Slashdot Mirror
KDE 3.4 RC1 Released
twener writes "The KDE project has announced the first release candidate of KDE 3.4 which brings many new features targeted for release at 16th March. Sources (requirements list, build script), an i486 GNU/Linux Live-CD (375MB) and SUSE 9.2 binary packages are available currently.
OSdir.com and tuxmachines.org have screenshots of this release. Source Code and a Live CD are available."
Windows Update = Fixing bugs
Maybe if you got more benefites from windows updates other than patching holes.
If you're going to Kompare apples to oranges go ahead. This is more along the lines of a service pack, or beta of a new version of windows.
The difference being, of course, you don't have to work with kde to do administration.
That said, I completely agree. I'll take it a step further and say they should snag the UI requirements that apple has available for software developers to ensure consistent look and feel on their OS ( I believe those are freely available ) and use that to redesign KDE once and for all. Or come up with their own list, but *stick* to it, and further, don't approve software apps for kde unless they follow the list as well ( note: not saying people can't develop for KDE without these requirements, just saying they won't be "officially approved" ).
The important thing is to make the entire thing feel consistent. If I right click in one window, I expect the same behavoir there as I do in any other window. That kind of thing.
Mod me down with all of your hatred and your journey towards the dark side will be complete!
I find it interesting that they released this as a live cd. As far as I can remember this is the first time a desktop environment released a live cd with their new releases.
This is certainly a trend I'd like to see more of. There have been times where updating to the latest version of kde or gnome could cause a headache that lasts several hours (yes even in debian where there are occaisional dependency problems especially in unstable). And there are of course some distros that take several weeks before packages are available. Of course compiling from source is an option, but remember if linux is ever going to be ready for the desktop, compiling from source has got to be just that, an option.
But with a live cd release you can check out the new features and decide whether it's worth the risk of a headache.
I'm still downloading the iso but I give KDE major props for releasing a live cd in addition to the source.
It seems the KDE team is emulating Microsoft's penchant for changing how major features of the interface work at frequent intervals.
I'm not a heavy KDE user, although I do develop programs under it. I can't say I've noticed anything that's changed much between releases, and certainly not "major features of the interface". Could you give an example?
Consistency is definitely good, but it's one of those short term benefits. If you want to keep on top of things you've got to be able to progress and evolve.
Take Windows for example... people certainly complain that the interface changes, but think about it. Does it really? It hasn't changed much since 3.1. If you used 3.1, you can use XP. You might be dazzled by the blue window colors, but basically everything is still the same.
Fact is, people don't like change, and people complain. What the KDE people are trying to do, are create a really awesome system. They're experimenting and trying things out. Let them! It's not like they're making any wild changes that totally blow your mind. After 10 or 15 minutes of using a new release you should be fine.
Come on, you can set aside 15 minutes to have fun and play around with a new system. Exploring rules!
Something tells me you might be the kind of person that would complain regardless: "Hey this is exactly the same as the last version, there's nothing new to explore here, this is stupid I'm not using it anymore!"
You can make it look like anything you want, what's your beef?
Want menus on top, like MacOS? Check.
Want clear background with no icons like FVWM? Check.
Want a Wharf-like sidebar with application buttons on the side instead of the panel like WindowMaker? Check.
Want to change the order and shape of the buttons in the window titlebar? Check.
Want the taskbar to sit at the top and not at the bottom, like AfterStep? Check.
Seriously, have you ever even USED KDE?
I lurk on the KDE optimize list and came across the funniest comment in an email there the other day that probably goes a long way to explaining the mess that is KControl. They were discussing reducing disk accesses for displaying icons (a worthwhile cause) by building a database of their location:
"Do you think that it can make sense to add an option in KDE Performance -> System to Cache icons location?"
It's just like they have some instinct to add options rather than taking decisions. Just profile the system with and without the cache on and if it helps, enable it. If building the database takes some time, spawn it as a low-priority background task. Don't push all that work off onto the user.
Those people are baffoons. How do you relearn: read screen , point at something with mouse, click on it?
Don't be surprised about open source programs rapidly changing. To a lot of people, their project is an educational tool. Interfaces and APIs are constantly being honed and retooled. Even the Linux kernel will trash APIs used for years and release totally new ones maintaing no backward compatability in the middle of a stable kernel release.(SATA,USB for example) KDE is going to use the latest techniques and abilities provided by Qt and KDE libs. Thats how those libs get tested and improved and ALL programs using those libs benefit from this.
In open source land, we don't have to hang on to old broken APIs and libraries. The best code *almost* always gets put into use and every project utilizing that code benefits, even if they have to make some changes to use it.
As a user, do you have to upgrade to this new whiz bang version? No. Is this policy the best for all projects, No. But,the end result in most cases is that developers of open source code are able to use the majority of their time creating (which is what we like doing) and less time 'maitaining' old stuff they might not find as interesting and hardly anyone uses anyway.
If the two methods of development, open source and closed source, were equal in all ways except for the fact open source developers had the freedom to trash old interfaces and closed source had to keep them around, who would end up with faster,tigher code?
I can definitely see your point of view. Progress does need to be made.
I think we're all inclined to complain about things we don't necessarily want, but I don't think I'm any more so inclined than anyone else.
With respect to Windows 3.1 -> XP. As another poster mentioned, you're not using KDE to admin the machine, whereas that's where I personally see a lot of change in Windows - the interface with regards to administration of the machine.
Where I (and some of my users - not nearly as technical as me) see change is partly in how everything looks. The widgets have changed a lot over the last two years. To less technical users that is a barrier - you and I might blow it off, but it can be an issue. Once you spend some more time, there are more things that have changed for what I can see as little reason - maybe someone's idea of organization changed over the years and so all of a sudden, things move around on control center.
As I've upgraded from KDE 3 -> 3.1 -> 3.2 -> 3.3, I've noticed that many times my settings get lost and it's not so easy to go re-establish them. Things like keyboard shortcuts, mouse onClick behaviors for the various buttons have moved around. Heck, at one release the Control Center started taking on different forms depending on whether I selected it from the menu or selected it off the "dock" (can't remember KDE terminology).
Am I opposed to exploring for 15 minutes to use it? No, certainly not - but when I use it so little, I am more accustomed to metaphors of other desktop environments, and using KDE 6-8 times per month becomes a chore - again - it's been easier to just use XFCE.
it's not that simple really. radical changes in the user interface require high learning curves to transition to. it's not often when these changes come around, and any major paradigm shift in guis should happen gradual to ease users into a new way of using the computer. it's difficult to do all at once because people will think it's just too 'different.' everyone really borrows ideas off each other in the interface world. pretty much all environments now are a mix of the attributes of others. gnome is heavily mac os influenced. kde is heavily windows influenced. windows is mac influenced and mac is mostly xerox with a bit of creativity and HCI behind it. it's a natural evolution for the open source choices to take the best of both and try and merge them.
- tristan
well, "intuitive" is a sham. there are no intuitive interfaces, no, not even the nipple (which is also learned.. look up "nurse latching" on google).. there are only learnable and familiar interfaces.
.. but sometimes i wonder if this seemingly common concept really isn't just a troll but a deep seated misconception.
but since people often can't find where this particular setting it, i'd agree that it's certainly not familiar. and while it's learnable, the learning curve is apparently a bit too steep.
the Control Center is one of the things that will be massively reworked for KDE 4.0. we've been holding off until 4.0 to do that for a couple reasons, one of the major ones being not to ruin the familiarity of the control center to people who have learned it. we happen to care about our users and their time investments =)
but having a setting for default web browser that isn't immediately findable does not make KDE not ready for the desktop. if that were the case, nobody would be able to use Windows or MacOS either, both of which have nicely hidden features that are difficult to find unless you are familiar with the system.
i know i know, don't feed the trolls
That said, I completely agree. I'll take it a step further and say they should snag the UI requirements that apple has available for software developers to ensure consistent look and feel on their OS ( I believe those are freely available ) and use that to redesign KDE once and for all. Or come up with their own list, but *stick* to it, and further, don't approve software apps for kde unless they follow the list as well
I'm beginning to come around to the point of view that perhaps Human Interfeace Guidelines ought to be enforced programmatically rather than as a document of requests of things you would like a developer to do. What do I mean? I mean, try and create a new layer of abstraction for programming so that GUI construction and layout becomes the responsibility of desktop environment rather than the responsibility of the programmer. The application developer has the responsibility for coding the various components of the application, exposing their functions in a particular way, and the relationships between components; then the HIG (a DE library, or formal code spec, whatever) takes the components, features, and relationships and constructs a GUI to make those components and features accessible to the user (via menus, buttons, list selectors, dialogs, etc.) structuring them, according to the given relationships, into a completely HIG consistent GUI app.
Okay, that is a very non-trivial exercise, and exactly how much work you can get the HIG to do instead of the programmer is not an easy question, however, thee are some real gains if you can actually do (at least some level of) this right:
(1) All applications coded with this will automatically be very consistent with all others, and complete HIG compliant - the developer doesn't have to worry so much about UI design (that work is pushed off more toward the people writing the HIG engine).
(2) Each DE can have their own programmed HIG, so an application coded with this system can be compiled against each DE and be fully HIG compliant for each different DE.
(3) It completely formalises the HIG - it isn't a document of reccomendations, but is required to be an actual formal layout engine.
(4) For the really "do it yourself" people you can code your own HIG engine and have your own completely unique look and feel that will be consistent across all the application coded with this system.
So perhaps widgets are too low level for application programming these days. It is my understanding that AWT worked something along these lines at least for layout of widgets), and well, obviously AWT isn't very popular. Then again, AWT was a little slow, and didn't provide the flexibility I'm talking about (it had, presumaly, its own hard coded HIG engine). As I said, this sort of thing would be very difficult, but perhaps it is worth considering.
Jedidiah.
Craft Beer Programming T-shirts
It seems the KDE team is emulating Microsoft's penchant for changing how major features of the interface work at frequent intervals.
I think in software we call that "active development".
While I'll be the first to agree that consistency is key to a stable system, Linux is an environment that, in my experience, is continuously shifting and improving. Yes, things move in KDE, but they also are moving most things in a generally consistent configuration - i.e. most configuration is now available in the Kcontrol panel. They're actively developing the software and with each release, KDE gets more and more complete as a desktop environment, but you have the hiccups that are inevitable when something is changing that much (overall for the better, in my opinion). The interface changes because they add features. I'd say it's a case of taking the good with the bad for the best end result.
-Jay
I've been using it for 5 or 6 years, and I also can't think of any truly major UI changes over that time. It's mostly been a gradual evolution of small improvements.
Certainly nothing as jarring as the Win2K -> WinXP changes, where in the defaults for navigation the directory tree mostly got replaced by wishy-washy wizards that try to second-guess what you're doing.
> but all this seems to be is a few more
;)
> appliKations thrown into the default KDE
> install
if you actually try it out, you'll find that there are many, many bug fixes and improvements in existing applications and libraries. it's much more than just "3.3. plus a couple new apps"
> Choose the best one, and if people like
> something 'different' allow them to install it
> themselves
this is currently something left up to the operating system vendor/integrator/distribution. most current mainstream distros of Linux do ship KDE broken up into applications. there has been an interesting amount of conversation within the development community on how to best aid in the process without losing the benefits of definition but possible improving selection and selectivity.
it's not a trivial issue to "solve". but i'm glad its seen as one of the more visible issues (apparently, anyways), since things could be much worse. notice how nobody complains about how DCOP or other core technologies don't work, for instance?
Yeah, where's that three-dimensional holograph display they have on Star Trek?
Seriously though...a desktop has only so many ways to go about being useful. Either you'd be copying OS X or copying Windows, or copying fluxbox or copying...you get the idea.
Personally, I dont think KDE is "copying" Windows, I think they're using good ideas like menus, context menus, icons, and the file-manager interface, and then adding bits and pieces that are unique to linux or using good ideas from other OSes of bygone times.
-Jay
"Fanboy"? What's your problem, with the namecalling? You agree that "it's _great_ that we have a shared mime spec". And in fact, any minimally workable MIME system that interoperates is better than sophisticated ones that don't interop. Because interop is a feature that unifies the desktop markets for app development. That's hardly "every single thing that hits the pages of FreeDesktop.org" - it's a primary strategic goal for all Linux developers, and one to which KDE has agreed. It's past time for your anal-retentive perfectionism - the perfect is the enemy of the merely good. stop designing windmills to tilt at, and get with the program.
--
make install -not war
Personally, I think KDE's interface hasn't really changed all that much. Every year I try it out, it feels the same.
? set_albumName=KDE_rc1&id=default3
For instance, look at this screenshot from 3.4 RC1:
http://www.tuxmachines.org/gallery/view_photo.php
That is one MASSIVE menu. The same redundancy I've been complaining about for years--"System," "Utilities," and "Settings"--is still there. Why are they even seperate menus? Why not remove all three menus and just have a link to the system configuration dialog? Oh, that's right, they have that too! That's four redundancies.
Why is there an "Edutainment" menu? Why is there an "Editors" menu? There should just be an "Applications" menu, and they let the user categorize their apps the way they want to. That menu is suffering from huge clutter overload!
And look at the apps. Basically, they have two names each. Instead of "AppName," you have "AppName (WhatItReallyIs)." Silly and redundant. If the original name isn't working well enough, rename it. Essentially, you're having to maintain two app names now instead of just one. When a name isn't descriptive enough, its icon should be--that's why Apple insists that OS X icons show the document type and some sort of action happening to that document or related tool, like the text editor showing a page with a pencil overlaid on it. Not all icons follow these guidelines, but they should, and the ones that do fit visually in the interface. Fishing through appnames with parenthetical descriptions is ugly and time-consuming.
Those are just a few examples. KDE is overloaded with buttons, tabs, sidebars, and input fields. A lot of that stuff is simply not needed but is only there because it seems like someone got happy with the form designer and stuck a bunch of stuff on all the forms to have multiple ways of doing things. You should have two or three really good ways of doing things, not seven ways that clutter up the really good ways.
What you describe is a whole field of academic research, running at least since the early 90s. They're called User Interface Manager Sytems and yes, it's difficult; that's why you haven't seen it in any new commercial interface.
The closer you can get in non-academic world are the XUL and XAML architectures, which could be a basis for this kind of system to be built on top of them.
Singularity: a belief in the "God" idea with the "demiurge" relation inverted.
> not actually doing the job of laying out the GUI
yes, this is exactly what XMLGUI does for menus, toolbars and even context menus (though not enough apps use it for context menus still). there are people playing around with it right now to extend it to simple dialog layouts as well.
additionally, there are APIs that apps use which allow them to say "configure the keyboard shortcuts" and the actual libraries handle all the details of that. it's exactly one line of code in a KDE app to configure shortcuts, or rearrange toolbars...
we can certainly take it further, and likely will, but we've been walking down that path already for some time.
this is, btw, one of the reasons why KioskTool works so well =)