Quickies: Amarok 1.4.7, MEPIS KDE 4, Desktop Survey, Lugradio Talk

Amarok 1.4.7 was released with improved collection backend, new streams, altered icon and bugfixes. *** The annual Desktoplinux.org Survey is under way. *** MEPIS released a KDE 4 Beta 1 live DVD using packages from Kubuntu. *** The videos from Lugradio Live are up including Ben Lambs' Conquering the Desktop with KDE 4. *** Finally, following the 10th anniversary of the free desktop last year, congratulations to another project which has gained double figures in age, but whatever did happen to those Scheme applets?

Dot Categories: 

Comments

by Morty (not verified)

It's not your English that's the problem in this case, but simple lack of knowledge regarding history and a little misunderstanding of the reason behind some of the thecnology. So the "after 10 years" comments are way off.

Like DCOP, back when it was created nothing exsisted which was fit for the use KDE needed. Later when Gnome needed similar thecnology they went for a solution based on CORBA, which KDE had alredy found unfit for the task. And that GNOME newer got a IPC usage comparable to KDEs use of DCOP underscores that. Years later when efforts to intergrate the desktop better with Linux, HAL was started. The need for an IPC was again apparent, and DBUS was created. And it was even modeled after DCOP. Besides KDE has for years already used DBUS for the same tasks as GNOME. From KDE 3.4(or was it 3.3) DBUS/HAL have been used for media discovery and detection just like on GNOME.

For khtml it has always seemed more a lack of giving back from people interrested in the thecnology. From Apples recent fork into WebKit, developed for a year behind closed doors rather than cooperating. To GNOMEs gtkhtml, orginally forked from KoHTML(khtml's predecessor) a long time ago.

Regarding the shared mime entries, it has not exsited for more than a few years and it has not been possible to incorporate in KDE earlier becouse of BIC issues. Had there been a mime standard 10 years ago, KDE would have adopted it like it does other standards that make sense. Like the menu entry standard, which KDE was able to support as soon as it was ready.

Solid is not intendend to be a library for other desktops, its a abstractionn layer on top of stuff like HAL, to to make it easier to use foe KDE coders and intergrate it better in the desktop. The standard here is HAL, and it's the choice of the the other desktops like XFCE and GNOME how they want to interact with it.

As for multimedia, KDE already tried to cooporate and make a standard years ago with aRts. To acomodiate this, aRts was early changed from using the Qt event loop to be based on glib. But in the end it seems like NIH syndrome and personal interests was bigger, and gstreamer was created from scratch. In truth the open source desktop is not any closer to a standard multimedia framework today than it was 10 years ago, but KDE has benefitted from haveing a common solution all those years. So to keep KDE both platform independant and not tie itself to a multimedia backend that make fade and die becouse of politics or supperior alternatives, Phonon is created. And as a benefit it makes it much easier for the KDE coders to use, as the style of the API is familiar with what tehy already are used to in Qt.

by djouallah mimoune (not verified)

thanks really kevin and Morty for your explanation, so it's all about politics and NIH syndrome. i hope initiative like lsb and portland will make the next 10 years better.

and sorry again for my language ( reading all those comments in digg, and slashdot has made a bad influence on my words ;)

by Morty (not verified)

It's not always about politics and NIH syndrome. Often there simply are no exsisting thecnology or standards or the ones that do are not possible to adopt to your needs(with reaosnable amount of work). And if that happens, it does not make it standard by declaring it to be so etiher. You have to look at the history of projects to see why things are done the way they are.

Imaging the loss of development pace with the loss of competition. And yes, GNOME and KDE are their biggest competitors, as they can be compared directly (opposite to Windows and OSX).
Withour GNOME, KDE's menus would still be clutters as they were years ago. Maybe no Kontact without Evolution. No ... without ...
If forces would have joined, progress wouldn't have doubled. I'd rather think progress would have been decreased.
Having a friendly competition pushes us forward to challenge even better with windows (which I doubt has a lot more than 90% marketshare left).

by Sergio fernandes (not verified)

If I am a kde user since ever, why I must read digg or the site promoting the survey to know that there is a survey???

I assume that if I use kde I want to vote on it. What's the trouble??

by Ian Monroe (not verified)

I think it is interesting that Miguel did have using language bindings as one of the main tenets in the creation of Gnome. And they still can't make up their minds about it. :) Maybe they should have just stuck to the Scheme bindings and avoided much internecine conflict.

by Richard Dale (not verified)

Yes I agree Scheme is a great language I wonder why it didn't take off for Gnome programming. There are still Guile bindings being actively maintained: http://www.gnu.org/software/guile-gtk/

"Qt also forces the programmer to write his code in C++ or Python.
Gtk can be used in C, Scheme, Python, C++, Objective-C and Perl"

Well since 1997 we've had Qt bindings for all those languages except.. Scheme. Maybe Qt/KDE Scheme bindings would be KDE's long lost killer bindings app?

"We plan to export the GTK API through a procedural database
(which will in fact be an object database) to allow easy
integration with scripting languages and modules written in
other languages"

This is interesting - I'd never heard of it before. It sounds a lot like the Smoke approach of a language independent dynamic introspection based binding of the complete static language C++ (or C for Gnome) api.

The magic bullet for bindings was even supposed to be CORBA at one time.

More recently it's been language independent virtual machines like the Java JVM or the .NET CLR (and Parrot - is that MIA?). Now we've got Qt bindings for both the Java JVM with QtJambi, and Mono bindings with Qyoto, I think we can have a lot of fun finding out whether that is a good idea. Simon Edwards, Arno Rehn, one of the QtJambi guys and myself, did a bit of playing around with that at this years aKademy and it does seem to 'nearly work', but isn't quite as easy as it sounds. I especially liked Arno's analog clock in VB.net although it apparently didn't quite work.

The latest Gnome bindings Great White Hope is Vala, which sounds pretty interesting to me and not a bad idea at all. But what if they'd just adopted the OpenStep based GNUstep instead, and radically improved that? I really think KDE would have much more trouble competing with a Cocoa clone, as there isn't a lot wrong with Cocoa/Objective-C even after all these years.

by Ian Monroe (not verified)

Vala seems to be mostly a "Mono is yucky" development. I've looked at the code for Banshee and it seems rather straight forward. Banshee itself is surprisingly responsive. Which really shouldn't be surprising, its just that I'm so used to the 10 years of experience of a VM meaning Java meaning AWT meaning clunky, memory hugging and ugly.

Though the Gnome guys probably know more about Mono then we do, maybe its not a horse you should bet on.

by Ian Monroe (not verified)

I can't really let your comment regarding Scheme go unnoticed, I just hate it so much. I had to use Scheme in a class, Lisp is such a horribly ugly language.

Granted it was a programming language class that we learned it in, so I suppose I shouldn't be surprised that a such an avid student of programming languages as yourself would be a fan. ;)

by Richard Dale (not verified)

When I was a student we used a language called 'pop-11' and Unix which was like lisp with a more Pascal like syntax:

http://en.wikipedia.org/wiki/POP-11

I think pop-11 was a better language to start learning programming than lisp (and ruby would be better today). My Philosophy professor, Aaron Sloman describes the differences like this:

"The closest comparable language with similar characteristics is Common Lisp, though many people (not all) find the Pascal-like syntax of Pop-11 easier to learn than Lisp's very terse syntax using very few syntax words. For the same reason some people find Pop-11 programs more maintainable. On several occasions I have met commercial programmers who were used to other languages, like Pascal, Fortran or C, and who had tried to learn Lisp and disliked its syntax immensely. By contrast when they tried to learn Pop-11 they found that it offered them a smooth transition from familiar programming constructs to more sophisticated AI programming. After that they found Lisp easier to learn.
Not everyone agrees on which is easier to learn or use: so it is good that both Pop-11 and Lisp should be available."

http://www.cs.bham.ac.uk/research/projects/poplog/primer

I used to think lisp was ugly then. But learning Scheme from the book 'The Structure and Interpretation of Computer Programs' changed my mind as it must be one of the best computer science books ever written.

by Boudewijn Rempt (not verified)

And even better, the complete text of the sicp is freely available: http://mitpress.mit.edu/sicp/.

by Ian Monroe (not verified)

The programming language class is an upper level class about the history, design and syntax of languages. So learning Lisp did make sense, Lisp is very simply defined (our text book had a Lisp compiler written in Lips) and in the 80s pioneered garbage collection and such. Java was written by a bunch of old Lisp folks.

Learning it as a first language would be pretty crazy. :)

by David (not verified)

A big congratulations to everyone involved in GNOME development for ten years of excellent work. The GNOME HIG may not suit everyone's needs but it has been a big step forward for Linux on the desktop. Well done!

by Dima (not verified)

I wonder how Safari got 0.3%...

Do people run Mac OS inside a virtual machine? Or do they run the Windows version in Wine, etc.?

And why?

by hf (not verified)

MacOnLinux? Or they compiled the WebKit sources with the Qt backend?

by a.c. (not verified)

In konqi, Go to tools->change browser ID->other->safari.
Why? Have to have fun, I guess. I know that I have ID konqi on a few sites as Mozilla or even MSIE so that I could view the site (home depot was HORRIBLE for this and they were just one that comes to mind; That attitude cost them about 20K from me; I shopped lowes because they at least worked). Don't have to anymore.

You have to hand it to the GNOME guys, to have been able to successfully create a good desktop environment. And while we all probably want everyone to use KDE (right? :P), GNOME's existence and presence has been really beneficial to everyone.

I saw this link from a dev from Ubuntu Planet:
http://www.gnome.org/press/releases/gnome10years.html

Seems like their going to make a scrapbook wiki documenting their 10 years of existence. I can barely remember how we celebrated and commemorated our 10th birthday last year. Aside from the official announcements on the website or blog posts on the Planet, did we have anything special during that time?

Yeah, congratulations to the GNOME project, for sure. They've come a long way since the early days.

It's nice to see that both projects seem to be getting back to their roots, at least to a certain level. I mourn the loss of a scripting framework for GNOME, for example, and wonder how in the world they took a light, speedy toolkit like GTK+ and managed to bog it down. OTOH, it's a decent enough desktop with usability in mind.

And KDE! Most of what I see about the upcoming KDE4 tells me that y'all are getting back to the usability roots. Hallelujah! I've missed the simplicity of the KDE 1.x days. The first GOOD non-Netscape browser on *n?x was, imho, KFM, and Konqueror continues that trend. I'm especially glad to hear that there's an effort underway to work WebKit into GNOME; glad to hear that, as they need to get that albatross...erm, gecko, off of their neck. :-D Had Netscape chosen to make their browser GPL only, and Gecko LGPL, it would have been no big deal that the Foundation keeps changing their mind about how their various copyrights and trademarks need to be enforced; unfortunately for projects like Debian, it has meant forking Firefox.

So yes, kudos to both projects for being around for more than a decade, and congratulations to both for continuing to improve by leaps and bounds, and to continue to converge toward a certain level of interoperability. May the progress continue; the true winners in the end will be the users. :-D