KDE getting ready to go native on Windows

KDE getting ready to go native on Windows

Summary: A native Windows port for KDE's graphical framework is under development and could help the Linux desktop environment attract Microsoft users, but some fear the move will harm Linux

The Linux desktop environment KDE is moving a little closer to Microsoft Windows as developers ready a release of QT, the KDE graphical framework, that will run natively on the proprietary operating system.

Ralf Habacker, a KDE developer who initiated a project to port the Linux desktop to Windows, said over the weekend that a native Windows port for Qt, the KDE graphical framework, will be released under the General Public License (GPL).

"The native GPLed port of QT/Win 3.3.3 will reach production state soon," said Habacker.

He said that people who have used the native port to build releases of Linux applications have only reported "minor problems".

At present, KDE fans who want to run the desktop on Windows have to use the Unix emulator Cygwin, which reduces performance. Habacker said in a recent post on the Sourceforge forum that performance is likely to be improved by running KDE natively on Windows.

The 'KDE on Cygwin' project team believes a Windows port is important as Microsoft users can try the open source desktop without switching their operating system, according to the project Web site.

"We think that KDE is a great desktop and has the opportunity to be a big player in the 'desktop environment' area," said a posting on the project's Web site. "Windows is the OS of choice for many companies. How does this square with the promotion of KDE? The answer is simple: build something that allows KDE applications to run atop Windows. This is the goal we are aiming for with this project."

But some KDE developers disagree that the desktop should be made available on Windows, and Habacker agreed there are differing opinions over the project. "Some developers like this idea, some do not," said Habacker. "The first group sees more the practical advantages, while the latter has more political objectives."

KDE developer Aaron Seigo said in a recent blog posting that making open source applications such as KDE, Firefox and OpenOffice available on Windows, means that users are less likely to switch to Linux on the desktop.

"By porting Free Software to Windows one increases the number of valuable applications on that platform," said Seigo. "Since application availability is a key factor in operating system usage, we can do the math pretty easily: if Windows has Microsoft applications plus the stable of Free Software apps while desktop Linux/BSD/etc has only the Free Software apps, why would anyone in their right mind switch to Linux/BSD (and incur the training and data migration costs) when they already have all the software they need and want right in front of them? They have no reason to. None. Ergo, they won't."

Topics: Apps, Software Development

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • If you really want Posix for Windows.: http://www.microsoft.com/windows/sfu/default.asp

    And it's free.
  • History repeating itself ...

    To those advocating that it's a GOOD THING to release the KDE framework for win32, as well as releasing OOo and the Mozilla family of apps, beware and read the history of OS/2 !

    When OS/2 Warp 3 was released in 1994 it was touted as a "Better windows than Windows ..." because of its optimized win16 sub-system. Same whenever Warp 4 was released in 1996, only this release also included native Java support.


    "Our win16/32s/java applications runs just fine on OS/2. There's no need for us to spend resources for a true native OS/2 port ..."

    Of course by the time the win32 API was truely past being usable under Windows 3.x it was too late for OS/2. This IMO was one of the major reasons OS/2 failed to gather enough of a market. Yes, IBM charged too much for the SDK. Yes, Microsoft played dirty with IBM during the Windows 95 launch, etc ...

    But had OS/2 a solid chunk of the marketplace this wouldn't have mattered as much.

    As one who live through these days, smug in my knowledge that OS/2 was soooo much better than Windows 95, take my word for it. Getting the vast majority of users to change to Linux when they are running OpenOffice.org and Mozilla or Firefox, among other opensource applications will be extremely hard to do.

  • I look at it from the other point of view, sometimes I still have to use Windows, but I don't like the UI, being able to run the KDE desktop under Windows would be nice.

    Another thing that the comments fail to note is that KDE is a graphics desktop and is (fairly) operating system independent, it has just come to fame as being associated with Linux, but it runs on most forms of *nix.

    I switched to Linux for several reasons, and none of them had to do with the availability of KDE or politics. There are a lot of good reasons to switch to Linux, and once there choose a graphical environment which suits you...
  • So I hope to be able to use http://www.reactos.com/ and KDE in a year or two?
  • There are a few *HUGE* differences between Linux and OS/2.
    1. Linux already has a fairly strong marketshare. OS/2 never did.
    2. These are Free/Open software projects being ported. If you use windows, but otherwise fully free software, when you go to get your next computer, you are more likely to get a Linux machine, why? because it's gonna be cheaper.
    3. Some of us have to use windows sometimes. if we can have the same software we use everywhere else available to us, great. it'll maintain our sanity
  • I can't argue with your 2nd and 3rd points, but as to your first, sorry but this isn't so.

    OS/2 Warp sold so well for general desktop usage that it topped PC Magazine's sales chart 2 or 3 months in a row.

    InfoWorld readers rated it so high that it won BEST PRODUCT like 5 times, so much so that it caused InfoWorld to literally change how they allowed voting on their site since one of their editors just couldn't believe that people liked OS/2 "that" much.

    EggHead Software as well as CompUSA carried all available OS/2 desktop apps and from what I understand they sold very well.

    There are *STILL* fortune 500 businesses that use it on the desktop. It also brought alot of money into IBM, but since they make literally billions and billions the profits didn't look as great.

    But in the end IBM OEM'd OS/2 out to Serenity Systems, which they now market as eComStation, or eCS for short. And again from what I understand IBM expects quite a lot of sales from their OEM partners, and SS has far exceeded them every year.

    All strictly Desktop usage.

    Unlike Linux. Desktop penetration was so dismal that RedHat and others officially stopped their desktop distros of Linux. Wonder why ?

    Please don't get me wrong, I like Linux. And I look forward to using it fulltime whenever eCS finally does close up shop. I am realistic. Linux has the momentum.

    eCS doesn't, unfortunately.

  • I don't remember OS/2 *EVER* doing particularly well outside the banking sector, You played hell trying to purchase a machine running OS/2 which meant you had to buy it retail, at a healthy price. Thus making it's use in homes pretty much out of the question. I ran both OS/2 v3 and OS/2 Warp, and they were great in a lot of ways. but Linux has something OS/2 doesn't have. a large number of ISVs. most linux ISVs are coders writing GPL software, but it's still a very vibrant communtity. I don't personally care if commercial company X decides to port to linux. Eventually they'll pick up on the choice. do it, or be marginalized. why do you think there's really only one commercial competiter to apache? that's the only company that can afford it.
  • The availability of cross platform applications has made it easer for me to switch clients to Linux.

    For example, I will first get them to standardize on FireFox, Thunderbird, OpenOffice and sometimes Sunbird. After they are comfortable with those applications switching out the underlying OS is relatively painless, since the tools they use most everyday is exactly the same.

    Having KDE will make it even easier, because some many other applications will be cross-platform, so the transition will be that much smoother.
  • I thing that it's great for projects like REACTOS, but I think all the effort in doing it would be better used developing KDE further.

    With new services comming into play like www.CosmoPOD.com any user can easily experience a Linux desktop enviroment.

    With the push of OS apps onto all platforms, hopefully one day the OS and common application stack will become ubiquitous.
  • I definitely agree that the availability of cross-platform software increases the likelihood that a competing operating system will gain ground on the desktop.

    Most people and organizations want smooth transitions. Risk and performance loss are in many cases lower when one incrementally changes software instead of an instant switch. Also people only change if they have the fait that it will work out. Getting familiar with cross-platform applications on Windows gives that fait. These are established facts in change management.

    Please keep in mind that most people (the not computer religious ones) only look at direct (perceived!) benefits for themselves. They will not switch to Linux or another platform, unless:
    1) they perceive Linux as a better platform for their needs
    2) they think the available desktops and applications suite their needs (better)
    3) they think the troubles of the switch is worth the advantage
    4) they think they will overcome the possible interoperatebility problems with other people / organizations

    Just having a better (or cheaper) desktop environment / applications won't convince most people. If after all people are satisfied with open source applications on an open source desktop environment on a proprietary operating system, they will not switch to for example Linux. But, according to the reasoning above they also would not if the open source software was not available on their proprietary operating system.

    Think from a strength perspective. The open source operating system has to have (perceived) benefits for the user. Don't try to lock them in to Linux because it has better or cheaper software on top of it. Make interdependent switching of every software component on their computer as easy as possible. Only then will people chose their software on basis of benefits in stead of lock in. There is no point in trying to make people switch because you like them to switch. Make them want to switch, ... every component. Only that will lead to the competition that makes open source software development a strong model. Such will lead to better software, fait and finally exponential adoption of open source software.
  • When I was young, similar things happened with the Mac...

    First of all, Apple introduced a floppy disk that used a DOS compatible format, and lots of people were upset about this "contamination" (after all, the Mac's proprietary 800k disks *were* technologically superior) and that Apple was finally going to surrender. But in fact, Mac acceptance went up because people wanted to be able to exchange data between Macs and "IBMs".

    Then Microsoft adapted Word, Excel and Access for MacOS, and lots of people were upset about this too (after all, Claris Works *was* technologically superior) and that Apple was finally going to lose its special flavour. But in fact, Mac acceptance went up because people wanted to be able to use the same programs on Macs and Windows boxes.

    Okay... you are getting the picture. The Mac is doing better than ever by now. By contrast, Gepard and Sinclair, Atari and Amiga all failed, and Acorn's Archimedes, though by far the best machine in this list, has never risen to more than a shadowy existence among freaks. Why? Because they all were isolated solutions, incompatible with anything else. Ever seen Signum II on an Atari ST? The most amazing text processor I have ever seen. But... compatible with nothing. Interoperability, that's what counts in the Real World.

    So I think offering a Windows version of KDE alongside with the Linux, Solaris and xxxBSD versions (I think that's it, or has it already been ported to AIX and HP-UX too? :-) ) is the best thing to promote open source in general. When average users will have got accustomed to the KDE desktop, OpenOffice, and Firefox, they won't be afraid of switching to Linux. And why should they? Lord, if there were no reason to except for KDE, we'd better drop all our developmental efforts! Linux is vastly superior in terms of security, stability and performance, not to mention price and policy. That's the reason why.

    BTW, I think there is one inaccuracy in this article: Cygwin is not an emulator but a software compatibility layer, just as Wine for the other way, or X11 for MacOS-X, so the performance tradeoff should be marginal. Sure, it will be easier to install a native version.
  • Linux needs more, better drivers. Not Windows help.

    All the Windows compatibility stuff is nice, but in my mind, Linux needs to be installed. tried, used, and liked by armies of techs.

    And the hard-to-configure drivers always stop that from happening.

    Troubleshooting the drivers just doesn't happen, it's too frustrating for the linux first-timer, usually an experienced Windows tech, who then just goes back to the old-and-confortable setting - reinstall windows.
  • Possible shortsighted comments by KDE developer?

    It is important to get powerful, secure, bug-free code (nearly) working on MS OSs: Firefox comes to mind. It will only be a matter of time before MS releases another OS version (Longhorn perhaps) that obsoletes most peoples' hardware. If the user has been quite pleased with OpenSource software and has many packages installed on their current system, they might just decide to switch over to Linux. Let's have a reality check here - Linux is not going to take over the desktop overnight. We saw that belief evaporate in the internet bubble of 1999 after the record-setting IPO of VA Linux. No, Linux will garner more and more converts over time and we must be patient. The best software will prevail, despite the fear, uncertainty and doubt (FUD) spread from Redmond. Let's face it, when the core of your business is threatened by a free alternative that is superior in many ways, that business will react in fear mode. Most computer users cannot sort out the truth from the FUD. Those that are tempted to explore beyond MS's windowed world will need our help to educate them, so that they can make their own choices. We must continue to encourage these folks with useful tools, and we must be prepared to welcome them with open arms.

    Here's to the continued success of Linux !!
  • As a small office desktop user our 'training' with Openoffice, etc. will make a switch to Linux easier when we are for$ed to by Microsoft.
  • I have to disagree.

    Our company is a Windows-only shop. We have some Linux machines as an evaulation. Having programs like Open Office and Firefox run on Windows helps in the migration to Linux because it gets users accustomed to the applications that they will be using most.
  • Isn't the point of the GPL to provide the freedom to port the software to whichever platform you please? I support the KDE port even though I prefer Gnome on my Linux boxes.

    At the office, however, it's strictly Windows...I started a new job a couple of months ago and found out that everything was M$ software.

    We even run Great Plains for crying out loud! (uncanny.... It does exsist).
  • Having the only ports of great free software has never been Linux's major advantage, you could even argue that it's a disadvantage - where is the incentive for Adobe to make Photoshop for Linux if most users would just stay with the Gimp.

    Linux's advantage is that it has a free, fast and stable kernel with native support for a huge array of hardware, and which is not coupled in any way to a specific environment, GUI, window manager, etc... When faced with the option of buying a (prohibitively expensive) Windows Terminal Server license for a large server, or just installing Linux, it's pretty obvious which you'd choose.

    In short, Linux shines when Windows is just not an option, due to limitations imposed by the one and only distribution available. The thing keeping many businesses from switching - Training costs. Training costs that would already have been paid if the same free software was already being used by staff under Windows.
  • I would disagree, as a very loyal Windows user, things are looking pretty good to me linux-wise these days. I've been 100% pro microsoft, and still favour the OS right now, but If the linux world can pull its act together and make things just a bit more easier to use, the idea appeals to me. I think having these open source options available to me and making them better than microsoft implementations will bring more people to linux when the time is right. Since linux will be free, and we will already be used to a good chunk of the software we'd be using on linux (OpenOffice, firefox, thunderbird, kde, etc.), it would make the transition more viable, and a bit less scary. With things like Mono, as a developer, I don't see a whole lot of advantages that windows has left except for the ease of use factor. Once that gets up to speed, all bets are off, unless Microsoft can one-up things with Longhorn. Linux has a chance, but they've got to act quick.
  • Well, I know OS/2 didn't get much publicity outside of the banking/business sector, but if they were selling *millions* of copies (as reported by PC Magazine) those copies had to have been used on the desktop for general purpose usage.

    Further, the initial advertising campaign for Warp 3 (The Totally-Cool Way to get on the Internet) was extremely successful. Plus, don't forget that at the time Warp 3 came out OS/2 was the ONLY 32-bit operating system for general use. Windows 95 didn't come out until almost 2-years later.

    All I know ultimately is that OS/2 was doing very well with Warp 3. It was only after MS pulled their dirty-tricks and IBM started to back away from supporting their product that things started to go downhill in a big way.

    Prior to that, OS/2 was doing okay. And had IBM continued advertising like their Warp 3 ads its success would have most definitely continued.

    What I find so ironic is that even today the WPS is very often still revered as a top-notch GUI and whenever a discussion takes place of Linux window-managers someone *always* seems to bring it up, lamenting the fact that todays GUIs still don't do things that OS/2 was doing 12-years ago ...

  • Are members of the Open Source community really suggesting that it shouldn't be ported because of political reasons?

    Surely this is totally against the principles of Open Source!!!

    If I want to run a particular operating system then it is none of their business. If I want to run a particular Open Source application on it then either their may be an existing port or I can port it myself. Of course the port would then be available to others.

    The complaints of these individuals runs against all that the Open Source community works for. As a small (or tiny) IT company owner I have to consider the options very carefully. The consequences of being tied in to software too deeply is not good. If the Open Source community is going to pick on a particular platform or technology then I am disinclined to use it.

    I don't want anti-Microsoft sentiment. I want solutions.

    There is no place for technological religion in my IT world.