute the fact that 4.0 has received a lot of attention over the fact that it was not complete or up to snuff. Nothing you say, no matter how arrogant you remain, changes that.
I’m done with you and this thread, don’t even bother replying – you are both condescending and pompous and, frankly, it’s just not interesting to discuss this with you. Also, you apparently live in a reality where no software is *ever* deemed complete by developers, just by “word of mouth.”
Because they are wrong, arguing about utterly the wrong things, they don’t get open source development and it is not up to them to look at the .0 version and then call it beta by their metrics and goals. The KDE people felt that they had reached their goals and milestones for this release, and it certainly wasn’t going to help the long-term stability and quality of the KDE 4 line by perpetually being in betas and alphas.
It’s up to the distros to screen users from that if necessary, and with KDE 4.0 and especially PulseAudio they totally failed.
You can call people as arrogant as many times you want and point to as many ill-informed and wrong articles regarding KDE 4.0 out on the innertubes as much as you like from people who like a rant, but I’m afraid they didn’t get to decide what went into KDE 4.0 in terms of functionality or feature completeness. Neither did I. The KDE developers did. Nothing you say will change that, and criticising KDE’s developers for achieving what they wanted with KDE 4.0 is totally and utterly irrelevant and a waste of time.
I’m not suprised considering that you’ve been rationally explained to be wrong, so the pompous and condascending bit is rather ironic.
That’s the way open source development works, and only to a very slightly lesser extent, proprietary software. Why on Earth do you think Windows 7 will be the real Vista release? You can only try something or read some reviews and see whether it does what you want it to do and then watch the software improve. The major version number will probably signify that it won’t crash, burn and eat all your children and pets, but it’s no guarantee of anything other standard you might hold it to.
[[This comment has been removed by OSNews staff]]
This is uncalled for. I disagree with the way KDE 4.0 was handled and I repeatedly stated this position of mine, but wishing suffering to people for this? No way, think about what you just said…
No, we really should. Especially in a news item about 4.2.
Some self-proclaimed “power users” whining endlessly about how we should feel sorry for them because they don’t like where KDE is going and how they themselves failed to read and understand simple instructions does not a disaster make.
I had sort of hoped that in 2008 computer users savvy enough to download and build KDE from source would be able to understand the simple statement “4.0 is not ready for users”.
Versions are arbitrary numbers, it doesn’t matter what they are.
This may come as a big surprise, but the average user don’t care about version numbers. It’s a geek thing.
The users use whatever comes with their distribution, be it 3.5 or 4, and if that is not in a state where it is good enough it’s the distro’s fault, not KDE’s.
And before someone comes running with the tired old “fan boy” mantra, I use both GNOME and KDE and they’re both good for me. I also like KDE4 much better than I ever did 3.5.
KDE 4.0 was NEVER to standard users to use. That was made very clear by the KDE people. The colossal disaster was that some distros replaced the stable KDE3.x with the new KDE4.0. So they forced normal users to use a development version of KDE.
“KDE 4.0 was NEVER to standard users to use. ”
My, that’s a big steaming pile of BS you have there.
The KDE devs working at SUSE emphatically and repeatedly argued that KDE 4.0 was ready for standard users, even to the point of arguing with all the reset of the openSUSE userbase who documented all of the gaping holes and massive, show-stopper bugs which made KDE 4.0 totally unusable for ANYTHING, other than a list of functions for applications devs to develop against.
Thats odd
Is that the same reason why the OpenSUSE website clearly mentioned, when they were realizing 4.0. that a lot of bugs and issues are still in it, and that its still a “show off” release
In fact there missing features page about KDE 4.0 still exists http://en.opensuse.org/What_features_is_KDE4_missing_when_compared_…
Obviously the page has been updated for 4.2, but when 4.0 was out, even on their forums, it was clearly said that 4.0 was not ready for home or average use which is the same reason why they have KDE 3.5 still supported and still as an option at the time (instead of Fedora who for example threw it out the window)
I agree but this should be detailed further:
– naming KDE4.0 instead of KDE4.0 Alpha/Developer Preview was a (big) mistake, I agree.
– Distributions including KDE4.0, was-it a mistake?
Well for me it depends on the goal of the distribution for Fedora is all about bleeding edge software so why should users expect functional software? for other ‘stable’ distribution a non-default installation of KDE4.0 with warning that this is software not-ready would be fine too.
Well, I’m a gnome user. I watched those videos and it actually looks like there’s some cool stuff there. I may end up trying it out. BUT man the theme in that video is so ugly. The little plasma things on the desktop look good, but the menus/buttons/other widgets in normal windows look horrible. Maybe (hopefully?) that’s not the default theme.
Are you suggesting that the Oxygen widget theme, as seen for example here [0], looks ugly? To each his own I guess, but for me it looks like a clear, calm and great looking theme.
[0] http://kde.org/announcements/4.2/screenshots/dolphin-full.png
Yes. It looks like it’s from the mid-90’s.
Edit: also, the theme doesn’t fit well at all with the plasma things on the desktop. They look great, but the normal windows next to them look completely out of place.
Edited 2009-01-27 21:05 UTC
I like the oxygen style, but I agree it doesn’t really match the plasma theme.
And I don’t like the window decoration style, so I switched to “plastik”.
But plasma is beautiful
True and this how another mid-90s UI looked like:
http://en.wikipedia.org/wiki/File:OPENSTEP_Workspace_Manager.jpg
Coming from a Gnome user…that’s quite a statement!
GTK has always looked old to me compared to KDE. I think it’s just the big buttons, blank spaces and default fonts…
Funny, I don’t remember seeing any desktop that looked like KDE 4.2 back in the 1990’s. Perhaps you need to install a copy of Windows 95 or maybe OS/2 to refresh your memory…
hehe, funny to hear you don’t like the widget style – most ppl don’t like the plasma style and DO like the Oxygen widget style…
Either way, it’s themable
I also think the Oxygen theme looks dated and unbalanced (the spacing buttons in the top right of the window / contrast). While not perfect, I think OS X and Windows 7’s themes are a lot easier on the eyes and should be studied for inspiration. I currently use the iLeopard theme on OS X which is the best I’ve seen on any operating system.
yeah, i really don’t like how it looks either. Has WAY too much empty space all over the place and i dislike how the control widgets look. Thank god its themable (is that a word?
) 
But don’t listen to me, I can’t stand the way most newer guis look. I don’t want lots of rounded edges, transparency (unless its helpful, which it is generally not), gradients, ‘shiny’ icons and crap. ugh. makes me want to vomit.
most modern guis don’t seem to care how much space they take up since everyone runs 20+ ” displays and really high resolutions (yeah, right..).. I’d rather keep my minimalistic UI and be able to see more data.
I suppose this is highly subjective, but have to agree that I am not that impressed with the Oxygen theme; it looks very heavy looking to me with little contrast and doesn’t go well with the black used for plasmoids, etc. Just my opinion, and my taste probably differs to most.
Some of the individual icons are quite nice, but on the whole the theme is disjointed, poorly laid out, and generally unappealing IMHO. Give me something like SUSE’s tweaked KDE4 theme any day of the year.
I used to be a Gnome user (couldn’t stand KDE 3.5), but KDE 4.1 really got under my skin and I just *had* to switch to it. I can’t wait to upgrade to Jaunty so I can get the new KDE 4.2 goodness 🙂
Try KDE 4, you might just like it (although Gnome is a little better integrated at the moment).
Considering how ugly Gnome is by default its rather interesting that a Gnome user sees fit to criticize KDE’s default theme (which is dead easy to change) without any sense of irony.
Just shows how taste is hard to discuss 😉
Many have fallen in love with Oxygen, but some don’t like it. Well, you can’t please everybody. At least Oxygen isn’t as IN-YOUR-FACE as the XP theme was
I especially liked the interview. Aaron was humble and honest. I don’t know but the whole tone of this “The answer” release is very low on marketing buzz words and hype. Very well done.
Personally I cannot wait for sonnet to get language detection. I will use 4.2 anyways
but I’ll switch once 4.3beta is out.
Not that bad, not that great.
Anyone try Antico (Built on Qt 4) as their desktop yet?
http://qt-apps.org/content/show.php/Antico?content=93778&PHPSESSID=…
http://qt-apps.org/content/show.php/Antico+Deluxe?content=95422&PHP…
I keep updating it on one of my systems. Nice project but still a long path to go.
And once again, the discussion is only about whether or not releasing 4.0 was a good ting or not. TBH, I’d say that whole discussion is off topic in this news.
Sheesh, some people will just never let go.
Many thanks to the KDE devs! Awesome work!
Watching those videos was indeed an eye-opener. 4.2 looks like it has a achieved a level of polish I think the devs can be very proud of.
I also want to take this opportunity to correct the misconception I may have created made when recently I mentioned 4.x’s “shortcomings” and subsequently got entangled in a huge debate about whether there were any or not. Namely, that on the whole I don’t think KDE 4 sucks, as a matter of fact I never have, and in fact I think KDE 4 is awesome.
Yes, I whined a bit that the theme is too gray, the preferences still too haphazard, and the icons still not allowed on the desktop by default
. But these nitpicks are absolutely nothing compared to the awesome benefits one gets: Subtle and useful desktop effects! Expose-like functionality! Custom desktops for every season! A ridiculously complete and awesome suite of applications! Awesome technologies for developers!
From this point on I will only sing KDE 4.2+’s praises. (After I try it out myself of course.)
i think 4.0 was only a disaster from an end-users perspective, from the perspective of a developer that realizes people don’t test beta/alpha software I don’t think it could’ve been more successful.
I suppose if you think alienating the user base into never trusting you is a good thing, then it’s a success.
On the other hand, most projects, both in the computer world, and in the rest of life, consider alienation of trust as a colassal failure, no matter how good the end product — and for good reason — if nobody trusts you, then they’re never going to use (or contribute to the further advancement) of your project.
I will never trust another thing Aaron Seigo is involved with in any capacity beyond “coding drone” — certainly nothing in which he is part of the leadership. For the short term advancement of his project, he has sabotaged untold numbers of Linux conversions by the newbies who blindly installed KDE 4 not realizing that it was still a crock of alpha-ware.
Then again he played a big role in the development of technologies which made KDE ready for the future. The few chased away by 4.0 will be replaced by the millions who will finally have a good alternative to proprietary software in KDE 4.x.
… not that it’s a bad thing or so. Quite a lot of spaces, expose, active corners and coverflow in there
Edited 2009-01-28 03:40 UTC
I think with spaces it may be the other way around. Virtual desktops in general, but Kompose in particular dates back long before spaces was announced, I know Compiz had it before Apple did, but not sure if they thought of it before them though.
Just as applicable to software. 🙂
seriously , i just dont realllly get it !
its not like osX 10.0 , windows Vista, Gnome 2.0, X 4.0, kernel 2.6.0 , kernel 2.4.0, pulseaudio whatever version, alot of Fedora releases, and probably a whole lot of other examples, where stable and great releases out of the box !
its always the same with all releases ! the inicial release is always missing alot of things !
Alright, I wanted to get this point across.
I find the notion that version numbers are meaningless and that, at the same time, that the KDE 4.0 release was beneficial to the project to be contradictory, even borderline delusional.
If release numbers (however arbitrary) were indeed meaningless, then why would slapping the ‘.0’ moniker on a product attract users, and thus contribute to the testing and adoption of the software?
Why would the hardcore testers pay any more attention to the released version of a product than to the development version? The people who want to test beta (or alpha or whatever other Greek letter-coded development stage) software would have been already informed through the ample coverage on blogs and news aggregators and the like of the need of their services, and would be already running it as it is on the repository, without the need of a ‘.0’ tag hanging from it.
The only people who would flock to download such a release are the regular users, who would rather wait until the project reaches the fabled status instead of just downloading development snapshots because they have different expectations on released software. It is to that (rather big) segment of the user base to whom the version number appealed, because they assign a meaning to it, and if it was truly meaningless there wouldn’t have been any need or gain to label KDE 4.0 as such.
If it was not for these assumptions on ‘.0’ software, the release itself would have been meaningless. Now quit denying it!
Bingo.
People could dance across that point and try to morph it into not saying users were tricked in doing Beta/RC release testing by saying that it was not intended for every-day-user, but we have to go back to even what Seigo himself said (and which I quoted a few posts above this one)… .0 usually has a definite meaning and that meaning attracts developers, testers, and users and that is why they did not want to keep doing Beta builds and RC builds… they knew that it would have still taken them a year to get to KDE 4.2 (or at best 6 months or so to get to KDE 4.1) and they did not want to stay 6-12 more months in Beta (or more because they feared that there would have been less contributors so they would have taken even more time to get to KDE 4.1 and KDE 4.2…).
The primary meaning of a version number is what the developers of the product assign to it. If they say .0 mean, this time, that it is not ready for production use then it is YOUR problem, and no one else’s, if you fail to follow their advice and stubbornly continue to think it will mean what you want it to mean.
Different projects and products use version numbers differently. Sometimes this changes over time.
Grow up and and deal with it.
Edited 2009-01-29 09:35 UTC
And there we disagree. Words have connotations, and it is common knowledge, an implicit understanding, that .0 implies some maturity in the software. If there was hope that with new labeling more users would come, it would only be because they acknowledged that a released version is somehow distinct to a development one, and that this difference is somehow appealing to regular users. I think that from the backslash one could learn what this difference is…
I think that the acceptance of this distinction would be the mature gesture in this case. In fact, one could recall some arguments being made in defense of the release based on the notion that KDE is an application framework, and that KDE 4.0 was sufficiently mature at the time. Still, end users have nothing to do testing application frameworks, and there should be no expectancy of them doing so, .0 or not.
Sorry, but that is not the case for FOSS software. End user involvement in testing application frameworks is a critical part of the way that FOSS is developed. x.0 releases are precisely the timeframe in which this occurs.
If you, personally, do not want to be involved, then don’t be. Wait for a few releases to pass by. Wait for the x.2 or x.3 release. This applied for KDE 3 just as much as KDE 4 … KDE 3.0, 3.1 and 3.2 were very much developmental versions, and you really didn’t want to run it in a production environment sense until at least KDE 3.3. Likewise for GNOME … 2.4 was the first version that you would possibly want to run.
Your lack of desire to be involved during those early testing phases in FOSS (x.0 and x.1), and your lack of understanding that this is the way that FOSS is developed, in no way invalidates the process.
Well then, Soulbender, if I call you a murderer, and in my mind, and publicize it around the world, when in my mind, I assign the word “murderer” to mean “person”….
that would be OK with you, because assigning new, random meanings is not deceptively contrary to clear communication?
Sorry, MORON, but you’re just another disingenous sophist.
Wrong, you know why dev’s slap on that 0?
It often means that they have remade a whole system, changed a lot of stuff and redefined everything.
Slapping on a 0 does NOT mean a system is stable, it almost ALWAYS means a software project has been completely overhauled.
If KDE didn’t go to the trouble of changing everything over to API’s and using a different paradigm, they would have just released 3.6,3.7 etc etc.
In the free software world x.0 releases have 95% of the time meant, “we are redoing a lot of things, a lot of things will change, do NOT expect it to be bug free” and 95% of free software products have shown that
Edited 2009-01-30 03:37 UTC
The version number 1.0 generally stands for “the first stable and feature-complete release”. Changing this expectation for every following .0 release to the opposite seems extremely odd and counter-intuitive to me.
If you redo everything and don’t have parallel stable/development release numbers, the only versioning scheme that makes sense to me is to give the whole thing a new name, possibly including a number. E.g. “KDE 4” version 0.1. Commercial projects often do that, and I don’t see why not.
Nobody expects a .0 release to be flawless, but it definitely shouldn’t be a beta version either.
No it doesn’t, and it rarely has (in the FOSS world). If you actually read what people have said in the comments, x.0 releases have never been “stable” (the only exception being 1.0) . Whether this be gnome 2.0 or any other piece of software, x.0’s have always meant the software has been remade or rewritten so therefore don’t expect it to be stable
Because you put a label of x.0’s being stable doesn’t mean the developers should follow that model
“(the only exception being 1.0)”
Which was my point. And no, GNOME 2.0 was not meant to be an unstable release, there has been a separate branch for this purpose (I believe it was 1.99 or something like that). Yes, .0 releases always tend to be unstable (GNOME 2.0 definitely was) because they are young, all I am saying is that I find it counter-intuitive and rather misguiding to assign a different meaning to 1.0 than to every other .0 release.
The main reason why is that 1.0’s are often seen as the first non alpha/beta release of a product, so they at least want the software product to be stable.
First impressions count, even in the software world, and if the first release of a product is very unstable or unusable it gets a bad reputation which is very hard to clear (even if you do fix the problems.
and that’s why you use qualifiers, like “alpha” “beta” and “developer’s release”… you moron3.0.
Yes that is one possibility, Jerk 1.0.
For Heaven’s sake, who was in charge to elect this horrible non-sense Vista-like theme in KDE4? There are of course other themes, but Oxygen is *ugly*. And where’s K3B/4 ? There is no K3B/4 !!! See what Linus said about *break everything*? I got the desktop, but where are the applications that were supposed to be using 4.x libraries entirely?
Here’s an analogy of a KDE menu:
About ->
————
About KBlock
About KDE
Enter your Address
Enter your Meal
Change System Date
Clean Hard Disk
Mop the Desktop
About KDE 3 (yes, we knew you’d want it)
Changelog
Visit Website
Pop Up MsgBox “WTF”
Useless Option
About QT
About Trolltech
The “3b” part of the name K3b stands for “burn baby, burn”. Hence, the KDE4 version will not be K4b.
K3b development seems to have hit a wall at the moment.
K3b is not part of the KDE desktop, it is rather a KDE application.
To burn CDs and DVDs on KDE 4.2, one can either use the KDE3 version of K3b compiled to work on KDE4, or one can use something else. My suggestion would be Brasero.
Even though Brasero claims to be an “application to burn CD/DVD for the Gnome Desktop” it is not particularly specific to the GNOME desktop at all.
http://projects.gnome.org/brasero/
PS: having said that, you would want to make sure it doesn’t try to install beagle, which is a Mono application.
It really is just a GTK GUI over the standard cd tools on Linux.
http://en.wikipedia.org/wiki/Brasero_(software)
I could write almost exactly the same description for K3b by replacing the text “GTK+” with “Qt”.
I have no comment on “prettiness” or “ugliness” or otherwise. Make up your own mind:
http://www.lampwebsitedesign.com/index.php?linux&release=KDE%20…
Edited 2009-01-29 00:24 UTC
I think Brasero and K3B use different backends because Brasero always errors out instead of burning a disc, while K3B does not. I’m a GNOME user who uses K3B because of this.
Hmmm. I didn’t have trouble with it.
OK, you can still use K3B with KDE4, but you must have Qt3 and the KDE 3 libraries co-installed. This causes no harm, but you do start collecting a lot of libraries.
Anyone know of a native KDE4 alternative to K3B?
There is a SVN version of K3B for KDE4, but I don’t know how well it works.
http://linux.derkeiler.com/Mailing-Lists/KDE/2008-11/msg00171.html
http://websvn.kde.org/trunk/extragear/multimedia/k3b/
The “3b” part of the name K3b stands for “burn baby, burn”. Hence, the KDE4 version will not be K4b.
K3b development seems to have hit a wall at the moment.
Hmm, good to know that. I absolutely hate mixing up toolkits, I never use QT apps on my GNOME desktop and as such I don’t wish to use any QT3 or GTK+ apps on KDE4 desktop..This just poses one issue: there is no QT4 burning application, is there?
I did try KDE4.0 when it came but it was a disaster. Of course the underlying tech showed promise..but the desktop itself kept crashing, it screwed up my GNOME mennu entries and file associations and so forth. I was planning to try KDE4.2 now that it is released, but there apparently aren’t RPMs available for my distro. So I guess I’ll have to wait.
I can’t see any problem at all with Qt applications on a GNOME desktop. For example, VLC and SMPlayer are both Qt4 applications (as opposed to being KDE applications) and they both run fine in a GNOME, XFCE or even lighter environment such as LXDE.
Qt applications are perhaps the very best applications in terms of being able to run on different underlying systems. For example, VLC on Windows is still a Qt application, and it looks and behaves almost exactly the same.
Likewise, I can’t see any problem with GTK+ applications on a KDE4 desktop. They too run fine in most distributions, because there is a KDE configuration tool available which sets the look & feel of the GTK+ applications to follow the KDE desktop settings. This works reasonably well, but one does have to tolerate GNOME’s awkward file selection dialog boxes when you open or save a file in such applications. Firefox, Thunderbird, OpenOffice and GIMP are all fairly standard GTK+ applications typically included on a KDE distribution.
I do agree about not having a mix of compiled-for-KDE3 and compiled-for-KDE4 applications. However, many distributions seem to be able to take a KDE3 application, such as K3b, and compile it for running under KDE4. AFAIK, this does not involve the KDE3 libraries, instead the applications are linked with compatible entry points to the KDE4 libraries. The result still looks like, and behaves like, a KDE3 application, but AFAIK it actually uses the KDE4 libraries. Amarok (less than version 2) and K3b are applications like this that, AFAIK, are KDE3 applications but compiled-for-KDE4 on some KDE4 distributions.
I can’t see any problem at all with Qt applications on a GNOME desktop. For example, VLC and SMPlayer are both Qt4 applications (as opposed to being KDE applications) and they both run fine in a GNOME, XFCE or even lighter environment such as LXDE.
Qt applications are perhaps the very best applications in terms of being able to run on different underlying systems. For example, VLC on Windows is still a Qt application, and it looks and behaves almost exactly the same.
It’s not the issue of not working correctly. It’s the issue of the application looking and feeling like it’s out of place. VLC does not in any shape or form feel like a GNOME application and as such I severely hate using it. That’s why I don’t even have it installed.
Make up my mind? That menu will never make up my mind. Instead, it will puzzle my mind. Whoever was in charge, convinced others, and made a big mistake. I believe KickOff is one of the reasons why so many users have left KDE (including Linus Torvalds).
Oxygene would be the second reason because it’s just flat, unconsistent, a breed of Vista-cross-OSX-Aqua thing. Third reason would be the crash-party show although there are many people saying it doesn’t *EVER* crash anymore. The fourth reason is the toolkit mix up and ugliness, Qt3 inside Qt4 based desktop, not mentioning GTK applications and the horrible way they behave.
Oh come on! How precious can you get!
Just set Kickoff to classic mode via a right-click, or alternatively remove Kickoff and put a Lancelot menu plasmoid on your panel in its place.
Problem solved.
Other “reasons” (really just preferences): all solved by a bit of settings tweaking.
http://randomtechoutburst.blogspot.com/2009/01/kde-42-im-tired-of-p…
In other words, “try the fish” (if this were a resteraunt) … or more literally, “try Bespin and Skulpture”.
http://www.linuxplanet.com/linuxplanet/reports/6651/1/
Edited 2009-01-29 22:26 UTC
http://www.internetnews.com/dev-news/article.php/3799116/Is+KDE+42+…
Of interest to people who may like to try out KDE 4.2, and assess the improvements for themselves, comes the news that “Fedora is planning on including KDE 4.2 in the upcoming Fedora 11 Alpha set for February 3rd”, and that “some features from KDE 4.2 were backported into the OpenSUSE 11.1 release last month”.
Also in the comments, one finds that “KDE 4.2 is now available in rawhide (Fedora development tree) and will be pushed as an update to Fedora 10 and Fedora 9 users shortly”.
Recent pre-releases of Kubuntu Jaunty Jackalope Alpha 3:
http://distrowatch.com/?newsid=05286
https://wiki.kubuntu.org/JauntyJackalope/Alpha3/Kubuntu
and Mandriva 2009.1 Alpha 2:
http://distrowatch.com/5290
both include KDE 4.2 RC1, so no doubt the next pre-release of these distributions will include the KDE 4.2 release.
Edited 2009-01-29 02:16 UTC
openSUSE users can find out how to get KDE 4.2.0 for all maintained versions of openSUSE here:
http://en.opensuse.org/KDE/KDE4#Factory_KDE_Project_.28KDE_4.2.x.29
http://randomtechoutburst.blogspot.com/2009/01/kde-42-im-tired-of-p…
I think this answers most of the KDE4 critics quite well. Your mileage may, of course, vary.
Hey, Lemur…guess what.
Users don’t like being lied to.
I’ve been withing the sphere of FOSS software since the early 80’s..and I’ve never seen such stupendously asinine behavior among any development group as what the KDE 4 team has done.
The whole FOSS movement DEPENDS ON TRUST.
Seigo and his underlings at KDE violated that trust.
There’s no excuse for it… PERIOD.
And likewise, there’s no excuse for morons and idiots like you defending that violation of trust.
“Users don’t like being lied to”
You weren’t lied to. Your incessant verbal excrement needs to stop. If it doesn’t, I for one am tempted to start speculating as to what your real motives are and possibly who might be paying you to post your offensive drivel since your overly shrill diatribes smell like the work of an immature and overly excitable astroturfer.
“I’ve been withing the sphere of FOSS software since the early 80’s..and I’ve never seen such stupendously asinine behavior among any development group as what the KDE 4 team has done.”
1. How old and grumpy you claim to be is irrelevant.
2. The only thing “stupendously asinine” here, are your own comments.
“The whole FOSS movement DEPENDS ON TRUST.
Seigo and his underlings at KDE violated that trust.
There’s no excuse for it… PERIOD.”
While KDE 4.0 was a disappointment to many end users, 4.2 is truly beginning to shine and the hard work they have done is starting to really pay off. No one has violated “that trust”. And since when did Aaron Seigo and the KDE team owe you anything anyway ? Its Free Software, it owes you nothing and nobody forces you to use it. So quit the self righteous indignation as if you’re owed something. If you think their work could be improved on then participate, start making commits and bless everyone with your supreme coding prowess and show everybody how it ought to be done. If you can’t code, then at least offer the KDE team some constructive feedback. Taking the very transparent and easily accessible work of the KDE team and accusing people of lying and violating your trust is completely ridiculous.
If you’re looking for people you who have “violated that trust” then talk to the people who pollute Free Software with highly questionable things like Mono and Moonlight. Go have a chat with the likes of Miguel de Icaza and his “we love all things Microsoft” pals at Novell and Redmond, who are serving no one’s interests but Microsoft’s.
“And likewise, there’s no excuse for morons and idiots like you defending that violation of trust.”
Morons and idiots ? Look in the mirror sunshine.
Akukis:
I happen to agree with much of your arguments of KDE 4.0 but would appreciate it if you could tone down your diatribe a bit. To compare coverage of the hard work of an open source development team to the work of the Nazi propoganda team is inappropriate and more than a little disrespectful.
In the opinion of many people, including some of the KDE team, 4.0 was a mistake or at the very least could have been handled better. And I expect it will be a debated part of KDE’s history for quite some time to come. But keep it civil.
The KDE team has made massive strides in the last year up to the 4.2 release. They have shown a keen awareness of what users have found lacking and have worked to add back features that are asked for. I have significant problems with KDE 4.X (including 4.2 final) that prevent me from using it as my primary desktop. At the same time I do feel compelled to acknowledge the great work being done.
The team went into the KDE 4 development cycle with good intentions and continue to work on it faithfully. I think with time the KDE 4 will live up to most, if not all, of the ideals that went into its planning. There will be some who are disappointed and move on to other environments. There will be others who like it and bolster the ranks of the KDE faithful. But based on history of other less than graceful development cycles by other products the woes will be forgotten by most soon after the product stabilizes again. 4.2 was not that stabilizing release but I think it will be coming soon. I am sure there will be some who never “forgive” and continue to carry a grudge against the KDE team and KDE in general, but it will be a small number.
I am drifting off the reason for this missive though. Regardless of your opinion of the product, the developers, or the advocates, be civil. Attacking distracts from the message you are trying to send. You are not the only one attacking but you have taken it up a notch beyond what is reasonable. At least for the usual fare of OSNews.