LWN.net Logo

KDE 4.1 Beta1 Release Announcement

KDE 4.1 Beta1 Release Announcement

Posted May 28, 2008 1:22 UTC (Wed) by mtall (guest, #52045)
Parent article: KDE 4.1 Beta1 Release Announcement

Interesting that this release has been called "Beta 1". Perhaps this is an acknowledgement of sorts that KDE 4.0 was a public alpha release. I thank all the developers, and at the same time I wish they would be more considerate next time they make a release that's allegedly for public consumption.

I've been a KDE user since the beginning, however the snafu caused by the 4.0 release (as present in Fedora 9) highly incetivised me to give Gnome a go. I had no preconceived notions about Gnome other than "they've got a different philosophy and use the GTK toolkit" (upto now I was a KDE user more out of habit than through a well considered evaluation).

After comparing both Gnome and KDE to MacOS X and Windoze, the overall impression is that neither Gnome or KDE can get their Scheisse together (to borrow a German word). Gnome seems to be a weird clone of Windoze, mainly as it gives the user a rather annoying set of defaults which are either not changeable or one has to dig through half-documented options via gconf-editor (I like the idea of the gconf-editor, but not its execution). A simple example of missing functionality: an option for a mouse-click based vertical-only maximisation of windows (a keyboard equivalent of sorts exists, but this is not the same) -- very handy when it comes to file selectors. Other simple things like "watching" a PDF file in Evince (to catch changes) are not present -- while I know there are issues with 100% reliability in re-reading a changed PDF, I'm more than happy to settle with the 95% reliability present in kpdf/okular. Another weird and highly anti-intuitive effect is the resizing of maximised windows when they're being dragged across spanning desktops on a dual-screen setup.

KDE has its own share of faults. For example, especially prior to 4.0, there is the problem with organising its multitude of options. 4.0 itself is another source of annoyance, but that's mainly due to being unfinished. However the main problem is that if I were to write any software specifically for KDE, it would have to follow the licensing of Qt. Even though the FSF recommends that the GPL be used for libraries, I find this to be too restrictive of my freedoms. (side-note question: is there an LGPL licensed GUI toolkit that will respect user-preferences set in either KDE or Gnome?)

All in all, a "better" desktop environment would be comprised of a cleaned up KDE with less restrictive licensing options. Alternatively, Gnome is barking up the right tree, but the human-interface guidelines, that Gnome is often citing, should really be revised.

As a final note, after so many years it is becoming tiring to deal with these type of issues. I know that OSS development does not have the engineering resources of Apple or MS, however the engineering resources of OSS development are not being used very effectively (why have three or more efforts for a desktop, if we include XFCE ?). It seems to be a case of too many cooks in the kitchen, many of which are convinced that their way is the "one true way".


(Log in to post comments)

KDE 4.1 Beta1 Release Announcement

Posted May 28, 2008 3:02 UTC (Wed) by JoeBuck (subscriber, #2330) [Link]

Everyone should know by now that Foo n.0, for any value of Foo and n, is something that needs to be approached with caution. If KDE 4.0 had issues with maturity and reliability (and I haven't used it so I don't know), that's just par for the course.

KDE 4.1 Beta1 Release Announcement

Posted May 28, 2008 4:01 UTC (Wed) by xbobx (subscriber, #51363) [Link]

> As a final note, after so many years it is becoming tiring to deal with
> these type of issues. I know that OSS development does not have the 
> engineering resources of Apple or MS, however the engineering resources of 
> OSS development are not being used very effectively (why have three or 
> more efforts for a desktop, if we include XFCE ?). It seems to be a case 
> of too many cooks in the kitchen, many of which are convinced that their 
> way is the "one true way".

I disagree.  One of the key values of open source is that people can work on whatever they
want/feel like/are passionate about.  This includes allowing different opinions about how to
go about things, and different projects to implement them.  If everyone were just working on
the same thing, it would no doubt turn away people with good ideas (intentional or not).

Another benefit is the programming equivalent of evolution.  As the parent mentions, (s)he
stopped using KDE and tried Gnome for a while.  This simply wouldn't be possible if there were
one all-powerful project.  You'd be forced to live with whatever mediocre project was The
Chosen One, and wouldn't have the ability to decide that you don't like it, because there's
nowhere else to go.

There really isn't the option of "everyone should do it exactly how I like it and stop wasting
their time on other things".  It's more like "everyone does it how they feel like doing it,
and you hope that somebody has the same tastes as you, or at least similar enough that you can
modify it to your liking".

KDE 4.1 Beta1 Release Announcement

Posted May 28, 2008 4:31 UTC (Wed) by mtall (guest, #52045) [Link]

You'd be forced to live with whatever mediocre project was The Chosen One, and wouldn't have the ability to decide that you don't like it, because there's nowhere else to go.

While I see what you're getting at, I don't think that the above generalisation is valid. Let's apply it to the Linux kernel: multiple BSD kernels exist, but frankly, compared to Linux their mind-share and "market-share" is tiny. Splitting already limited resources across two or more projects is simply not a good idea. The Linux kernel would not have developed so rapidly if not for the concentration of developers.

Prior to KDE/Gnome we had a multitude of window managers and docking applications, none of which could be considered as integrated desktop environments. Freedom to all, but they were a pain in the butt to use (even so called "power-users" have finite amounts of patience). The original KDE/Gnome schism came to being due to licensing issues of Qt (which were originally very restrictive). The Qt licensing issues were lessened, though in retrospect it was not that wise to start off with the Qt toolkit in the first place. (While Trolltech is a large contributor to KDE, to them it is/was a significant form of advertising).

Had KDE started off with GTK (or some other unencumbered toolkit), it is quite likely that by now we would have had a desktop equivalent of the Linux kernel, without fragmentation along the lines of Ubuntu/Kubuntu, or issues with synchronising user preferences across two desktop environments.

KDE 4.1 Beta1 Release Announcement

Posted May 28, 2008 10:20 UTC (Wed) by epa (subscriber, #39769) [Link]

You're probably right, but see the original KDE announcement:
BTW: Usually these postings get a lot of answers like "Use a Mac if you want a GUI, CLI rules!", "I like thousands of different widgets-libraries on my desktop, if you are too stupid to learn them, you should use windoze", "RAM prices are so low, I only use static motif programs", "You will never succeed, so better stop before the beginning", "Why Qt? I prefer schnurz-purz-widgets with xyz-lisp-shell. GPL! Check it out!", etc. Thanks for not sending these as followup to this posting :-) I know I'm a dreamer...

KDE 4.1 Beta1 Release Announcement

Posted May 28, 2008 11:02 UTC (Wed) by pointwood (guest, #2814) [Link]

"Interesting that this release has been called "Beta 1". Perhaps this is an acknowledgement of
sorts that KDE 4.0 was a public alpha release. I thank all the developers, and at the same
time I wish they would be more considerate next time they make a release that's allegedly for
public consumption."

It's called beta 1 because this is the first beta release of KDE 4.1. Nothing strange about
that at all. There was a beta 1 of KDE 4.0 too. The 4.0 release have been discussed a lot all
over the net. I believe the biggest complains was the lack of polish and features of Plasma.
The KDE developers warned about this quite a bit and for a long time. Maybe they should have
called it "KDE 4.0 - developers release". I do agree that KDE 4.0 wasn't really ready for
"normal users", but it has been said often that most users probably should wait for KDE 4.1 or
maybe even KDE 4.2 before switching from KDE 3.5.x. 

"I've been a KDE user since the beginning, however the snafu caused by the 4.0 release (as
present in Fedora 9) highly incetivised me to give Gnome a go."

Fedora is generally well known to live on the bleeding edge. If you prefer a more stable
environment, you might want to use a more conservative distro.

"However the main problem is that if I were to write any software specifically for KDE, it
would have to follow the licensing of Qt. Even though the FSF recommends that the GPL be used
for libraries, I find this to be too restrictive of my freedoms. (side-note question: is there
an LGPL licensed GUI toolkit that will respect user-preferences set in either KDE or Gnome?)"

Where exactly have you gotten that information? Most of the KDE libs are LGPL not GPL licensed
and there are lots of exceptions to the QT GPL license, a quick google search provided me with
this list: http://doc.trolltech.com/main-snapshot/license-gpl-except... 

Golden rule of the internet: The persistence of a piece of information is entirely based on
how wrong it is. If the information is seriously wrong or outdated, it simply can't be killed.

Someone more fluent in english should reformat that into a rule that's actually understandable
:p

"As a final note, after so many years it is becoming tiring to deal with these type of issues.
I know that OSS development does not have the engineering resources of Apple or MS, however
the engineering resources of OSS development are not being used very effectively (why have
three or more efforts for a desktop, if we include XFCE ?). It seems to be a case of too many
cooks in the kitchen, many of which are convinced that their way is the "one true way"." 

It's not a matter of too many cooks in the kitchen. That's completely flawed. It's not one
kitchen - there are thousands of kitchens, each with x number of cooks and unlike a real
kitchen, it mostly works pretty well in the open source world with lots of cooks in the
kitchen.

KDE 4.1 Beta1 Release Announcement

Posted May 28, 2008 12:28 UTC (Wed) by mtall (guest, #52045) [Link]

It's not a matter of too many cooks in the kitchen. That's completely flawed. It's not one kitchen - there are thousands of kitchens, each with x number of cooks and unlike a real kitchen, it mostly works pretty well in the open source world with lots of cooks in the kitchen.

Each cook could be a "chief" over a particular sub-domain, but for making a meal you need one head cook -- people go to a restaurant to eat, not to admire how many sub-cooks were involved in the process. The same applies to open source development. A successful development typically requires the concentration of many developers onto one task, and having a focused leader is more than helpful: in the case of the Linux kernel, there is a "strong man" who is the trusted person to carry the project along. At the same time Linus is also open to criticism and input from other people, as well as having "lieutenants" responsible for sub-systems (this is not an advocation of a cathedral style of development).

Would there be any point in having two incompatible Linux kernels? How about two incompatible GCC forks? How about two X servers? Splitting of resources into unnecessairly competing projects amounts to having slower progress in each project. While this has been partially alleviated by the freedesktop organisation when it comes to KDE/Gnome, it's rather unfortunate that we have two large projects trying to achieve the same difficult goal.

The differences between Gnome and KDE are ultimately superficial -- it's entirely possible to move over to one toolkit, or have a wrapper which abstracts Qt and GTK. Furthermore, the degree of configurability (a source of difference between KDE and Gnome) can also be controlled - simply have one switch to allow "advanced configuration". For the disbelievers, look at the linux kernel: a large project with many disagreements along the way, yet it ends up being a cohesive effort.

KDE 4.1 Beta1 Release Announcement

Posted May 28, 2008 13:49 UTC (Wed) by pointwood (guest, #2814) [Link]

Linux isn't exactly the first operating system either so by your theory Linus Torvalds should
never have developed Linux.

Forks can be good too (not saying that they always are or that GNOME is a fork of KDE). Just
look at all the good stuff that have come from OpenBSD (which is a FreeBSD fork if I recall
correctly).

You mention Linux as an example where Linus Torvalds is the "strong man". That is just one
example and that's not exactly how all OSS projects work. We're talking about KDE and GNOME
here - who's the "strong man" for those two projects? The fact is that they work very
differently than the Linux kernel.

Competition is good and as you mention, KDE and GNOME hae different philosofies. Just as
Windows and OS X are different. Some prefer Windows, others prefer OS X. Yes, there are
problems and overhead by having two "competing" desktop environments, but I believe there are
advantages as well just like it's rarely a good thing that a company have a monopoly like
Microsoft have. Also, you seem to think that if we closed down the GNOME project today, all
the GNOME developers would start working on KDE. That's simply not the case - these people are
mostly doing it because they think it is fun and interesting. They are not getting paid for
it. The people working on KDE is doing it because they prefer that to GNOME (etc.) for
whatever reason. That's also why it's actually a complete waste of time to discuss this
because you/we have no way of changing it.

KDE 4.1 Beta1 Release Announcement

Posted May 29, 2008 13:36 UTC (Thu) by aftk (guest, #52302) [Link]

> The differences between Gnome and KDE are ultimately superficial
> -- it's entirely possible to move over to one toolkit, or have
> a wrapper which abstracts Qt and GTK.

There is a partial solution to this: see the GTK-Qt Theme Engine

KDE 4.1 Beta1 Release Announcement

Posted May 28, 2008 12:09 UTC (Wed) by philipstorry (subscriber, #45926) [Link]

mtall said: 'All in all, a "better" desktop environment would be comprised of a cleaned up
KDE'

I'd regard that as a worse desktop environment.

It's down to personal taste, I suppose.  I left Windows for Debian/Ubuntu, and Gnome was a
major factor in that.

After many many years in Windows - from version 2.11 onwards - I'd become sick and tired of an
interface that was getting more and more cluttered.

Microsoft were throwing sticking-plasters at it - things like the Most Recently Used programs
list on the Windows XP Start Menu - but they were just that, sticking plasters.  They didn't
reliably solve the problems, they just covered it up 90% of the time.

KDE didn't make the grade for me because it felt just enough like Windows that it was
comfortable and intuitive, right up until it was different and frustrating.  Its similarity is
its downfall for me.
Gnome, by contrast, was just what I wanted - a simpler interface which was easy to use yet
retained 90% of the power I needed.

Clean up KDE all you like, but I suspect if you removed its clutter it would lose lots of its
fans.  I'm not going to use KDE until it's as clean and simple as Gnome, and that's just not
going to happen.

Of course, this is all just a matter of personal preference.  I wanted to go back to the idea
of a simple clean GUI, with the "10,000 options and ultimate flexibility" being left for a
command line.  Gnome gives me that.  KDE users typically want those options and that
flexibility in their GUI, which I find makes the GUI harder to use and - crucially - less
pleasing to use.

It's good that we have both options.  Especially now that KDE programs can run under Gnome and
vice-versa without any major hassles.  I see no need for the two to mimic each other or to
converge, and would rather that they didn't.

KDE 4.1 Beta1 Release Announcement

Posted May 28, 2008 17:21 UTC (Wed) by jospoortvliet (subscriber, #33164) [Link]

All in all, a "better" desktop environment would be comprised of a cleaned up KDE with less restrictive licensing options. I know this is a flamewar thing but the GPL was created for a reason. The very concept of freedom, or 'total freedom' is flawed. If I have total freedom, I also have the freedom to limit YOUR freedom. Hence only 1 person on this beautiful planet can be truly free. That's why the only limit to the personal freedom of one person should be another person's freedom. And we need laws to protect those freedoms. In other words, freedom needs to be restricted to protect it. Of course, finding a prober balance is hard, and power-hungry people will always try to take advantage of the resulting compromise - one way or another. The GPL is a beautiful compromise - it pretty much only restricts the freedom to restrict other peoples freedom. Why would you want to take that away from KDE and Qt? I think it's great a GPL-ed Qt forces people to either protect other peoples freedom (write MIT/BSD/GPL etc), or contribute to further development of Free Software by paying Trolltech to improve Qt. I wouldn't want it any other way. It's actually pretty cool one of the largest contributions to Free Software (Qt under the GPL) is mostly being paid for by companies developing proprietary software on Windows...

Copyright © 2012, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds