Moved by Freedom – Powered by Standards » 2010

Why I’m not moving to KDE (yet)

July 25th, 2010

3 comments

In previous posts, I had mentioned that I’m using KDE more and more. This is very much true today, but somehow I haven’t entirely migrated to KDE yet. I’m of course talking about the KDE 4 branch. I thought it would be interesting to share my reasons why my primary desktop on GNU/Linux is still Gnome.

Before the release of the KDE 4 and its subsequent versions, I was only seldomly using KDE. For some reason it felt odd to me; I never quite got used to its looks and style. Things changed a lot with KDE 4. I have been testing KDE ever since the 4.0 release and have been using it regularly in a “production-mode” ever since. By production mode I mean that I’m using it at work over extended periods of time (one full day of work or one half-day). Yet I never fully transitioned fully to KDE, using instead Gnome as my main and stable desktop. I evaluate the usage ratio in the following way: Gnome is around 60% of times, KDE 40%. There are some reasons for this that I would like to share here. They fall into two broad categories: Things KDE could improve and things Gnome still has the upper hand on. For what it’s worth, I’m using Arch Linux, which means the KDE and Gnome versions are the latest stable and pristine versions released straight out of their respective projects (no distribution specific tweaks).

Things KDE could improve:

I see two main areas for improvements, which does not mean KDE fails in these two fields, but simply that I find things could just work better. The first one is stability, which really means quality and the second one is the applications. Don’t get me wrong: There has been a huge leap towards quality between the ill-fated 4.0 release and the 4.4.5 one. KDE is stable, very stable in fact, but some details just remain a bit clunky. The Plasma desktop could be more stable as it behaves sometimes in a funny way: Plasmoids would not show up after logging into the system for no clear reason. Talking about Plasmoids, these can be sometimes buggy and the Plasmoids Installation and selection interface could really be clearer (although it does look quite elegant).

On the side of applications, I am still looking for better replacements of some of the software I use everyday, namely: Claws-Mail, Rythmnbox, and Nautilus. I know I can use them on KDE (although it would not make sense for Nautilus) but obviously GTK+ software does not play that well on KDE even with Qt styles. I don’t think I would trade Claws-Mail for any other email apps (and no, I don’t like Thunderbird), while the Kontact suite that ships with Kmail does not seem to meet my expectations (handling tons of emails while using MH mailboxes). Rhythmnbox is somewhat of a question mark. I like this one, but believe it too could use some improvements, and Amarok is an obvious choice on the KDE platform. I have been using Amarok regularly but I find it quite difficult to configure and not really user-friendly, although I understand there are lots of fans of this media player out there who could help me…. The same line of thought goes for Nautilus vs. Dolphin. I am quite sure that Dolphin is very powerful but it lacks this sense of simplicity that Nautilus conveys. Perhaps toying around with its default settings could be the key. Nautilus obviously has its drawbacks too.

There is another type of applications KDE is lagging behind: Web browsers. I really like Firefox, which is cross-platform, but I have never used it exclusively. In fact I do like to use platform specific browsers as they are supposed to provide a more diverse and integrated experience. Unfortunately, Konqueror is really outdated, Arora (using the webkit) is way too unstable, which leaves me with reKonq. I think this last one still needs improvements (especially in the interface and bookmark management) but it is so far the most powerful browser for KDE.

Things Gnome does better

Gnome does certain things better, some of which I already outlined above. I can add to this list a general feeling of stability and simplicity. Obviously, you don’t have to use Claws-Mail to stick to Gnome: but you could be using Evolution, or Thunderbird and still have a better experience on that one. Another advantage with Gnome is that it does convey this sense of complete control over the interface that KDE 4.0 traded away for a much more innovative -and beautiful- experience. A well-configured Gnome, as a result, will always feel faster and more effective than a well-configured KDE. Faster access to your data, to any of your applications, better sense of desktop room and predictability have so far kept me to fully leave Gnome.

Gnome, however, will very soon change its interface significantly: I have given a try to the Gnome-Shell interface, and I must say that while I find its concepts quite relevant and useful, I keep on having the feeling it’s a great interface for… Netbooks and small devices. I think I will in fact use KDE more.

Last but not least, I felt this post would not be complete if I wasn’t quickly highlighting the fact that Qt apps for the KDE platform tend to become more and more powerful and feature packed. I would not be using KDE so much if it weren’t for Okular and Gwenview. The document viewer is simply a swiss-army knife for documents reading and annotating, while Gwenview provides a much better experience than F-Spot will probably ever deliver.

As you can see, I’m still somewhat undecided on these matters, but can probably make up my mind quickly if any of the areas discussed above actually gets improved. Stay tuned!

Links for the end of July

July 22nd, 2010

No comments

Can one sponsor sustain a FOSS project on the long term? A crucial question that comes with my crucial answer, this time on FOSSBazaar.
After several articles in the press reporting on the Board of the OpenSolaris sabotaging itself, I wanted to clarify that the OpenOffice.org has fortunately a quite different situation with a clear ideas on who our community manager is, who sits at the Community Council, who does this and who does that. Of course not everything is perfect, far from it. But we do talk to Oracle, although we would hope to be kept in the loop as to where Oracle would like to go in the future with OpenOffice.org…
“Rotten to the Open Core”: a great post by Dave Neary setting the record straight on this insipid debate. And for what it’s worth, let’s remember Free Software means something and is not just nice and fancy brand to be used by everyone.
Steven Vaughan-Nichols is an incorrigible cynic. I’m glad OpenOffice.org is supposed to survive the slow death he predicted to others.
lpOD O.9.2 has been released. Grab it here!
Ars Aperta will be giving conferences and talks at the OpenWorld Forum at the end of September-beginning of October. Please come and visit us!

Is there really an Open Source “Tea Party Movement”?

July 6th, 2010

6 comments

A few days ago ZD Net’s anchor Dana Blankenhorn came up with an interesting theory. Because of the recent Bilski judgement by the Supreme Court of the United States of America, the Free & Open Source Software community seems to be divided and looking for scapegoats. One of his “leaders”, Florian Mueller, is actively pointing at IBM as being the source of all evil. By evil, he means software patents, but perhaps also the latest iPhone blunder, the gigantic oil spill in the Gulf of Mexico and probably AIDS as well.

To the credit of Florian Mueller, he’s been involved in the fight against software patents for over 10 years. He also seems to have worked for the Real Madrid Football Club before landing back in Brussels to defend the poor millionaire Monty Widenius against Oracle. Now Florian is all about IBM and claims that IBM essentially shot down the debate on software patents at the Supreme Court by addressing a “Friend of the Court Brief” in which IBM was essentially explaining the Court should not outlaw software patents right away, as these were valuable intangible assets that could badly hurt US companies. Florian Mueller has also been behind the infamous “Open Mainframe” initiative, targeted again at IBM and involving himself in an existing anti-trust case.

I disagree with Dana Blankenhorn on several points. First, Florian Mueller is by no means a leader of the Free & Open Source movement. Florian had his shot several years ago when the debate on software patents in Europe emerged and was a defining moment for the European community of digital rights and innovation proponents. It was in a sense a “rite of passage” for many. Since that time, some went their own way, some others maintained the flame and vigilance that is tested again these days. Florian was part of the former group. He went to work for the Real Madrid Football Club, and we somehow lost his track. Sometimes after the OOXML standardization odyssey, Florian took several planes to Brussels. He went to see many people, including many of my own personal friends and colleagues. His big project was to crush IBM, and Oracle, and anyone who was benefiting from software patents. Well, not exactly anyone: Microsoft was thoroughly avoided each time, sliding through the raindrops, but leaving everyone with an odd taste in the mouth. Soon enough, Florian’s campaigns, backs and forths appeared to many as they have always been since his come-back: an over-inflated bag of wind.

Let me explain: There is no division when it comes to software patents inside the Free Software & Open Source community. Software patents are evil. They stifle innovation. When used by patent trolls “non-practicing entities” they are at the core of a despicable racketeering system that can at times be seen as proxy war between some large software vendors. When piled up by software vendors, they are used as weapons of mutually assured destruction. That is, by the way, what the Open Invention Network is doing, piling up patents and neutralizing them, and it’s one of the reasons why Canonical recently joined it, as Ars Aperta has done a few years ago. Let there be no doubt: Software patents suck. We do not want them in Europe, no matter how high the pressures are. In this game, every major software vendors has patents on software: IBM, Oracle, Google, Microsoft, Computer Associates, Novell… heck, I even think that Red Hat has some. Is this a situation we wish to solve? Certainly. Does the fact of owning software patents mean you stand against Free & Open Source Software? Hardly. Does IBM own Software patents? Yes it does. Does IBM help Free & Open Source Software? It did more than most of the others, Red Hat excepted. Does IBM benefit from Free & Open Source Software? It certainly does, it even makes billions of dollars out of it, and they’re not the only ones. But the question is perhaps asked the wrong way, so let me rephrase it: There is a project, there is a team, or only one person, there is code. And when the code is available online, in the form of source code and binary bits, that you can access, study, modify and redistribute the software under these two forms with no constraints attached on the code, then you have Free & Open Source Software. That’s where contributors or sponsors come in. IBM is no different than any contributor or sponsor. There is given project with available code. IBM decides to contribute to it (the same is true for anybody else), period. Free and Open Source Software does not have bow to anyone to strive, because it does not have to “join” one camp or another. It is its own camp, which IBM, Oracle, Google, Red Hat, Novell, etc. agree to join for contributions or not.

In this sense, Florian Mueller has been rattling his saber for almost a year now, launching “fatwas” and anathemas on selected vendors (IBM and to a lesser extent, Oracle) while refusing to address the very big elephant in the room: Microsoft. Now this is not a rant against Microsoft I’m engaging into, but truth be told, Ballmer must have a crack at watching its competitors diving into various anti-trust cases in which they’re involved. And he sure must be very grateful to Florian Mueller, although Florian’s actual impact is very much limited to his own buzz: There is a reality distortion field that seems to be on around Florian’s weblog. It does not go beyond it. Florian has no troops, no clout, no beef, no legitimacy, no credibility among the Free & Open Source Software community. The man seems to have enough money to drum along though and in this sense, he’s acting a bit like the infamous “Tea Party Movement”: You create a pretendedly grassroots movement in the hope people will believe it’s real until some more people actually joins your cause. In this case, however, I’m afraid this “Tea Party Movement” is stuck at level 1: The Astroturfing stage.

Make no mistake though: we all stand against software patents. But Florian Mueller’s tactic is strange, extremely partial, and leaves some big questions unanswered: Who does benefit from an anti-IBM campaign? Who does benefit from Oracle not merging with a dying Sun? Who does benefit if everybody thinks Google infringes your privacy?