The KDE Project is happy to set the first beta of KDE 4.1, codenamed Caramel, free today. KDE 4.1 is intended to meet the needs of a broad range of users and we therefore respectfully request you to get testing Beta 1. Beta 1 is not ready for production use but is in wide use by KDE developers and is suitable for testing by Linux enthusiasts and KDE fans.
Highlights of 4.1 are a much more mature Plasma desktop shell that returns much of the configurability that was missing in KDE 4.0, many more applets and look and feel improvements, the return of Kontact and the rest of the KDE PIM applications, and many improvements and newly ported applications. The feature set is now frozen, so the developers look forward to using June and July to metamorphosing your bug reports into rock solid code, completing documentation and translating everything into your language. A series of Bug Days where users can contribute quality assurance to the release will continue towards 4.1's final release on the 29th of July, so watch the Dot for details.
For more details, see the release announcement and info page or if you are at LinuxTag, see KDE 4.1 being presented in Berlin this Friday.
+1, please don't release another version of kde4 without proxy support, I'd really like to use kde4, but no internet access from kde apps is a deal breaker for me.
Apart from this, I'm really looking forward to trying it. Thanks to all the KDE devs.
I do not use proxy.. but: +1
I do not understand how a system like KDE can ship without such important feature.
+1 (same for me)
+1: I don't use proxy, but it does need to be implemented.
+1 need proxy support.
Is there any trick to make it works ?
This patch works fine:
Have fun :)
[09:44] the patch is wrong
[09:45] really really wrong
[09:45] it makes support for SOCKS5 work, but it breaks proper HTTP proxying
[09:45] well, it would work for HTTPS, but HTTP would be ... stupid
[09:45] instead of sending GET and POST commands to the proxy, we'd be sending CONNECT
Which bug is the patch attached to?
http://bugs.kde.org/show_bug.cgi?id=155707 I believe
Yes, you are right, its in comment #46
FYI, this patch is included in Fedora's current KDE packages (the 4.0.4 update for Fedora 9 and the Rawhide 4.0.80 build).
Last I knew, proxy support was included in someone's summer of code project.
Hopefully that was one that made it in. :P
Any news about proxy support?
Proxy is still not working
The skills needed to "add nice GUI effects" are completely different to those needed to add proxy support. Don't try to spin this as a "KDE guys only care about eye-candy hurrrrr".
Well, but I guess it's much more complicated to implement GUI effects (the math!) then proxy support.
Not as long as it's just dragging and dropping GUI elements around in KDevelop ;-)
for proxy, you have to know quite a bit about network internals - and apparently, there simply is nobody who does. If you know someone...
Not at all - not to detract from the achievements of the guys writing the kwin_composite effects, but most of the effects are just a few hundred lines long and mainly use OpenGL to do most of the 3D "math" stuff. The "core" stuff by Seli, Rivo et al was probably pretty hard-core, though.
Proxy support requires an intricate knowledge of networking stuff and, even if it isn't "hard" as writing kwin effects (I think it's harder, but there we go), it's certainly more "niche" than this kind of graphics stuff and less people are likely able to write it.
Proxy support does not require the same amount of prior knowledge to learn than graphical stuff. The data structures simply are not going to be as complicated, unless you write highly scalable apps that has to communicate with hundreds of servers or databases but I don't think that's the point of proxy support in your *desktop*.
The amount of algorithms and data structures you have to know to write a 3d engine is sickening. I always wished I had the time to learn the stuff but it is just too big and time consuming and unrelated to what I do nowadays.
Hell, at least using proxychains kopete works .... techincally all it does is replace some libc's calls (probably connect/read/write or such) with versions that do the same but using a proxy (http(s)/socks) connection.
I suppose that this method and probably the code too can be used by kdelibs ... Or hopping that some poor soul at Qt will add that support there which is probably even more "right way", in some non graphical library, the networking one ...
Unfortunately i dont have much knowledge of internals of KDE (KIO part? ) to put my hands in there and then one has to count for the time it takes to compile all ....
Whatever, i digress. Hope that the proxychains or Qt ideas got the attention of some right person though ;)
I was wondering why plasma widgets can crash plasma with simple actions like remove widget.. ?? After the first crash plasma starts again, but with the second one, kde doesn't start again...only remove .kde* dirs helps.. :((
Please, do not remove the complete configuration directory. Try to delete Plasma's configuration first and see if it works. And file a bug report, the developers might know better where to look and can help you on that problem.
There are already some bugreports about this:
I hope they will be resolved soon.
http://bugs.kde.org/show_bug.cgi?id=158052 is solved with 4.0.2.
I don't know what changed in the last few days, but my last build from yesterday was downright snappy. Konqueror renders quickly, kmail loads suddenly.
Give your khtml devs some love.
No More Desktop Icons in 4.1 = No longer a KDE user.
Reading the article and comments rather than just the title would be even better.
And you should know you *can* have icons on the desktop... but that in fact, Plasma will not display *the contents of ~/Desktop* over the wallpaper by default. For that you can use the folderview widget, or drag items to the desktop from kickoff or dolphin.
The main difference is that now it won't be a display of a specific folder on your filesystem.
Mindless "BAWWWWW"-ing by mindless anons which won't read the above message ahead!
The folderview widget shows icons in a containment. Breaking ~/Desktop is a deal breaker for me. Aaron has talked about doing this for a while. He has said that he hates the concept of ~/Desktop, and repeatedly I've asked what is wrong with the concept, and how it can be improved. I've never seen a good explanation. So the preferences of a few are now the mandate of everyone. Why remove flexibility and choice for no good reason? If some people want to use folderview widgets, then let them, but why destroy ~/Desktop at the same time? This is not the KDE I knew and loved.
This isn't the first time such a decision has been made to remove functionality and choice, and it won't be the last. If I wanted minimal choice, I'd run Gnome.
Please read Aaron's post _fully_ before complaining...
I did read his post fully, and commented several times. And I've asked Aaron about this every time I've seen him bring it up here, or on his blog. I doubt you fully read my posts or comments.
Your comments were answered several times in that discussion. In short: this change enables better functionality in the future and you can still have your precious icons. This argument is about nothing, basically.
No Janne, they weren't.
Not once has anyone explained why ~/Desktop is broken and must be abandoned. You repeat over and over again that this someday will be better. You explain that you can see ~/Desktop in a folderview, but you continue to ignore my questions.
1 - What is so broken that we have to throw it out?
2 - Why throw out the old before the new is ready?
3 - Why can't we have a folderview applet, and still allow people to have icons natively on the desktop like before? Can't they exist side-by-side?
4 - Why make such a change right before the 4.1 freeze?
You also insist no functionality has been removed, which just isn't factual. I can't place files directly on my desktop. I either have an applet window cover my desktop (in which the icons also act differently now) or I replace the desktop containment with folderview, and have no wallpaper (and the icons still behave differently).
I had been championing for Linux installs on our new desktop rollout at work. I am no longer doing so. I do a slew of desktop work on the side for all my friends and family. I can no longer advocate KDE desktops to people. I certainly won't use it anymore myself.
Just a few days ago I was in an airport showing off my KDE 3.5.9 desktop and watching people's jaws drop (technically it was my wife's KDE install on her laptop). Now I'm supposed to show people a desktop without a wallpaper, with huge ugly black borders around icons, a whole icon fiasco, no proxy support, etc.?
This is the release we were supposed to wait for to switch for our everyday desktop?
Making this kind of change right before the 4.1 release was horrific decision making at best. These kinds of changes should be developed in an individual developer's git tree (KDE should seriously consider moving to git) and fully fleshed out into a mature, functional concept before integration at all, and should then be integrated as an option.
In the much the same way with Kickoff, something new and unpopular was made the only choice (not just a default, but the only functional menu) until enough uproar was raised to half-ass an implementation of a classic menu.
If the future of KDE is sacrificing choice, flexibility, features and stability for slower, less productive, buggy, restrictive interfaces just so something supposedly looks nicer (and I get widgets!) I might as well adopt Vista.
I feel a bit foolish for buying in so wholely into the hype of KDE 4 the past few years.
1. In the Plasma view of things, file icon plasmoids on the desktop containment would make absolutely no sense, since the Plasma desktop is not a file icon dumping ground, but rather a container for much more featureful items. (Which could include a widget that displays files in a folder!)
2. The old implementation was ultra-confusing, since files were represented as icons plasmoids. With folderview Mr. Upgraded-from-KDE3 has his files right on the desktop, and they work as expected, with the only difference being that they are set aside into a box. No confusion with the new implementation, plenty with the old.
3. Confusion, see above. Not having icons directly on top of the desktop isn't a big deal anyway, and anybody should be satisfied with the current implementation until the folderview containment has an option for wallpaper, thus being suitable to be used as a main desktop containment.
4. To save 4.1 from the confusion of the current implementation. I know many, many people would be dismayed to find out that they would have to live with the current way ~/Desktop is displayed. Right before the feature freeze would be as good a time as any.
KDE 4.1 will be viable for every day use. ~/Desktop icons are displayed in a plasmoid by default and work just as expected. Nothing prohibiting daily use here.
>In the much the same way with Kickoff, something new and unpopular was made the only choice (not just a default, but the only functional menu) until enough uproar was raised to half-ass an implementation of a classic menu.
I should remind you that this was done before the release. It's not like KDE 4.0 was released without a choice. If people want to improve on this alternative choice, then let them do so! (I personally would like to see the Traditional Menu get more features, but I will have to wait until somebody improves it)
>If the future of KDE is sacrificing choice, flexibility, features and stability for slower, less productive, buggy, restrictive interfaces just so something supposedly looks nicer (and I get widgets!) I might as well adopt Vista.
Good thing it isn't. The infrastructure is all in place for numberless different, and innovative Plasmoids and containments so that your desktop can become a completely customizable and more productive environment. That is the future of KDE. With Vista, you'd have extreme luck if you'd ever get any new features to your desktop. You'd be restricted to one type of start menu, and you would be unhappy.
For what it is worth, thanks for your response.
I still can't imagine why this change is necessary, especially right before this release. More and more distros have focused on Gnome, and if KDE 4 isn't ready to be an every-day desktop now, I fear KDE will lose ground even more.
Given that the desktop in KDE 4.x is a containment, I'm not sure why icons had to be removed, only to implement a completely new containment rather than code the desired improvements into the existing system.
There are some concepts out there that appeal to me, such as automatic filtering of displayed icons based on tasks, but again this seems like a lot of work that could be simplified into expanding already existing multiple-workspace concepts.
Icons are certainly in a weird state in KDE 4, and I don't see how this improves the state of icon behavior. It seems like staining the drapes so you don't notice the stain on the rug. Why spend developer resources on something new, when that time could have been spent working directly on how to better interact with icons?
Icons are now plasmoids as I understand it, which is part of the reason they have that odd border now. Icons are displayed in an odd manner, and handled in an odd manner. They advantage seems to be that I can rotate them, and resize them so my desktop looks silly. I'm not sure how much of an advantage that is. It certainly doesn't boost productivity, and it makes my desktop look less "clean".
And the odd icon-around-icon implementation just doesn't seem intuitive, but perhaps that is because we think of an icon as a symbol, not a widget itself. When people want to interact with an icon, they either click on it (default action, often open) or they right click for more detailed options. If you want to switch to a more graphical interface here, you could use a radial menu which would address a few concerns instead of hovering producing these small icons that are hard to deal with.
Having a right click produce a clear, decent-sized radial menu would solve one problem with many context menus. To select one item in a typical context menu, I have to hover past many other options, which involves then expanding the menu outwards over every option I mouse over, sometimes that involves some computing to figure out which sub-options are available. In a radial menu, you can more directly move to the option you want.
Even better, when you right click, you get a radial menu which shows these new icons as common choices around the icon, and the option directly to the right would open a traditional, full context menu.
How does a radial menu replace the overlay icon? The overlays are there so you don't need a right mouse button (like with touchscreens). A radial menu could be combined with an overlay icon - the icon is a way to access the menu.
Anyway, the change in icon behavior you are angry about is done because it would be relatively hard to make icons work like they used to - and yet it would, in a sense, be a waste of time. After all, plasma is about trying new things, trying to innovate. Innovation is difficult, often doesn't work and takes time.
Now you probably don't want us to do anything new and original, but we think we must. So we're gonna ignore those who are holding us, and The Free Desktop, back. Sorry about that, but we aspire more than just appealing to a few geeks. We want to overtake Apple and MS - and for that, we need to be clearly better.
And to be clearly better you need to to things different. Try new things. Innovate. So that's what we are trying.
If you can be constructive, and tell us better, new things to solve old problems - we will listen. If you can only scream 'I want it to work EXACTLY like KDE 3.5.x - then just go to KDE 3.5.x and never look back at 4.x because it WILL NOT HAPPEN. We sure want the same features, but NEVER the same THING.
We're going for the future, and we're leaving the old behind. If you can't keep up, stay with 3.5, nothing else I can say.
Why do people continually insist I am against anything new, especially when you are directly responding to a post where I suggested something new?
When Win 95 launched, people had the option of loading Program Manager if they wanted. When Windows XP had complex theming, they gave the option of going back to a classic look, and again with Vista.
Again, I've only said about ten times that I'm not opposed to choice and new ideas. I'm opposed to the philosophy of forcing people into a new way of doing things against their will.
OK, I'll try to be as specific as possible here....
Was the old method with desktop-folder horribly broken? Well, from user point of view, not really. I mean, it worked right? The problem with it is that it's inflexible and stsatic. So you have a folder, and the contents of that folder are displayed on the desktop. And that's the only way that feature works. If you want it to work some other way, tough! Not gonna happen!
What this change does is that it gives you, the user a lot more flexibility on this matter. You just want to display a bunch of files and folders on your desktop? Fine, you still have that. No-one has taken that feature away from you. So take a deep breath and repeat that fact: no-one has taken your precious icons away from you.
With this new system you can do things that you couldn't do before. You can display icons from several folders if you want to. And those folders can be remote (hell, they could be on the opposite side of the world!) or local. You could have filters that only show certain kind of files. In the future you could have automation that changes what files and folders are displayed, based on what you are doing with your computer. That was not possible with the old system.
"3 - Why can't we have a folderview applet, and still allow people to have icons natively on the desktop like before? Can't they exist side-by-side?"
Pray-tell: what would you accomplish by that? Could you please tell me what functionality "icons on desktop" gives you that this new system does not give you? You seem to be focusing on the applet, while you ignore the fact that you can use it as a containment as well.
"You also insist no functionality has been removed, which just isn't factual. I can't place files directly on my desktop."
yes you can.
"I had been championing for Linux installs on our new desktop rollout at work. I am no longer doing so"
Why? because you incorrectly believe that one of your favourite features has been removed, and you therefore feel that everyone else will hate Linux as well? And last time I checked, there are other GUI's besides KDE available for Linux.
"ust a few days ago I was in an airport showing off my KDE 3.5.9 desktop and watching people's jaws drop (technically it was my wife's KDE install on her laptop). Now I'm supposed to show people a desktop without a wallpaper, with huge ugly black borders around icons, a whole icon fiasco, no proxy support, etc.?"
Those black borders are determined by the theme you use. Don't like them? Then use some other theme. And aren't those black borders part of the theme for the plasmoids? Do you still have those borders if you are using the folderview as a containment instead of plasmoid?
"Making this kind of change right before the 4.1 release was horrific decision making at best."
No it wasn't. It was part of the development-cycle and it was done before the feature-freeze.
"If the future of KDE is sacrificing choice, flexibility, features and stability for slower, less productive, buggy, restrictive interfaces just so something supposedly looks nicer (and I get widgets!) I might as well adopt Vista."
The old desktop & icons was exactly what you describe. It was inflexible, featureless and offered the user no choice. Either you used it in the only way it worked, or you didn't use it at all. Now that users are presented with a system with more features and flexibility, you whine?
"I feel a bit foolish for buying in so wholely into the hype of KDE 4 the past few years."
No, you are being foolish, because you incorrectly believe something to be true, and when dozen or so people line up to tell you that you are wrong, you cover your ears and shout "lalalalalala, I can't hear you! KDE 4.1 does not support icons on the desktop! I can't hear you!".
"Not once has anyone explained why ~/Desktop is broken and must be abandoned. You repeat over and over again that this someday will be better. You explain that you can see ~/Desktop in a folderview, but you continue to ignore my questions."
It's stupid to have hardcoded "Desktop" on your home, what you cant change to other name. On KDE this is possible, but on GNOME, it's not. (Yes, it is GNOME problem). You are always stuck to the english name of Desktop, if you change it to "desktop" or "desk" on KDE and you login to GNOME, it greates new one named "Desktop". If you have other localisation, you have all folders on your own language but then you have "Desktop" there as ghost what you cant ripp off.
Why you want to use desktop as place to store files? It's just a one folder. This function adds you A POSSIBILITY(!) to have TWO desktop in one desktop on multiple plasmoid desktop what are on multiple virtualdesktops. You get more control what you are doing on your computer, you are not forced to have your desktop files/icons on ~/Desktop. You can change those containers so you can have keep multiple project files on desktop, all pointing to correct folders on your home directory. Currently you need to add shortcuts to do that.
And this is first step. When Nepomuk comes, you could set smart-folder to your home, then set that as container to your desktop. No more needing to search stuff, place them to correct folders, when you always find what you need, in one container.
Why to force desktop to be a ~/Desktop? Why it cant be flexible and powerfull tool what allows more control to your files, icons and information?
Why it should be just a folder what has plasmoids/widgets/gadgets top of it? Why to stop there? Why not doing desktop as plasmoid/widget/gadget?
This is just a first step, dont blame it if you cant see what future can bring to you. Many said when they heard from iPhone that it's silly or stupid idea and Apple does not sell those so many. What happend... Very simple technology what brought new way to use the phone.
This is samekind technology, it brings new way to use your desktop, to use your whole desktop enviroment, whole computer! But it does not replace the old one. If you dont like it, dont use it. You can still keep the icons on desktop, you can still keep it as ~/Desktop.
It's up to you what you want! I tested that and I liked it much more than KDE3 style, I can see great potential for it and flaws of old style. I wouldn't suprise if Microsoft, Apple or GNOME would copy this idea.
When you use this new thing a week, you notice the difference. It's like you would be getting virtual desktops to Windows desktop!!
"Why to force desktop to be a ~/Desktop?"
Here is the kicker. If you don't mind browsing your files in an applet instead of directly on the desktop, then have a folderview applet as you please. See, we both win. I'm not against a folder with filtering and such. I'm opposed to not being able to store things directly on my Desktop.
~/Desktop has a few advantages.
1 - It is simplistic to understand. As a career IT guy, I've had to support tons and tons of end-users who aren't that computer savvy. I've always told people that Linux has this undeserved reputation of being hard to use. It is far more flexible and powerful, but that shouldn't intimidate new users. New users should be able to get their feet wet without being scared off. Given that the file structure of Linux is considerably different from Windows (a disseration for another day, but do we really need to stick to Unix roots here and have binaries in 20 different folders?) people need to know where they can find things. Desktop passes the Grandma test.
2 - Desktop is fast. Many apps default to saving to the Desktop. I don't let things linger there, but I throw things there when I'm in a hurry to deal with later.
3 - Desktop is a reminder. When those files stare directly at me, I know what I have to deal with. I could drop things in /home/foo instead, but then I have to pull up a window to look at those things, as opposed to having a built-in reminder right in front of me.
4 - No one is forcing you to store things there if you don't want. Again, you can have any folder you want in a folderview applet, or you can have a completely blank Desktop. Do as you will here. But don't take my choice away from me.
Now, if the change were transparent (the new containment was fullscreen, transparent, had a wallpaper) and in practice offer the exact same functionality and more (as it perhaps might in 5.2) then no one would raise a stink. Expecting me to set the current containment as a desktop with no wallpaper, or have an ugly applet cover my Desktop is not the same thing.
So you are complaining that a feature that the developers plan to implement in the future hasn't already been implemented? Sheesh, and I just went ahead and wasted my time explaining stuff.