New partnerships with server makers Dell and Egenera will be among the displays of Novell’s newly bought Linux status at the LinuxWorld Conference and Expo beginning Wednesday. Novell on Jan. 21 will ship its exteNd 5 suite for development of service-oriented Web applications, featuring boosts in security, productivity and Linux platform support.
The Open Source Development Labs plans to unveil at the show a new multicompany effort to improve desktop Linux technology.
Woh! This is the first time I’ve heard about that. I wonder what they’re up to and if the OSDL decides to choose one DE over another.
I’ve said it before and I’ll say it again. KDE’s technology is so much better than Gnomes, but the QT license is an issue. It’s not just the money either. Companies don’t want to be tied down to a small company in Norway for a toolkit if they’re going to start basing their entire product line around it. It has to be the license, because both QT and KDE are superior (technologically) to Gtk+ and Gnome but corporations keep on choosing Gnome for the desktop.
To me, Gnome is just a bunch of random apps that happen to use the same toolkit and don’t have any integration into the surrounding DE. Does anybody even use Bonobo anymore and don’t most of the hardcore Gnome developers consider it a failure? It’ll be interesting to see what Novell has up their sleeves.
IMO, one of the DEs has to become the dominant one. Don’t get your panties in a bundle. That doesn’t mean there won’t be choice. Obviously, you will always be able to run fluxbox or whatever, but ISVs need a standard toolkit to run. The funny thing is, I think KDE is a much better desktop than Gnome, but if I have my choice in which to write code on linux, I’d choose mono in a heartbeat which is tied with Gnome Gtk+. Oh well, if that means I run a KDE desktop with some Gnome apps(maybe with some of those new event hooks for gtk+ apps in KDE), then so be it. It’s not going to kill me if it’s ugly.
> … Novell’s move puts SuSE on a better footing in comparison to Red Hat. “We think this is a very healthy thing, because it provides a good balance between the two,” said Scott Handy, vice president for IBM’s Linux strategy and market development.
Very sensible.
> And at the Linux show, Microsoft’s booth will once again feature the company’s Servers for Unix software, now free.
Good to see these guys being forced to interoperate. They owe credit to OpenBSD, of course, but what the heck.
> The Open Source Development Labs plans to unveil at the show a new multicompany effort to improve desktop Linux technology.
More and more interesting.
> Red Hat likely will show off its application server software, which it began testing late in 2003.
Eagerly awaiting this one. Will Sun certify them as J2EE compliant? So Redhat’s now going to provide the entire software stack.
> Handy now believes business issues such as partnerships and software support have replaced technological obstacles as the primary Linux challenge.
Amen, brother.
Although the article has nothing to do with KDE vs. GNOME, i too would like to post here OT.
(what follows is the rantings of a incurable GNOME freak)
As much as I love the functionality of KDE and certain KDE apps I abhore the user interface of KDE. I think I must be brain damaged, but no matter what options I tweak, no matter which configuration I choose, I always find myself fleeing from KDE within an hour of having launched it. I know it is totally irrational. For some reason the “whole”, as in “the whole is greater than the sum of it’s parts”, of KDE just turns me off big time.
The reason this bothers me is that certain KDE apps like K3B, or Konqueror-as a file manager,are simply the best of their class for Linux period. I could go on for days talking about the wonders of K3B-that app is simply amazing-simple to use and unbelievably powerful- makes anything in the windows world look like a joke. The same goes for Konqueror as a file manager. As a web browser it is *ok*, but I will never prefer it over Gecko. Nautilus is light year away from the functionality of Konqueror.
KIOslaves are incredible and I wish that every Xwindows app could tie into it. I understand that KDE is like LEGO on steroids- it’s all about parts and and the myriad of ways in which these parts can be integrated. Konqueror is just a wrapper for some large number of parts. I guess m problem with KDE is that every KDE app is an implementation of the same parts which every other KDE app uses. Of course this is it’s strength, it’s integration, it’s consistency etc.
But it is boring at least as far as regards my irrational aesthetic tastes. I have never felt home in KDE, and I trully regret this and my writing this is a genuine form of soul searching, grappling to understand that which eludes me.
What I really wish would happen is that aplication writers would provide the option of integrating the sensible parts which KDE provides for the Linux desktop. I wish that someone w/could create a a gtk+ file manager which was based on the Konqueror’s functionality. I wish when I would type in :///burn in this app, or selected it from the spartan menus, that a bonobo-ized version of K3B would appear. I wish that every GNOME app would have ties-ins for KIOslaves.
I wish it was possible to abstract the wondrful functionality of KDE apps from their GUI appearance. Of course I also wish that KOFFICE would read and write files from other OSS office apps. I wish Gedit could use Kate syntax stuff and offer the same network transparency. I alo wish that all things multimedia in KDE would be replaced by the superior Xine/Mplayer/Gstreamer stuff.
Note I am not asking for the creation of an exclusive desktop, ie. KNOME, or GDE. My suggestion would still require both GNOME and KDE existing as seperate indepenedently developed DE’s. But if the “parts” of KDE, which are the building blocks of all KDE apps could be used beyond KDE and in particular in GNOME apps with gtk+ visuals I would be in seventh heaven…
I know what I am saying here is quite absurd and probably totally unrealistic. But I long for a day when each of the major open source projects renders their core components as “parts” which can be easily integrated into any and or all of the different DE’s. As it stanbs I don’t mind loading up K3B from my GNOME desktop. IF I have any serious file-manager work to do using GUI tools I always choose Konqueror. I always make sure and manually add my KDE menu to my GNOME menu, and a system install without both KDE and GNOME is lacking for me.
Often I use XFCE4-for it gives me the aethetics which I love(gtk+) and is insanely fast and small-I use menumaker and have everything GNOME and KDE in my XFCE4 right-click menu, which dynamically includes new apps as they are installed. I wish the Ximian cups system could become the defacto standard for all Xwindows apps.
What would really be awesome would be a a new meaning for Gentoo’s USE keywords and the applciation this beyond Gentoo. If I compile and and tell .configure to use KDE then the program would be compiled with sensible integration of the core functionality of the KDE parts which are appropriate to the application. If I compile and tell .configure to use GTK+ then I have a HIG-compliant interface for the application which is tied into my Desktop theme and my desktop fonts.
This would promote really close work between the KDE and GNOME guys, and freedesktop.org would be the place where something like would happen. Of course what I am writing here is purposefully ignoring the programming/tookit constraints which probably make my wishes utterly impossible.
What would be necessary would be a kind of “third level” between the UI and core functionality. Most UI folks want a kind of Zen-based oneness between UI and functionality. Most Xwindows apps have historically been written with the UI as a mere afterthought. But If such a “third level” was possible, one could mix and match elements from the different UI’s with the different functionalities exposed in the API’s….Ok go ahead and shoot me, just some food for thought.
here is a summary:
all apps should have tie-ins to KIOslaves
all multimedia apps should have tie-ins to Xine/Mplayer/Gstreamer
all apps which need printing should tie into gnome-cups-manager
all major oss projects ought to adopt a three tier design:
tier 1: toolkit + HIG
tier 2: cross-toolkit/toolkit-independent functionality
tier 3: deep API functionality stuff, system integration
Now laugh at me, call me a clueless bozo, shoot me down, or just ignore me, enough ranting for one day…..
You’re not crazy, you’re right on. KDE has the technology and trimming down options/menus is easy. I think the reason KDE’s technology is so superior is because they used c++ for it. I’m no big fan of c++, and try to stay away from it when possible, but for doing a desktop and Application framework it’s just the way to go. No matter how you try and spin it, c is just too clunky for this sort of things.
I think things might be starting to go in the right direction. Gnome should just ditch whatever component tech they’re relying on(Bonobo) and use the underlying technology of KDE.
People should check out Kdevelop too see why kparts are such a good thing. Anjuta is just floundering in the wind and is basically dead. It’s just too much of a pain to develop Gnome apps in c. There’s lots of bindings but there’s no component technology because nobody is using bonobo. Mono might be the only salvation for Gnome unless some corporate interests have some tricks up their sleeves.
You can use KIOSlaves with any X/command line app… KIOSlaves can be mounted directly into the file system using Fuse
Google for it.
Read this link and it may provide you with what you are asking for….(not just yet as its early days but it have great possiblities)
http://dot.kde.org/1073599985/
And i wish people would stop the QT licencing rubbish
Ironically, companies like QT. There are lists of QT projects posted here on a regular basis. For all that QT might be controlled by a Norwegian company, corporations are used to buying toolkits and language stuff from companies. It’s the way it always has been for them, from all the little things they need to buying Visual Studio from MS.
It’s hard for them to get into GTK, to be honest, because it doesn’t actually fit into this model. Where do they download the definitive GTK version? Where do they find the web site with all the documentation, who do they call to request new features (for a fee!) and when they have issues, what is the phone number to talk to the tech support or the programmers themselves? At the end of the day, the majority of QT’s license is effectively for support; companies like it, and Trolltech are happy to sell it
Sidenote: The major re-write for GTK 2 will leave companies feeling a bit edgy, to have that amount of redesigning to do. QT, in general, remains stable throughout releases (doesn’t do the Java trick of deprecating and then removing stuff) and the changes that coders have to make to recompile for a new version are minimal, and usually add functionality.
And yes, QT is just a much nicer toolkit to use, it has to be said. It does everything you need, in obvious ways, with good documentation and acts intuitively. In this way, it saves so much development and maintenance time, it pays back any sort of major development, immediately.
Regarding the GTK issue. To get legitmacy in the commercial software world they should create a “GTK Foundation” which encompasses the whole GTK world, GTK+, GTKMM, GTK for Perl, GTK for Python and numerous others out there.
All of them need to reside on the same server with periodical releases, and all of them being feature complete, that is, one language doesn’t have more features available than another.
Behind the scenes books need to be written, support provided, consultancy services. This will enable a handful of fulltime developers to be hired to manage and develop the “unsexy” but very important areas of GTK.
If the commercial software world see that there is an organisational structure behind it and that they can get support when they require it, it will releave alot of managers.
What Qt provides to companies is a point of reference. When these companies require support they know who to turn to, and they know that the money they pay actually goes towards services that they could possible need in the future so actual fact, the price they pay is very small relative to the services provided and the possible problems that they may face and require assistance to rectify them within a timely manor.
The first poster said:
>>The Open Source Development Labs plans to unveil at the show
>>a new multicompany effort to improve desktop Linux
>>technology.
>Woh! This is the first time I’ve heard about that. I wonder
>what they’re up to and if the OSDL decides to choose one DE
>over another.
I think he has a point. Another clue you can find in the article which OSNews posted where Linus said the home desktop is 5 to 10 years away:
http://australianit.news.com.au/articles/0,7204,8407881%5E15841…
In her he says one sentence which made me curious namely:
Question: Anything you’re particuarly looking forward to?
Linus answered: I’m mainly following the desktop stuff, so the GNOME meetings…
Linus works for OSDL isn’t it?
Right on. I missed the connection there between the announcment of the OSDL doing desktop stuff and Linus checking out the gnome meetings.
@iaian peters
Whether you like it or not, people are having issues with the QT license. Even big companies are having issues it seems or why would they choose the technically inferior Gnome desktop over KDE. It can’t just be menu clutter.
By the way, you can download the managed and unmanaged SDKs for free from Microsoft. No, you don’t get the IDE, but you don’t have licensing fees with it either.