The KDE Release Team has decided to release KDE 4.0 this coming January. The release was originally planned for mid-December. The KDE developers want to solve a couple of essential issues before releasing. Having solved some of those issues, among which were glitches in the visual appearance, and in Konqueror, the KDE community hopes to have a KDE 4.0 that will live up to the high expectations for it. Read on for more details.
Meanwhile, the progress towards KDE 4.0 is astonishing. Most parts, such as the KDE Development Platform and a lot of applications are considered stable and well-usable.
Some parts of the desktop experience do not yet meet the KDE community's quality standards and expectations for a stable release. There are also some issues which need to be addressed upstream, for example a bug in certain codecs of xine that cut off audio fragments prematurely. The developers are confident to be able to release a more polished and better working KDE 4.0 desktop in January. The changed plans involve releasing on January 11th, 2008.
At the same time, the release team's call for participation is repeated. To make KDE 4.0 a success, your effort is needed. An overview of current showstoppers can be found on Techbase, KDE's knowledge platform.
This is also a call to the wider Free Software community, and also to companies working with KDE. If you have the resources to contribute, assistance in fixing the remaining bugs is most welcome.
Comments
Exactly. No need to rush, better deliver a better product. The reputation of the project will only be greater because of it; people will quickly forget about the delay if the product really kicks ass.
This week I tried RC1 from Ubuntu packages, but it seemed more like a beta than close to release:
*) The screen locker crashed.
*) Konqueror wouldn't accept cookies.
*) The system tray was in the center of the screen, always, and on top of everything.
*) When you tell it to shut down, it asks again what you want to do.
*) When I zoomed out and then back in, the wrench at the top right was partially off the screen.
I am not sure where to report these bugs - I put the crash into bugs.kde.org, but I am not sure if that's the right thing to do at this stage of development.
RC1 is over 5000 revisions out of date, so verify the bugs with a more recent version, first.
bugs.kde.org is fine; there is also the Krush wiki:
http://techbase.kde.org/Contribute/Bugsquad/KrushDays
and IRC channel, #kde4-krush on irc.freenode.net
All but one of those issues are fixed already. The one that's not fixed yet is that when logging out from the menu is chosen, it pops up the logout dialogue instead of directly logging out, hardly a serious issue IMO.
> This week I tried RC1 from Ubuntu packages
There is one issue too: Ubuntu's KDE 4 still has packaging bugs. So the bugs you see are not nessecairy KDE bugs. It's a bit unfortunate, since it gives KDE a bad name.
Ubuntu gives Linux a bad name on the whole, if you ask me.
Why? can you support your opinion in-depth?
Anyone can criticize, but opinions is better in pair with supporting explanations imho.
I think this was a good decision. I checked out openSUSE packages yesterday and it still has many issues.
the opensuse packages are really out of date... the panel isn't working correctly, the systray also not, on my system it has no background (yes i've cleared my .kde4 folder) ... i would recommend to try the KDE4 vm image to get a good overview of the progress KDE4 is making. You'll be suprised!
I'm using the KDE 4 repo for openSUSE (and get updates nearly every day), and all of these issues are fixed. I sometimes switch my session from KDE 3 to 4 by pkilling Kdesktop, replacing KWin3 with KWin4, and starting Plasma. I'm not pkilling Kicker, this causes issues with the tray icons of KDE 3 apps, but beyond that KDE 4 is working quite well.
Someone asked me for the repo URL. It is
http://software.opensuse.org/KDE:/KDE4/openSUSE_10.3/
The current version is 3.96.2-8.2 (compared to the old-as-the-hills 3.93.0.svn712059-6 in the default repo)
I think this is the right time to give a big virtual hug to the packagers at openSUSE. You deliver me a great Linux experience!
The service is very good but some of the packages don't work properly.
I have the same problems that you mentioned.
I think the same. Quality should be the highest priority and it would be very sad if kDE shouldn't live up to it'S name by releasing an unfinished "product".
The openSuse packages have some issuses that are related to SuSE and not necessarrily KDE.
> The openSuse packages have some issuses that are related to SuSE and not necessarrily KDE.
Like?
There are missing plasmoids, my background was missing for almost a week before it reappeared after a few updates and some things don't work as well as those that are compiled from source. Sometimes KDE4 on openSuse doesn't even start after an update so one has to wait for the next update etc..
There are rather few plasmoids contained in kdebase, for more plasmoids you have to install the extragear-plasma and playground-base packages.
I installed the plasma-extragear but there were still missing plasmoids. However I didn't install the playgroud-base packages, now I followed your advise and those plasmoids that I missed are there again. Thanks :)
This was obviously a good call. It's also obvious that it's nonsense to name these Release Candidates! Everyone knows they are not ready for release, so they're Betas. In any case, not much to do about that right now, good luck guys and thanks for all your work.
You know what a release candidate is good for as well? Testing if it is a release candidate!
Only now it's obvious that deadlines were missed and a schedule extension is necessary. Notice how the delay announcement is shortly after the release candidate has been found incomplete.
But you will never know, if you are not making deadlines, what people have sitting in their checkouts, and another Beta wouldn't be a hard enough deadline, or else things would already be there.
Yours,
Kay
> Only now it's obvious that deadlines were missed and a schedule extension is necessary. Notice how the delay announcement is shortly after the release candidate has been found incomplete.
No, I think it was plainly obvious beforehand. Look at the list of current showstoppers. Plasma:
> goal: A basic panel is there; current work is to flesh it out into a more capable panel system. System tray and taskbar both need work as does a desktop containment with legacy support for the Desktop folder.
This is clearly not ready for release, you don't need millions of testers to realise that. It is not the surprise that you are making it out to be.
> But you will never know, if you are not making deadlines, what people have sitting in their checkouts, and another Beta wouldn't be a hard enough deadline, or else things would already be there.
Nobody is arguing against deadlines, only against unreasonable ones that don't reflect the state of the code. And the idea that a release candidate is meant to prompt people to commit code is ludicrous.
> And the idea that a release candidate is meant to prompt people to commit code is ludicrous.
Well, so is reality. Actually It is quite clear that the code would have been far less ready if the original schedule stated the January release date in the first place. Honestly, who cares about the integrity of the term 'release candidate' and its correct usage. Whatever - let's rather finish this thing. Then we can call it 4.0-stable.
Thanks for waiting with the release. I prefer a stable KDE, over a rushed one. "Gut Ding braucht (wirklich) Weile" - Colloquial German for "Good things take time"
If I want rushed incomplete software I'll just stick with Micro$oft products. -That was a low blow :D
Seriously though. Please resist peer pressure and make it stable. The Google release party isn't till January anyways. :p This way it's more likely that KDE 4.x will make it to Kubuntu 8.04.
Also please make sure that KDE plays nicely with "eye candy" apps like Compiz fusion, etc. This will be the future in years to come, so if it works nice now and is future proof, there will be less struggles later..
Please make sure that the "desktop paradigm" is innovated further. Don't be afraid to be a step ahead of the competition, in the end, that's what makes open source great.
Compiz-fusion is not the future...
* Compiz and Beryl are the past.
* Compiz-fusion is the present.
* KWin >= 4.0 with builtin support for compositing and effect plugins is the future.
Yes, but we're not at the future yet. With the bugs and speed issues of Kwin at the moment, I think many will be using compiz-fusion at launch.
I still think they should have worked to incorporate existing compiz-fusion core into Kwin, and allowed Kwin to just use compiz-fusion plugins.
If you watch the talks given the last two years at Akademy by Lubos Lunak, he explains why this was not done.
Could someone please sum up these statements?
Living in the US, I can't make it out to Akademy.
Neither can I, but you can do what I do, which is to watch the videos recorded at the conferences:
http://home.kde.org/~akademy06/videos/
http://conference2007.kde.org/conference/programme.php
(some of them have some encoding problems, you just have to put up with the frame freezing)
Basically Beryl/Compiz Fusion just exposes its internals for plugin writers (instead of having a good API), therefore re-using its plugins it's not possible.
Some reasons not to use Beryl/Compiz:
* The "API" for plugins was using internals directly. So it's not usable outside Beryl.
* Effects happen to be a small portion of the window manager code. (like 10% for the total code)
* Window management is quite a black art. It was better to add effects then to throw all good code away.,
I didn't suggest that you throw out Kwin. I suggested you include the 10% of effects code from Compiz-fusion into Kwin, and try to stabilize a plugin API that both Compiz-fusion and KDE can use.
Frankly, re-implementing the effects would be far easier than a) persuading the Compiz Fusion guys to re-architect everything so that they have a neutral API and port all their plug-ins to it just so KDE can use them and b) actually doing the changes once you've somehow managed to persuade them and c) porting KWin and all its effects to the new API.
I read here on the dot (I believe) some time ago that kwin implemented compiz like functionality in a trivial amount of code. Compiz is something like research and kwin is something like realistic implementation of said research. Compiz is huge and buggy whereas kwin has all the kinks worked out, is lean, and is fast. This is mainly because it has been around for over a decade (right?). Last I checked the devs did make sure it works with 4.0 anyway ;) Guess that goes along with the whole "future proofing" idea as well.
I don't think the kde devs have as a high priority making kde work with compiz. KDE 4 will have its own composite manager that we hope will be competitive with compiz. Furthermore, compiz is very gnome centric and doesn't play very well with kde. Finally, the responsibility is on the maintainers of compiz to make it work well with kde, not the other way around.
I know people are eager to get their hands on "KDE 4.0", but if you don't care about stability then you can use betas (which will be exactly the same code as what the "KDE 4.0" was going to be, regardless of what it is named), and if you do care about stability, then you should welcome this delay.
I think this delay is a big positive step for the KDE project and I'm very glad they did it.
" If I want rushed incomplete software I'll just stick with Micro$oft products. -That was a low blow :D "
or if you want to be a permanent beta tester using sub quality software stick with open source software.
And if you want to be a troll, keep hanging around and post this kind of useless comments.
André, the truth hurts doesn't it?
I think this is a good decision and I would encourage to delay it even further, if necessary. At the moment especially the desktop experience with hardly any configuration options, does not seem suitable for daily use. I hope that can be resolved and that also katapult will be available (the new menu really, really sucks, sorry for that).
I do not like the "new" menu either and prefer the "90's" one. For a simple reason, you cannot see what is under the submenus while moving the mouse. I do not like the search text box (it should be an option), because I use either the mouse or the keyboard and I use only the mouse when navigating under the menu.
Someone please tell me that this horrible piece of unusability I just installed is the result of Kubuntu's mucking about, and NOT what I have to look forward to with KDE4! If this is what KDE4 is going to be, I'll switch to plain vanilla blackbox.
1) Konqueror is nearly unusable. Typing in the text field is painfully slow. Resizing fonts with Ctrl+ or Ctrl- results in garbled rendering and the inability to use this edit box. I am using Firefox just to post this.
2) I added the analog clock widget to my desktop. Stupid me. It's stuck in the middle of the screen with no way to move it. No combination of keys/mousebuttons can get it to move anywhere but the center of the desktop. The rotate (circular arrow) button does nothing. Is it supposed to?
3) I can find no way to change the wallpaper. It is missing from system settings. Neither is there a desktop menu of any kind. I tried dragging an image to the desktop, but all that did was result in an immovable widget (see above).
4) Even more lack of configuration! There are so many configuration settings missing that I feel like I'm using a GNOME parody instead of KDE. If non-newbies are expected to hand edit configuration files, the least KDE could do is document them!
5) And the worst, that stupid menu! I have to keep scrolling up and down, and clicking back and forward, to find anything. And since the "Favorites" is directly above the K menu button, it's nearly impossible to avoid moving the mouse over it. And true to form, there is no way to configure this menu. Do the developers really expect me to type in the search field every time I want to run an application? Once again, the default menu listing is item description followed by item name, leading so such bizarre results as "Board Game" (In larger boldface font) as the name of game. At least they could have put "Board Game" under "Board Games" instead of "Strategy & Tactics". Geez.
7) There's lots of little stuff, but I'm going to write them off as merely known bugs waiting to be fixed, or additional Kubuntu shenanigans.
On the bright side, Dolphin4 isn't the piece of crap the Dophin3 preview was.
I hate being so negative about a desktop I love, but this RC1 is a complete load of crap. Can we at least stop pretending that this will be suitable for use by non-developers?
i also think, that RC1 doesn't look or behave like a Release Candidat, it still should be called "beta", in my opinion, but maybe the backends and interfaces are already very stable, so that developers can say, "we are close to get something what can be named a release candidat".
well... they have a release party planned so you can bet it'll be called 4.0 even if it's still really a beta. i'm not going to touch it until 4.1
sounds like what my professor said about Vista, that he won't touch it until SP2. I don't think this reflects positively on KDE.
Do you take what some anonymous poster on a website says as absolute truth? Or do you make your own mind up?
I'm not saying the KDE 4.0 would come out bad, or at least I didn't mean to. if you'd notice what I said in a different comment on this page (I don't really expect it, but then you might be a better person than myself), you'd notice that I do not think Vista came out nearly as bad as people make it sound like, although I wrapped it in a lot of talk about how I recognize that Vista is inferior to my distribution. what I meant to say, and I guess I phrased it poorly, is that it reflects badly on the public perception of KDE, making it sound as bad as the image of Vista. if things like that keep being said, people might think of KDE as being as buggy, incomplete, and poorly designed as they think Vista is. I'm talking not about the state of KDE, but about KDE's public image. I didn't try KDE since the second beta, so I don't know what it is really like at the moment.
He is wrong, then. Vista is ready now, as long as you don't need software that doesn't work well with it (quite rare these days)
Yeah, and if you don't need your resources for actually running apps instead of just the OS. And if you don't need your battery life.. But hey aside from that it's great!
This is really not the place for this kind of discussion, but anyway... my battery lasts as long as it did under XP. About system resources, well, usually those who moan more about those are precisely those who have more to spare. RAM is damn cheap these days, and besides, when mid-range PCs are coming out equipped with 2GB anyway, no one sane should really give a damn about that.
And what if you get sent something like big.jpg, that consumes 1.75GB of your RAM? I'd upgraded my RAM to 2GB a matter of days before getting sent that. Good thing I use an OS that doesn't grab huge amounts of my RAM, eh?
Typed on a laptop with 256MB RAM, using 138MB with stuff like Kontact open.