DEC
31
2000

Do Interoperability Technologies Help or Hurt KDE Development?

In a recent story at Kuro5hin.org, the author considers the implications of some recent developments announced on the dot -- namely, XParts and QGtkWidget. These projects permit non-KDE programs -- particularly Gtk applications -- to be used within KDE applications. The author opines that these projects will end up hurting KDE development. In particular, the author wonders why a Linux developer would now develop for KDE when they can write for Gtk and have the application work both under KDE and GNOME. What do you think, will these projects encourage Gtk development at the expense of KDE/Qt development? Even if so, does it nevertheless make KDE stronger as a competitor against proprietary desktops? Is it important for KDE developers to create projects that would make KDE apps work inside GNOME apps (such as a KParts container that works with Bonobo) or with Gtk widgets (essentially a GtkQtWidget)?

Comments

Looks like this is starting to be a flamewar about Gnome again.
Everybody is whining about how KDE/QT is superior and how commercial and bad Gnome is (while it is not).
Get a life will ya! You don't even know what's really going on!


By Anonymous at Mon, 2001/01/01 - 6:00am

It may hurt commercially, but who cares?
KDE will live on, and so will Gnome!
We aren't power-hungry marketing-weenies, trying to pull as much users/developers to our side, are we?
Both Gnome and KDE's goals are to create the best desktop and development environment.


By Anonymous at Mon, 2001/01/01 - 6:00am

PLEEEAAAASE don't Slashdot this article!!!!
There's already enough flamewars here and I don't want more!
Especially because /. is full of *ignorant* trolls!


By Anonymous at Mon, 2001/01/01 - 6:00am

Please behave yourself too ;-)


By GlowStars at Mon, 2001/01/01 - 6:00am

I'm surprised (and kind of disappointed) that dot.kde.org picked up this article. Anyway...

Personally I think it's good for KDE. But that's not really the point. This is Open Source, ladies and gentlemens. If we take QGtkWidget for example: There wasn't some KDE focus group somewhere that sat down and discussed "hmm, KDE really needs the possibility to include gtk widgets". No, a guy (in this case bero at redhat) wanted to create QGtkWidget for whatever reason, and so he did. Fine. That's the nature of Open Source -- if you want to create something for it, feel free.


By matt at Mon, 2001/01/01 - 6:00am

Amen to that.


By caatje at Tue, 2001/01/02 - 6:00am

I believe it's good to have interoperability.
Users should be free about which desktop to
use and programmers should be free about which
desktop architecture they prefer to program for.

Personally, I think most opensource developers
"who do it for the hobby" will never choose to
develop on an architecture that they don't feel
comfortable with or isn't their taste. So if a
developer decides to program primarily for KDE
or primarily for Gnome, you can't stop him / her.

What you can do for the users is to make Gnome
programs run well on KDE and make KDE programs
run well on Gnome, so the users have more good
applications to choose from on both desktops.

Best regards,

Eric


By Eric Veltman at Mon, 2001/01/01 - 6:00am

Since most Gnome and KDE apps are still developed by hackers in their sparetime, XParts and QGtkWidget wont hurt KDE, as nothing in the hackers eyes have changed. The reason why people code for KDE is because they like it. They don't care about market share and all that crap. Coding for Qt/KDE is still easier and more fun than coding for Gtk/Gnome and that's what matters.

All in all this mostly good news for KDE not for Gnome. This gives more options and possibilities for Qt/KDE developers as well as KDE users.


By Erik Engheim at Mon, 2001/01/01 - 6:00am

"Do Interoperability Technologies Help or Hurt KDE Development?"
It HELPS, of course!

This is in fact what we can consider as a MAJOR FEATURE of KDE environment.

Just have a look at XPart example: this does not only mean your are able to play with the HTML
engine your want. This means without recompiling Konqui, you are able to upgrade its web browser
functionaltity.
So what? So if KHTML had a deep problem (let's say is no more upgraded, or have a
deep architectural defficiency), Konqui (and all KDE apps using KHTML) are not dead anymore because
it's easy for them to replace KHTML by any other Gnome component.

More globaly, this means Interoperability Technologies is very important for KDE, not from a
component but more from an application point of view. Those who are developping applications would
find KDE a better environment because if needed, it would be possible for them to easily use,
replace or upgrade each component they use by any other better KDE or Gnome component offering the
same functionnality.

Happy new year,

Nicolas


By Nicolas at Mon, 2001/01/01 - 6:00am

The same problem happened about OS2 wich was able to use windows 3.1 programs, resulting in more programs and users for windows due to the fact that OS2 users would anyway be able to use windows user's work


By renaud at Mon, 2001/01/01 - 6:00am

That is completely different! This is the open source world!
KDE apps runs fine in Gnome and vice versa.
We're talking about some widgets here.

FreeBSD is able to run Linux binaries.
Then why isn't FreeBSD dead yet?
Think about it.


By ac at Mon, 2001/01/01 - 6:00am

The real problem about OS2 was its lack of win95-compatibility. We had to use the new versions of the standard-software-packages (like Corel Draw and Office). Many users prefered the 0S/2-Desktop, but they were forced to use Windows95 because of OS/2s missing compatibility.


By wmueskens at Thu, 2001/01/04 - 6:00am

This is really stupid of you to have hatred toward other programmers work! How could you people be so cruel? Instead of helping each other we tend to fight...why?

Remember "Cooperate We Survive, Fight we Die!".

Hatred is not the Answer...


By Asif Ali Rizwaan at Mon, 2001/01/01 - 6:00am

Hey Folks, do you relly think that this PR Gag about Bonobo-components used in KDE-apps has some relevance?
Does anybody know which features can be used at all? How do you thing Bonobo-Components can be used in KDE-apps at all.

How would you for instance use an evolution-calendar component in a KDE app which doesn't know about the evolution-API? Or a red carped-component?
Does anybody of you have the slightest idea of what componentisation is about and how it is used? How can you think that GNOME-components can really be used in a KDE-app?

I think you should all go home and do your homework. Read some bonobo-code etc... Making a quick hack to embed a bonobo-component indo a KDE app and use it as PR is one thing. Actually use a bonobo-component is another.

Wake up!!!!!

Jacob R. Kreutzfeld


By Jacob R. Kreutzfeld at Mon, 2001/01/01 - 6:00am

They wrap a Bonobo component using an XPart "component" with it's own API.


By TrollKiller at Mon, 2001/01/01 - 6:00am

I am really tired of this media attitude that GNOME and KDE are competing! Listen up : KDE AND GNOME DO NOT COMPETE! The whole concept of trying to get people to use KDE instead of GNOME is stupid in this sort of system. Why? Because we have the same goal! Our goal is : "Make Open Source Software userfriendly and powerful". The only reason there are 2 projects in the first place is partially because of liscensing problems, but mostly because of programmer preference! Interoperability technologies help because they encourage more programmers and users to use Open Source! Also, these sort of interoperability technologies might eventually lead to a project merge, which would be great for everyone! Great for the users because now they have twice the developers working hard on their software, and great for the developers because they can use either of 2 APIs and still have a result thats compatible with developers who prefer the other API! I fail to see any sort of downside here, other then hightened memory and CPU requirements.


By Carbon at Mon, 2001/01/01 - 6:00am

GNOME is a comercial desktop, even though it's
under the GPL, the main GNOME mantainer own's
a company that sells GNOME. He's more interested
in money than GNOME, not only that, GNOME is
inferior and slow, the only reason people use
it is because of RedHat, they want people to
use GNOME so that they don't know how to use
KDE, and will continue to buy RedHat with GNOME. This is why so many big companies chose GNOME, because they know they can influnce development with money.
It wouldn't suprise me if in 2 years GNOME
only runs on RedHat and Solaris. This would kill
linux also - GNOME is a piece of crap is completely unusable and is horribly designed, probably beyond repair. RedHat and Solaris dont care because they're pushing for server market share. I wouldn't be suprised if RedHat chose GNOME to push linux off of the desktop and get more opensource developers working on server aspects of linux. This won't just hurt linux it'll eventaully kill it off entirely. Linux
is about free software, and the GNOME people are trying to force people to buy it. The only way
to get people to use linux is to kill GNOME off entirely, it's what's kept it unusable for so long. This is KDE vrs GNOME because GNOME will eventaully destroy linux if noone does anything. RedHat and the other companies don't care as long as they can get a few $s out of it.


By _ at Tue, 2001/01/02 - 6:00am

Dang! Seeing posts like this on /. is rare. But you, you are really funny! I don't know what you smoked, but you did make me laugh! Thanks!


By Anonymous at Tue, 2001/01/02 - 6:00am

Unlike the companies involved with KDE, the main GNOME companies are committed to only releasing free software. It was only the growing success of GNOME that forced Trolltech to first open source the unix version of QT, and then to GPL it. The Kompany, Caldera, etc sell proprietary software. Not that there is anything necessarily wrong with this, but I find the attitude of the GNOME companies much more admirable (and certainly much more advantageous to the free software movement.) In fact, the GNOME project as a whole seems much more idealistic than that of KDE. Certainly, in the past, they've taken a stand on principles, rather than doing that which is merely convenient. This seems to be a consistent differance between the two projects.


By Nexus at Tue, 2001/01/02 - 6:00am

Have you looked at the business model that Helixcode and Eazel have? I have, and I find it amazing that at least Helixcode is projecting $120 Million in revenue per year starting this year based on subscription services. Advertising has to be part of this model, how many people have to be paying them $5 per month to make that kind of money? At least with my company we are doing valuable open source work that everyone can take advantage of, we have a model for making money that we are clear about and we do our best to work with the community to make sure that what we are doing is acceptable. What piece of proprietary software do we, or Caldera sell? As far as I know Caldera doesn't, and neither do we. I love people talking about theKompany, but I do wish you would have your facts straight. If you have a question about what we are doing, you are always free to ask, I respond to every piece of email I receive.


By Shawn Gordon at Tue, 2001/01/02 - 6:00am

Caldera recently bought out SCO's unix business. This is what they said about why they wouldn't open source it: "Ownership [of code] is not a bad thing; it's actually a good thing," said Love. "It protects [code] quality."
Compare this with Red Hat's acquisitions.(Cygnus etc) "Every line of code that we write, we put back as open-source software," McNamara (general manager of Red Hat's enterprise business unit) said. "We believe the operating system should be part of the public infrastructure and not the account of one company." As for theKompany, you are keeping most of the stencils for Kivio proprietary, and from your past criticism of the Eazel/Helix business model I assumed you felt it necessary to keep some things proprietry to be successful. My comments weren't, however, intended as a criticism of theKompany, merely as a rebuttal of the absurd statement that the GNOME people were only in it for the money.
The business model used by Helix/Eazel is still unproven; it may not be successful, though I hope it will be. Still, it's a noble effort.

Of course, as a company heavily involved in free software, I wish you success, even if one day you may have to change the fourth letter of your company's name. ;)


By Nexus at Wed, 2001/01/03 - 6:00am

Just came back from Slashdot?

And how do you know Miguel is more interested in making money?
Do you know him personally?


By TrollKiller at Tue, 2001/01/02 - 6:00am

Boihh, how does it feel to be controlled by paranoia? I knew that smoking crack is bad for you mental state, but I didn't expect to be that terrible.
Poor guy, I'm affraid there is no help for you anymore, so go and bang your head.

Jacob R. Kreutzfeld


By Jacob R. Kreutzfeld at Tue, 2001/01/02 - 6:00am

But what about Motif. It´s possible to use this on most linux computers. Perhaps that is the best alternative.
I have just downloaded it myself and will soon find out if it´s good for the average homeuser.

regards!


By roger at Fri, 2001/01/12 - 6:00am

On on Sunday December 31, @11:46AM, Dre asked:

> In a recent story at Kuro5hin.org, the author considers the implications of some recent
> developments announced on the dot -- namely, XParts and QGtkWidget. These projects
> permit non-KDE programs -- particularly Gtk applications -- to be used within KDE
> applications. The author opines that these projects will end up hurting KDE development.

In my opinion the only way this would hurt KDE development if it included the need to make the code structure worse - something I cannot see at all.

The only way to harm the KDE project is making KDE worse, not making KDE even more open-minded and able to cooperate with other projects.

> In particular, the author wonders why a Linux developer would now develop for KDE when they
> can write for Gtk and have the application work both under KDE and GNOME.

Shouldn't this be the decision of the developer herself/himself?

There is no way to prevent a developer from choosing FLTK or GNOME or whatever they like - and I am glad about this!

Anybody contributing to Free Software should have the right to choose his/her favorite toolkit/desktop environment/window manager... right?

If the chance to use Bonobo components from insite KDE results in a developer deciding to code Bonobo components now, what it wrong then?

There is an increasing number of some hundreds HIGHLY motivated developers loving Qt and KDE - continuously contributing many good ideas and a lot of very good code.
I see no desperate need to prevent anybody from prefering another project: KDE is by far the #1 desktop today and continues to find skilled friends throughout the world, more and more and more.
No need to handcuff somebody who might be attracted by 'the others'.

> What do you think, will these projects encourage Gtk development at the expense of KDE/Qt development?

Perhaps a few people, perhaps. Perhaps nobody at all, perhaps...

Please consider that for somebody being attracted by the idea of converting from Qt and KDE development to GNOME development there must be meet one very important requirement:

A skilled software developer would only decide to start GTK development and forget about Qt in case she or her is convinced to like GTK more than Qt.

:-D

Considering this simple question those of you knowing both toolkits (as I do) will surely admit there really is no need for any handcuffs. ;)

> Even if so, does it nevertheless make KDE stronger as a competitor against proprietary
> desktops?

KDE is that much appreciated because of its superior quality, its many skilled contributors, its great developer-documentation, easy to use toolkit...

Being more open will be another advantage for KDE: since there is no magic in the way Bonobo components can be plugged now. Everything is open and IMO very fair. This cannot be compared to idiotic ways a 'pseudo marked-leader' in the proprietary software world would act - nothing about hiding interfaces or surpressing open standards - the contrary is true: KDEs new ability supports interoperation at its best and both the GNOME and the KDE team should be happy about this.

Karl-Heinz   Keep coding...


By Karl-Heinz Zimmer at Tue, 2001/01/02 - 6:00am

Hi there,

I think interoperability is one step in the right direction. I really dislike the gnome/KDE flame wars. Personally I believe that the two projects joining their powers and developing together the *perfect* environment would be best but I guess the barriers for that are to big. Either side wouldn't see enough of their code and development in the combined thing. Plus competition might not be that bad, since both sides put a lot of effort in the development to try to be better than the other.

Personally I prefere KDE2 over gnome. But I guess that's just because when I compared the two gnome was in a very early stage and just looking like a childrens toy with huge icons. This is not the case anymore I know but I got used to KDE and thats the main point. It's the same with distributions. In my early days I tried Slackware, Debian, DLD, Suse and Redhat not in that particular order but RedHat was the last and because distributions got better every increase in version number it would be unfair to compare an older distribution with another newer. Since some years I only use RedHat because I got to know their setup and got used to it. I think that's all there is. You get used to something get to like it, get to know it and then you just don't want to change.

I admit, I've never written a line of GTK/Gnome code but I started to develop with QT and KDE just because it's KDE I'm using. Both QT and KDE-Libs are easy to learn and extremely powerfull. In addition to that they are very good looking (my oppinion).

Interoperability would bring more apps to each of the both sides as long as either side would provide such a support. This can only be a good thing. Not perhaps for one side or the other but for Linux as a whole. People will have the ability to choose and get all the apps of both sides. Developpers can chose the environment they like and provide apps for both sides. Since I believe Linux will never be theoperating system for the dumb user who isn't interested in what's going on within the system you will always have a userbase that is at least a little skilled and at least as skilled enough to be able to use both app-types at the same time. And if you had let's say drag and drop or other things working cross environment wouldn't this be just very nice?

So putting it all in one phrase, I'd say go interoperability!

Greetings and happy new year,

Richard


By Richard Stevens at Tue, 2001/01/02 - 6:00am

Use the better KDE GUI and run that Gnome application you must use.

Wouldn't you love to use Linux, but run Microsoft Word or Excel ??? Same thing.


By KDE2 Man at Tue, 2001/01/02 - 6:00am

...or use the better Gnome GUI and run KDE programs you must use.


By ac at Wed, 2001/01/03 - 6:00am

ac wrote:

> ...or use the better Gnome GUI and run KDE programs you must use.

Dear aonymous, I got an even better idea:

Why not send all trolls to jail and continue coding in peace?

Karl-Heinz


By Karl-Heinz Zimmer at Wed, 2001/01/03 - 6:00am

Do Interoperability Technologies Help or Hurt KDE use? would be a much better question, with a simpler answer: They help!

Come on people, what do you want? To rule the world? Yeah, that would be nice, but I just want good apps as a user and a possibilities as a developer and interoperability just helps on both counts.

Interoperability is a Good Thing.


By Luis Coelho at Wed, 2001/01/03 - 6:00am

Exactly!
I don't even think, that it would be "nice" to rule the world. Who needs this? The developers, to be proud? This could be the case for some individuals. But from a users POV? No...
Why do i care about what other people are using??
I want the best thing for ME. I used to love KDE but recently i switched to IceWM and fall in love with it. Than i had to realise, how stupid it is to use Gnome or KDE apps without having one of the desktops running. So i avoided them as much as possible. Why should i use kcalc, if i can use xcalc? Well, kcalc looks better, but wouldn't a Qt calc or a Gtk calc look better two? I can see that newbies could like to have all their apps look the same way, that's why i think it is a good idea to make KDE Frontends for apps. But i don't like that a lot of KDE applications are reinventing the wheel, just to have something less functional, bigger, slower, desktop dependent.
That's why i support anything that's not Gnome nor KDE dependent. Qt is a great Toolkit, so why are there almost no Qt apps but a whole bunch of KDE apps? I honestly think that recreating all important apps for a desktop environment is the wrong way. They should rather try to cover existing applications. Imagine a simple Qt calculator. If this Qt calculator would now use a Qt RC file to set a unique theme for all Qt applications (this could be configurable in kontrol-center), what would be the disadvantage of this calculator to the KDE calculator?
Of course, there are a lot of cases where the KDE libraries are usefull (KOffice for example).
So better embedding of other applications into the KDE desktop would be a great thing. I would be glad if everyone would help to make alle applications as independent as possible, so that we can remain the power of choice.
I still feel bad about kparts and bonobo. Wasn't it possible to stick together and write something desktop independent?
Another thing is the KDE system tray. It makes KDE applications feel really bad in other windowmanagers if they use the system tray! But this is not only KDE's fault, others are doing the same (gnome...). I can't understand it. Of course it's more trouble to create something compatible, a kind of standard, so that application programmers can support ANY system tray, not only the one of KDE or gnome, but wouldn't it be worse it?
The KDE2 system tray isn't even compatible with KDE1 apps. D'OH!

This wasn't intended to be a flame, i just hope that some people will keep this in mind when they develop new concepts and that hopefully everyone stays away from thoughts like "how can we force the people to use our software" as the parent article implied...

Spark

Disclaimer: I feel that my english wasn't the best and i had to guess a few times. I hope everything was understandable. Well.. and i always forgot to write capital I instead of i. ;)


By Spark at Thu, 2001/01/04 - 6:00am

Wow! You're the 1st person who consider KDE apps "bloat"!


By ac at Thu, 2001/01/04 - 6:00am

But it's true. Especially if you don't use the KDE Environment. They are very well designed and most of them make sense. But the KDE libs are still bloat for a calculator. ;) Qt itself does almost the same, but without the overhead.


By Spark at Thu, 2001/01/04 - 6:00am

Believe it or not, but I also switch from KDE to IceWM. For some reason I couldn't get KDE to compile, IceWM on the other hand was up and running in no time. Also, for the very seem reason as mention above, I don't use anything that is desktop depenent, it just doesn't look as good without the desktop environment.


By Dave Kok at Fri, 2005/07/15 - 5:00am

I hope you are aware that 1) KDE is a *D*esktop *E*nvironment while IceWM is a *W*indow *M*anager (KDE's window manager is Kwin) and 2) you are replying to a four years old flaimbait.


By ac at Sat, 2005/07/16 - 5:00am

.. KDE doesn't really want to crush Gnome, but take market share from Windows/Mac.
What is the point of fighting over a 3-5% market share, when you can go after the other 95-97%?

Both KDE and Gnome should have the policy of helping interoperability(both ways).
To be frank, I use both KDE and Gnome, and I really don't care which ends up bigger, as long as they together have a big impact in the market.

If KDEs goal was to crush Gnome, I would stop using KDE.. fortunately I think both camps think bigger (and more noble).

A merging of the two projects are close to impossible, but interoperability is possible, and should not be hurt by fear of loosing the battle.


By Gaute Lindkvist at Thu, 2001/01/04 - 6:00am

Why does this war exists? I'm tired of listenning KDE fans saying KDE/Qt programming is easier, KDE2 is faster, Konqueror is the best browser for Linux, ... while on the other side GNOME fans say GTK/Gnome/ libs are more powerfull, GNOME is faster, Nautilus and Evolution are better and so on. I think programs from one side sould run on the other side so that when people could benefit from both projects. The ideal was that both projects could merge into one and end this wasting of resources in competition between them after all both share the same enemy M$


By Paulo Jorge at Thu, 2001/01/04 - 6:00am

I think this thread does more damage.

Face it, kde isn't something you can or should personify.

It doesn't have opinions, it can't be harmed.

Its nothing "official" nor important.

Like gnome it is just a pile of source code.

On the other hand, the people using, developing, promoting and debating it are (on the whole) the ones who have opinions and who may consider themselves affected by what happens to kde.

Perhaps the question they really meant is "will it harm me?" because it certainly won't harm kde whether you embed gtk or not. Maybe you should question their fears?


By Michael at Fri, 2001/01/05 - 6:00am

The replies to this can be simply summarized: way too many people have their political, ethical, or other senses bothered by the fact that somebody sat down and instead of doing mouthwork did codework.

Creativity is about being free to express yourself.

Open Source and Free Software is about being free to use whatever you want and need to accomplish your task.

Freedom of Speech is about being free to say whatever hell you want to.

Common Sense is about using the freedom productively. Get a cold shower, pals.

If discussions like this will be dominated by the people who don't have a clue what they're talking about, prospective open-source developers will eventually get scared off... I'm starting to get scared-off, as nobody knows what kind of the war the code I'll develop may start.


By Kuba Ober at Mon, 2001/01/08 - 6:00am

Pages