OSDL to Bridge GNOME and KDE

Trax88 writes "Open Source Development Labs is previewing work that will attempt to make life easier for software companies by bridging GNOME and KDE. The effort, called Portland Project, began showing its first software tools on in conjunction with this week's LinuxWorld Conference & Expo. Using them, a software company can write a single software package that works using either of the prevailing graphical interfaces. Working with Freedesktop.org on unifying interface issues, they plan to release a beta version of the software in May and version 1.0 in June. Ultimately, advocates hope that it will be part of a larger but separate effort called Linux Standard Base, which is designed to make the operating system easier for software companies to use."

ask slashdot

[Douglas Simmons]

Slashdot ought to ask its visitors what their favorite features between the two that are not shared so this OSDL project can get more guidelines from the right demographic. Ask Slashdot is a powerful resource to collect knowledge, perhaps more than any other system in the galaxy.

Cross platform?

[gentimjs]

Will this help users of non-linux systems, like myself running KDE on solaris/sparc whom are upset that all of Sun's bundled tools are gnome-specific and load up a billion gigs of dependant libraries when I try and launch them?

Lowest common denominator

[TheCoders]

It's hard to tell exactly what this project is going to deliver, but it looks to me like an abstraction layer that will run on top of whatever GUI toolkit is available, rendering with native widgets.

This has been attempted before, and it usually doesn't catch on. There are plusses and minuses to both toolkits (as there are in any GUI toolkit). The problem that arises when you try to combine them is you end up with a superset of the negatives and none of the plusses that would lead you to choose one over the other. Essentially, it's the "lowest common denominator" problem. If a certain feature is present in one toolkit but not the other, then guess what? It's not going to make it into DAPI. If similar tasks are accomplished differently in the two toolkits, the Portland project is going to have to choose one, and shoehorn the other to fit. Either that, or introduce a third way of doing the same thing.

People view the existence of two competing desktop standards a "problem." I disagree with that. As a developer, if I see a certain application already exists on my platform of choice, I'm not going to make another one, even if mine would have been better. On the other hand, if I were a KDE man, and there was an existing app for Gnome, but one that I didn't really like, then there's a little more incentive to make a native KDE version, in the mold of what I really want. In the end, it's the users who win, because they can pick and choose between both apps.

So for now, pick one and go with it. Don't fall into the trap of trying to conquer both worlds at once.

Re:Merge ?

[chrismcdirty]

Gnome is generally not configurable. KDE is configurable out the wazoo. That's why. Gnome seems to be very resource hungry. KDE has the option to run extremely light. That's why. Because I prefer KHTML/Konqueror to Gecko/Firefox. That's why.

More Standards! Yippee!

[Teclis]

Just what I needed. ANOTHER computing standard to learn. Which standard is next to join the act? Maybe the next one will be more standard than all the other standards.

look *and* feel

[eddeye]

There's a difference between looks like kde and works like kde. Will the menus/config/keybindings be in the right place/format? Will the application handle dcop messages properly? Cross-platform toolkits usually abstract away the differences between platforms. It might translate the function calls and provide the right look, but that's only half of getting the proper look-and-feel.

The ubuntu openoffice-kde package does a nice job, but it's obviously not a kde application. I hope this toolkit gets it right because I would kill for a KDE version of firefox (damn these infernal gnome save dialogs!).

Re:Merge ?

[MAXOMENOS]

What you suggest is a very difficult task.

The two desktop environments do basic tasks very differently. One of the main reasons why I no longer use Firefox on Linux is because I hate the GNOME file browser that Firefox uses by default. To me, all it does is make my job harder. For the sake of a more sensible file browsing interface, I am willing to tolerate Konqueror's relative slowness at loading web pages. Who's going to negotiate those differences?

The two desktop environments use very different core libraries with different licensing schemes (Qt is GPL, gtk is LGPL). These licensing schemes may carry big implications for those who use them (for example, you can base wxWindows on gtk without a problem, but can you do the same with wxWindows and Qt?)

There may also be major architectural differences that make a merging nontrivial.

Basically, what you're proposing is a huge project. The Portland Project has a much more limited scope, and I think it's much more achievable.

Re:Will they be able to deal with KDE sound apps?

[SirTalon42]

KDE4 will (possibly, not 100% final yet) use an abstraction layer instead of directly using a sound system. This way the end user can decide which to use, and there are plans for gstreamer, xine, nmm, and arts so far I believe.

Re:Merge ?

[CastrTroy]

I was a KDE user for a while. It was always slow and kludgy. I recently switched to Gnome on a whim, and I have to say It's about 10 times faster than KDE. What do you do to get KDE running faster than gnome?

$0.02

[molarmass192]

I think GTK is admirable, but GNOME has regressed over the last 2-3 years to the point that it's no longer usable for me. The dumbing down of the GNOME widget set cornered me into a Fisher-Price user experience that I disliked greatly. Let's face it, I'm sure only a tiny tiny slice of Linux users are technophobes. Catering to such a tiny user base is a death wish for any but the most specialized of projects. If GNOME doesn't make an about face, it will eventually become nothing more than a fringe player with KDE owning 95%+ of the desktop pie. I have faith that GNOME can turn about and drop this "simplicity" crap, the question is will it?

Real problem is a single set of guidelines

[wysiwia]

Because the real problem is not so much the used framework but to use a single set of guidelines. The main obstacle of the Linux desktop is the usability, the look&feel of the applications. If one just uses 2 different applications on Linux, one most likely has to learn 2 different ways how to work with. If one uses 10 different application one doesn't have to learn 10 different ways but quite possible 5 to 7.

So I created wyoGuide (http://wyoguide.sf.net/ [sf.net]) exactly for this, to finally have a single set of guidelines. And I designed wyoGuide to be cross-platform guidelines since no serious developer codes for a single platform these days. wyoGuide can and should be used on any platform with any framework and any language. Sure I do provide sample code written in C++ with wxWidgets but I'd love to put up others sample code as well. So far nobody familiar with other's framework volunteered.

To stress this point again, the Linux desktop won't become a success unless it can't be agreed on this single set of guidelines. It's possible that everybody sits together and designs yet another set but the outcome won't be much different than wyoGuide. On the other side wyoGuide is still work in progress and I'm open to any suggestion to make it more suitable for anybody.

If somebody doesn't believe me just read the LXer article here (http://lxer.com/module/newswire/view/54009/index. html [lxer.com]) and follow the links to the sources. Or go and read the guidelines themselves at http://wyoguide.sourceforge.net/guidelines/content .html [sourceforge.net].

What I'm curious about is how the Portland project handles this info, the knew it since December 2005 (http://lists.osdl.org/pipermail/desktop_architect s/2005-December/000349.html [osdl.org]), they seems to already have forgotten. I've also informed Novell and posted it to LinuxQuestions, almost no reaction. So what else can I do?

O. Wyss

Re:Merge ?

[davidsyes]

Firefox?

What I LIKE about FireFox is that it respects the Gnome code enough to let me use SCIM to input Kanji and other foreign characters into the search engine right in the browser. KDE/Konqueror WON'T let me, and it **appears** I have installed all the requisite stuff. The Gnome apps, even running in Konqueror DO let me use foreign characters. OO.o refuses to play ball, too.

What I DON'T like about Firefox is the lack of a Konqueror-like page archiver. I find myself copying or cutting the URL and pasting it into Konqueror. Also, I don't like the file browsing/saving method. I'm addicted to KDE/Konqueror's.

But, I haven't honestly USED Gnome as a workspace, though I intentionally install it because I am sure it has some framework goodies that enhance my use of KDE while I want run Gnome-based/friendly apps that supposedly have a KDE twin.

X11 virtual terminals, please!

[Maljin Jolt]

There is no need for mixing KDE and Gnome alltogether.

As I already mentioned in another slashdot discussion some time ago, I run KDE on vt7 amd Gnome on vt8. (And Fluxbox on vt9 just for OpenGL 3D accelerated games but that's another story.)

Just try it: On KDE 3.5.x, click "Switch User:Start New Session" on K menu. You will get your favourite login manager running on a new terminal. Pick another deskop you have installed. Switch back and forth with Alt+Ctrl+F7,F8,F9... And don't forget you still have your framebuffer consoles on Alt+Ctrl+F1..F6.

Comment removed

[account_deleted]

Comment removed based on user account deletion