The slowdown of GNU/Linux

I was just reading about the GTK+ toolkit which is pretty much out of commited maintainers. Its previous major developer (who actually developed stuff instead of just maintaining), Owen Taylor, has moved to Mugshot development instead…

This piece of news is just the tip of the iceberg. The whole “Gnome and peripherial ecosystem” is dry for 2 years now. By now, gnome 3.0 was supposed to be released, and Red Hat/Novell engineers had big plans for it. Instead, Red Hat moved away from the desktop, Novell only does bits and pieces now, and there is no GRAND plan about the future of the project. There is no real roadmap, not even a solidified idea as to where they would like to see the project move to. Mostly fixes get into major versions of Gnome for the last 2 years and most of the major additions are third party developers deciding to bring their small utilities into the platform. KDE is not much better either. There is no major distro that pays engineers to work full time on it anymore, so KDE is as dry as Gnome is. I am not saying that nothing is moving forward, it’s just that it’s not moving not nearly as much as it should have been, and certainly not as much as things were in 2001-3.

I think that in part, the reason for all this slowdown is that the ‘Linux in the desktop’ bubble has been already broken and so very few third party programmers want to spend serious time on Gnome/kde. Linux’s high times were between 2001-2004 in terms of desktop hype and enthusiasm. Personally, I find the whole thing pretty sad, because I really like the Gnome desktop.

Update: Thom has commented on the issue in his blog.

Post a comment »

billg wrote on December 20th, 2006 at 2:03 AM PST:

Shows what happens when the primary meme is the freedom available to a developer, rather than a compelling reason to build something. Historically, a salary has always been an excellent way to keep minds focused on the task at hand.

If nothing changes, Linux risks reverting to its status as a coder’s plaything. No one wants to be dependent on software that’s written and supported by developers who can walk away when they get bored and not be replaced by anyone.

Adam wrote on December 20th, 2006 at 3:48 AM PST:

The buzz has died down because both KDE and Gnome got to a point to be usable and beautiful “enough” for everyday use. It’s now fully functional as a desktop, albeit not as tweaked, perfected, and integrated as would be ideal. It’s exciting to watch things develop; much less exciting to watch things mature. Hence, the big buzz and rapid development has tapered.

Moulinneuf wrote on December 20th, 2006 at 5:40 AM PST:

Thanks for the GNU/Linux. ;-)

I will disagree that it was a buzz , there certainly was a great deal of false general desktop marketing put behind Red hat and SuSe. But they still do invest in it but from OpenSuSe and Fedora.

The big desktop player this day is Ubuntu. Also as you pointed out in a previous comment its still the blame game with the OEM that is in effect where the GNU/Linux distribution are waiting for the OEM to pay them to install there software on there hardware.

I don’t think its development is slowing down , but some big projects are receiving less funding and less developer and less code and are making less of an impact. Since you follow them closely you do not see that they are not as important as in the past as other developers and projects are filling the gap.

People are now used to see new distribution come out almost every month and are waiting for the revolutionnary stuff. who know’s what’s coming maybe a completely new desktop … like portland …

Jeff Rollin wrote on December 20th, 2006 at 11:36 AM PST:

Hate to break it to you, but the KDE devs are planning big things for KDE4.

billg wrote on December 21st, 2006 at 1:53 AM PST:

All other issues aside, Linux is hard to find and worrisome to install for most folks using a PC at home. No one sells it or gives it away in the places where they shop for computer stuff. If they do learn that they need to download it, their next hurdle is to decide which of the hundreds of different Linux versions to look for. If they actually acquire a CD or DVD image, they need to burn it, quite likely also a new experience for them. Then, they have to take the scariest step of all: reading about partitioning and learning that they can easily wipe out all data on their drive.

For most people, Linux offers nothing that mitigates that hassle and risk. Before any new OS sweeps away Windows, it needs to to deliver an order of magnitude improvement that parallels the improvements that, say, Mosaic and Netscape brought to the web. Right now, if Windows is Lynx, Linux wants us to believe it is a better Lynx. Ho hum.

Marketing a product, even a free product, needs to involve something other than standing up a website and getting a blurb on Distrowatch. (And people should understand that all the bitter public catfights that take place online between developers, fans, kibitizers, etc., amounts to a generous amount of negative marketing.)

If volunteers went door-to-door with laptops demonstrating Linux, offering to install it on each person’s PC, I’d bet the vast majority would say “No, thanks.”

‘Tis really a pity, this state of affairs. I used Linus for almost a decade, until Ibought my way into OS X. Linux is older than OS X. It ought to be comparable in terms of reliability, standardization enforcement, design, and usability. It could be all that, and more. But, it isn’t. I’m no longer interested in listening to people try to explain that away.

KCorax wrote on December 21st, 2006 at 2:34 AM PST:

@Jeff I just yesterday attended a workshop for open source and spoke with KDE devs. Not a single person could:
– explain what Plasma is, it hasn’t even started !
– say with confidence that Phonon *will* work. We agreed however that arts is dead
– explain why akonadi (a PIM database) exists
– explain why they introduced yet another search engine and not just connect to beagle

@Kristian Gnome is many years behind the Vista Shell. It’s pathetic to see Microsoft dropping the ball for 6 years and having the Gnome desktop getting *simply* up to par. It’s just as pathetic as having Gnome people cutting down on features with the pretext of keeping the UI clean and simple.
From the little I’ve seen I’m much more compeled to run KDE 4 when that arrives than Gnome 2.x, even if the new icons cause me a headache.

Kristian Poul Herkild wrote on December 21st, 2006 at 10:08 AM PST:

Baah… the “ecosystem” of Gnome isn’t the least bit dry. Just because there is no revolution doesn’t mean there is no evolution.

It has played a major catch-up in the early years and are even ahead of Windows Vista in regard to functionality. Rather than spewing out a lot of new functionality every 6 months the development is concentrated on adding in third party applications (to the extent they can be considered an integrated part of a modern DE), and maintaining and enhancing existing functionality while making sure that it still works on older systems.

Why add in a lot of visual crap if it is not needed?

The fact that mostly fixes are getting in Gnome is all the evidence we need to prove everything is working just fine.

Gnome3 can be 5 years away. And who cares? The functionality and ideas one might expect from Gnome3 is already present to some extent in Gnome. And more will be integrated with every 6-month cycle. It’s a slow thoughtful evolution rather than a system-breaking revolution every 5 years. This means Gnome just works.

Jeff Rollin wrote on December 22nd, 2006 at 1:26 AM PST:

@KCorax: Plasma: that’s odd, because Wikipedia seem to know all about it:

Phonon: Well saying that arts is dead is one thing. I doubt anyone can say with confidence that anything will work unless it has actually been tried. That’s why people build prototypes instead of sending civilian passengers up in the Airbus A3XX.

Beagle: Given the MS-Novell deal, if KDE’s search engine does not use Mono then that’s enough for me. Regardless, you could just as well have said, “why didn’t GNOME just switch to Harmony”. The KDE people probably feel like they want their “own Beagle” just as they have their own WM.

I think it’s probably obvious why we need a PIM. It’s worrying that the KDE people couldn’t answer your questions, but maybe what we have is a failure of communication.

Whether you think having the GNOME desktop “up to par” is “pathetic” in 6 years rather depends on how much you thunk they were behind in the first place. Personally, I agree that dumbing down was a bad idea, and it’s probably absorbed much of the momentum.

Nevertheless, KDE 3.5 and GNOME 2.16 are different beasts to KDE 3.0 and GNOME 2.0

This is the admin speaking...
Eugenia wrote on December 22nd, 2006 at 1:37 AM PST:

Jeff, I have requested on my blog to use HTML to represent links, because this has an impact on mobile browsers (it stretches their rendering wide). Thanks.

Bigshot wrote on December 22nd, 2006 at 8:15 AM PST:

firefox 3 is very similar to firefox 1
linux 2.6.19 is so fat away from 2.4

take care of your mind…
a number is only a number… we prefer new feature instad of a new number…

This is the admin speaking...
Eugenia wrote on December 22nd, 2006 at 8:30 AM PST:

Nobody talks about numbers here. We don’t care how things are called. We are simply looking about major new releases and features.

Bigshot wrote on December 22nd, 2006 at 10:05 AM PST:

In italiano si dice che tu sta bellamente vaneggiando…

If you don’t know how is the open source (but i’d prefere “free software”) world you MUST NOT write about this…

shut up nOOb

johnSchaffer wrote on December 22nd, 2006 at 11:17 AM PST:

So, in other words… just because your favorite project has gotten stuck, you decide to throw mud to the one you don’t like. How typical.

Everyone who has a clue knows Gnome has always been technically mediocre, and that one day it would collapse under its own weight because its foundations would not allow it to grow beyond a certain size. Also, anyone can see that KDE development is more alive and thrilling than ever, and that KDE4, with the solid foundations that Qt4 and KDElibs provide, is already in an advanced state in KDE’s svn for anyone who cares to give it a try.

Comments are closed as this blog post is now archived.

Lines, paragraphs break automatically. HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

The URI to TrackBack this blog entry is this. And here is the RSS 2.0 for comments on this post.