“Despite the head start that KDE enjoyed, the large number of KDE users and developers, and Linus Torvalds personally endorsing KDE, GNOME has won the desktop environment battle. The final victory came with the third piece of a corporate trifecta, giving GNOME the official nod from Red Hat, Sun Microsystems, and finally Novell. The question is, will the triumph of GNOME lead to the rise or downfall of the Linux desktop?“ Run Forrest! Run!
I like gnome with a customized sawfish window manager.
http://perfectwm.blogspot.com/
Well being GNOME better or not articles like this one just divide the Linux community and now that KDE and GNOME had noted that working together bring better results, taking sites is ridiculous.
Edited 2006-04-28 14:26
This article is a complete troll. Gnome has more corporate backers simply because gtk and the whole platform is LGPL. That allows companies such as vmware, adobe, etc to use gtk in their applications without worrying if they have to open source it or not.
KDE isn’t a bad desktop environment although it follows different ideas versus gnome on what a desktop environment should be like. KDE offers ultimate control and flexibility where Gnome offers sensible defaults and the “Just Works TM” mentality.
I think anyone who has been using Linux more than 4 years will have noticed that Desktop Linux is picking up serious steam from some big names. Both desktop environments will benefit from this and the future. This helps the “Linux Desktop” as a whole.
Concur.
While ensuring interoperability in the middle and back end of the system is certainly desriable, declaring anyone the winner at the front end is specious.
The strength of FOSS is that the user isn’t getting flocked.
If some sheep prefer a flock, we must gently educate them out of this.
I completely agree.
There is not really a “war” and neither side has “won.”
The article seems to imply that KDE will cease to exist, and only Gnome will matter. What a load of cr@p.
BTW: I use IceWM.
“The article seems to imply that KDE will cease to exist, and only Gnome will matter.”
I think the article implies that KDE will remain vastly popular with the GNU/Linux power user due to its “technical superiorities”, while GNOME will remain popular with the general masses (well, as long as GNU/Linux adoption increases).
Still with the new LSB 3.1, it is a bit too early to make such a definitive statement (such as “GNOME wins”). Then there is KDE 4.0, which looks to really change the desktop experience (or so we are lead to believe). Regardless if a “winner” does emerge, KDE will not be going anywhere.
Although I prefer Gnome at the moment I must say that they overact with their “Just Works” mentality.
For example you cannot set text any more in Gnome screensaver GLText which makes this screensaver useless for me. I had to install XScreensaver to adjust these settings.
They cut to many features. Why is there no option to leave the file save dialogs unfolded!? Or why can’t you unmount devices with a right click in the places sidebar in Nautilus?
It would be OK if they remove these features in application dialogs if they INCLUDE them in GConf! But I can’t see this approach – there are nearly the same options in GConf available as in application dialogs…
It would be great if Gnome developers include more “advanced” features in GConf. Otherwise this whole approach makes no sense to me.
They cut to many features. Why is there no option to leave the file save dialogs unfolded!? Or why can’t you unmount devices with a right click in the places sidebar in Nautilus?
I second that !
The gnome-screensaver dialog most definitely did cut an important feature (configuring an individual screensaver).
However, the file dialog and not being able to unmount drive from the place sidebar in nautilus are not “cutting features.” The GTKFileChooser API included in GTK2.4 (or was it 2.6?) allows people to develop any number of different file choosers that suit different needs. The fact that no one is really doing that is a little odd, but nonetheless the ability is there.
Also, the nautilus places sidebar is most definitely not that way to follow any “just works” ideology. It just hasn’t been done. I haven’t hear anybody turning down such a patch, and in fact a patch was _accepted_ to 2.14 to allow editing of the bookmarks in the places sidebar.
There is a big difference between “haven’t had time/resources/thought to implement it” and “we cut that feature because it’s bunk.”
>”However, the file dialog and not being able to unmount drive from the place sidebar in nautilus are not “cutting features.” ”
It is being worked upon, so be patient. Nobody “designed” it to be a “feature”, and soon it will disappear (Gnome ver.2.16?).
For example you cannot set text any more in Gnome screensaver GLText which makes this screensaver useless for me. I had to install XScreensaver to adjust these settings.
I agree with you.
Why is there no option to leave the file save dialogs unfolded!?
This shows something really interesting that the GNOME project realized (Havoc P. actually). GNOME’s behaviour is broken and instead of asking the developers to fix it, users ask them for an option to have the correct behaviour.
The right design in this case is to remember the state of when the dialog was last used. I believe there’s a bug against GTK+ for this, but if you don’t find it please file another one.
Or why can’t you unmount devices with a right click in the places sidebar in Nautilus?
This is a bug. In fact it will be fixed in the next release, as a patch has been written already.
http://bugzilla.gnome.org/show_bug.cgi?id=316184
It would be OK if they remove these features in application dialogs if they INCLUDE them in GConf! But I can’t see this approach – there are nearly the same options in GConf available as in application dialogs…
Like I said before, notice that only one of the issues you mentioned is actually a design decision. When you run into them don’t assume it’s one, but file a bug.
Gnome has more corporate backers simply because gtk and the whole platform is LGPL. That allows companies such as vmware, adobe, etc to use gtk in their applications without worrying if they have to open source it or not.
Maybe VMware chose GTK but for example Parallels is QT-based and so is Skype. I think there are many other companies that use QT and don’t have to share their sources in any way
Yes, because they pay Trolltech lots of money for a proprietary license to QT. If you dont believe me, here is a website for you:
http://www.trolltech.com/products/qt/pricing.html?cid=18
Please note that would all be free with GTK
‘free with GTK’ as in no upfront cost, but Qt is very nice set of libraries for developers to work with and saves time over the length of a project, so the ‘cost’ of GTK is measured in person hours, but is not 0. And if the app is run on multiple OS then Qt is even more of a money saver.
In the end, it all comes down to horses for courses. There are areas in which both are appropriate and other areas where they both don’t represent such a good value proposition.
On a different note, why do people imply that writing a KDE app has overheads (in system services etc) that a GNOME app doesnt? A GTK app might not require all the GNOME infrastrucure to run, but the same is true that Qt apps don’t require the KDE infrastrucutre.
“Gnome offers sensible defaults and the “Just Works TM” mentality.”
just my 2cents worth, but I have to disagree on the sensible defaults. I know the gnome people have a working set of ideals to implement, but when I do set up gnome for myself I spend a_Long_Time changing its behavior to do what I want it to, most of it just finding the switches. It’s really not worth it to me personally whenkde is bright and shiny out of the box and the defaults I do wish to change are right in front of me.
That said, I prefer fluxbox over all of ’em in the end
“when I do set up gnome for myself I spend a_Long_Time changing its behavior to do what I want it to…” — Well, you’re doing better than I am:
I’m a PowerUser environmental modeler (having written over 200,000 lines of LGPL and GPL environmental modeling code), and I still can’t make Gnome behave to my notions, even when I’ve spent hours at it. Life is too short!
(For KDE, it takes me about 2 minutes.)
I’m a PowerUser environmental modeler (having written over 200,000 lines of LGPL and GPL environmental modeling code), and I still can’t make Gnome behave to my notions, even when I’ve spent hours at it. Life is too short!
(For KDE, it takes me about 2 minutes.)
Ok, this comment was amusing:) What actualy is a PowerUser in desktop concept?
If you were meaning to say you do things like remote administration, coding, heavy art drawing… I would be in the same group too.
And it takes me exactly 3 options for desktop to behave exactly as I need (rollup instead of maximize, allways group and disable screensaver)
While KDE poses too many distracting and unusable options everywhere by default, and disabling them is a pain in the ass.
btw. I like SimpleKDE concept very much on the other hand. And is as easy to set up as Gnome
Edited 2006-04-28 18:43
KDE offers ultimate control and flexibility
Whenever I see something like this, I just crack up in laughter. KDE flexible? Puh-lease. Even IceWM offers better configurability, not to mention FVWM or a host of other WMs.
How is KDE not flexible as compared to Gnome?
KDE and Gnome are not the only WM/DEs available. “Ultimate” does not mean “better than one (or a few) of the others.”
KDE offers ultimate control and flexibility
Whenever I see something like this, I just crack up in laughter. KDE flexible? Puh-lease. Even IceWM offers better configurability, not to mention FVWM or a host of other WMs.
Are you using your own version of KDE thats different from everybody elses? Its a well known fact that KDE is highly flexible and can be customised to levels that don’t exist in other DEs.
Before anyone else replies to this I’ll point out that the parent said
Are you using your own version of KDE thats different from everybody elses? Its a well known fact that KDE is highly flexible and can be customised to levels that don’t exist in other DEs.
WM need not apply.
In fact, AFAIK there are only 4 DEs on linux: KDE, GNOME, XFCE, and e17.
Oh and how about my prediction for the future:
NO DE will win since usability standards are converging and that cross-platform nature of apps will mean that which DE will matter a lot less.
Afger all, I switched my family from IE to Firefox and NOBODY NOTICED!!!.
Articles aren’t human. The author of the article is the troll. This article is flamebait depending on how someone’s views the Linux DE.
I don’t recall Desktop Environments being a race with winner-takes-all consequences.
I guess I should dump my fluxbox and install Gnome. Hate to be using a losing Window Manager.
=
On a more serious note, the corporations can pick whatever they wish, and if they happen to generally consense that Gnome will be default for their distro, that is fine. Most of them also come with KDE or others to choose from, too. Will this help Linux adoptation? Probably not. It is already doing fine, in my opinion, and I don’t particularly care if it reaches 5% desktop useage, as long as it suits my needs well.
Settling on a single desktop is just one step in a long, twisty path toward getting desktop Linux into the mainstream. GNOME has won the corporate battle and needs the support of the broader community. Because many people view KDE as technically superior and there are some egos at stake, that may be a bitter pill to swallow. Even if the community does coalesce around GNOME, it in no way guarantees success, but the continued fragmentation of the desktop guarantees it will languish.
I aggre witH this part.
Edited 2006-04-28 14:38
It’s true that KDE suffers because of licensing issues, but with KDE soon to be included in the LSB Desktop, with freedesktop.org and Portland and Tango all working on cross-Desktop compatibility, and with most major vendors offering KDE as an optional install, and especially having won “Best Desktop” in the 2005 LJ Reader’s Choice awards, can it really be said to have “lost?”
“It’s true that KDE suffers because of licensing issues…”
KDE does NOT suffer from ANY licensing issues. KDE and QT is GPLed software.
Edited 2006-04-28 18:18
Well, it suffers because of popular misconceptions, then. And because non-GPL development requires a paid Qt license. (I’m a KDE user; definitely not trying to slight the big K here.)
KDE does Not suffer from ANY licensing issues…
Well, not exactly. Free applications written using QT widgets don’t require a license but commercial applications do require one.
I could also make one suggestion to the poster: if you are going to put things in caps indicating you are yelling correct information you might want to actually *read* the license first.
Directly from KDE’s site:
…the use of the KDE libraries is free of charge for development of commercial applications. However, the KDE libraries rely on the Qt library. Thus you will probably need to obtain a license of Qt from Trolltech.
Please, be aware of the advantage that an investment in Qt licenses brings, as it won’t be limited to commercial KDE development, but it will also allow general commercial Qt development and will eventually enable multi-platform software development.
However, developing free software applications for KDE does not require a commercial Qt license.
Edited 2006-04-28 18:28
And why should I care as a KDE user and free (as in freedom) software supporter if commercial applications using QT require a license?
You don’t have to, but you must understand that many do care this issue, and they are more of what you think.
Edited 2006-04-28 19:09
The fact that it is GPL is the issue. While the GPL is a great licence, it is rather restrictive. Very few people, corporations, or whoever likes to feel chained by a licence.
This isn’t to say that the GPL doesn’t have it’s place or that I disagree with QT using the GPL (it is their choice, after all) but that because QT uses a relatively restrictive licence versus a rather relaxed one (such as the LGPL or BSD licences) it does suffer from licencing issues that will disqualify it for use from a larger potential developer base.
Edited 2006-04-29 16:53
This is utterly rediculous in the exetreme. The war isn’t over, nor has gnome won. If you like Apple you’ll like gnome and if you used Windows you like KDE.
geesh..gnome fan boys
I personally don’t like gnome
So, if you hate Windows you hate KDE too? non sense since most of the Linux users hate Windows as well.
The reason why someone should use Linux is becaue they like it better, not because they dislike Microsoft.
The reason why someone should use Linux is becaue they like it better, not because they dislike Microsoft.
The reason I use Linux over MS is precisely because I have an intense dislike for windows, a hatred you might call it. Why? Not sure, but it’s there stronger now then ever. Maybe it’s the way they oprerate monopolistic practices and all. Or maybe they brainwashed people in to think they havea great product.
The reason I use Linux over MS is precisely because I have an intense dislike for windows, a hatred you might call it. Why? Not sure, but it’s there stronger now then ever. Maybe it’s the way they oprerate monopolistic practices and all. Or maybe they brainwashed people in to think they havea great product.
Agreed ! Except that my dislike for Microsoft came after I really started enjoying Linux. Before, well I was a brainwashed person also.
I use Non-MS software at work because the entire company is Non-MS (and no, we aren’t a tech company). However, through that experience, I’ve found that if worse came to worst, I would be able to tolerate not using Windows at all.
But for now, I still use XP at home.
So, if you hate Windows you hate KDE too? non sense since most of the Linux users hate Windows as well.
You are great at taking comments out of context, I’ll grant you that. I never said that what I said was that “if you used Windows you’ll like KDE” not if you liked windows. I personnaly hate Windows and like KDE
Yep, for reasons that havent been true since windows 9x. Most people only say they hate windows and like linux to get attention at this point.
Edited 2006-04-28 15:56
Yep, for reasons that havent been true since windows 9x. Most people only say they hate windows and like linux to get attention at this point.
Believe me I don’t say I hate windows to get attention, if I wanted attention I’d make sure to put in a $ when I spell it or call it winblows.
I most genuinely have a hatered for windows, so much so that I changed university because of thier total reliance on windows and not admit to better products.
If you genuinely hate (hate is avery strong word) Windows, you’ll be able to list of a number of reasons…
Proceed.
If you genuinely hate (hate is avery strong word) Windows, you’ll be able to list of a number of reasons…
Proceed.
1. Virus
2. Mal-ware
3. Spy-ware
4. Lockin
5. Inferior piece of crap
MS Windows doesn’t lock you into anything any more than Mac OSX locks you in. The applications you use might lock you into Windows.
If you use Word and can’t move off the platform, that’s your own damn fault. Not the OS you chose. You can use plenty of cross platform applications on windows. OOo is one of them, firefox is another. Don’t try to say server platform either…you can run apache, php, perl, etc…cross platform
Inferior is vague…almost troll like in vagueness.
Virus’, there are linux virus, there are mac viruses. I’ve never had ANY on Windows, but I know what i’m doing.
Mal-Ware, same thing, I know it’s there, but I haven’t had any.
Spy-Ware…see Mal-Ware.
It seems to me that you hate windows because you made some stupid mistakes…
Thats where you dead wrong, there are no virus for Linux that will do the damage that Windows viruses do. So you don’t think you got any then your better than most corporations who have security up the ying-yang and still get virus.
I garentee you’ve gotten some, you just haven’t noticed. Yes windows does very much lock one in, weather you think so are not.
Prove it.
I just ran three different virus scanners (MS, Anti-Vir and Symtantec’s) and all three came up clean.
I just also ran Spybot S&D and Adaware Plus. Nothing but a few cookies.
I did all I can to show you that my system is clean. You want to spread FUD, go ahead. I’ll sit back and tell those people ask me if Windows can be secure that it can and inform them how.
You can go ahead and spread FUD and continue to prove that linux zealots really don’t know what they are talking about.
I said that there were viruses for MacOSX and Linux. I said nothing about the severity of the damage that can be caused.
For the home user, loss of personal data is more important than loss of system files. Linux does NOTHING to prevent that from viruses and it’s a pain to install linux again just like it is for windows.
Don’t bring up multiple users, chances are that there aren’t multiple users on typical windows boxes as well as joe user’s linux box. You are not a normal user, nor am I.
I wouldn’t say I hate windows and I don’t use linux (I actually use windows because I can’t get networking to wokr on Linux), but here are a few things that annoy the hell out of me on windows.
1. Very few customization options for the desktop. It hard to get a suitable desktop experience with the tools that MS provides out of the box. You have to use a range of different tools to get the desktop working the way you want. Even something as simple as removing the recycle bin icons from the desktop. There is loads of other stuff as well, things like not being able to move rearrange programs in the taskbar, no virtual desktop manager.
2. ‘The user is dumb mentality’. Why the heel are their folders in program files which can’t be deleted moved in an easy manner (I am tlaking about the frontpage folder, Xerox and some other MS shit). To get rid of them you have to turn off windows files and folders protection, but that can cause other issues.
3. Brain dead OS tools. Why doesn’t windows explorer have no tabs? And there seems to be no simple way to add this functionality. Why don’t they have an extensions system like Firefox?
4. Piece of shit media player. What sort of trash is WMP? The all in one solution is such a bad option. Why don’t hav to seperat players. A good version of MPC for ppl like me and WMP for joe blow.
5. Lack of management tools. Why doesn’t windows have a directshow filter manager?
There is also other stuff as well, if you want me to continue….
“1. Very few customization options for the desktop.”
Same can be said about gnome. Metacity is a terribly limited window manager and can’t really be extended to do much out of the box. You can replace Explorer.exe with something else or you can extend it with apps like what are provided by Stardock.
“2. ‘The user is dumb mentality’. Why the heel are their folders in program files which can’t be deleted moved in an easy manner (I am tlaking about the frontpage folder, Xerox and some other MS shit). To get rid of them you have to turn off windows files and folders protection, but that can cause other issues.”
I have no answer to this other than to stop people from screwing up their system. And apparently you can remove them by turning off Files and Folder protection as you stated, so it can be done.
“3. Brain dead OS tools. Why doesn’t windows explorer have no tabs? And there seems to be no simple way to add this functionality. Why don’t they have an extensions system like Firefox?”
Windows Explorer is the shell and file manager. Firefox is a web browser. I’m confused. Internet explorer is a web browser and it does have tabs (IE 7) and you can install other IE based browsers that have tabs (Maxthon, etc).
“4. Piece of shit media player. What sort of trash is WMP? The all in one solution is such a bad option. Why don’t hav to seperat players. A good version of MPC for ppl like me and WMP for joe blow.”
Install iTunes, wmp, foobar, there are many. Choice exists if you don’t like the default.
“5. Lack of management tools. Why doesn’t windows have a directshow filter manager”
I’ve never thought about this. I guess I just don’t care because it doesn’t ever come up when I use my computer.
BSD for unix lovers..
Linux for the Windows haters..
Blaah…
I’d like Gnome if they’d dump that thing they call a window manager…
KDE is a far nicer environment, and gnome apps work great under it
.
KDE also has a hard time getting 3rd party apps because your app has to depend on a huge library, and its startup depends on kde services. With Gnome apps you just depend on the libraries you select (often you can just make it a gtk app). This makes your program available to a lot more people!
No one has anything on kmail yet though
.
>The war isn’t over, nor has gnome won.
The war has never started! The war is just something which is used from some trolls and users who have nothing better to do than wrangle with other about senseless topics.
There was no war, there is no war and there will never be a war between KDE and GNOME. We have two great Free Software Desktop Environments and many great Free Software Window Manager and everyone can pick whatever he wants.
I had Windows hose my hard drive on account of bad graphics drivers (I am not shitting you on this, hand to God), so I went with Linux.
Originally, I installed SuSE 8.1, which ran with KDE. I adored it. It was such a great time! Using the computer was fun (and a little challenging) again!
Over time, I installed the GNOME desktop on SuSE and found that I really preferred a lot of its features. After SuSE 9.2, I moved to Ubuntu’s Hoary or Warty or whatever-the-hell was their first release, GNOME based, obviously, and it was great.
I don’t like KDE because I came from Windows. In fact, quite the opposite happened: I came from Windows, and found I liked GNOME.
A lifelong Mac-using friend of mine, on the other hand, insisted right-out that we install Kubuntu on his non-Mac box, rather than the GNOME default Ubuntu. He liked KDE because it was prettier.
I’ve also dabbled in fvwm (yay CS computer labs) and e17 and XFCE. They’re all very functional and have their place (even the non-DE fvwm). To each, his own, though, I say.
I keep starting to write a comment about this article and I find myself trolling badly so I am just going to shut up and move on, enjoying my beautiful KDE desktop with majority of other Linux users.
Seth
Edited 2006-04-28 14:39
where did you get that number? Seems pretty high to me.
Lol you still managed to troll. It isn’t 70%, it’s less.
The 70% figure I think was around 2-3 years ago, its around 50-60% today, with a rise in the more minimalist window managers – XFCE has really taken off, as those disappointed with the direction of KDE and GNOME have looked for a plce where a desktop environment is just that, rather than (in their words) a ‘bloated desktop’.
What GNOME 3.0 however, requires is a feature complete framework – HAL/DBUS/Cairo and the likes fully feature complete and integrated into GNOME, from top to bottom – – get to a stage that all the infrastructure is setup, and its just a matter on each release, tweaking things, bug fixing, and maybe adding some new features/applications.
I don’t remember exactally where, it is a rough figure, however it is the one I remember quoted as the general usage statistic. It isn’t as high as it sounds though really considering that the KDE is far more popular outside of the states. I’ll change my post to majority if that will make you happy though
Seth
Wow, this article will serve one purpose alone… let the pointless yelling commence
And just because I can’t help myself, no GNOME has not won anything. It just happens to currently be the default option on a couple of the more popular distros
A far cry from the holy grail it may be but default option on the distros of choice is not a bad distinction to have.
I’m not sure I’d even call them the distros of choice, just a couple of the larger commercial distros. Gentoo is a dsitro of choice for lots of people and is DE neutral for example. Linspire is a commercial distro wich has a KDE DE and has had some success, kubuntu shares equal billing with ubuntu with reguards to commercial support also.
Windows have not won either, its just the default one..
E17 will be the nuclear option to end all wars.
second that, but i really doubt its ability to gain much in the way of corporate support- it’s just way too cool
E17 is unstable, but I don’t care. I love it and using it. But it’s not going to have corporate support like Gnome/KDE does. Because it’s not a complete DE rather WM!
The link is currently dead. Maybe an indication of better things to come. lol
🙂
This kind of articles should be bundled with free popcorn. It’s all we’re going to make out of it.
The basis for the article is when the author jumped to the conclusion that corporate desktop use would dictate all desktop use. While it has been true that many people were introduced to technology at work and later purchased the same thing for use at home, there is a very big difference between an enterprise product and a consumer product. The top three enterprise Linux options may have defaulted to GNOME, but that does not exclude KDE.
I actually think that Novell is going to be in the best position moving forward with respect to DE development. OpenSUSE will be the basis for future Linux products coming out of Novell. Long-time SUSE users will most likely help keep KDE as advanced as possible while the team from Ximian helps make it one of the premier GNOME options.
In addition, distributions that try to aim at new users tend to be focused on KDE. Will someone who wants to use Linux at home after being exposed to it at work be choosing Red Hat or NLD? I think it would be just as likely that they pick up Xandros or Linspire.
Honestly I think whatever desktop environment is predominant, linux and *BSD win more and more adepts.
I don’t really care what company choose since FOSS is about freedom of choice.
Nor is Linux tied to either DEs. This just shows how the person who wrote the article misunderstands the Unix world. One can use Gnome, KDE or any other WM on any Unix-base (or Unix-like) OS. They success of these DE/WMs is independant to a large degree of the popularity of the underlying OS, which is how it should be.
Declaring that “GNOME has won” and “KDE has lost” just because some distros have made a default choice is meaningless, especially since KDE 4 is just around the corner. This article is a bad troll, and should not have been linked to here.
I think one of the reasons why those big Linux Distro vendors choose GNOME is the improvement seen from GNOME 2.12 to GNOME 2.14 is impressive, while in KDE’s side, things almost remains the same. As KDE 4 is on its way, I believe when it releases, such situation will gonna be changed.
Gnome is simple, clean and functional having said that gnome is just like windows. It gives the same amount of flexibility as windows does. Gnome developers are trying hard to give less and less features, that simply works and looks good. Thus they have to fix less problems and teach less to use Linux.
Why is it bad?
It is bad because it stands completely against the Linux. Linux is all bout customization. That is why it is open, KDE on the other hand follows the same principle of Linux.
I think at the end gnome is gearing toward a very boring and featureless desktop manager that simply provides 2 functions to corporation. Where as KDE will be just like Linux.
How does fiddling with Window manager features teach you to use Linux? In GNOME the terminal is just a few clicks away. It’s no less Linux than KDE. It offers enough configuration without going overboard. I like it, it gets out of the way just like a usable desktop environment should. To say it’s a barrier to getting to know Linux is silly. I’m no less able to compile source code or run any application than I am under KDE. And taking a recent example, how is adding things like Deskbar “trying hard to create less and less features”?
Knowing you way around kcontrol no more makes you an expert at Linux than knowing how to launch the control panel in Windows. A Linux distribution is a layered operating system (kernel/gnu tools/X/WM/etc.), not just a desktop environment. And neither GNOME or KDE get in the way of you learning about those different layers if you want to.
So arguments about GNOME being less Linux seem just plain silly to me. I’ve been using Linux since about 1997 and have gone through lots of Window Managers, from FVWM to WindowMaker, blackbox, fluxbox and then GNOME. None of those environments will stop you learning about Linux, and usability doesn’t automatically make something less Linux.
you the man homer
It is amasing how regularly some people have an urge to argure G vs. K. Is there really so muh steam in all of us? I wish there was a way to moderate down the authors. He would get enough “flaiming troll” scores and would wanish from the face of OSNEWS forever…
On the other hand, with ads and all, all this sensationalism may be actually encouraged here? Lets rename osnews into flamer’s_cove.
I can’t RTFA, but if the implication is that KDE is doomed because GNOME has more corporate support, well it is not exactly new.
And it haven’t prevented KDE from thriving until now. I wonder why people keep fantasizing about the demise of either major linux desktops.
If you want to ignite a proper flamewar, you have to raise highly subjective matters that people cannot possibly agree about, not something that cannot withstand any exposure to facts.
Edited 2006-04-28 16:02
That’s the dumbest article ever. It’s like the author got up in the morning and just automatically put on the asbestos longjohns. I find Gnome stifling, it’s like using OSX’s ugly sister
But developing closed-source applications using QT requires you to purchase a QT license from TrollTech, because the alternative would be to GPL all the code that links to QT…..
With GTK apps you can just link to them merrily. It’s been said that it’s a bit ironic, given GNOME was developed to produce a fully Free desktop vs KDE’s use of the QT toolkit under the original license, that KDE is using a GPLed toolkit to GNOME’s LGPL. I’m not sure it’s really desirable to lock out closed-source vendors, though, but of course the licensing is the decision of Trolltech. But hey, they’re both free, they compete, they improve, they provide choice. So it’s all good 🙂
Nb. the KDE libraries IIRC are LGPLed, so *if* you have the commercial QT license you can develop closed-source apps with proper KDE integration.
We’re talking about desktop enviroments and afair this means kde/gnome/xfce not anywminexistance.
WM’s are a different topic.
PS: I prefer xfce, but I’m really enjoying all that takes place in the DE developement both in KDE and Gnome. Diversity _is_ a pro.
http://www.linuxforums.org/forum/coffee-lounge/52464-desktop-enviro…
So? Is Gnome winning?
That actually quite a small poll. The linuxquestions.org community has well over 300,000 users and in a similar poll KDE was tops with 60+ %.
Anway figures don’t mean much, because both desktops are good and cater for people with different needs.
Edited 2006-04-28 16:49
That everytime some artical like this comes up KDE users just dont like it. Just because more people like/use KDE dont mean NOTHING. If people like Adobe, Sun, Novell back GNOME for there way then thats good, dont mean KDE will suffer. Personally I think KDE goes off and does it’s own thing to much rather then using GNU or freedesktop spec, KDE4 will move away from that even more.
It’s not about that DE YOU prefer, it’s about what the people like Adobe (which are big players in software) think suits there needs, gnome fits that so whats the problem?
Edited 2006-04-28 16:10
That everytime some artical like this comes up KDE users just dont like it. Just because more people like/use KDE dont mean NOTHING.
Actually, it means a lot. That’s called market share.
Personally I think KDE goes off and does it’s own thing to much rather then using GNU or freedesktop spec, KDE4 will move away from that even more.
KDE is a participant in freedesktop.org. I don’t understand what you’re trying to say.
It’s not about that DE YOU prefer, it’s about what the people like Adobe (which are big players in software) think suits there needs, gnome fits that so whats the problem?
Actually, it is exactly about what DE most users prefer. And, while we’re at it, Adobe has used Qt for some of their products before. The fact that they chose Gtk for Acrobat Reader means very little, as most Linux users don’t use AR to read PDF files, but rather more efficient alternatives such as KPDF.
“Actually, it means a lot. That’s called market share.”
What market share?, I guess KDE devs designed it knowing it would get more “market share” because it’s Windows like.
“KDE is a participant in freedesktop.org. I don’t understand what you’re trying to say. ”
right, right, thats why installing gnome which uses the freedesktop spec menu breaks the KDE menu.
When I installed gnome, it didn’t break the KDE menu, so at least on Kubuntu, your assertion doesn’t hold up. all the Gnome apps appear exactly where they should go. Maybe it’s your distro that is not up to the specs, or maybe you did something wrong.
What market share?, I guess KDE devs designed it knowing it would get more “market share” because it’s Windows like.
Market share as in, more users. I can’t figure out what you’re trying to say in the second part of your sentence.
right, right, thats why installing gnome which uses the freedesktop spec menu breaks the KDE menu.
What are you talking about? I have both KDE and Gnome installed on Mandriva and Ubuntu, and menus work just fine in both DEs.
You really should research more before posting.
Ofcourse there not broken in Mandrava and k/Ubuntu, go and install gnome from source on something like slackware and it breaks the KDE menu. I’ve seen it for myself so I dont need to do research because the prebuilt distro’s fix it for you.
Dont assume evrything is alright because it’s alright on your prebuilt distros.
Edited 2006-04-28 18:16
Okay, so it’s broken on your Slackware distro. Big freakin’ deal. That’s still irrelevant to the topic at hand.
one big thing wrong w/ this article is, these 2 WM’s are not in competition… then how pray-tell has gnome won the battle? Ask anyone who works with it for real, it’s not like the windows world, sorry.
For some reason, I just don’t care to customize a desktop environment. It just doesn’t make sense to me because I just use it to surf the web, check my email, and launch professional applications. Of course, I would expect to be able to fully customize those professional applications (Eclipse, etc.) since I will be doing real work in it, but for my desktop, I just want it to be able to launch applications, do some simple file manipulations, and then just get out of the way.
Seriously now? I’m a GNOME user. Won a desktop battle? WTF? I mean even the article title sounds completely biased. Yeah right, maybe when GNOME is on 95% of newly manufacture PCs/macs out there I will consider the headline normal.
There’s no battle either with other desktop environments. Sorry I can’t read the article, seems it has been OSNewsed.
I really don’t get it. The article summary says it won the war because three big corporations are backing GNOME. Yeah, so? What makes that a proof that it has won over something else? Desktops are not only for corporative uses, as far as I know, there are lots of home or corporate users using KDE or other environments out there. Am I missing something?
the file selection dialog allow me to type in the path. Having to click through the path is SOOOOOO annoying.
That’s probably the biggest reason I prefer KDE.
You can. Just type / or ctrl-L. Or use search-as-you-type and enter to navigate.
Gnome does that. Hit ‘/’ in any file selection dialog and it pops up a dialog to enter a full path with tab completion. Also, if you are already in the directory you want, simply start typing the file name you want and it selects the file.
Link dead.
Considering the title and the expressions such as ‘won’, ‘victory’ and ‘triumph’, I believe it’s either a nice and very well thought troll or a clumsy article.
An in depth analysis of the corporate support Gnome succeeded to gain would have been interesting. Since the link is dead, can’t say much about the article.
Anyway it looks like a nice attempt to excite trollers and other Gnome and KDE zealots.
I’m a KDE user with a brain. I do not think kde is perfect and I’m glad other projects such as Gnome are out there. I’m quite confident in the future of KDE and I don’t need no zealot to claim ‘KDE is the best DE and we’re going to rule da world’ 😉
linuxboxadmin.com – boring name – is down. Couldn’t be much of a site if a few connections from OS News knock it over.
KDE here, though I use Gnome on my laptop via Ubuntu. In the meantime, if corporate-style distros want to standardize on Gnome they are entirely welcome to.
It is the apps that count for me, not the DE. Strictly imho, Konqueror easily wins over Nautilus, Kontact over Evolution (which always seems buggy and with an awkward interface), Kate over Gedit, Amarok over Rhythmbox, K3b over Gnomebaker, to name a few examples.
I’m baffled how anyone can claim the Linux desktop battle has been won when one looks at programs like these. In many cases, the Gnome offerings are modest programs with a lot to be modest about. OK, they might not be needed on a slimmed-down corporate desktop. But then the jury is still out as to whether there is really a market for this corporate desktop Linux stuff. The commercial distros would like us to believe it, but that’s another matter.
If you want a fully loaded home machine, KDE is hard to beat, in my experience. Gnome is fine on my lappy – reasonably quick, simple, no bells and whistles because on my lappy none are wanted.
I wouldn’t say Gnome has won just yet.
KDE 4.0 is looking both good and a well thought out desktop environment.
http://www.linuxdevices.com/news/NS5209121793.html
Mixtures of KDE and Gnome work just fine, and are working together better all the time. freedesktop.org has made a strong push for interoperability. Qt is moving to adopt the event loop from glib, so that you can mix Qt/KDE and Gnome/gtk+ widgets in the same program.
As a developer, you need to choose what you will build on top of, but your choice does not restrict your users; I run mostly Gnome, but I use several KDE applications, k3b probably most often. It just works, it doesn’t matter.
Wait, the whole point of Linux is for ‘competition’ and ownership, something called Capitalism NOT Communism.
What is this Gordon Gecko mindset that capitalsim automatically means a communist monopoly?
If one side is better in the FOSS world then the other then usually becaseu the other side has ownership can switch tactics and compete therfore creating better quality products in the market.
So if one iside is a little beter then the other it doesnt mean a monopoly will occur in FOSS. If one side is allot better, then the bad side loosed market share.
Communism has too much permeated our Western cultures.
What criteria do they use to declare gnome the winner???
I’m not sure. It’s either:
1.) because for-profit business organizations like Red Hat, Sun and Novell have decided to use Gnome.
2.) because he said so.
Honestly, this seems like it’s probably more a reaction to SuSE being bought and changed by Novell than anything new.
KDE vs GNOME. OSS is not about rival desktops. OSS is about diversification : Where is CDE? Why JDS runs on Gnome instead of OpenLook? Most of the time I use WindowMaker+gnustep as a desktop and gnome or kde tools. I have no personal preference between Gnome/KDE. Both have served me equally well and this discussion is about OS/Distributors not about losing time. LessTif needs more hands and there is a lot of Polishing for Desktop Linux to be done and autoconf is not perfect and … and … and … . We need more hands and better documentation. BSD/Linux/Haiku/Syllable/AROS/OpenSolaris/OpenDarwin
/Plan9 . Stop this stupid discussion right now !!!
QT apps that are distributed outside of the organization in which they are developed in that the company wishes to keep closed source, need a paid qt license. A company can develop an app for internal use using qt, and so long as it doesn’t get distributed outside the company, the gpl does not apply.
As long as we can choose it doesn’t matter what DE is on your PC.
Well if people start to behave more rationally and take in account their recources linux & co is even more on the rise ( also on the desktop).
A default XP install is a lame bird with hardly any fuctionality nor security.
Getting XP on par with linux & co will cost you rapidly as much as the hardware itself ( or more).
I just norton internet security 2006 with norton ghost and nero,bang dang –> Euro 300.Allthough i must say they are worth it.
Would be nice to have something like ghost for linux,saves a lot of bandwith 🙂
Just have a look at the pace new distros are showing up on distrowatch and more important their origin.After the server rooms the desktop is the new territory to conquer.It’s more a matter of when and how than if or maybe.
I know there are plenty of small differences between the two DEs, but to me none of them seem that significant. They certainly don’t seem enough to justify the fanaticism some people show for their DE of choice. To me they both seem to follow a very similar design, with a panel/taskbar and windows that work in a very similar way. With a bit of tweaking they can look and feel almost identical to each other. Personally I’m equally happy using either of them and to me developing both of them always seemed like a waste of resources.
Personally I think the state of desktop Linux would be better if there hadn’t been both KDE and GNOME. Even though consistency between apps written for different DEs has improved, with features like copy and paste finally working between many GNOME and KDE apps, the remaining consistency issues do still let down Linux usability when compared with other operating systems. Hopefully it will get better in the future whether or not a particular DE “wins” and becomes dominant.
There are differences, especially at the application level. The differences are largely philosophical with respect to designing user experiences. The GNOME developers tend to follow the KISS/minimalist school of thought. The KDE developers tend to follow the everything-plus-the-kitchen-sink school of thought.
When I first began using Linux I loved the KDE approach. It made me feel like a “power user”. I spent weeks tweaking and I enjoyed it! Especially since Windows never gave me such freedom. Needless to say, I was misguided. Spending anything more than 2 minutes tweaking any software program is a bloody waste of your time. It is embarrassing and counter-productive. And no, it doesn’t make anyone power user, contrary to popular misconception! You are exempted if you have an obsessive compulsive disorder.
I learned all these after I grew older, got more responsibilities and became a developer myself. I do not have time to tweak anything! I’d rather spend my time hacking. I do not have time to wade through configuration options and endless menus. I have boring reference manuals to grok. I just want to get the job done, and any application that allows me to do this with no frills, minimal hassle and the least obtrusiveness is a killer application in my books. Oh, and I am a power user!
There’s much power in simplicity. It’s exponentially tougher to design an application that is simple, yet powerful, than one that is clobbered with unnecessary, superfluous, annoying features and endless options. It is much tougher to design a user experience than it is to pack widgets into a container, make every pixel configurable and call it an application. Geeks will never understand and appreciate this, especially the younger ones. I didn’t too, until I realized there is more to life to spending hour tweaking, tinkering and configuring my computer. Of course the geek in me still wants to do all these, but I just can’t find the time, and sometimes the energy, to do this.
This is where I begin to appreciate GNOME’s KISS/minimalist philosophy. GNOME developers spend much time analyzing what feature is worthwhile and what isn’t. They spend an obscene amount of time debating correct behavior and sane defaults. They try to picture users, like me, who just don’t have the time, patience, or excitement to tweak, tinker and configure anything because they know 90% of users don’t enjoy those activities either. They focus on adding only features that yield value, are used more often and worthwhile.
It is no wonder that the corporate desktop is bound to be captured by GNOME. Why? Well, because GNOME developers focus on the ergonomics of the user experience of their applications, while KDE developers focus on the length of the features list of their applications. KDE developers just add the features and then throw their hands up and say do what you wish. Geeks will jubilate at this freedom. Many users like me will curse out loud.
Let me give an example. I had to use Kate yesterday, and I needed to turn on text wrapping. First, it took a few minutes wadding through configuration options to find the option. And then there were about three more options (if I remember correctly) on various types of text wrapping behaviors, among many other trivial but related options. There were no tool tips to give me more hints as to the differences in the behaviors of the text wrapping styles. So here I was wasting my time testing out all the text wrapping styles to see which one I needed. The most annoying thing was that I couldn’t even tell the difference between any of the styles.
In my text editor, Scribes, a GNOME app, there’s just one option to toggle text wrapping on/off. There is a tool tip to tell you exactly what is going to happen when you toggle it on. In fact, when you toggle it on, the feedback is visual and instant, so you see exactly what’s going on. And it will take you less than 3 seconds to find it and configure this option. Can anyone tell me why I need more than one type of text wrapping behavior? Annoyances like these abound everywhere in KDE and KDE applications, that it becomes frustrating for me, and I imagine Home and Corporate users, who have better things to do with their time than bask in the idiosyncrasy of any software.
I’m not saying KDE sucks. In fact, it is impressive. And there are lots of people who appreciate this extensive configuration options even though I feel they misconstrue it for power. However, to say there is no difference between either KDE and GNOME is just wrong. This is also the reason I believe GNOME will always be the Corporate Desktop Darling. When KDE developers begin to understand that 90% of users don’t configure software applications and instead focus on delivering an unparalleled user experience, they’ll win me back. Go to any computer laboratory and see how many users configure, tweak or tinker. This observation is key in designing ergonomic, optimal and efficient user interfaces and experiences. The GNOME folks figured this out a long time ago beginning with GNOME-2.0, lets see what KDE4 has in stock for us. No doubt KDE will attract a lot of Geeks and Window converts, but like me, with time, they’ll get overwhelmed and frustrated and most likely switch before long. This is a trend I’ve noticed from anecdotal experience.
Nb. the KDE libraries IIRC are LGPLed, so *if* you have the commercial QT license you can develop closed-source apps with proper KDE integration.
I was always wondering about this…
How can KDE use GPL’ed code and license the libraries as LGPL? Does that mean, the KDE code is GPL, but the API is LGPL?
Also, can I make a program that uses only KDE API, not Qt? Will that mean that I can make it closed-source and not buy a Qt license?
KDE libs are licenced under the LGPL. Other parts of KDE (such as apps) are licensed under the GPL. Where’s the problem? There’s no single license for the whole of KDE.
Also, can I make a program that uses only KDE API, not Qt? Will that mean that I can make it closed-source and not buy a Qt license?
No. When you link the lgpl kde libraries with the gpl version of qt, the whole load of crap becomes gpl’ed at that very moment.
No. When you link the lgpl kde libraries with the gpl version of qt, the whole load of crap becomes gpl’ed at that very moment.
Exactly! Which is why it makes no sense that the KDE libraries are GPL since they link against the GPL version of Qt. Hence, they are effectively GPL and any program that links against them must be.
Exactly! Which is why it makes no sense that the KDE libraries are GPL since they link against the GPL version of Qt. Hence, they are effectively GPL and any program that links against them must be.
PUH-lease. KDE libraries are LGPL’d for the benefit of proprietary apps using a proprietary Qt license, and effectively GPL for Qt GPL apps. Sort of like, er, Gnome.
Did you GET the memo? Licensing is old news. Even the LSB caved. You guys are supposed to be hiding behind “HIG” as your shield now.
I’m just gonna go ahead and send you that memo. M’kay?
PUH-lease. KDE libraries are LGPL’d for the benefit of proprietary apps using a proprietary Qt license, and effectively GPL for Qt GPL apps. Sort of like, er, Gnome.
Did you GET the memo? Licensing is old news. Even the LSB caved. You guys are supposed to be hiding behind “HIG” as your shield now.
I’m just gonna go ahead and send you that memo. M’kay?
You act as if this is a close and shut case. It is not. I have a legitimate concern. My legitimate concern (as a developer that has used the GPL version of Qt!!!!) is is that the supposed LGPL license of somehow makes it so that commercial apps can link against them. I won’t dispute they as long as they’re linking against versions of the KDE libraries built against the version of Qt they licensed.
However, if you can explain to me in exact legal terms how an LGPL library + a GPL library does NOT equal a GPL library. In that case, since Linux distributions ship Qt as GPL and link the KDE libraries against that GPL library, then the KDE libraries are effectively GPL’d as shipped.
Now, if you can explain to me that the KDE libraries do not depend on or use or link against the Qt libraries, then there isn’t a problem and we can all live in bliss.
Now, please, stop trolling and explain this instead of dismissing me as a troll, instead of as a free software and commercial developer with a legitimate concern!
is that the supposed LGPL license of somehow makes it so that commercial apps can link against them.
Yep. kdelibs is LGPLed. The rest of KDE is released under the GPL.
I won’t dispute they as long as they’re linking against versions of the KDE libraries built against the version of Qt they licensed.
Bollocks. Doesn’t matter. The licenses are compatible.
However, if you can explain to me in exact legal terms how an LGPL library + a GPL library does NOT equal a GPL library.
Because they’re two separate pieces of software? A LGPL component (kdelibs) can use a GPL component (Qt – the two LGPL and GPL licenses are compatible) and GPLed and other licensed components can use that LGPL component (kdelibs). Not rocket science. That’s what the LGPL is for……
In that case, since Linux distributions ship Qt as GPL and link the KDE libraries against that GPL library, then the KDE libraries are effectively GPL’d as shipped.
No, kdelibs is not GPLed. It is LGPLed, and it can use a GPLed Qt component because the LGPL and GPL are compatible licenses. Qt is also dual licensed.
Now, if you can explain to me that the KDE libraries do not depend on or use or link against the Qt libraries
Right, so the root of the problem is that you don’t want KDE to use Qt at all? Glad we got that sorted.
Now, please, stop trolling and explain this instead of dismissing me as a troll, instead of as a free software and commercial developer with a legitimate concern!
The person you’re responding to wasn’t trolling at all. You don’t have a legitimate concern.
The person you’re responding to wasn’t trolling at all. You don’t have a legitimate concern.
Unfortunately, you didn’t provide me any proof or references to your claim that GPL + LGPL != GPL for libraries. If that was the case, then I could write an LGPL library that merely links to a GPL library and thereby circumvent the GPL anytime I wanted to use one in proprietary software. In fact, by the same conclusion you have, someone could write an LGPL stub for the Qt library and thereby use it freely in a commercial application.
Do you see why I’m concerned?
The person you’re responding to wasn’t trolling at all. You don’t have a legitimate concern.
Unfortunately, you didn’t provide me any proof or references to your claim that GPL + LGPL != GPL for libraries. If that was the case, then I could write an LGPL library that merely links to a GPL library and thereby circumvent the GPL anytime I wanted to use one in proprietary software. In fact, by the same conclusion you have, someone could write an LGPL stub for the Qt library and thereby use it freely in a commercial application.
Do you see why I’m concerned?
I’m not sure if I understand what exactly you’re trying to convey here so I will try to explain what the parent poster is saying.
The KDE libraries are LGPL because they they are just that: libraries. In order to use its functionality, your average KDE/QT application must link to it and not the other way around. And since it doesn’t draw widgets nor anything like that (probably low level stuff that makes KDE work, I guess), it is perfectly clear of Qt code. The LGPL was created in order to allow commercial development using FOSS libraries. The catch here is that if you make improvements to the library itself, those changes have to be made available to everybody else. But your application can remain closed.
In fact, it has been said that you can develop applications that takes advantage of KDE features provided by these libraries even if you don’t use Qt.
Basically, what that means is that LGPL/BSD/Closed source app + GPL lib = GPL app while LGPL/BSD/Closed source app + LGPL lib = remains the same license.
So, answering your rhetoric question, I don’t think that it’s possible to “hijack” Qt into a closed source application the way you described. But it is perfectly acceptable to write a closed source application that links to LGPL library such as KDE libs. In fact, this is desirable as it would allow ISVs to create commercial offerings tightly integrated to KDE.
The KDE libraries are LGPL because they they are just that: libraries. In order to use its functionality, your average KDE/QT application must link to it and not the other way around. And since it doesn’t draw widgets nor anything like that (probably low level stuff that makes KDE work, I guess), it is perfectly clear of Qt code. The LGPL was created in order to allow commercial development using FOSS libraries. The catch here is that if you make improvements to the library itself, those changes have to be made available to everybody else. But your application can remain closed.
Thank for you actually reading my post! That is precisely what I was getting at. If the KDE libraries do not link against Qt at all (as you guess), then yes, I agree that they can be LGPL and commercial applications can use them without a problem under that license. Thank you, thank you, thank you for listening! I didn’t think I was speaking that vaguely!
I am familiar with and understand the licensing terms of the GPL/LGPL/BSD/CDDL etc. I had no confusion on those points. I was just confused about the KDE libraries.
Now the only thing I would wish for is a cheaper Qt license for small developers and then it would be perfect!
You act as if this is a close and shut case. It is not. I have a legitimate concern. My legitimate concern (as a developer that has used the GPL version of Qt!!!!) is is that the supposed LGPL license of somehow makes it so that commercial apps can link against them. I won’t dispute they as long as they’re linking against versions of the KDE libraries built against the version of Qt they licensed.
However, if you can explain to me in exact legal terms how an LGPL library + a GPL library does NOT equal a GPL library. In that case, since Linux distributions ship Qt as GPL and link the KDE libraries against that GPL library, then the KDE libraries are effectively GPL’d as shipped.
Now, if you can explain to me that the KDE libraries do not depend on or use or link against the Qt libraries, then there isn’t a problem and we can all live in bliss.
Now, please, stop trolling and explain this instead of dismissing me as a troll, instead of as a free software and commercial developer with a legitimate concern!
Fair enough, my tone was maybe a little more snippy than it should have been but the licensing issue has been addressed several times and I don’t know why I’m letting this thread irritate the hell out of me. Besides, I just watched Office Space for about the hundredth time last night so it seemed a witty enough reply with just the right amount of sarcasm for dramatic effect.
To be sure, it’s a legitimate concern, but not a difficult one to reconcile. To clarify, LGPL is mostly an unrestrictive license, so the licensing of your app depends on the licensing the LGPL libraries are linked to. GPL + LGPL does equal GPL. So:
Qt GPL + KDE Libraries = GPL application
Qt proprietary + KDE Libraries = your choice
Even if you write an app that links only to KDE libraries, you still are obligated under Qt’s licensing since KDE is linked directly to Qt.
From the KDE Myths site linked from kde.org:
Myth: You can’t write closed source KDE applications because, while the KDE libraries are LGPLed, Qt is under the GPL.
Facts: You can write closed source applications for KDE, if you aquire a Qt license from Trolltech to do so.
While the GPL requires that all programs that link against GPLed code also be GPLed, the LGPL makes no such restrictions.
Therefore, the KDE libraries (which are LGPLed) do not restrict the licensing of code they are linked against. The limiting factor becomes Qt itself (commercial, GPL or QPL). {emphasis added}
If you develop using a commercially licensed Qt, you can distribute your KDE program under a commercial licence of your choice.
If you develop using Qt Free edition, your program must also be Free (it must be released under a license that is compatible with either the GPL or the QPL).
URL: http://kdemyths.urbanlizard.com/myth/59
From the KDE Myths site linked from kde.org:
Yes, I’ve seen that and I still don’t believe it. You think the KDE website themselves would willingly put such statements on their own web page as their own claims if they believed them.
Second, the below file (just one of many from kdelibs) clearly includes code from Qt:
http://websvn.kde.org/trunk/KDE/kdelibs/kioslave/http/http.h?rev=51…
This means that if you link that code against the GPL version of Qt included with a Linux distribution, then it becomes a GPL binary. The source code of the library itself obviously remains under the GPL, but the resulting binary is GPL.
This is what I’d like to see people explain their way out of. Nothing but an official statement reference from an appropriate member of the KDE organization will comfort me otherwise.
This means that if you link that code against the GPL version of Qt included with a Linux distribution, then it becomes a GPL binary. The source code of the library itself obviously remains under the GPL, but the resulting binary is GPL.
Yes, this really is true. Instead of moderating me down, why doesn’t someone provide me a link that states in plain black and white text that linking a LGPL *binary* to a GPL *binary* does not make both fall under the GPL.
Did you GET the memo? Licensing is old news. Even the LSB caved. You guys are supposed to be hiding behind “HIG” as your shield now.
Also, “Licensing” is not “old news.” Licensing is still an important consideration. If it wasn’t, would we still be having flame wars over this day?
Also, what’s with the “you guys” comment? Are you aware that I don’t even run GNOME? Are you a omnipotent being that somehow knows instantly whether a person is a GNOME developer?
I bet you’d be surprised if I told you I have never developed a GNOME app. Why the heck do you people automatically assume that because someone disagrees with a license that they are automatically part of the “opposing camp”? Seriously, stop the trolling.
As a KDE user, I used to buy into this sort of junk. I used GNOME for a while.. Stopped hating it, but sstill use KDE because there are some common criticisms that sstill sstand for GNOME — though they’re far less exaggerated.
There will always be people that think like this, there will always be Qt, there will always be KDE, and there will always be disstros that favour nothing. I don’t really care if 10/10 mainsstream distributions ship with Gnome as the default desktop. I’ll always have the option to use KDE, and it really doesn’t matter if you don’t.
Now, I’m a Gnome user, and I like it’s behaviour (if not the slowness and the excessive resourse usage).
AFAIK the KDE user base is 40% larger than the Gnome user base, according to most sites supervising the “war”.
The user base for KDE is quite different from the user base of Gnome, so you can’t really talk of winners and losers.
BTW link dead or so .
Troll ? Yep .
GNOME & KDE are both very good desktop environments but there are more & people will pick what they like most for whatever project or product they have in mind .
So stop trolling … both are very good .
For me WindowMaker is the best
Who made up the rubbish of KDE trying to be Windows like & GNOME Mac-OS-dektop like .. ?
& Yes .. rubbish trying to make a flame article when all sides are actually cooperating
… there is no war & no victor .
Can’t we all agree that KDE and Gnome are both horrible? Ugly, big, plasticky trash that somehow manages to make XP’s default theme seem almost pretty. Seriously, try using the default settings on a low resolution monitor (by today’s standards) like 1024×768 and open a KDE terminal: how much real estate does that use for a terminal? (Declaration of loyalty: openbox3 WM, and no DE under linux.) Come on, people, let’s all pull together (yes, this is rhetorical, and tired of all the nonsense about DEs).
On another note, in reference to the actual article, if corporate adoption was what mattered what has MS not suffocated all the alternatives?
hear-hear … agree absolutly … the rescource wasting in little fancy things at every corner (& edge line & bar … ) is painful but with a customised slicker look they can be nice to use
Could someone please get rid of window decoration
… things should be there when the user needs it & not all the time IMO – an adaptive desktop 8)
Don’t get me wrong, I don’t necessarily dislike fancy effects (I am writing this on an iBook with OS X in all its translucent glory, but on linux I prefer to use Openbox3, pybar, and, well, that’ll do for me), I just don’t think that either KDE or Gnome are particularly good out of the box, and the default settings are almost unusable: between them they *can* suffer from terrible fonts, title bars that are far too large, pointlessly huge icons in menus, bad colour schemes, intrusively large panels,…
Furthermore, I’m specifically thinking of the last set up I’ve used (previous ones are no better), one that is on a corporate version of linux (Suse Pro) installed on my machine at work, not some unconfigured built from source version.
Edited 2006-04-28 21:11
That’s not what I’m asking…
KDE libraries use Qt. Qt is licensed as GPL. KDE libraries are licensed as LGPL. How does that work?
And, can I use only LGPL’ed KDE libraries in a closed-source application, and not buy a Qt license?
LGPL and GPL are compatible. You can mix and match components of each but you have to abide by their licenses. Therefore you have to pay trolltech to use the non-gpl lisense for their libraries. KDE Libs depend on Qt so you can’t write a closed source KDE app without going through trolltech, but the KDE libs themselves don’t place any further restrictions on you, like any other LGPL library.
Other components of KDE that are GPL are the same… they can not be close sourced but can inter-operate with LGPL code.
Edited 2006-04-28 20:41
Remember, in reality, the only clear winner is microsoft windows. It’s installed on 90% of the desktops.
> LGPL and GPL are compatible. You can mix and match
> components of each but you have to abide by their
> licenses. Therefore you have to pay trolltech
> to use the non-gpl lisense for their libraries.
Almost accurate. Actually Qt is covered by three licenses: The GPL, the QPL and the commercial License. All three in combination make it possible to write either:
– any kind of _OpenSource_ Qt(/KDE) applications without the need to buy a Qt license
– closed source Qt(/KDE) applications on the other hand need a Qt license to be bought.
The latter isn’t an issue most of the time: Many proprietary applications like Google Earth, Adobe Photoalbum, Skype and various others are using the commercial Qt license. One of the reasons is that the money you spend for Qt is small compared to the additional development time that the usage of other toolkits would imply.
Trolltech’s small business program
http://www.trolltech.com/products/qt/smallbusiness.html
makes it even easy for small startups to cover the costs.
Kin Korder Kfor Kthe KKDE KX-client Kto Kbe Kconsidered Kas Ka KSERIOUS Kalternative Kfor Kthe KGnome KX-client, Kmeaning, Kto Kbe Kconsidered Ka Kmature Kalternative Kfor Kthe KWindows or KGnome Kenvironment, Kfirst Kget Krid Kof Kthe Kdebilitating KK.
Now, how does that read? Pretty desturbing Knot?+
Do you also have problems with xclock, xscreensaver, xterm, xkill, xlock, xpdf…?
Kyes Khell Ki Kdo, Kyuou Kbastastard :>)
This is the dumbest article that I have ever seen. Gnome was slowly dying and this is exactly why Novell picked it over KDE. No single company can control KDE and this is simply not good enough for Novell. However, while Gnome 3 is nowhere on the horizon KDE 4 is well under development. Not only that but the KDE team is finally putting their house in order and creating a real SDK. This is one of the few OSS projects that is actually introducing some sort of inovation in the community and I am affraid this is the only future that the comunity has. Yes, Novell does bring inovation with XGL but everything they do is to make profit and this is a huge problem in the OSS world. Not because it goes agains principles but because redesignning a framework because it could be done better is not worth the investment in the world of Novell. And at the end you end up with the same qualtiy that Windows offers. Only while MS can rely on their monopolistic position no Linux distribution can. Hence the only future for the community is inovation and bold decision and this is not something that Gnome offers. Even XFCE4 has more inovation than Gnome and this is really sad.
Not that I think that both desktops couldn’t be improved in terms of usability, but …
> I had to use Kate yesterday, and I needed to
> turn on text wrapping. First, it took a few
> minutes wadding through configuration options
First of all enabling text wrapping should rather
be considered a kind of action that you are supposed to find among menu items. So that step shouldn’t have taken minutes.
> And then there were about three more options […]
> There were no tool tips to give me more hints
> as to the differences in the behaviors of the
> text wrapping styles.
According to the feedback that we got from our users KDE decided to restrict the usage of tooltips mostly to toolbar-like widgets. Configuration dialogs are usually covered by the “What’s this?” help. If you click the question mark in the upper right window corner and point to the option in question you get an extensive explanation. In this case I tried it and the explanations shouldn’t leave any questions open.
No need for trial & error or wondering what the different settings would do … 🙂
> In my text editor, Scribes, a GNOME app, there’s
Kate is an _advanced_ editor that is mostly targeted at developers. For your purposes KEdit or KWrite might have been a better fit.
> Can anyone tell me why I need more than one
> type of text wrapping behavior?
Well for developers (who are the target users of Kate) it might be necessary partly due to different coding style specifications among projects.
However I also expect that there’s room for improvement here and will talk to the people working on Kate/KWrite about the issues you mentioned.
I am a developer. 🙂 I found a lot more issues, but I just gave up and tried to find a Ubuntu liveCD to use an editor I was much comfortable with and that needed minimal configuration.
Edited 2006-04-28 22:54
I am a developer. 🙂
If you’re satisfied with the feature set of scribes, fine. But there are a lot of developers who expect more.
Scribes has 97% of the feature set of Kate and then more! There were about 3 features that I found in Kate not present in Scribes of the top of my head.
– Project manager (IMO outside the scope of a text editor)
– code folding
– Tabs (Will not be implemented in Scribes)
Outside those 3 features, Kate does not do anything more than Scribes does, at least from my very frustrating experience with it yesterday. There are quite a few features present in Scribes that I miss in Kate, if I remember correctly. Snippets and basic string manipulations like selecting words, line, sentences, paragraphs, opening new lines and your typical VIM/Emacs text processing operations. I doubt any of these are in Kedit/Kwrite either.
Edited 2006-04-28 23:19
I’ve never used scribes. I don’t like Scite. As someone who uses TextEdit a lot at work, Kate is a sharp piece of work and the reason I don’t bother with jEdit.
Kate is very good. I was just voicing out my frustration with KDE and KDE applications, in general, to corroborate the difference in philosophy between KDE and GNOME developers. I’m not saying Kate or KDE sucks. I’m just saying there are differences in the way GNOME and KDE approach creating user experiences. I chose kate as an example because I just used it yesterday.
Edited 2006-04-28 23:39
Not like I am a Kate fan, but it’s quite easy to blame an alternative when you are used to an application. As for using an Ubuntu LiveCD for development, it must depend on what you are doing…
Remember that Kate isn’t just an editor. The core is a KParts, which makes it useful outside its own MDI application (like for Kdevelop).
I tend to agree that the KDE folks should take a hint at the KISS principle, it just seems to me that some GNOME devs are forgetting the first S! That said, I don’t believe one or the other is “winning”. You have your tastes, I have mine. Although I prefer to use GNOME, KDE just suits better my needs for development.
>>Not like I am a Kate fan, but it’s quite easy to
>>blame an alternative when you are used to an
>>application.
My blame lies in the fact that Kate made me waste so much time trying to activate a rather trivial feature, because of the structure of its interface as well as its developers decision to inadvertently confuse me by exposing both useful and needless options in its interface. An activity that should have taken a few seconds took me minutes. I don’t think toggling text wrapping on/off should be a time consuming activity for an advance user like myself, or any user for that matter.
>>As for using an Ubuntu LiveCD for development, it
>>must depend on what you are doing…
Python development.
>>Remember that Kate isn’t just an editor. The core
>>is a KParts, which makes it useful outside its own
>>MDI application (like for Kdevelop).
That’s all fine and dandy. But as a user this should not be my concern.
>>I tend to agree that the KDE folks should take a
>>hint at the KISS principle, it just seems to me
>>that some GNOME devs are forgetting the first S!
I disagree. I think GNOME could do more to make their apps even more simpler. Getting rid of menus (not context menus) will be a good start that allows developers focus much more on the important elements of their interface as well as its ergonomics.
>>That said, I don’t believe one or the other is winning”.
I agree, as long as both projects have contributors no one is winning.
I am a developer. 🙂
If you’re satisfied with the feature set of scribes, fine. But there are a lot of developers who expect more.
——
Are you trying to say that there are developers who waste time more on configuring their tools than coding?
Are you trying to say that there are developers who waste time more on configuring their tools than coding?
Are you trying to say that emacs and vim are a waste of time?
I’ll admit there are just too many comments to read but from reading the first page it seems that a lot of KDE advocates are real annoyed at the fact that GNOME has won the Linux Desktop Wars. I don’t use either GNOME or KDE now but have favored both at one time or another but I have to admit that GNOME has definitely won and the only people who don’t see that are the KDE advocates. This doesn’t mean that KDE is going away or that no one will use KDE but it is obvious that corporations have made their choice. IF the Linux Desktop does ever hit the mainstream what desktop do you think people will be using? GNOME. Commercial software will be made with the GTK toolkit and possibley integrated into GNOME. KDE will probably be installed by some more adept users but there won’t be many people who change the default. Most people don’t even know how to change their theme in Windows I doubt think they will change their entire desktop. The beauty of open source is that KDE will never go away and people will always have a choice. With the integration that has gone on at freedesktop.org it will be a seamless transition for people who want to move from GNOME to KDE.
This doesn’t mean that KDE is going away or that no one will use KDE but it is obvious that corporations have made their choice.
This comment underscores why the article is pointless. Corporations have not made their choice. Red Hat doesn’t focus on desktops, they’ve made that clear. Novell’s only existing desktop base, such as it is, is predominantly KDE due to Suse and NLD 9. And Sun? Come on. Sun is a weather vane when it comes to linux or anything related to desktops that doesn’t involve java.
What is evident is that Gnome is the dominant DE for downloadable, community-oriented distros. KDE is the dominant DE for those distros that people actually pay money for, such as (the now defunct) Libranet, Xandros, Linspire, (the now Novell owned) Suse etc.
I could try and point out that targeting a market that has proven it is willing to pay for software may be more desireable than targeting one that values being able to download for free and espouses GTK as being ideal due to it’s prix-free LGPL status, but the sad fact of the matter is that nobody knows which way the vendors are going to go since the applications simply aren’t there yet.
I could also try and point out that Trolltech has sold literally thousands of development licenses even for companies developing Windows-only applications, despite the fact that Windows is free-as-in-beer to develop for, but the idea that developers may look beyond price tag when making development decisions seems offensive to many and often leads to curious posts about how wrong it is to use a proprietary framework to lure vendors to develop proprietary applications regardless of the advantages it may bring them.
I could also point out that calling Gnome the “corporate desktop” is as foolish as calling KDE the “corporate desktop”, since neither really exists on corporate desktops in significant enough numbers to warrant consideration.
But really, there’s no point. Both sides seem intent on declaring victory in a battle that really hasn’t even been fought yet.
All versions of NLD are Gnome-defaulted.
All versions of NLD are Gnome-defaulted.
There was only one real version of NLD, and it was a KDE by default though it offered Gnome as an equal option.
SLED will be Gnome by default if the user doesn’t make a preference, still offering KDE as an option, but it’s not even released yet.
Despite that, all Novell cares about in the end is mono adoption and they’ve gone to great lengths to make sure KDE isn’t treated as a second-class citizen. They don’t care if you’re using Gnome or KDE, as long as you’re using mono and they’ve made sure you will be regardless of the desktop you choose.
This comment underscores why the article is pointless. Corporations have not made their choice. Red Hat doesn’t focus on desktops, they’ve made that clear. Novell’s only existing desktop base, such as it is, is predominantly KDE due to Suse and NLD 9. And Sun? Come on. Sun is a weather vane when it comes to linux or anything related to desktops that doesn’t involve java.
First of all RedHat sells workstations and Novell sells desktops. Linux is not mainstream on Desktops anywhere yet. IF it does catch on it will be in business first and all the major players are using GNOME. In fact GNOME doesn’t just run on Linux so your argument about SUN and Linux holds no water. GNOME runs on Solaris too.
What is evident is that Gnome is the dominant DE for downloadable, community-oriented distros. KDE is the dominant DE for those distros that people actually pay money for, such as (the now defunct) Libranet, Xandros, Linspire, (the now Novell owned) Suse etc.
Forgive me for breaking the bad news but sales of Linux desktop software is miserable and the fact that KDE distors are sold more often now (if that is even the case) makes no difference at all.
I could also try and point out that Trolltech has sold literally thousands of development licenses even for companies developing Windows-only applications, despite the fact that Windows is free-as-in-beer to develop for, but the idea that developers may look beyond price tag when making development decisions seems offensive to many and often leads to curious posts about how wrong it is to use a proprietary framework to lure vendors to develop proprietary applications regardless of the advantages it may bring them.
You are arguing a point I never even made.
I could also point out that calling Gnome the “corporate desktop” is as foolish as calling KDE the “corporate desktop”, since neither really exists on corporate desktops in significant enough numbers to warrant consideration.
Considering GNOME is the only desktop in position to land on corporate desktops at this point I wouldn’t say it is foolish to call it the corporate desktop. There is a very, very small chance that a tiny company can make KDE the default desktop but it’s not very likely, especially when large, well known companies are offering GNOME.
But really, there’s no point. Both sides seem intent on declaring victory in a battle that really hasn’t even been fought yet.
You don’t understand that it has been fought already and in fact needed to be fought before anyone adopted Linux on the corporate desktop. Businesses don’t want a choice like that.
First of all RedHat sells workstations and Novell sells desktops. Linux is not mainstream on Desktops anywhere yet. IF it does catch on it will be in business first and all the major players are using GNOME. In fact GNOME doesn’t just run on Linux so your argument about SUN and Linux holds no water. GNOME runs on Solaris too.
Red Hat sells workstations, yes, but they’re not targetting mainstream corporate desktops. The desktops Novell is selling right now are KDE-based or default to KDE, SLED hasn’t been released yet so it’s impact on the market and whether people will joyfully embrace Gnome because of that is theoretical. And Sun running Gnome on Solaris is immaterial, Solaris is less likely to attract corporate desktop users than linux is.
Forgive me for breaking the bad news but sales of Linux desktop software is miserable and the fact that KDE distors are sold more often now (if that is even the case) makes no difference at all.
How is the fact that KDE is the dominant choice in paid commercial distributions irrelevant to a discussion about Gnome leading the desktop? I think it’s a reasonable indicator of the standards people will expect when they’re paying for a product versus downloading one for free.
You are arguing a point I never even made.
You arbitrarily decided that software development on linux will be GTK based because it’s free. I’m pointing out that sufficient evidence exists that developers will pay for Qt even with a “free” alternative. Absence of a license fee does not mean GTK saves developers cost, and if developers try both and find that GTK increases time spent developing applications versus Qt, then GTK actually carries a higher price for them than Qt. It’s not always about raw dollars.
Considering GNOME is the only desktop in position to land on corporate desktops at this point I wouldn’t say it is foolish to call it the corporate desktop. There is a very, very small chance that a tiny company can make KDE the default desktop but it’s not very likely, especially when large, well known companies are offering GNOME.
How? What? Where? When? How? KDE leads in desktop deployment. KDE is leads in paid desktop deployments. KDE was chosen in large scale deployments like the one going on in Munich. Novell’s strategy for reaching desktops remains to be proven, and given how quickly they retreated from the backlash from their customers over the Gnome decision and given how much effort they’ve put into things like splitting mono from Gnome for the benefit of KDE users and creating KDE-front ends for what were formerly Gnome-specific tools, it’s evident that even Novell doesn’t have 100% confidence in Gnome’s superiority for the desktop.
You don’t understand that it has been fought already and in fact needed to be fought before anyone adopted Linux on the corporate desktop.
Wow, this attitude that the only thing holding back linux was people’s failure to acknowledge Gnome’s rightful place never ceases to amaze me. As if the only way to lure corporate customers from Microsoft lockin is to lock them into a DE that a significant majority of the linux community choose freely to not use.
The arrogance from the community is amazing. The devs, projects like freedesktop and Portland, the LSB et al. have all acknowledged the importance of interoperability between the two DE’s because they will both find a place within the userbase. Why go to the trouble when “the community” seems to have arbitrarily decided that Gnome should lead and all others should follow, right?
Businesses don’t want a choice like that.
I see, they told you this, did they?
Bah.
Red Hat sells workstations, yes, but they’re not targetting mainstream corporate desktops. The desktops Novell is selling right now are KDE-based or default to KDE, SLED hasn’t been released yet so it’s impact on the market and whether people will joyfully embrace Gnome because of that is theoretical. And Sun running Gnome on Solaris is immaterial, Solaris is less likely to attract corporate desktop users than linux is.
Don’t discount workstations. They will have an impact on which desktop becomes the standard. So Sun and Redhat definitely matter. Novell has been a stricly GTK+/GNOME based company. It doesn’t matter what they are doing now. What matters is the fact that GNOME has been picked by all the major players for the future. I highly doubt a small company will come in and take over with a KDE based distro. I guess anything is possible but it’s not likely.
How is the fact that KDE is the dominant choice in paid commercial distributions irrelevant to a discussion about Gnome leading the desktop? I think it’s a reasonable indicator of the standards people will expect when they’re paying for a product versus downloading one for free.
I think you’re missing the point. Name one relevant company that uses a KDE desktop standard. Name one large Linux distributor that has KDE as a default. I said from the very beginning that IF desktop linux takes off that it will be GNOME based. This is based on what is happening with all the large Linux companies. If Dell starts selling desktop linux (or any other computer company for that matter) what distro do you think they are going to use? Dell already has deals with Redhat. Who has the infrastructure to handle the requirements of supporting large numbers of users? Redhat, Novell, and Sun, to name a few.
You arbitrarily decided that software development on linux will be GTK based because it’s free. I’m pointing out that sufficient evidence exists that developers will pay for Qt even with a “free” alternative. Absence of a license fee does not mean GTK saves developers cost, and if developers try both and find that GTK increases time spent developing applications versus Qt, then GTK actually carries a higher price for them than Qt. It’s not always about raw dollars.
You put words in my mouth. I never claimed that people would use GTK because it is free. You just decided that I claimed that. I never gave a specific reason. If you want one I’ll give it to you. If big software vendors are going to make applications for linux who do you think they are going to make their apps for? The easy guess would be Redhat, Novell, and SUN.
How? What? Where? When? How? KDE leads in desktop deployment. KDE is leads in paid desktop deployments. KDE was chosen in large scale deployments like the one going on in Munich. Novell’s strategy for reaching desktops remains to be proven, and given how quickly they retreated from the backlash from their customers over the Gnome decision and given how much effort they’ve put into things like splitting mono from Gnome for the benefit of KDE users and creating KDE-front ends for what were formerly Gnome-specific tools, it’s evident that even Novell doesn’t have 100% confidence in Gnome’s superiority for the desktop.
The point of the article and my point is that all the big companies have standarized on GNOME. We are talking about the future here. Corporate linux desktop isn’t here yet but IF it comes it will be GNOME.
Wow, this attitude that the only thing holding back linux was people’s failure to acknowledge Gnome’s rightful place never ceases to amaze me. As if the only way to lure corporate customers from Microsoft lockin is to lock them into a DE that a significant majority of the linux community choose freely to not use.
That’s a ludicrous statement. As I said before freedesktop.org is doing an amazing job and it won’t matter which desktop you use in the future but most will come with GNOME as default.
I see, they told you this, did they?
It’s pretty apparent. You don’t see many corporations with a mix of Win98/ME/XP/2000 and Mac9/X now do you?
I find it funny that most of the arguments against me in this thread are things that I never even said.
Don’t discount workstations. They will have an impact on which desktop becomes the standard. So Sun and Redhat definitely matter.
Many people like you said this five/six years ago – and you’re still saying it. We’re still talking about KDE, and people are still talking crap about what will become a desktop standard.
Many Unix workstations have been replaced by Linux, but most of those are using a free distribution like OpenSuse, Kubuntu, Debian or similar. The market for SLED and Red Hat workstations, and for people paying a subscription for them, is miniscule. Red Hat has sensibly admitted as much.
I think you’re missing the point. Name one relevant company that uses a KDE desktop standard.
You explain to me how half to two-thirds of Linux users use KDE and then we’ll have a discussion about how relevant these companies actually are. Red Hat has been using Gnome since the dawn of time, and KDE is still the most popular desktop and Red Hat still even ship it!
Most people in the Windows world see Sun and especially Novell as irrelevant to them and Red Hat sits unseen in their server room. Since when did CDE or any desktop a company like Sun used have any relevance to what people used on their desktop?
It’s funny how people latch on to these irrelevant Linux companies and claim that what they’re using is some sort of standard.
I said from the very beginning that IF desktop linux takes off that it will be GNOME based. This is based on what is happening with all the large Linux companies.
If you’d actually read previous posts you could have saved some RSA strain. The large Linux companies have no market for their desktop offerings and no potential market. They have been trying for years, and nothing has happened. I daresay that there are more users of Linspire and Xandros than there are of the NLD.
Unless these companies come up with some creative way of funding any desktop activities, or actually come up with some technology that is actually good enough and work out what’s required, then they will continue to be irrelevant and we’ll have yet another article and discussion about this in 2010.
If Dell starts selling desktop linux (or any other computer company for that matter) what distro do you think they are going to use?
They’re pretty ambivalent about what distro they’ll use. If someone comes up with some good technology and users go for it in a big way then they’ll ship it. Michael Dell has said as much. At the moment they have absolutely no confidence in the current crop of distros, and none in the foreseeable future, as well as the spectre of Windows.
The point of the article and my point is that all the big companies have standarized on GNOME.
The real point, which has gone right over your head, is that in the face of KDE’s popularity, and nothing on the horizon from any of these Linux companies of any relevance, their apparent choice of Gnome means zilch.
If big software vendors are going to make applications for linux who do you think they are going to make their apps for?
Vendors will only develop more software if there is a market and the development tools are good enough for them to use. If those two things aren’t in place forget it, and no, the names of Sun, Novell and Red Hat do not hold sway.
Corporate linux desktop isn’t here yet but IF it comes it will be GNOME.
People have been saying that for years and years and years, and I’m sure many people would like that to be the case – but it isn’t.
Many people like you said this five/six years ago – and you’re still saying it. We’re still talking about KDE, and people are still talking crap about what will become a desktop standard.
If you haven’t noticed now is a lot different than 5/6 six years ago and we are talking about now and the future in this discussion.
Many Unix workstations have been replaced by Linux, but most of those are using a free distribution like OpenSuse, Kubuntu, Debian or similar. The market for SLED and Red Hat workstations, and for people paying a subscription for them, is miniscule. Red Hat has sensibly admitted as much.
Just about as miniscule as KDE based corporate desktops.
You explain to me how half to two-thirds of Linux users use KDE and then we’ll have a discussion about how relevant these companies actually are. Red Hat has been using Gnome since the dawn of time, and KDE is still the most popular desktop and Red Hat still even ship it!
First of I would love to find out how you got that statistic and how in the world you could even trust it to be accurate. If you want to talk about Redhat (by far the most widely used distro) then find me the plethora of users that run KDE on Redhat. You won’t find that many in comparison.
Most people in the Windows world see Sun and especially Novell as irrelevant to them and Red Hat sits unseen in their server room. Since when did CDE or any desktop a company like Sun used have any relevance to what people used on their desktop?
Your point? GNOME isn’t CDE and the fact that “the Windows world” sees Novell as irrelevant is in fact irrelevant to the discussion itself.
It’s funny how people latch on to these irrelevant Linux companies and claim that what they’re using is some sort of standard.
Yeah, you’re right Redhat is an irrelevant Linux company.
I can see the big misunderstanding in our conversation. You think that my views are in perfect alignment with the author and other GNOME advocates when they are not. You assume things about me that I never said based on what the author and other posters have mentioned.
In fact it is quite easy, unless you are a KDE fanatic, to see what is going on here. We can agree that Novell, Redhat, and SUN have decided to ship GNOME as the default, correct? You may argue against Novell but it’s clear that after a lot of Suse developers left Novell and the Ximian guys really took over that Novell as a company is gearing itself towards GNOME. Now take my original statement “IF the Linux Desktop does ever hit the mainstream what desktop do you think people will be using? GNOME”. My reasoning behind this is that all the big Linux distribution vendors (and even Sun where GNOME can be used on Solaris) have decided to default to GNOME and create largely GNOME-centric software. Now IF, as I said before, Linux becomes popular on the corporate desktop then GNOME has the upper hand. If a company uses Linux on their servers already and want to explore the option of desktop Linux who do you think they are going to call? I doubt it is going to be Xandros. I can guarantee you that the majority of companies would prefer the integration and consolidation of having a standard distro across the board (It worked wonders for Microsoft in the server market). Redhat and others will most certainly offer discounts for bundles of servers and clients. If that happens GNOME will become the standard desktop and applications will be coded to GNOME APIs.
That is my point and if you can’t refute that without parrotting the same irrelevant nonsense about virtually non-existant corporate KDE desktops over and over again then why are you wasting your time posting?
They’re pretty ambivalent about what distro they’ll use. If someone comes up with some good technology and users go for it in a big way then they’ll ship it. Michael Dell has said as much. At the moment they have absolutely no confidence in the current crop of distros, and none in the foreseeable future, as well as the spectre of Windows.
My point was that any Linux company smaller than Novell or Redhat was going to have a hard time supporting the amount of users that a company like Dell can create.
The real point, which has gone right over your head, is that in the face of KDE’s popularity, and nothing on the horizon from any of these Linux companies of any relevance, their apparent choice of Gnome means zilch.
No, you did miss the point. As I said originally I qualified what I said with a big fat “IF“. KDE’s poplurality means nothing now. If Linux suddenly becomes relevant on the desktop and its userbase quadrules, the default desktop is going to be used by 90% of the new users. Linux is still very much in the minority so if there is a linux breakthrough, the companies largely controlling Linux’s direction will probably play a big part in it and that will have a huge affect on the overall usage of GNOME considering it is the default.
Vendors will only develop more software if there is a market and the development tools are good enough for them to use. If those two things aren’t in place forget it, and no, the names of Sun, Novell and Red Hat do not hold sway.
This is true but it isn’t an argument against anything I said. IF Linux makes it in the mainstream it WILL matter.
People have been saying that for years and years and years, and I’m sure many people would like that to be the case – but it isn’t.
It only isn’t the case because Desktop Linux hasn’t made it to the mainstream yet. Talk to me again if it ever does.
First of all RedHat sells workstations and Novell sells desktops
In this day and age there’s no difference.
Forgive me for breaking the bad news but sales of Linux desktop software is miserable and the fact that KDE distors are sold more often now (if that is even the case) makes no difference at all.
I have yet more bad news for you. In the market that exists now KDE matters. You’re trying to dicuss a mythical, non-existant market that people have been talking about for about six years that has never materialised and won’t as it stands.
Considering GNOME is the only desktop in position to land on corporate desktops…
Sigh. Gnome has not landed on corporate desktops at all. The only thing to focus on right now are people who are actually using desktop Linux and build it up, which is what’s happening with KDE.
Continuing to talk about a corporate desktop market that doesn’t exist will ensure the failure of Novell in particular (not that hard considering what’s happneing at that company right now) and Gnome and these debates can be brought to some sort of conclusion. Red Hat and Sun are not in the desktop market.
There is a very, very small chance that a tiny company can make KDE the default desktop but it’s not very likely, especially when large, well known companies are offering GNOME.
Are their Gnome offerings good enough in view of what they have already? That’s the only thing that matters, and the answer is a big fat no.
You don’t understand that it has been fought already and in fact needed to be fought before anyone adopted Linux on the corporate desktop. Businesses don’t want a choice like that.
Unfortunately, the desktop that you and certain others would like people to choose is not going to be Gnome for various fundamental reasons already pointed out to death here. Telling everybody that Red Hat, Sun and Novell uses Gnome holds no sway when it comes to desktop Linux adoption.
Unless I see Gnome’s userbase go through the roof then you can live in that corporate fantasy world all you like.
In this day and age there’s no difference.
That only helps my point.
I have yet more bad news for you. In the market that exists now KDE matters. You’re trying to dicuss a mythical, non-existant market that people have been talking about for about six years that has never materialised and won’t as it stands.
I have bad news for you. The KDE desktop market is virtually non-existant as well. Give me one good reason why a company like Dell would pick Xandros over Redhat.
Sigh. Gnome has not landed on corporate desktops at all.
You have really poor reading comprehension skills. I didn’t say Gnome was on corporate desktops now.
Continuing to talk about a corporate desktop market that doesn’t exist will ensure the failure of Novell in particular (not that hard considering what’s happneing at that company right now) and Gnome and these debates can be brought to some sort of conclusion. Red Hat and Sun are not in the desktop market.
First you claim workstations and desktops are the same thing then you say Redhat and Sun are not in the desktop market. Make up your mind.
Are their Gnome offerings good enough in view of what they have already? That’s the only thing that matters, and the answer is a big fat no.
Yes it is. You make all kinds of claims without a shred of evidence to back it up.
Unfortunately, the desktop that you and certain others would like people to choose is not going to be Gnome for various fundamental reasons already pointed out to death here. Telling everybody that Red Hat, Sun and Novell uses Gnome holds no sway when it comes to desktop Linux adoption.
There goes your poor reading skills again. I never said I would like to have GNOME as the standard desktop. I just said it is what will likely happen. As for Novell and the like they will definitely determine what desktop will be chosen. They are the large companies that offer GNOME based distros. Name one large company that offers KDE. Name one KDE based distro that has the manpower to actually handle a hardware supplier’s needs.
Unless I see Gnome’s userbase go through the roof then you can live in that corporate fantasy world all you like.
Do you actaully read anything I say or do you just froth at the mouth?
That only helps my point.
No it doesn’t actually – it just shows how far Gnome and these companies have to travel to actually make an acceptable desktop. You can’t pass off an old workstation as one.
A workstation in the traditional sense isn’t enough these days, which is something Novell has discovered. They thought they could use a cut down Linux ‘workstation’ that would perform limited functions like a Unix workstation would do – and charge for it. They came out with lots of rubbish about the NLD ‘not replacing Windows’. However, they found out that people just demanded more functionality of it until it needed to become a replacement for Windows.
Those people who are actually able to use desktop Linux, and possibly replaced their Unix workstations, are already doing it with Kubuntu, OpenSuse or Debian for free and don’t need any help from Novell whatsoever. Result? The NLD and SLED don’t matter one iota to anyone.
You have really poor reading comprehension skills. I didn’t say Gnome was on corporate desktops now.
Ahhh, sidestepping the actual issue ;-). Gnome isn’t on corporate desktops now (which means the demand is complete fresh air!), it isn’t in the process of being put on corporate desktops nor is it likely to be in the future on that basis for reasons already pointed out. This was the point out of all of that:
“I have yet more bad news for you. In the market that exists now KDE matters. You’re trying to dicuss a mythical, non-existant market that people have been talking about for about six years that has never materialised and won’t as it stands.”
Which you didn’t reply to :-).
First you claim workstations and desktops are the same thing then you say Redhat and Sun are not in the desktop market.
And how many ‘desktops’ do they actually shift and sell in the grand scheme of things? I’d call that ‘not being in the desktop market’. That’s why even Red Hat themselves say they aren’t in it and try and describe their ‘desktops’ as ‘workstations’ ;-).
You want to be a slippery weasel, and guess what? You fail ;-).
Yes it is. You make all kinds of claims without a shred of evidence to back it up.
Wow. Pot meet kettle. There’s been ample evidence, which of course you don’t want to read. Gnome is not a developer’s platform for one, developers create the applications, applications attract userbase. Easy.
I never said I would like to have GNOME as the standard desktop.
Nice try, but there is a long line of comments as evidence to the contrary. Please don’t try and take the moral high ground regarding how you don’t want this to happen but there is some logical explanation which is detached from your emotional attachment as to why that will be the case :-). Your arguments, based on what flimsy assumptions there are, have been quite clearly taken apart.
As for Novell and the like they will definitely determine what desktop will be chosen.
Speaking of evidence…. Where’s the evidence for that, because it isn’t what’s actually happening in the desktop Linux world, and people at Novell and elsewhere have been claiming that for years. Guess what? Hasn’t happened. Where’s the evidence that that hasn’t happened? People like you are still arguing the contrary! Arghhh, you’re killing me :-).
Like I said fifteen thousand comments ago, you can tell us that all these corporate companies are using Gnome but it isn’t making a blind bit of difference. There’s been about six years of evidence for that.
Name one large company that offers KDE.
*Hammers nail into thick skull* It’s userbase that matters…… That was also a point hammered home several thousand comments ago. Perceived endorsements from companies that have actually been using Gnome for years have made not one jot of difference. Hint: That’s why you’re still arguing to the contrary! ROTFL.
Anyway, Novell and Suse do for one. Despite protestations to the contrary all their major paying distributions still use KDE, YaST uses Qt and just about everyone who uses OpenSuse uses KDE. I don’t see a replacement for YaST and the graphical configuration tools people need written with Gnome and GTK…. I don’t even see them in Red Hat either.
Besides, as I hammered home several thousand comments ago how are about half to two-thirds of Linux users obtaining KDE then? I don’t see endorsements from companies which don’t matter mattering to those people.
Name one KDE based distro that has the manpower to actually handle a hardware supplier’s needs.
And considering that that is largely a kernel issue, the kernel is open sourced, projects like wireless on Linux were started privately without the distributors and that it has little to do with Gnome or KDE – that matters how?
Name me one distributor anywhere that has actually handled any hardware supplier’s needs.
Do you actaully read anything I say or do you just froth at the mouth?
Sounds like froth to me. Since everything you’ve ‘said’ is based on a very large assumption that Red Hat, Novell and Sun actually matter and will dictate what people use (there is about six years of evidence against that), and that there is no corporate Linux desktop market (which you’ve admitted!) but there will somehow magically be one in the future, based on these companies which have never mattered to anyone, using Gnome(!) – foregive me if I’m inclined to disagree :-).
ROTFL. I’ll take that as a yes to everything I’ve written. I’ll re-iterate, in case you had trouble reading:
“Unless I see Gnome’s userbase go through the roof then you can live in that corporate fantasy world all you like.”
That corporate world has no relavance to what’s actually happening, and most importantly, what’s actually required to make desktop Linux adoption beyond what it is now.
Nuff said.
No it doesn’t actually – it just shows how far Gnome and these companies have to travel to actually make an acceptable desktop. You can’t pass off an old workstation as one.
Then don’t claim a desktop and a workstation are the same thing. Remember you made that claim, not me.
Those people who are actually able to use desktop Linux, and possibly replaced their Unix workstations, are already doing it with Kubuntu, OpenSuse or Debian for free and don’t need any help from Novell whatsoever. Result? The NLD and SLED don’t matter one iota to anyone.
Yeah I bet companies would love that kind of setup, totally unsupported software.
Ahhh, sidestepping the actual issue ;-). Gnome isn’t on corporate desktops now (which means the demand is complete fresh air!), it isn’t in the process of being put on corporate desktops nor is it likely to be in the future on that basis for reasons already pointed out. This was the point out of all of that:
I’m not sidestepping the issue, you are. My whole point was about the future and how GNOME is being positioned to take over.
And how many ‘desktops’ do they actually shift and sell in the grand scheme of things? I’d call that ‘not being in the desktop market’. That’s why even Red Hat themselves say they aren’t in it and try and describe their ‘desktops’ as ‘workstations’ ;-).
You want to be a slippery weasel, and guess what? You fail ;-).
You fail miserably again. I was using your own words and now you are arguing against yourself. I call that a pretty miserable failure.
Wow. Pot meet kettle. There’s been ample evidence, which of course you don’t want to read. Gnome is not a developer’s platform for one, developers create the applications, applications attract userbase. Easy.
Are you really trying to say that GNOME doesn’t have applications? Are you nuts. There are more GNOME/GTK apps out there than KDE/QT.
Like I said fifteen thousand comments ago, you can tell us that all these corporate companies are using Gnome but it isn’t making a blind bit of difference. There’s been about six years of evidence for that.
Like I said in my very first comment, I never claimed any corporations were using GNOME now. You just decided to make up half of what you think I said.
*Hammers nail into thick skull* It’s userbase that matters…… That was also a point hammered home several thousand comments ago. Perceived endorsements from companies that have actually been using Gnome for years have made not one jot of difference. Hint: That’s why you’re still arguing to the contrary! ROTFL.
Userbases change and that is what I am talking about. The reason endorsement by large linux vendors hasn’t made a difference yet is because they haven’t made a difference in the desktop market yet. I thought I couldn’t be more obvious.
And considering that that is largely a kernel issue, the kernel is open sourced, projects like wireless on Linux were started privately without the distributors and that it has little to do with Gnome or KDE – that matters how?
Name me one distributor anywhere that has actually handled any hardware supplier’s needs.
You missed the point yet again. It’s called support. Dell and other manufactureres are not going to mess with small Linux vendors that can’t support a large userbase.
Sounds like froth to me. Since everything you’ve ‘said’ is based on a very large assumption that Red Hat, Novell and Sun actually matter
I find it funny that you think Redhat, Sun, and Novell don’t matter. Redhat and Novell are the biggest Linux vendors and Sun ships GNOME with its operating systems.
there is no corporate Linux desktop market (which you’ve admitted!) but there will somehow magically be one in the future, based on these companies which have never mattered to anyone, using Gnome(!) – foregive me if I’m inclined to disagree :-).
Well I can see your reading skills haven’t gotten any better. I did admit there isn’t a corporate desktop market right now…in my very first comment! What I didn’t say is that there would magically be one in the future. I said IF. Do you know what that word means?
http://dictionary.reference.com/search?q=if
I’ll admit there are just too many comments to read but from reading the first page it seems that a lot of KDE advocates are real annoyed at the fact that GNOME has won the Linux Desktop Wars.
This always makes me chuckle. It seems to me that many Gnome advocates are really, really annoyed that there never was a desktop war, Gnome hasn’t won anything other than some apparent approval (which it has apparently been doing for years) from companies that simply don’t matter and never have with desktop usage, and quite frankly, Gnome just isn’t good enough to actually matter.
This “desktop war” was not thought up by any KDE person, OK? Articles and comments like this are just an act of desperation that KDE, after all this time (years later) and many, many, many comments and articles like we’ve got here, has not “gone away” (in their words).
There are some hard and fast reasons why KDE has not “gone away” – it’s good to develop for, has good structure and this comes out in the applications and desktop which attracts users and developers. If you don’t have that then you can forget it, no matter how much corporate approval you may have or may think you have. I think it’s time some people dealt with that in their minds.
This always makes me chuckle. It seems to me that many Gnome advocates are really, really annoyed that there never was a desktop war, Gnome hasn’t won anything other than some apparent approval (which it has apparently been doing for years) from companies that simply don’t matter and never have with desktop usage, and quite frankly, Gnome just isn’t good enough to actually matter.
First of all I am not a GNOME advocate and you would know that if you actually read my comment. I don’t even use GNOME and in fact I used both at one time or another and to claim GNOME isn’t good enough shows cleary that you are a KDE fan blinded by your own bias.
This “desktop war” was not thought up by any KDE person, OK? Articles and comments like this are just an act of desperation that KDE, after all this time (years later) and many, many, many comments and articles like we’ve got here, has not “gone away” (in their words).
Did you have a point to go with that statement?
There are some hard and fast reasons why KDE has not “gone away” – it’s good to develop for, has good structure and this comes out in the applications and desktop which attracts users and developers. If you don’t have that then you can forget it, no matter how much corporate approval you may have or may think you have. I think it’s time some people dealt with that in their minds.
What? I don’t have anything against KDE and I never have. It is a great desktop and I haven’t argued against anything you said in that entire paragraph. It seems like you are desperately trying to ratonalize your use of KDE over GNOME. I don’t see a need to considering I stated that they are both excellent desktops and neither are going to go away soon.
Your comment just reinforces what I said before. KDE fans are so ready to freak out about GNOME being chosen by the big dogs when there is no reason to. It won’t change a thing when it comes to YOUR desktop so there is no need to get worked up.
I don’t even use GNOME and in fact I used both at one time or another and to claim GNOME isn’t good enough shows cleary that you are a KDE fan blinded by your own bias.
You don’t even use Gnome and yet my claim that Gnome isn’t good enough (which I do use), which I’ve given some evidence for (mainly it’s not a developer’s desktop), means I’m blinded by my own bias? Right…..
“This “desktop war” was not thought up by any KDE person, OK? Articles and comments like this are just an act of desperation that KDE, after all this time (years later) and many, many, many comments and articles like we’ve got here, has not “gone away” (in their words).”
Did you have a point to go with that statement?
Have a wild guess. You wrote “KDE advocates are real annoyed at the fact that GNOME has won the Linux Desktop Wars.” If that isn’t a silly comment I don’t know what is.
Your comment just reinforces what I said before. KDE fans are so ready to freak out about GNOME being chosen by the big dogs when there is no reason to.
And you’re not a Gnome advocate in any way? Hmmmm…..
Sorry, but your attempt at trying to appear rational and somehow ‘above’ all of the stuff you’re complaining about has failed miserably.
You don’t even use Gnome and yet my claim that Gnome isn’t good enough (which I do use), which I’ve given some evidence for (mainly it’s not a developer’s desktop), means I’m blinded by my own bias? Right…..
Ok let me clarify things. I use GNOME at work, so I’m pretty sure I know its capabilities. I got my brother into KDE and help him with it often. I am exposed to both desktops on a regular basis although I don’t use either one as my desktop.
Have a wild guess. You wrote “KDE advocates are real annoyed at the fact that GNOME has won the Linux Desktop Wars.” If that isn’t a silly comment I don’t know what is.
You haven’t explained why it is a silly comment. It’s your own personal belief but you haven’t made any real arguments against it. My argument is based on the comments I see on this forum and the reality in the Linux Corporate Desktop world.
And you’re not a Gnome advocate in any way? Hmmmm…..
I admit that I like GNOME, but I also like KDE. If I had to choose I would take GNOME because it is more coherent and their regular releases get noticeably better each version.
Sorry, but your attempt at trying to appear rational and somehow ‘above’ all of the stuff you’re complaining about has failed miserably.
How can you say that? Because I don’t agree with you? That seems to be your only basis for attacking me. I have no bias because neither is my desktop of choice but I have experience with both. I would have to say that you failed miserably at making a point but you did succeed in reinforcing mine, yet again.
Best thing to do with flamebait like those articles is ignore them. Neither KDE or Gnome is going to win.
IMO KDE is designed so it can be used by anyone, but also so it can appeal to power users. Gnome on the other hand is designed so it can be used by anyone, but the developers seem to have decided to stop short of appealing to power users. That’s not to say Gnome users can’t also be power users, but it does mean their desktop environment will be less accomodating to that. Because of this, using Gnome for me is like going through life with only one arm, it takes me longer to get stuff done in Gnome because I preffer having more malleable tools, but on the other hand the same limitations make Gnome better off for someone who doesn’t want to have to do too much learning to use their computer.
No one desktop environment is going to appeal to a large enough percentage of Linux users to kill off the other, the developers already know this and they cooperate, but for some reason some people insist on trying to spread animosity like this article does.
My computer uses KDE. And I use Fedora. This seems to contradict the story. Use what you like. No distribution can stop you from using KDE and RedHat isn’t exactly known for making smart decisions.
Was the article just meant as a trolling attempt, or what was the point? I can’t see GNOME and KDE developers joining as they both have quite different ideals and tastes. That just won’t happen. And I actually want things to stay that way. Atleast that way there will be not one, but two, completely different desktop environments which still let you do the same things. How can this be a bad thing? Sure, it means not every Linux desktop is the same, but they aren’t the same even between those who use f.ex. GNOME only.
I am a long-time GNOME user and can never even think about switching over to KDE, but I don’t really have anything bad to say about KDE either. Neither has won, I just use GNOME because it fits better to my taste.
And licensing is part of the answer of why and another reason is that the libraries were written in C which makes it easier to bind to.
People around OSNews like the status quo and are happy with both KDE and Gnome, but the fact is that Linux will never, ever move out of the hobbyist/corporate-info-desk market without a clear, dominant set of libraries that are going to be included almost all linux desktops.
But there are corporate interests that want something more than “choice”. And about choice…you can never take away KDE or anything else, but there needs to be a standard toolkit on Linux.
So you guys can scream troll all you want (god..how many times have we heard that before…and like it has any meaning left), but the fact is that people around here aren’t who the target is.
And no, I’m not a Gnome fanboy. When I do run a linux desktop it’s KDE because I feel it’s just a better environment, but I have to look at the bigger picture.
Frankly, I think the game is already over whether Gnome won or not. OSX is already taking people away from linux because linux failed to put together that cohesive feel that OSX put out.
And licensing is part of the answer of why
Your desktop has to be good enough and good enough to develop for. Regardless of any licensing, if you don’t have that it simply isn’t going to get used.
nd another reason is that the libraries were written in C which makes it easier to bind to.
Hmmm, right. Here’s the cluestick. Someone actually has to write that low level C stuff, and then create bindings for people to develop with on top. Does that sound like a good usage of time? How about you just get yourself an object oriented language coupled with a good toolkit, like say, C++ and Qt so you don’t actually need to do that?
you can never take away KDE or anything else, but there needs to be a standard toolkit on Linux.
People have been saying that for years and there isn’t one. The likelihood of GTK becoming a standard for anything is not high, simply because it’s crap. It’s awful to develop for and simply doesn’t have the logical structure to develop for the Gnome environment that people want. That’s even in the article.
Novell, Sun and Red Hat will have to spend millions, or even billions, making a development platform that will fuel widespread adoption of Gnome on a par with what’s in the Windows world – which they’re not doing. Where’s that money going to come from?
So you guys can scream troll all you want (god..how many times have we heard that before…and like it has any meaning left), but the fact is that people around here aren’t who the target is.
Scream all you want, but the vast majority of people who use desktop Linux are people around here. People download OpenSuse, Kubuntu, use Debian or something similar. Do you use SLED or the NLD, and do you pay for it? Do you use the Java Desktop System? Do you know anyone or any business out there who actually does? I’m guessing no, but people still persist in this fantasy and thinking that these meaningless distributions and systems are where it matters.
And no, I’m not a Gnome fanboy.
OK.
OSX is already taking people away from linux because linux failed to put together that cohesive feel that OSX put out.
OS X is going nowhere. It still doesn’t have the critical mass that makes it relevant, but it is more relevant and you do see more of them than desktop Linux.
“Hmmm, right. Here’s the cluestick. Someone actually has to write that low level C stuff, and then create bindings for people to develop with on top. Does that sound like a good usage of time? How about you just get yourself an object oriented language coupled with a good toolkit, like say, C++ and Qt so you don’t actually need to do that? “
Bindings for C++, Python, Perl, and Java are officially supported by GNOME. Even if developing the original libraries in C was harder than it would be doing them in C++ (which is simply not true), that would be something 3rd party developers could care less about; it’s not their problem.
The reason people don’t use C++ so much in the GNOME world is that it has the same problems as C. If you want to move away from C then something like C# or Java would make more sense, not C++.
The likelihood of GTK becoming a standard for anything is not high, simply because it’s crap. It’s awful to develop for and simply doesn’t have the logical structure to develop for the Gnome environment that people want.
So GTK is crap because you say so?
Novell, Sun and Red Hat will have to spend millions, or even billions, making a development platform that will fuel widespread adoption of Gnome on a par with what’s in the Windows world – which they’re not doing. Where’s that money going to come from?
Red Hat, Novell and Sun already have (on GNOME) a suitable replacement for Windows, in fact for a good while now.
Scream all you want, but the vast majority of people who use desktop Linux are people around here. People download OpenSuse, Kubuntu, use Debian or something similar. Do you use SLED or the NLD, and do you pay for it? Do you use the Java Desktop System?
That’s smart. Now all you have to do is prove that the majority of people have the same profile as OSNews readers.
>>The reason people don’t use C++ so much in the GNOME
>>world is that it has the same problems as C.
That’s not true. C++ has its own baggage of issues that are entirely different from C. The reason people don’t use C++ in GNOME is because it has become evident that GNOME has better and more productive languages for desktop application development, like Python. C++ will never become popular in GNOME, period.
So GTK is crap because you say so?
I think you need to get out more and start looking at what developers, both of prorietary and open source software, actually want from the tools they use.
Red Hat, Novell and Sun already have (on GNOME) a suitable replacement for Windows, in fact for a good while now.
No, they don’t. They don’t have it with KDE either, but at least they’d be able to get far closer with what’s at their disposal.
Note this very carefully: Commercial open source applications DO NOT require a QT commercial license. Read this for more info: http://www.trolltech.com/products/qt/pricing.html?cid=18
This is text quoted from that site:
This is how it works: In return for the advantages you realize from using a Trolltech product to create your application, we require that you do one of the following:
Either: Contribute to the continued development of the product by purchasing commercial licenses from Trolltech. This option secures you the right to distribute your application under the license terms of your choice.
Or: Contribute to the Open Source community by placing your application under an Open Source license (e.g. the GPL). This option secures all users the rights to obtain the application’s full source code, modify it, and redistribute it.
Don’t spread FUD around and scare away developers who want to use Qt for commercial open-source development.
Dont spread FUD. Developing commercial but open source applications doesnt require any licensing fees. Licensing fees are required for the commercial license which allows your application to be closed source, in this case Qt is licensed under QPL.
See this for more info: http://www.trolltech.com/company/model.html
A quote from this site:
This is how it works: In return for the advantages you realize from using a Trolltech product to create your application, we require that you do one of the following:
Either: Contribute to the continued development of the product by purchasing commercial licenses from Trolltech. This option secures you the right to distribute your application under the license terms of your choice.
Or: Contribute to the Open Source community by placing your application under an Open Source license (e.g. the GPL). This option secures all users the rights to obtain the application’s full source code, modify it, and redistribute it.
Why you shouldn’t use the Lessor Gnu Public License
http://www.gnu.org/licenses/why-not-lgpl.html
Gnome has corporate sponsors and interest because they can use the code and not give back to the community. What do you want more of, free software or proprietary and commercial software?
Keep Linux free, use KDE!
“But really, there’s no point. Both sides seem intent on declaring victory in a battle that really hasn’t even been fought yet.”
Interestingly enough, neither side has actually declared a victory or even a war to begin with.
The only ones insisting on such nonsense are usualy clueless fanbois with nothing better to do with their time than to write trollish, inflammatory articles in some vain attempt to validate themselves.
Edited 2006-04-29 10:11
in the real world, competition is a good thing that spurs growth and innovation.
that people want KDE and GNOME to merge. The very reason there are more than one “desktop environment” is because *people are different, prefer different approaches and like different things*. Merging would be the worst possible choice since instead of two very different approaches we would now have a single bland, one-size-fits-all environment. Catering to the smallest common denominator is a surefire way to ruin the uniqueness and quality that is the very reason for people to chose either one. Of course, this will somewhat hamper such things like market-share but really, do we want gained market-share at any cost?
It’s like saying every artist should disrergard their artistic integrity and start being like Britney Spears, Backstreet Boys or Chrstina Aguilera because, hey, that stuff may not be great music but it sure sells and selling is all we really should care about.
“Well, not exactly. Free applications written using QT widgets don’t require a license but commercial applications do require one.”
So you are saying that the same companies that are now developing for Windows and paying insane amounts of money for MS devtools licenses would have a hard time paying the, in comparison, miniscule QT license?
Come on, that’s just silly.
So you are saying that the same companies that are now developing for Windows and paying insane amounts of money for MS devtools licenses would have a hard time paying the, in comparison, miniscule QT license?
Come on, that’s just silly.
That logic is flawed at best. “MS devtools licenses” aren’t required to develop any Windows application to my knowledge. You can use mono for .NET or sharpdeveloper with the free .NET SDK. You could use Metrowerks CodeWarrior, and all kinds of other commercial compilers or software packages. Not only that, with the new express editions of Visual Studio — you can create commercial applications for the Windows platform using “MS devtools” for free. The issue at hand here is that adopting the GPL version of Qt for a free desktop platform *requires* developers that develop closed-sourced applications to pay a licensing fee to TrollTech. Windows and OS X do not require this, nor do many other platforms.
That logic is flawed at best. “MS devtools licenses” aren’t required to develop any Windows application to my knowledge. You can use mono for .NET or sharpdeveloper with the free .
That really is a flawed argument. Despite the fact that the SDK, Mono and SharpDevelop may be free the poster’s point was that even with this the vast majority still buy Visual Studio and MSDN licenses because to get any serious work done you need serious development tools with investment put into them.
The issue at hand here is that adopting the GPL version of Qt for a free desktop platform *requires* developers that develop closed-sourced applications to pay a licensing fee to TrollTech.
And developing for Windows, Office and other Microsoft software doesn’t require you to purchase licenses for Windows, Office and other software from Microsoft?
Windows and OS X do not require this, nor do many other platforms.
So Windows and OS X are given away for free? Windows and OS X do not have different business models where they are given away for nothing like desktop Linux. Charging people for development to put back into Qt that creates a desktop environment like KDE, as well as the software they’re using to increase productivity or make profit from, is a sensible idea.
Note also that any money payed to Trolltech helps to keep Qt in good condition not only for you as a commercial developer but also for KDE developers developing that software you’re using for nothing! You don’t just pay money into a Windows or Office pit – there is some serious gain here.
Investment needs to come from somewhere, and nowhere have I seen anyone address that issue regarding desktop Linux. Instead you get fanboys who think that their corporate backers in Novell, Sun and Red Hat are simply going to pour money that doesn’t exist into development tools that are good enough to help widespread Gnome adoption – for no appreciable gain out of the goodness of their hearts.
Note also that any money payed to Trolltech helps to keep Qt in good condition not only for you as a commercial developer but also for KDE developers developing that software you’re using for nothing! You don’t just pay money into a Windows or Office pit – there is some serious gain here.
Investment needs to come from somewhere, and nowhere have I seen anyone address that issue regarding desktop Linux. Instead you get fanboys who think that their corporate backers in Novell, Sun and Red Hat are simply going to pour money that doesn’t exist into development tools that are good enough to help widespread Gnome adoption – for no appreciable gain out of the goodness of their hearts.
Well said.
I would add to that Trolltech’s self-sustaining development model helps ensure KDE/Qt continuity without relying on corporate backing that could dry up as quickly as a bad quarter occurs, or fall victim to competing agendas, but that’s probably not relevant to a licensing discussion so I’ll refrain from bringing it up.
Oops… too late…
That really is a flawed argument. Despite the fact that the SDK, Mono and SharpDevelop may be free the poster’s point was that even with this the vast majority still buy Visual Studio and MSDN licenses because to get any serious work done you need serious development tools with investment put into them.
While that may be, at least you don’t have to pay thousands of dollars per year for each developer to have their copy of Visual Studio to develop applications. Additionally, the other point is that it’s still NOT a requirement. Licensing Qt would be for KDE.
Let’s see, $1500 for an entire development suite *once* or a few thousand *per year*…how is that a fair comparison?
While that may be, at least you don’t have to pay thousands of dollars per year for each developer to have their copy of Visual Studio to develop applications.
It’s not just Visual Studio but an MSDN subscription you have to compare it to – and there is far better development support. You also need licenses, and dictate licenses to your customers, which Microsoft’s development tools hook into – SQL Server, Exchange, Sharepoint etc. It ain’t cheap.
Additionally, the other point is that it’s still NOT a requirement.
It is for people who want to develop for a living. Mucking about with cheaper development tools just isn’t an option. You may use a lot of free and open source stuff around the edges, but the core nucleus of your development tools will be bought and payed for.
Let’s see, $1500 for an entire development suite *once* or a few thousand *per year*…how is that a fair comparison?
I don’t follow. You can pay for Qt once, or you can choose to pay per year for some support, documentation, tools… etc.
It’s not just Visual Studio but an MSDN subscription you have to compare it to – and there is far better development support. You also need licenses, and dictate licenses to your customers, which Microsoft’s development tools hook into – SQL Server, Exchange, Sharepoint etc. It ain’t cheap.
Here are the facts plain and simple no matter the justifications and excuses:
* FACT: Developing closed source applications for the KDE platform requires a Qt license (they don’t even have to be commercial).
* FACT: Developing closed source applications for the Windows and OS X platforms DOES NOT REQUIRE a *development* license (the OS license does NOT count, as you need that to run any application to begin with, that’s a user license NOT a development license — we are only discussing development licenses).
* FACT: Developers do not have to purchase MSDN subscriptions, MS Visual Studio or anything else to develop applications for the Windows platform (it is their *option*).
* FACT: Developers do not have to pay Apple for XCode or MetroWerks for CodeWarrior to develop applications for OS X.
You can try to justify this however you want. The point is that it’s NOT a requirement. It doesn’t matter what developers “usually” do. The point is that as a developer I have a *choice* when developing applications for a platform on what *development tools* I license. The KDE platform does not give me that choice if I want to develop commercial applications for it. The KDE platform forces me to license tools from *one* vendor.
At the very least, on the Windows platform, I could choose to license my *development* tools from MetroWerks, Borland, SUN, Microsoft or others. With KDE I only have one choice: Trolltech.
You can try to justify this however you want. The point is that it’s NOT a requirement. It doesn’t matter what developers “usually” do. The point is that as a developer I have a *choice* when developing applications for a platform on what *development tools* I license. The KDE platform does not give me that choice if I want to develop commercial applications for it. The KDE platform forces me to license tools from *one* vendor.
Your choice is to use KDE/Qt or not. If you as a developer choose to purchase development tools to aid with Windows development despite the lack of mandatory requirement, then you obviously see value in doing so and one would assume you expect to ultimately be more profitable in doing so, offsetting the cost of the tools.
If you as a developer see value in the advantages KDE/Qt may offer and feel that it will make your development more efficient hence more profitable, then it’s not unreasonable to attach a price tag to that.
In either case, it’s your choice to do so, no denying that.
I fully agree that forcing a license fee to develop for linux would be inherently wrong, which is why I fully support the idea of Gnome/GTK despite my choice not to use it. I also have high hopes for linux’s future as a platform for commercial development, which is why I’m glad the dust has settled and the LSB agreed that Qt needs to be included as a standard linux library.
The reason this seems to be such a contentious issue is this unstated assumption that a) Gnome has won the desktop battle so b) GTK will be the de facto development environment, which is based on the flawed argument that GTK is “free” just because developers don’t have to pay for it or fund it.
Some developers prefer professional supported development tools and don’t mind paying for them, forcing GTK on them if they want to write standards-based linux applications would be a good way of encouraging them to stick to Windows and OS X.
But to the point of all your “FACTS” about licensing cost versus Qt, you overlook the one point that many GTK proponents do, and that’s one of sustainability. GTK inherently relies on corporate funding for development and advancement, and if GTK really is going to fuel enterprise application development, then it will need ever increasing funding for development and advancement.
Apple and Microsoft are well established companies, there really isn’t a “gamble” in choosing to develop on one of those platforms, and they are sustainable because of the money those two organizations make on license fees that get pumped back into development of the platform. So there’s no need to charge a license fee to cover development costs, that cost is incrementally passed on to the end user in the license fee they pay for.
You as the developer are adamant that you should not have to pay to use GTK, so who sustains it? You want to keep all the money you make for your time invested developing applications. Users likely aren’t going to pay between $100 – $300 for desktop linux they way they do for OS X or Windows, even if that price is buried in the cost of their hardware. So that avenue of funding is not available either.
Red Hat’s core revenue comes from service contracts on it’s server products, is it going to funnel extra funding into accelerating GTK development for desktop application that they won’t directly profit from? Novell’s financial position is shaky right now and there are rumours flying about possibly being acquired, can you be sure Novell will be in a position to keep pouring money into GTK despite their best intentions? What if Novell tries to force mono into Gnome, Red Hat may have caved with Fedora but will they acquiesce with their enterprise products and hope Microsoft doesn’t call their bluff about patent countersuits?
Furthermore, there’s the flawed logic in Red Hat and Novell’s enterprise strategy. They want to see the large ISV’s porting applications yet they’re basing it around a lowest-common-demoninator approach to the development framework. That’s great for the community, but not necessarily compatible with the way commercial developers make business decisions.
This whole thread was originated under the flawed assertion that Gnome has somehow won the enterprise desktop war, and then that spun off into GTK becoming a de facto development framework because of Qt’s “cost”, as always seems to happen. Again, it’s all supposition based of flawed arguments and hypotheses of things that remain to be proven.
Qt carries a one-time price. But Trolltech is an established company with a proven product, Qt is cross-platform and while KDE may be their “flagship” high-profile project, their framework is actually more widely deployed in Windows-based applications from a commercial point of view. Qtopia is becoming widely adopted on PDA platforms. Basically, Trolltech will likely successfully exist with or without KDE. That provides stability and sustainability, and in addition to the advantages their framework may offer for development, THOSE are the factors commercial developers will look at. At least, the successful ones anyways. That’s not to say everyone will choose Qt, just pointing out that many factors beyond license fee come into play. You pay an upfront fee but you know what you’re getting in return for that fee. The only true question about Qt’s viability from a development standpoint in the past was whether it would be recognized as part of the linux stadard set of libraries, and has since been addressed.
That’s not to say there’s anything inherently wrong with GTK. Certainly it’s an equally viable platform for development and many people choose to use it, as is perfectly both acceptable and their right to do so. Nobody would argue that. But to call GTK “free” is simplistic and ignores the factors that real-world business people use when making decisions. UNTIL Red Hat and Novell have actually achieved some sort of saturation point with Gnome on enterprise desktops and are profitable in doing so (all the while competing with the Debians/Ubuntus etc. of the world) then there is absolutely no assurance that GTK is sustainable from a business point-of-view. Sure there’s always the community to keep it alive and running, but that doesn’t really give commercial developers the warm and fuzzies.
The thing that has been holding back companies from making the investment into porting to linux is likely more related to shaky and evolving development standards rather than licensing issues. These companies are going to be looking at a long-term investment, they’re not going to want to shakeup their toolkit environment every year or so to try and hit a moving target of standards. That is why sustainability and stability are more vital than development licensing, in the big picture. The Gnome versus KDE thing will become less relevant now due to LSB 3.1 and the eventual impact of Portland, but GTK vs Qt will still be a business decision that the vendors will have to make. At least now they can make it freely based on what works best for them, which ultimately benefits everyone.
Your facts are correct, but as is often the case with facts they don’t always show the big picture. Freedom carries a price one way or the other, sometimes you know how and where that price is being paid, and sometimes you don’t. The community can look blindly and assume there will always be someone to pay the bill, commercial companies generally want a little more assurance than that.
I don’t follow. You can pay for Qt once, or you can choose to pay per year for some support, documentation, tools… etc.
While that may be $3300 and then you can’t get any updates is still more expensive and less equitable.
In comparison, I could buy Visual Studio *once* and get updates for the remainder of that product’s lifetime. So, once again, it’s still a lot more expensive.
The point is that Trolltech is selling an SDK, and the SDKs for core functionality on the Windows platform are free. Trolltech’s is not, nor is it even “low cost” by comparison.
Let’s compare, I could:
Use Windows Platform SDK, DirectX, and other SDKs from Microsoft for *years* without ever paying a penny while still getting free updates, documentation, and some tools *OR*:
I could pay Trolltech $3300 per developer, and then $1020 every year after that. Hrm, $3300 + ($1020 * 4) for four years…yes, what do you know, $7380 > $0 for four years of additional support plus the initial year of licensing.
Remember, I’m strictly comparing the cost of *development* tools, not the OS. Every user has to have an OS so that’s just an assumed cost. You can’t argue that Linux is always free either, if you want to develop certified applications for the RedHat Enterprise Linux platform (for example) you’ll have to purchase a subscription and maintain it from RedHat.
The point is that Trolltech is selling an SDK, and the SDKs for core functionality on the Windows platform are free. Trolltech’s is not, nor is it even “low cost” by comparison.
*Bashes your head of a wall*
Did you not read what was written? :-
“So Windows and OS X are given away for free? Windows and OS X do not have different business models where they are given away for nothing like desktop Linux. Charging people for development to put back into Qt that creates a desktop environment like KDE, as well as the software they’re using to increase productivity or make profit from, is a sensible idea.
Note also that any money payed to Trolltech helps to keep Qt in good condition not only for you as a commercial developer but also for KDE developers developing that software you’re using for nothing! You don’t just pay money into a Windows or Office pit – there is some serious gain here.”
Remember, I’m strictly comparing the cost of *development* tools, not the OS.
And there you fail. You need Windows, Office and other Microsoft software to develop anything for – as well as fully fledged and quality development tools. Those are things your customers will need to buy as well just to make things worse.
A free SDK means absolutely nothing to a developer who is looking to actually make money from his software.
You know, this is why this thing called open source software has supposedly caught on – Microsoft license fees for nothing *rolls eyes*.
You’re just not understanding that things here are the opposite to what they are in the Windows world, certainly when it comes to using Qt and developing for KDE.
Ahh. Another “Gnome has won this mythical corporate desktop battle that doesn’t exist amongst companies that don’t matter” article. I thought these were becoming more scarce, but apparently some people still believe this tosh. I really don’t know how this got posted really.
A good half to two-thirds of all desktop Linux users today are KDE users, and the corporate market which Novell in particular seem to think exists (at least Red Hat have said there’s no such thing) doesn’t. How many companies have you been into where you’ve seen the NLD or Sun’s Java Desktop System? None I’ll bet. The vast majority use desktop Linux through OpenSuse, Fedora, Kubuntu or just plain Debian or another distro with KDE on it. Few people actually use desktop Linux in businesses, but no one uses those corporate offerings.
Gnome uses Gnome on their servers, but the lack of good graphical tools to do just about anything tells you that they’ve made the wrong decision ;-). The development platform is everything, and if you want people to develop for desktop Linux and “pick one” like the author says, then I’m afraid there’s only one choice ;-). What are Novell doing? Who knows? Who cares? And who cares what desktop environment they use on their server OS whose market share is in freefall. Sun? Who are they and what have they done?
This is just an article written by someone who wishes things to be a certain way, but they’re just not. The reasoning, or lack of it, is the same wishy washy stuff we’ve had from the same usual suspects for years. Nothing has changed.
Gnome uses Gnome on their servers…
Hmmm. Typo. Should read “Red Hat uses Gnome on their servers…”.
There is no desktop battle.
There is no war.
Nothing of that happens.
April 1st is long over.
Why fool around?
And licensing is part of the answer of why and another reason is that the libraries were written in C which makes it easier to bind to.
That’s just plain wrong. It isn’t easy to create bindings for the entire Gnome api, and it isn’t particularly easy to do it for KDE either.
But with C++ you have more type info available and a clear cut object model, and so it is possible to automate bindings generation to a greater extent. The Qt runtime introspection works very well with language bindings, and so for instance when you debug a Ruby Korundum application in the KDevelop debugger, it uses Qt properties if available, in the Ruby inspect() methods for the classes.
I have done some extensive coding for GNOME to port several applications we used in our business away from Windows so that It would work on Linux desktops. Initially we went with Redhat (not my choice but a political decision). After some time of writing the applications we kept on getting into issues where the code was getting so big and a pain, we as developers found that we constantly hit walks with GNOME. After a group of us decided to pitch in some over time we wrote the entire application for KDE (based on some recommendations from some other application developers ) Its got more functionality, its a fraction of the size and have since moved all desktops in the company across to Kubuntu and SuSe. Overall for US in OUR situation GNOME was a horrific nightmare to develop and manage our inhouse bespoke applications. Moving forward we are putting a plan together to start making the applications web based so that we can start to hvae more offices and better scalability, less support.
Your comment is hard to believe because on average a KDE application is larger than a GNOME one. What issues did you run into when developing for GNOME? What libraries gave you problems?
Evidence? Examples?
1. Why is it that Gnome developers like to make the Gnome GUI so bland and outdated looking with it’s default color theme? I realize the GUI can be themed just as easy as on KDE but isn’t “first impressions” considered important by Gnome when marketing a product?
2. Why try to create an OSX knock off where Apple has less market share than Windows? Isn’t ease of migration a consideration by Gnome developers? One of my reasons for choosing KDE over Gnome is less time spent learning my way around the GUI due to closer resemblence to Windows which currently has a larger global market share.
3. Where are applications that can actually compete with products such as Apple’s iTunes so as to attract consumers away from Microsoft and Apple? Please don’t mention Banshee which Novell is touting as being great for the desktop as it can’t compete with what’s currently offered in Amarok http://amarok.kde.org/ Also unless I’m mistaken I haven’t been able to find anything on Gnome that can compare with software such as KPilot for syncing Palm devices like the Treo.
As it is now I just can’t understand why companies such as Novell which are trying to market their distribution to both Enterprise and Private markets take the Gnome approach. Gnome being farthest from what most consumers are used to due to Microsoft desktop monopoly over the past decade. So how does offering Gnome ease Windows users migration to Linux? After trying each release it feels more like an attempt at offering what developers want, not what they believe consumers are asking for or need. Now I’m not a developer but I do know what I want to use and that is something that will provide ease of migration either from one desktop to another or one application to another with out losing out on features I’ve come accustomed to.
Where both Gnome and KDE fail is to provide basic features that both Windows and OSX customers are used to such as VOIP with Webcam chat capability for users of GTalk, MSN Messenger, Yahoo Messenger, etc. It’s really silly for developers of products like Kopete and aMSN to provide capability for a third party MSN Messenger user to see the Linux user via webcam in the chat window but have no VOIP unless the Linux user asks the Windows user to open NetMeeting.
Until all, not just some Gnome and KDE developers learn to develope products for what consumers both need and ask for then Linux will be like a player on the side bench while Microsoft and Apple play the game. The good news is out of the two desktops I see KDE as being the successor for Linux users partly due to some of the points I’ve raised here.
Edited 2006-04-29 17:18
Heh, the naysayers said the same things about Linux on the server/data center years ago. Now they are choking on their own words.
1. Why is it that Gnome developers like to make the Gnome GUI so bland and outdated looking with it’s default color theme? I realize the GUI can be themed just as easy as on KDE but isn’t “first impressions” considered important by Gnome when marketing a product?
Good question.
3. Where are applications that can actually compete with products such as Apple’s iTunes so as to attract consumers away from Microsoft and Apple?
Bad question, More pertinent would be where is Apple’s iTunes for Linux program. (At the very least you should ask where is a non Apple windows version of iTunes.)
I don’t think most of your points really allow for the problems of reverse engineering things to fit with other company’s protocols. Surely it’s as important to ask why apple does not ship with its operating system a chat tool capable of interfacing with MSN? Or why don’t MS ship a jabber client (I could be wrong here and perhaps they do, it’s been a while since I used windows).
Your comment is hard to believe because on average a KDE application is larger than a GNOME one.
Are you sure about that? In my experience a C++ program has less line of code than an equivalent C program or rather it largely depends on the functionality you need to implement.
Do you have any statistics which show that a KDE application has on average more lines of code as GNOME applications?
Edited 2006-04-29 18:00
>>Are you sure about that? In my experience a C++
>>program has less line of code than an equivalent C
>>program or rather it largely depends on the
>>functionality you need to implement.
What were you writing? What APIs were you using?
>>Do you have any statistics which show that a KDE
>>application has on average more lines of code as
>>GNOME applications?
I was talking about binary size, not lines of code, and even then I doubt GNOME applications have anymore lines of code than KDE apps. The last time I checked Epiphany vs Konquerors, Konqueror had a lot more lines of code than Epiphany did, which debunks your LOC myth.
Edit: I forgot Konqueror is both a file manager and browser, which may be the reason it contains more lines of code. However, I still maintain the LOC argument is bogus especially since when one is writing a desktop application one will by and large be using APIs and other abstracted components.
Edited 2006-04-29 18:34
How is there a meaning way to make any such comparison. As you’ve realised, konqueror and epiphany have different features/functionality. Unless you can find two applications that implement the exact same features in the same way, any comparisons are meaningless.
As you will see from above, I suspected you were pulling this comment out of your arse. Now I know. No analysis or evidence.
I didn’t make the comparison, someone else did. I merely said the comparison was bogus. You can’t automatically construe that GNOME applications are larger than KDE applications because GNOME is written in C. That’s just BS. Most of the C++ projects I’ve looked are almost always larger than their C counterpart, especially since many of them like to abuse the OO paradigm. And OO code is almost alway more verbose than procedural code, in my experience.
Where both Gnome and KDE fail is to provide basic features that both Windows and OSX customers are used to such as VOIP with Webcam chat capability for users of GTalk, MSN Messenger, Yahoo Messenger
Interesting. I knew that MSN Messenger is begin developed by Microsoft, but I didn’t know that GTalk and Yahoo Messenger where being developed by Microsoft or Apple as well.
I always assumed GTalk to be developed by Google and Yahoo Messenger by Yahoo.
Or maybe you are wrong?
Anda_skoa,
Re: “Interesting. I knew that MSN Messenger is begin developed by Microsoft, but I didn’t know that GTalk and Yahoo Messenger where being developed by Microsoft or Apple as well.
I always assumed GTalk to be developed by Google and Yahoo Messenger by Yahoo.
Or maybe you are wrong?”
I guess I could of worded that first sentence differently in the paragraph regarding lack of features for Linux users of chat messengers. Though I thought anyone reading further in the paragraph should of understood what I was referring to. It’s true that a Linux user cannot really use both VOIP with Webcam to chat with MSN Messenger or Yahoo Messenger Windows users. As it is now Linux chat messengers such as Kopete and aMSN Messenger only provide text chat with limited support for webcam. They do not offer VOIP unless you ask the third party to open NetMeeting on Windows which defeats the whole purpose of using one chat messenger or even just claiming ease of migration to Linux from Windows. If small developers who either lack funds or experience to provide full featured messenging then it’s time for developers like Novell, Red Hat and Mandriva that have both the money and resources to do something about it. Otherwise Linux chat messengers as they are today are just sitting on the bench while other messengers on Windows and OSX play the game.
Edited 2006-04-29 18:27
I guess I could of worded that first sentence differently in the paragraph regarding lack of features for Linux users of chat messengers.
That might have helped
The way your original wording is it looks like you were thinking KDE or GNOME would be responsible for the lack of VoIP components of instant messengers while from the the list of Windows messengers only one is being developed at the desktop environment vendor and others are third party which could also deliver a Linux version if they felt like doing it.
Btw, I thought that GTalk was SIP compatible?
What were you writing? What APIs were you using?
gnome_vfs 2.0, working with MIME types. However the problem could have been that I got really confused about all the different URI datatypes because there seemed to be no overloaded functions for some of them.
I was talking about binary size, not lines of code
I assumed that you did, however I also assumed that the poster you were replying to was talking about code/project size.
I still maintain the LOC argument is bogus
I still maintain that depends on the needed functionality, since stuff like overloaded operators, compiletime type checking without macros and niceties like STL standard algorithms tend to help in quite some cases.
>>gnome_vfs 2.0, working with MIME types. However
>>the problem could have been that I got really
>>confused about all the different URI datatypes
>>because there seemed to be no overloaded functions
>>for some of them.
There is just one URI datatype, if I remember correctly, which represents a URI object(structure) that has attributes. Some functions will even allow you to pass URI strings as arguments. But GNOME-VFS has good libraries for MIME type manipulation, I wonder what problems you were having, or why you needed to write more lines of code when all you have to do is call the GNOME-VFS functions.
>>I assumed that you did, however I also assumed
>>that the poster you were replying to was talking
>>about code/project size.
Yes, that equally seems bogus to me. How does using C++ make your project size automatically smaller?
>>I still maintain that depends on the needed
>>functionality,
Agreed.
>>since stuff like overloaded operators, compiletime
>>type checking without macros and niceties like
>>STL standard algorithms tend to help in quite
>>some cases.
Perhaps, marginally, be in reality, I doubt it does much. In GNOME you use GLIB/GObject with provides nice higher level data structures, containers and algorithms that make writing desktop applications or any app for that matter easier.
However, if you can explain to me in exact legal terms how an LGPL library + a GPL library does NOT equal a GPL library
Good question, a KDE library binary shipped like this could theoretically require someone who wants to do GPL-incompatible development to replace it with their own compiled version.
However I find this not very likely, as it would imply that all public domain, BSD, X11, Apache or LGPL with exception licenced libraries would always be just LGPL licenced binaries on Linux as they all link at least with LGPL licenced glibc.
However I find this not very likely, as it would imply that all public domain, BSD, X11, Apache or LGPL with exception licenced libraries would always be just LGPL licenced binaries on Linux as they all link at least with LGPL licenced glibc.
It’s been fairly well established that FSF treats libc as “special.” Not only that, proprietary + LGPL still equals == properietary + LGPL as long as it’s dynamic linking (from what I understand).
In this case, the thing I really am wondering about is people have always said the GPL + anything == GPL (assuming you’re distributing, don’t be nitpicky folks).
I have no doubt that if someone took the *source code* to the KDE LGPL libraries, built them against their commercial version of Qt and then linked agains that there would be no problem. I’m just trying to figure out how they could dynamically link against the ones on Linux that are linked against the GPL version of Qt wihtout becoming GPL…
what an incredibly stupid title
Yep.
But it’s an amusing flamewar. Or so I think.
It’s not as amusing as the “Torvalds: ‘Use KDE'” discussion. That one was much worse than this is.
There is just one URI datatype, if I remember correctly, which represents a URI object(structure) that has attributes
Yes, sorry, I wasn’t clear enough. I mean that some functions require such an URI object while others seem to work on string URIs and I didn’t find the respective overloads that take the other type.
why you needed to write more lines of code when all you have to do is call the GNOME-VFS functions
In order to call the functions you’ll need to pass properly typed arguments and at least my C compiler seemed to be incapable of converting them automatically, very likely because there are no overloaded cast operators or automatically invokable constructors.
However I admit that this was my first attempt at GNOME VFS usage, so I might have overlooked the respective overloaded functions that convert internally or have missed some kind of C compiler switch that lets it know about inheritance and conversion functions.
It’s been fairly well established that FSF treats libc as “special.”
Well, then wxWidgets, Java AWT or SWT which have their own licences but are shipped linking with GTK+
I’m just trying to figure out how they could dynamically link against the ones on Linux that are linked against the GPL version of Qt wihtout becoming GPL…
Maybe they are just not shipping with such a KDE library binary but either include their own or do not include it at all, i.e. assuming it is installed on the target system.
Well, then wxWidgets, Java AWT or SWT which have their own licences but are shipped linking with GTK+
How is this relevant?
Maybe they are just not shipping with such a KDE library binary but either include their own or do not include it at all, i.e. assuming it is installed on the target system.
Only one problem with “assuming it is installed on the target system.” They would still have to build against the ones in the distribution in that case (more than likely anyway) to guarantee compatability possibly. It’s really a mess, and I wish someone with the authority to do so would clarify this 😐
As long as you have a billion distros with no standard desktop libraries, then Microsoft continues to win.
Microsoft must laugh as KDE and Gnome developers fragment the desktop. But that’s all historical and obviously any kind of grand unified merge of Gnome and KDE is long gone.
Oh well, maybe one day someone will come along and put a desktop operating system on top of the kernel.
Except that KDE owns Gnome totally and hard.
Who cares if they have choosed Gnome, they probably didn’t understood better, or it was that the DEVELOPERS and not COMPANIES decided what was good for KDE. Or maybe that it’s easier to get commercial closed source apps which uses GTK than QT.
Hugs for the KDE developers.
Say what you will, but I’m still going to use KDE.
Last time I check KDE pissed all over it?
It’s got commercial support because it’s LGPL and the companies can do whatever they want to it, with little disregard to the original developers.
I’ll stick with KDE thanks. You can actually *DO* stuff with it. Gnome’s gone back to being nothing more than a diluted version of KDE1 (hell, it parallels with Win 95, minus 90% of the functionality).
Give me one good reason why a company like Dell would pick Xandros over Redhat.
Xandros might actually be interested in such a deal, while Red Hat is likely not?
Name one large company that offers KDE.
Might depend on the definition of large, but if you count Novell and Red Hat as large, they do.
Novell maybe more than Red Hat, but using Red Hat at work (actually CentOS, AFIAK) I can confirm they offer KDE because that is what I get after logging in.
FACT: Developing closed source applications for the KDE platform requires a Qt license (they don’t even have to be commercial).
I guess this depends on your definition of “KDE platform”.
If you mean using the KDE application development framework, then it is correct.
If you mean developing applications to run on the KDE desktop and use KDE desktop services, then it isn’t.
However, I admit that it is a lot easier to do that when using the KDE application development framework.
Given current development efforts, doing it without using the KDE application framework will become easier with KDE4, due to goal of having shared service infrastructure with other desktop environments for improving the interoperability.
I guess this depends on your definition of “KDE platform”.
If you mean using the KDE application development framework, then it is correct.
I think that any reasonable developer would expect that to be the case. If I’m going to spend time and effort developing applications for a platform then it had better well be integrated smoothly.
If you mean developing applications to run on the KDE desktop and use KDE desktop services, then it isn’t.
Of course, and I will gladly concede that point.
Given current development efforts, doing it without using the KDE application framework will become easier with KDE4, due to goal of having shared service infrastructure with other desktop environments for improving the interoperability.
I’ll believe that when I see the level of integration they can provide. I think in the end, *if* KDE becomes the defacto platform (which it has not *yet*, nor has *GNOME* in my mind), then users will be able to tell which apps are Qt and which are not.
However, I sincerely hope that is the case. If these frameworks can help other applications be integrated to a level where users (for the most part) can’t tell a difference between a Qt app and an alternative one, then developers will finally have a true choice and it won’t matter and I will finally be happy.
Not directed at you:
I wish people would stop acting like I don’t like Qt. I DO like Qt. I’ve used Qt a few times in the past to develop applications. It’s a great library, I just want people to admit that their platform will cause developers to make a financial choice if they want to develop applications for the KDE platform. No one seems to be willing to admit this except for a few brave souls.
At the moment, due to my limited resources, if I had to choose whether to develop a small shareware application for the KDE platform or the GNOME platform, I would have to pick GNOME. Because I don’t have $3300 in upfront costs for Trolltech just to make it an “integrated” app with KDE. I would choose wxWindows or GTK.
Going even farther, if I wanted to develop that app for Windows, Linux, and OS X all at once. I would have to pay an initial license fee of $6600 to Trolltech, instead of $0 to wxWindows or the GTK project.
In fact, as many shareware developers will tell you, they would be lucky to recover that in the first year that their application is launched.
I’ll believe that when I see the level of integration they can provide.
It’s already there.
It’s a great library, I just want people to admit that their platform will cause developers to make a financial choice if they want to develop applications for the KDE platform.
People do that now in the real world. It’s not as if it’s somehow this massive obstacle to get over.
At the moment, due to my limited resources, if I had to choose whether to develop a small shareware application for the KDE platform or the GNOME platform, I would have to pick GNOME. Because I don’t have $3300 in upfront costs for Trolltech just to make it an “integrated” app with KDE. I would choose wxWindows or GTK.
Then you don’t develop software for a living.
I would have to pay an initial license fee of $6600 to Trolltech, instead of $0 to wxWindows or the GTK project.
This is just so laughable and confirms to me when people have not done development for a living.
I think it’s a dead certainty that you’re not going to spend nothing on your project for the entire duration no matter what you do. There is equipment to buy, bills to pay, salaries and wages to pay as well as othet costs. The longer you develop and the more you get bogged down in any difficulties the more these things are going to cost. Add up these additional costs versus a toolkit which will ensure your development goes smoothly, compare to the cost of Qt and see what happens.
In fact, as many shareware developers will tell you, they would be lucky to recover that in the first year that their application is launched.
Then I would say your product and market isn’t viable. In the Gnome, KDE and desktop world it’s an open source environment. You’re going to have very, very great difficulty selling shareware there even if the userbase increases to that of Windows. You’re going to need to think carefully about what you sell, how useful it is and if there’s actually a market for Windows, Linux and Mac versions.
Of course, I could completely turn your argument around and say that there is a financial implication for people to consider when buying your shareware software. I don’t like your license, and I mean, how dare you try and charge for software that will almost certainly do very, very little for me :-). How does it feel now?
I’ll believe that when I see the level of integration they can provide.
It’s already there.
It is? Proof please
You’re saying it isn’t ready, and other people say it will be ready when KDE4 gets here, and yet others say it’s a few years off. Who should I believe?
Of course, I could completely turn your argument around and say that there is a financial implication for people to consider when buying your shareware software. I don’t like your license, and I mean, how dare you try and charge for software that will almost certainly do very, very little for me :-). How does it feel now?
That argument rarely holds for games, or game tools. Which is what I would be developing if I were to develop another application for KDE/Qt. Also, you make assumptions that I “don’t develop software for a living” — which is wrong and an insult to me personally. I have been commercially developing software for the last six years.
Yes, there are costs in software development. But the point is that the lower the costs, the more I have for other resources. Would I want to spend $3300 on a Qt license for a year, or would I rather pay $3300 to an artist to make better graphics for my application?
Would I rather pay $3300 for web hosting, equipment, and miscallenous items, or spend it on a toolkit that I shouldn’t have to pay for?
The argument that Qt would give me a better development experience is something I won’t argue with having used it. *However* I don’t like the fact that I’m *not given the choice* of whether or not I use it. I’m *forced* to use it if I want to develop integrated applications for the KDE platform.
Then I would say your product and market isn’t viable.
Then I have trouble beliving that you have developed or know anyone who has developed a lot of small shareware applications
I’m not saying that you don’t, it’s just in my experience, as well as others I know, it sometimes takes a few years before an application really takes off. Or, like some developers, you rely on a suite of small applications to provide a larger level of income.
You’re going to have very, very great difficulty selling shareware there even if the userbase increases to that of Windows. You’re going to need to think carefully about what you sell, how useful it is and if there’s actually a market for Windows, Linux and Mac versions.
Again with the non-relevant details. It really doesn’t matter what your philosophy on development, licensing, or software is. What matters is this:
KDE’s adoption of Qt as their primary platform SDK causes closed-source (not even commercial!) developers to have to make a financial committment to a single entity (Trolltech) to thave the right to develop truly integrated applications for the KDE platform. KDE is therefore also encouraging a monopoly by ensuring that only one vendor can be the commercial licensor of software for the KDE platform.
The operating systems that run on 90%+ of the desktop computers in the world (according to recent public estimates seen on this website) do not require developers to license the platform SDK to develop applications.
It is? Proof please
I suggest you find out how KDE actually works before making silly comments. You don’t get out of it by demanding me to provide you with evidence that you should have found out about before commenting.
For one, KDE actually has a common base (kdelibs and kdebase) that actually ensures that all KDE applications look the same (not superficially through a HIG) and that they all sing from the same hymn sheet technologically. The KDE environment has central DCOP interfaces that all applications use, applications can use DCOP interfaces themselves for others to use, and KParts ensures that the same component can be embedded in multiple applications easily. Look at KOffice or Kontact for good examples of that. Integration. The infrastructure is there and is only going to get better. That’s why I make a big thing of a desktop being a developer’s platform and being made of the right stuff.
But you’d know that, right?
That argument rarely holds for games, or game tools.
Developing games is an expensive business and you’ll need some top notch development tools if you want to go there. Guess what? Games companies pay through the nose for them.
Yes, there are costs in software development. But the point is that the lower the costs, the more I have for other resources.
The other costs are substantially more than the costs of any development licenses.
Would I rather pay $3300 for web hosting, equipment, and miscallenous items, or spend it on a toolkit that I shouldn’t have to pay for?
You mean you wouldn’t rather spend money on the tools that will be used to develop the software that is going to be the absolute focus of your whole business? Especially considering that this investment will be a small proportion of your total costs? Remind me not to use your software.
Again with the non-relevant details.
It was actually very relevant to the drivel about shareware you came up with.
KDE is therefore also encouraging a monopoly by ensuring that only one vendor can be the commercial licensor of software for the KDE platform.
Wrong. That’s why we have this thing called open source software. People can see how KDE works, and people can see how Qt works, and if demand for KDE development increased you’d see an awful lot of open source alternatives. You can already have GTK take on the look and feel of a KDE environment through QtGTK, and further unlimited integration of GTK with KDE would be more than possible. Of course, many people want to pretend like hell that this isn’t possible ;-).
Trolltech would have a decent circular relationship with the KDE community (which Trolltech doesn’t own) where license fees would go into maintaining Qt so that KDE developers can create a top-notch desktop and applications which would be a platform for people to buy Qt….and so on and so forth. This is the only way desktop Linux is going to work on a wider scale and have a sensible, working model.
However, a monopoly? No. That’s Microsoft, remember?
The operating systems that run on 90%+ of the desktop computers in the world (according to recent public estimates seen on this website) do not require developers to license the platform SDK to develop applications.
The company that develops the operating system that runs on 90%+ of the desktops in the world does not invest in a desktop and applications given away for free for everyone to use. They make money from applications that do little appreciable for people whilst locking them in. Like I said, that was the point of open source software, right? Even then, as already pointed out the Microsoft’s SDKs by themselves are next to useless. You’re going to need Visual Studio or similar at some stage or another. Microsoft do not give away SDKs out of the goodness of their hearts.
You’re desperately trying to have this all ways, but you just can’t.
If I’m going to spend time and effort developing applications for a platform then it had better well be integrated smoothly.
Definitely, however integrating does imply using the native API. An application on Windows can integrate nicely without having to touch WinAPI, for example by using wxWidgets or GTK+ or Qt, which in turn access the Windows services.
While there are of course some parts of KDE that would require linking kdelibs, e.g. some widgets, a lot (maybe most) KDE services are out-of-process already by design, for example KIO slaves, everything that uses DCOP for interaction, etc.
KDE is, like other desktop environments, a name for a collection of things: the desktop environment, applications and a development framework.
KDE developers will obviously suggest to use the framework if you want to create applications for the desktop environment, but it isn’t required.
At the moment, due to my limited resources, if I had to choose whether to develop a small shareware application for the KDE platform or the GNOME platform, I would have to pick GNOME.
Actually I think it will be more likely that you’ll end up doing what all other ISVs do as well, not pick a desktop environment at all and maybe integration through plugins or an integration project like then Portland project.
And, let me be upfront, I don’t believe anyone is going to do shareware on Linux. Either sell full/complex applications or you will be confronted with a free software competitor sooner than than you’d like.
Anyway, Trolltech is known to have negotiated delayed payment terms for startups.
Anyway, assuming I would be developing a close source program, for example because that is required by some contract like a NDA, but doe not plan to have it suitably priced, I’d say using wxWidgets will be a very wise choice, since it is available on many platforms and might even become available in editions with Linux desktop integration, e.g wxGNOME or wxKDE.
Actually I think it will be more likely that you’ll end up doing what all other ISVs do as well, not pick a desktop environment at all and maybe integration through plugins or an integration project like then Portland project.
Yes, I know about the Portland Project. However, until it is reality, it is just a nice dream. A really nice one that I hope comes true, but we’ll have to wait and see.
And, let me be upfront, I don’t believe anyone is going to do shareware on Linux. Either sell full/complex applications or you will be confronted with a free software competitor sooner than than you’d like.
There are shareware titles out there for Linux, just like this one:
http://www.jollygoodgames.com/turbosliders/linux.htm
Games are an area where I think free software will never compete successfully. Most of the ones that manage to, only do so because they can use resources from commercial titles.
( wikipedia shareware definition page: http://en.wikipedia.org/wiki/Shareware )
Not only that, I think most people would agree that some software is so good that free software will likely never be at the same level. The argument of whether a particular product is viable for the market is up to the developer, it shouldn’t be up to the platform. If I want to try wasting my resources into oblivion, I should be free to do so
Anyway, assuming I would be developing a close source program, for example because that is required by some contract like a NDA, but doe not plan to have it suitably priced, I’d say using wxWidgets will be a very wise choice, since it is available on many platforms and might even become available in editions with Linux desktop integration, e.g wxGNOME or wxKDE.
Yes, and that is what is unfortunate. For example, one of the things that I maintain a Linux port of is a game runtime engine. It is not open source, but it has always been zero cost to users. The reasons for it not being open source are many, and I don’t own the code anyway. I’m under NDA with the author to maintain the Linux port. It has an editor, which is thankfully written using wxWidgets.
I guess the real conclusion is that I can only hope that the Portland Project really does provide a new level of integration for non-Qt applications. With that in place, *some* commercial and non-commercial closed source developers would finally have a viable option for developing for the KDE platform.
Yes, I know about the Portland Project. However, until it is reality, it is just a nice dream. A really nice one that I hope comes true, but we’ll have to wait and see.
The Portland project is mostly about offering integration without knowing which desktop one is integrating with.
My point was that no ISV is going to choose any desktop environment to directly integrate to, not even one they might think to be free of cost.
And I also mentioned Portland because it shows that in order to integrate it is not necessary to link to the desktop’s base libraries.
Even without additonal integration options offered by Portland’s there are, especially in the case of KDE, a lot of options not requiring linking KDE or Qt since KDE itself uses a lot of out-of-process services. (I have listed examples before IIRC)
Games are an area where I think free software will never compete successfully.
I am not experienced in this area, so I am quite surprised. Does it make Quake a worse game the moment its engine gets open sourced by Id Software?
Isn’t the idea, the gameplay, levels and artwork what really makes a game?
The argument of whether a particular product is viable for the market is up to the developer
Yes, definitely! I didn’t mean to suggest it wasn’t an option, I just wondered if it could ever be a viable one.
I guess the real conclusion is that I can only hope that the Portland Project really does provide a new level of integration for non-Qt applications.
Portland project ist not about providing a new level of integration for non-Qt applications, it will also provide a new level of integration for Qt applications, or rather provide a new level of intergration for all non-desktop specific applications (or in other words toolkit-only applications).
Heck, in the long run it might even improve integration of KDE applications running in GNOME and GNOME applications running in KDE.
With that in place, *some* commercial and non-commercial closed source developers would finally have a viable option for developing for the KDE platform.
*some* already have, *some* others would have but do either not know about it (a lot of KDE related technology is not dependent on Qt but got the stigma of being it) or chose to ignore their options and *some* others might not want to integrate anyway.
Anyway, thanks for the game link, I think I will test it
> Are you really trying to say that GNOME
> doesn’t have applications? Are you nuts.
> There are more GNOME/GTK apps out there
> than KDE/QT.
I don’t want to disturb the flamewar you guys are having here. But would you mind to share the statistics that you draw that conclusion from? Because judging from some more sophisticated research that takes into account several sources and indicators I can’t help to doubt your claim — although coming up with really objective numbers is really hard and can’t be accomplished in a satisfying way.
Just a very plain and simple example:
The most popular software repositories for KDE and Gnome are probably KDE-Apps.org and GnomeFiles.org.
Both repositories seem to be equally maintained and both of them share the aim of delivering a comprehensive database of applications for the respective toolkit. So one could expect that both sites are doing equally well (especially given that GnomeFiles.org is “attached” to a major news site which should favour GnomeFiles.org quite positively)
If you go through the last month’s entries you might realize that the ratio of established quality software vs. “v.0.1 / Improvement / script / addon” – software that is in early stages of its development is quite comparable. Same goes for double entries, unmaintained and outdated software.
Still KDE-Apps lists 1952 applications while Gnomefiles lists about 1194. On KDE-Apps in the last week there were 53 new software releases while GnomeFiles had 43. And this ratio and trend is quite consistent with other site data I have seen.
For proprietary and in-house applications it’s of course much harder to guess and much more so a subject to speculation. Given that Trolltech targets mostly the latter target group to make money I doubt that the comparison would be less favourable in that segment though.