staring into the abyss

I suppose I can’t just leave my last post standing there as-is. I’ll start by listing a bunch of things I consider facts about the GNOME project. I don’t want to talk about solutions here, I just want to list them, because I don’t think they are common knowledge. People certainly don’t seem to talk about them a lot.

core developers are leaving GNOME development.

The most recent examples are Emmanuele and Vincent. Both cite the need to look for something different, there is no hard feelings.

GNOME is understaffed.

This is hard to explain in a short and concise way. For anecdotal numbers: GTK has 1 person working full-time on it (me). Glib doesn’t even have that. I think evolution is in a similar situation (a complete email client). We can also try Ohloh’s statistics for GNOME (they include 131 packages, including GStreamer and NetworkManager). You’ll see a sharp drop off of committers on the first page already which suggests around 20 full-time developers at most.

GNOME is a Red Hat project.

If you look at the Ohloh statistics again and ignore the 3 people working almost exclusively on GStreamer and the 2 working on translations, you get 10 Red Hat employees and 5 others. (The 2nd page looks like 6 Red Hat employees versus 8 others with 6 translators/documenters.) This gives the GNOME project essentially a bus factor of 1.

GNOME has no goals.

I first noticed this in 2005 when Jeff Waugh gave his 10×10 talk. Back then, the GNOME project had essentially achieved what it set out to do: a working Free desktop environment. Since then, nobody has managed to set new goals for the project. In fact, these days GNOME describes itself as a “community that makes great software”, which is as nondescript as you can get for software development.
The biggest problem with having no goals is that you can’t measure yourself. Nobody can say if GNOME 3 is better or worse than GNOME 2. There is no recognized metric anywhere. This also leads to frustration in lots of places.

GNOME is losing market- and mindshare.

I don’t want to point out Linus’ bashing, but a bunch of very pragmatic facts that all together lead to fewer GNOME users and developers:

  • Distros are dropping GNOME for other environments instead of working with GNOME.
  • Previous supporters of GNOME are scaling back their involvement or have already dropped GNOME completely.
  • Most important desktop applications have not made the switch to GNOME 3. From talking to them, it’s not a priority for most of them.
  • The claimed target users for GNOME are leaving desktop computers behind for types of devices GNOME doesn’t work on.


#1 Anonymous on 07.27.12 at 04:03

LibreOffice already has a GTK+ 3 port, and Firefox has one currently in development. I don’t know about GIMP and Inkscape.

As for tablets and smartphones, look at the current effort to make a KDE tablet; I think the same kind of project would succeed brilliantly with GNOME. Someone (preferably several someones) needs to step up and do it, and that gets back to the point of your post, but nonetheless I don’t think smartphones and tablets will pose an unaddressable threat to GNOME’s future.

#2 Mark on 07.27.12 at 05:42

Do you have any suggestions?

#3 Oleg on 07.27.12 at 06:49

@Anonymous: From my POV KDE Vivaldi only reinforces the idea that you can’t build interface simultaneously for touch and keyboard. You can prepare foundations for both of them (touch events etc.), but the UIs must be substantially different. Removal of features from e.g. Nautilus because they are not touch friendly is sheer madness^W^W rather questionable, because the aforementioned tablet/smartphone users would not use Nautilus at all.

#4 Robert Ancell on 07.27.12 at 06:50

Thanks for listing these. People do certainly talk a lot about them, at least in the circles I hang out in. It is good to get them more into the public.

“GNOME has no goals” – this is by far the most important issue I think. GNOME has been free-wheeling for too long.

Two other symptons showing at GUADEC this year:
– The number of participants
– The number of sponsors; none of them above silver level.

#5 Jasper St. Pierre on 07.27.12 at 06:59

I don’t think targeting smartphones and tablets is all that useful. Xan talked about this in their “Future of GNOME” talk, and I was completely unconvinced.

iPhone and Android won the Smartphone market, it’s done, they’re moving on, Google with Glasses, Apple with secret R&D.

Nobody would buy a GNOME phone; it’s a relatively unknown brand, and it doesn’t have a Facebook app or Google+ app or Angry Birds and Words with Friends on it.

It would be a waste of our time. I don’t know where GNOME should go, but chasing the industry coat-tails of the already-saturated mobile market is misguided and doomed to fail. Just look at Moblin/Maemo/MeeGo, OpenMoko, WebOS, Windows Phone, etc. etc. A ton of efforts from people better staffed and well-known than us, and they still didn’t care.

If we really wanted to play in the mobile market (we shouldn’t), our best bet would be to just write apps for iPhone/Android, instead of writing a new mobile OS.

#6 Greg on 07.27.12 at 07:28

“GNOME is a Red Hat project.”
“GNOME has no goals.”

Perhaps the goals are known only to Red Hat?

I would suggest making the actual users the target users, before they all switch to Mac.

Oh, wow. Gosh. I hadn’t looked at the about page in forever. Are the target users chimps, small children, or cats?

#7 Sandra on 07.27.12 at 07:47

Not to mention Unity being ported to Fedora.

#8 bochecha on 07.27.12 at 07:53

> “GNOME has no goals.

I first noticed this in 2005 when Jeff Waugh gave his 10×10 talk. Back then, the GNOME project had essentially achieved what it set out to do: a working Free desktop environment.”

That was in fact exactly the point of Jon, Jimmac and Alan’s talk at GNOME.Asia: Beyond Dead Reckoning.

I think they are presenting it again at GUADEC this year.

[WORDPRESS HASHCASH] The poster sent us ‘0 which is not a hashcash value.

#9 ebassi on 07.27.12 at 07:54

Benjamin, I didn’t leave GNOME – especially considering that when I was at intel I wasn’t actually working on GNOME at all.

#10 Matěj Cepl on 07.27.12 at 08:04

I was repeating many times and for couple of years already that one of Evolution and Thunderbird should be abandoned in favor of the another one (I would vote for killing Evo, actually), but nobody ones to leave his project, so now we are in the situation that both programs are slowly dying (with TB it is official now).

#11 Michael on 07.27.12 at 08:39

Well, the part of distro shipping other environnement is a bit inexact. Mint does it because they didn’t want to ship Unity, and they ship gnome, just not gnome-shell.

And for gtk3, i can understand your frustration but how long did it took to switch to gtk 2 ? You also speak of cross desktop application, and I am not sure that gtk 3 is that mature on Windows/Macs ( chicken and egg problem, cross platform apps are not using it, so no feedback, no bugfixes, ). That’s also linked the fact that gtk3 is understaffed, yes, like most low level stuff.

Your points are valid and you are right that something should be done, but on the other hand, I fear that you are maybe suffering from
( ). You look like totally desillusionned on gnome, and that’s worrying.

#12 Crazed Geek on 07.27.12 at 08:46

The Linus G+ link in the last bit is dead. This has the full text, I believe:

#13 RORVI on 07.27.12 at 08:50

Well,firstly,for a tablet i would suggest Ubuntu or Fedora(and here is an example of fedora running on a tablet :
Mostly,i think developers should focus on improving their project,like the response time,a deeper level of integration with social networks,a better graphical experience(better effects,more customization,more options for the users in general). Also,i don’t yet fully understand why linux hasn’t been optimized for smartphones(in my opinion i think this would be a big hit for the large corporations such as Apple;and yes i know that Android is basically linux,but i’m reffering to distros such as Ubuntu,PinguyOS,FreeBSD,Puppy,CentOS, Red Hat). Mostly,i think we will have the freedom we wish for in the smartphone market when we will be able to buy a smartphone just like we buy today a computer and then go home and install whatever operating system we want on that piece of hardware. So,i think that there are plenty of directions towards which gnome could head to , but it’s all up tot the developers(also,i forgot to tell,but another idea could be to replicate the Windows interface,so that Linux can take over the computer market…or make a more intuitive interface for the average user,because Linus Torvalds said that Linux isn’t dominating the computer market mostly because of the small number of preinstalls…).

#14 Steven Don on 07.27.12 at 08:51

Your Google+ link to Linus’ bashing has an incorrect protocol prefix (http://https//) and even when that is removed, the URL points to a non-existent post in another profile.

#15 Marko on 07.27.12 at 08:55

As a user, I started looking for alternatives (Xfce, KDE) when GNOME tried to force it’s inadequate window manager on me. Before, I could either run other windows with the gnome panel, and the default was also quite usable (not as good as KDE, but good enough, and the whole environment was cleaner). The current system with two level app/window hierarchy is unusable for me, and if there’s not a trivial/obvious way to turn it off immediately and completely in core system (no extensions), the GNOME is gone from my desktop. On a tablet the current default might be almost acceptable, but for desktop it’s not.

Also, I don’t like the top menu bar, otherwise I might consider using a Mac.

#16 Wandspiegel on 07.27.12 at 09:02

This would have worried me back when Unity was crap a year or so ago. Now Unity is actually good, it doesn’t matter if Gnome dies a slow death; we’ll still have multiple good alternatives that are still being enhanced.

#17 Claudio on 07.27.12 at 09:05

Just wanted to point out that you’re missing several talks and discussions that actually addressing some of the problems you’re mentioning.

#18 zu on 07.27.12 at 09:09

keep working on gtk
gnome is the issue, other wms use gtk. its a good toolkit.

#19 Rafael on 07.27.12 at 09:14

This just made me so sad. If not for Gnome, I probably wouldn’t have gone past a month “trying” out Linux, and it’s been six years now.

#20 Ivan Drinchev on 07.27.12 at 09:44

You have dead link for ” Linus’ Bashing “.

I haven’t seen anything better than GTK+ in the open source community. Since QT is out of the game, do you have any suggestions for the next Graphical Toolkit?

#21 Peter on 07.27.12 at 09:48

I think there’s some misleading statements here. You mention distros moving to other environments and linux Cinnamon but it’s all based on gnome but isn’t using the default shell. It uses all the other gnome technologies. I don’t see that as “moving away from gnome” just like there’s some components of the existing shell that would need adjustment for tablets to be finger friendly.

OLPC is using gnome in their laptops. There’s nearly 4 million of those deployed globally. They also use most of the gnome stack in the sugar UX.

Evolution has a number of developers also, two at least at Red Hat and Novell/SuSE use to have more than that although I don’t know what is the case there now.

OLPC employs a few developers to work on GNOME tech as do a number of other companies such as collabora / codethink and others such as webkitgtk etc. Intel still has a team of people working on the dev team with gnome based tech such as clutter even though they’ve had departures.

I do happen to agree though that there is certainly a lack of leadership and direction though. I think the org needs to lose the one UX to rule them all because that’s clearly pissed a lot of people off even if I do like it for a laptop UI I don’t think it suits other form factors.

#22 Zhou Dong on 07.27.12 at 09:53

GNOME is not a good product any more. It has not many tendril as other softs. Soft with tendril is better.

#23 a on 07.27.12 at 10:18

The link to Linus’ post is broken. It’s “https://” not “https//” and even when that’s changed, it says the post could not be found

#24 Max on 07.27.12 at 10:24

Tablets and smartphones are a strategic goal. Go for it.

Even though a free desktop is only part of what’s needed – you also need a kernel and a video card driver and/or server – it would be a pivot part for a transformation from “monolithic” to “assembled” tablets and smartphones.

Can I imagine a future where a customer buys a tablet or smartphone and, as we do nowaday with PCs, chooses which operating system to install on it, and which desktop environment she prefers?

#25 Philip on 07.27.12 at 10:46

Really? 1 developer on Gnome and 20 on Gtk? Amazing… I thought the numbers were tenfold higher. (Such low numbers make it actually more interesting to contribute IMO.)

#26 Juan José Sánchez on 07.27.12 at 10:53

Xan López and myself tried to do an analysis of the current status of GNOME (quite aligned with what you say here) yesterday in our “A bright future for GNOME” GUADEC talk. Taking as starting point the good and bad things the project has, and current industry situation and opportunities, we tried to propose some generic and not that generic ideas about how we could move forward and make GNOME more relevant.

The executive summary is that we think that we need an attractive and ambitious plan for the next 2 years, with ‘More GNOME 3′, ‘GNOME targeting new form factors’ and ‘GNOME as an OS or final product/UX’ as the 3 main pillars. The plan would a way to keep resources inside the project and hopefully recover/attract more, and we would need the support of the board, release team, and the main maintainers, developers and specially of the companies currently investing in GNOME to make it feasible.

The idea is to continue this discussion on Monday, during the full-day GNOME OS BoF at GUADEC. It is really a pity that you cannot be here and contribute to the discussion face to face.

Thanks for this post, anyway. We need honesty and to acknowledge the situation as the first step before deciding what comes next for the GNOME community.

#27 Zorglub on 07.27.12 at 11:20

Glad I not the only one seeing the trend.
To me, one of the aim of Gnome-Shell was to attract more developer, especially webdev, and it’s a big fail. It’s driving core dev away and upsetting end user.

Unity is on the same path. Here some end-user feedback : 30 Ubuntu installed here at work (not a tech company), users juste plain hate Unity those environments. We had to revert to Gnome Classic for most of them.

Some ideas and redesign are very good, but the no compromise rule goes way too far.

#28 Anonymous on 07.27.12 at 11:23

“Make things as simple as possible, but not simpler.”

That’s the major mistake the gnome team did. Being arrogant to almost everyone and simplifying the GUI to the point that the whole system became user unfriendly and generally unusable. Now we have a lot of forks, the one uglier then the other. Hello unity! Code bloat, more spaghetti code, shell extension nightmare. The “Gnome” simply made a suicide.

The worst thing is, there is currently no sane desktop alternative for linux…

#29 juan234 on 07.27.12 at 11:36

When GNOME makes such a fundamental change in direction / philosophy from v2 to v3, should we really be surprised when users reject it?

Gnome 3 should have been a fork. Power linux users do not want their desktop dumbed down for touch interfaces. I’ve never done any Desktop Environment development before, but I’m so mad about all this that I have been investing significant time in getting up to speed with Cinnamon so I can contribute in some fashion.

#30 grubz on 07.27.12 at 11:39

i have a suggestion… switch to xfce.

[WORDPRESS HASHCASH] The poster sent us ‘0 which is not a hashcash value.

#31 Jimmy on 07.27.12 at 11:45

Your statement that, “Nobody can say if GNOME 3 is better or worse than GNOME 2″ contradicts very directly with fact that, as you put it, “GNOME is losing market- and mindshare”.

The community has spoken. They’ve spoken very loudly, for a long time now, even while GNOME 3 was in development. It’s too bad that nobody involved with GNOME was listening at the time, nor apparently listening now.

Almost nobody likes GNOME 3. It’s the pariah of the open source desktops. It’s clearly inferior to GNOME 2 in basically every single way. GNOME Shell is among the worst open source disasters of all time. The complete dumbing-down of so much of the environment is absurd to us long-time GNOME users. What was once a usable, productive environment has been turned into a horrid experience.

So it’s no wonder that users, and people otherwise involved with GNOME, are fleeing quickly. Nobody wants to use such a failed project, and nobody wants to be associated with such a failed project.

#32 Bas Burger on 07.27.12 at 11:50

Not too long I picked up the need to program a gnome-shell extension and noticed 2 important negative issues.

Let me first state that GJS and the gnome-shell are good efforts to gain more developers but if you really want more developers i suggest you provide the developers world some good PHP bindings to the introspection.

Now the 2 negative things I noticed, the first one is a complete lack of unambiguous documentation, to gain some insight I have to collect documentation from all sorts of places and often there are differences enough to make you search for the right way to do things.

Try to complete the libraries, In the short time programming for gnome-shell I noticed several incomplete parts of libraries,

The reason why PHP never has problems people using it, is that the developers let people use PHP, even encourage them by making things easy like this:

My main conclusion is that if you want people to use your stuff, you actually have to enable people to use your stuff,
for various reasons people are not able to use your stuff even if they want to.

Success of finding answers, I hope you do find them as I would like to continue using Gnome.

#33 Benjamin Otte on 07.27.12 at 11:53

@Jimmy: (singling you out because you’re the last one to say that – a lot of comments point at this issue)
You assume 2 things with your statement: (1) That GNOME3 wants to retain GNOME2 users and (2) that market share should actually be a goal of GNOME. Both is reasonable to assume, but I don’t think it’s necessarily true.

#34 Jimmy on 07.27.12 at 12:00

Max, it’s the focus on netbooks/tablets/phones that has been the greatest killer of existing desktop environments.

It’s absolutely stupid to think that the same environment should be used on small devices like phones and tablets, and on netbooks and laptops, and on desktops and workstations.

The screen sizes (and number of screens) are hugely different. The input methods are vastly different, as well. The way people use such environments is different. Trying to have GNOME support all of these is a disastrous idea.

Notice that the most successful open source desktops today, KDE and Xfce, have put limited emphasis on mobile devices. This has served them well, because it hasn’t absolutely destroyed their usability on desktops and workstations.

Putting the tablet hype aside, the fact is that the vast, vast majority of people still use desktops day-in and day-out for any productive work. This traditional environment is the one that truly matters. This is the environment that GNOME should focus on.

#35 Jimmy on 07.27.12 at 12:10

Benjamin Otte, what exactly is the point of GNOME 3, then?

If retaining GNOME 2 users doesn’t matter, then why call it “GNOME”? It’s clearly an environment that is completely unlike GNOME 2 in terms of features, functionality and usability. So why call it “GNOME” at all? Why make it sound like it’s a successor to GNOME 2, if it really isn’t?

And market share is essential to the sustainability of any open source project, especially one as complex as a desktop environment. Having fewer users means that it’s much less likely that some of those users will also be software developers. When a software project lacks software developers, it rarely moves forward at a reasonable pace, and rarely provides any value.

If GNOME 3 wasn’t about improving upon GNOME 2 and maintaining, if not growing, its existing user base, then its only goals must have been:
1) Alienate as many users as possible.
2) Alienate as many developers as possible.
3) Provide an unusable, widely-disliked desktop environment.

If those were in fact the goals of GNOME 3, then I’d say that they succeeded completely in fulfilling each and every one of them.

#36 klapauzius on 07.27.12 at 12:51

Face it: Both Ubuntu (with Steve Jobs-wannabee Shuttleworth and Unity) and GNOME only want one thing: To become like Apple. Glitzy, colourful, animated desktop animations. Icons that hop around, notifications popping up all over the place. The feeling that “I AM IMPORTANT !” “I AM COOL !” “LOOK AT ME !” “LOOK AT ME NOW !” “DAMN, LOOK AT ME NOOOW!”. That’s OS X think.

Users want a working, fast, stable, low-footprint, easy-to-use Desktop environment. GNOME 2.x was pretty much perfect in the end. But declaring it finished, going on fixing bugs, adding a little bit here and there (like XFCE does) doesn’t make you important. It’s not cool. It’s not something you can blog about or post it on Facebook.

GNOME 3.x is simply unnecessary. There’s no need for it in the userbase. Dump it.

#37 Fewt on 07.27.12 at 13:06

I plan to fork GNOME 2, similar to MATE but with the goal of improving what we already have. You are welcome to jump ship, I can use all the help I can get.

#38 Patrick Avella (@patrickavella) on 07.27.12 at 13:11

Developer entropy affects every open source project, and I’ve heard the concerns you’ve listed here echoed out about many open source projects, current and past.

I do really hope the Gnome project pulls through. I love gnome 3. I’ll be keeping my hopes up. Keep up the great work that you do!

#39 Alexander von Gluck on 07.27.12 at 13:16

The biggest mistake was redesigning the wheel with Gnome 3 and completely dropping Gnome 2.

Gnome “worked” as a desktop, for those who didn’t want shiny bubbly KDE 4 in their soup… they used Gnome. Gnome was fast, feature-full, and stable.

The other nail in the coffin was Gnome 3 coming out at the *exact* same time as unity. Suddenly desktop users had their entire world shaken up… with no alternatives.

All reasons why I switched to XFCE. How you you be confused about the failing of a project after making a drastic change few people want? You’ve alienated your base.. You listened to HIG of 12-year olds without any concept of engineering, and turned out the mess that is Gnome 3.

It’s not about being shiny and redoing every aspect of gnome, it’s about getting work done.

#40 Robert Swain on 07.27.12 at 13:18

Regarding other desktops quoted, whether forks of Gnome 2 or Unity, neither are as slick or polished as Gnome 3. As a GStreamer developer, from my perspective, moving to Gnome 3 has produced a lot of modernisation of the UI and platform that now makes it, for me, something that is there, looks good and otherwise gets out of the way while I’m doing my work. I think Gnome 3 and shell kick ass. Unity does not.

The lack of developers for the desktop however is an interesting issue. And the lack of goals.

#41 romu on 07.27.12 at 13:23

Hi there,
I came here from OSNews as a Gnome enthousiast.

I agree with some oints expressed here and disagree with others. Please take this a long term Gnome user, with some software development skills but not at all on Gnome.

First, to me, Gnome 3 (the Shell indeed) is the most DE ever made. To me, it’s really a beautiful piece of software from a user point of view. Have been able to integrate all dock, virtual desktops, applets in the single “Super” key, is really great. This is to counter balance the previous negative feedbacks. And, for using Gnome & Unity on a daily basis, Gnome Shell is far superior to Unity in my opinion as all is very smoothly integrated and Gnome Shell seems to perform well with less hardware resources, but this is needs more analysis.

But I agree with some previous comments, Gnome should not try to make the universal graphical interface. There will never have Gnome smartphones or tablet, too late. So commit now to a full “desktop” orientation appears the best thing to do in my opinion.

As I said I’ve some software development experience. And as I have some software ideas, I’m looking at how I can develop Gnome apps. My preference comes to Javascript because if I don’t really love its too free syntax, it seems an ubiquitous language.

I’ve read all tutorials on the Gnome Develop web site. And know, I’m at the point where I don’t even know how to write the code to just create a simple text file. So I’m trying to digg into the Gnome source code to get some examples, but honestly, I’m getting close to give up. Unfortunately searching the web doesn’t work as there is too few very simple tutorials for Gnome which covers a wide range of use cases.

Another comment from a developer point of view, is the lack of a good development environment. I’m not a big fan of M$, but just consider how you can make a simple working “hello world” windows app with 3 mouse clicks ! Making possible to write apps in Js with GJS is a step forward to ease the Gnome dev, but Anjuta doesn’t support Js debbuging (I failed trying to make it work), and Gedit, even if simple and stable, lacks some modern IDE features.

My 2 cents.

#42 Luis Medinas on 07.27.12 at 13:31

People, face it – GNOME was a UNIX Desktop Environment, not sure what it is now but it clearly lacks direction. I don’t know if i belive anymore on Linux Desktop besides Ubuntu and Canonical now. Yes they don’t contribute for FOSS in general with code as Benjamin mentioned before but at least they got it! Look at the current feature they announced… the WebApps, Epiphany has that feature for 1 or 2 releases ago but the point is that nobody integrated that feature in the Desktop.

I dont think Qt is out of game already because it’s just a project like GTK+ and i don’t see the whole Qt community let it die because Qt is really good.

As a GNOME Developer (not sure if i still am) i lost the will to contribute to GNOME in MY FREE TIME.
Sad to see things like they are so i share the same opinion as you Benjamin.

#43 JO on 07.27.12 at 13:36

IMHO I agree that “The worst thing is, there is currently no sane desktop alternative for linux…”

I don’t like Unity at all so I’m hoping Red Hat will do something to secure the future of the Linux desktop.

Start from scratch but make it good. Preferably using Gtk because of the number of applications… or Qt.

Maybe put some effort into the classics like Window Maker or OpenBox?

Don’t radically change UI’s or limit options because of tablets and smart phones. Extend from the bottom and create two different clients if necessary. Are developers and old timers supposed to drag and drop files by touching the monitor on a desktop or laptop computer? This is silly.

The Evolution client should be buried as soon as possible. Don’t spend developer time on web browsers or mail clients, there are plenty good alternatives.

#44 danw on 07.27.12 at 13:37

I would say that GNOME 3 represents the first time since 2.0 that GNOME *does* have goals. The problem is, GNOME has never been a community of people who all wanted the same thing (see, eg, the “we need to innovate!” / “no, we need to stop innovating and stabilize!” pgo flame war from a few years back). And so as GNOME’s goals become clearer, those users, developers, and corporations who wanted something else are leaving to try to find that something else somewhere else.

#45 Richard on 07.27.12 at 13:42

I feel most of the Linux DE projects could do with a little self reflection to decide where they are going and what they want to be.
I do sometimes feel GNOME has become a project where Design is considered before Function
Conversely, KDE still to me seems to focus on Function before Design
Any the others are niche players slotting into specific design or functional goals

Perhaps the bigger question is “What future does the Desktop have?”
Linux, Windows, OSX, there is a justified questionmark on how personal computing might look over the next 5-10 years.

I think its likely that both traditional desktops and laptops will fade into a more secondary role, the preserve of ‘creators’ (be that of code or digital media), where as it seems destined that tablets, smartphones and similar appliances will become the primary devices of the unwashed masses, who require their computers primarily for consumption and communication

(I don’t think this is a good thing, it’s just how I see things moving)

If that’s the future of the desktop, is the direction of GNOME, or KDE, or for that matter any of the Linux DE’s really that bad? They all generally cater for their developers before the general public.
GNOME 3 is a fine choice for a developer/techie who wants a clean, clear UI with minimal distractions (a role I personally think it excels in)
There’s plenty of other viable options for people who want or need something else

#46 Lars Vogel on 07.27.12 at 13:51

I always assumed Unity was using GTK+ as their underlying engine and that hence Unity was a good thing for Gnome. I guess I was wrong.

#47 Hussam Al-Tayeb on 07.27.12 at 13:51

Robert Swain, sure it is a modern desktop. It’s probably a very nice thing for home users too. But it’s a step in the wrong direction for open source desktops.
Linux is not longer some toy. it’s a professional operating system. Do you think someone is going to run gnome-shell or unity at work?

#48 Test on 07.27.12 at 13:52


What about sitting down and defining new goals for Gnome 4 and ask the question:

-What do users want.
-What do developers want.

And establish what Gnome 4 can give them and try to do that. Maybe this is not the Shell by it self? Maybe it’s desktop shell for the desktop and mobile shell for the mobile?

Maybe it’s unified composition manager for all DE? I don’t know but i think first step is asking what user and developers want and then start working on it and maybe find new sources for income like gaming industry?

#49 eric-yorba on 07.27.12 at 13:56

What worries me the most is the trend toward “simplicity,” which is not only an ill-defined goal, but is often an excuse for laziness. And yes, maintaining features is hard, especially when you’re on a moving target like the Linux desktop; but just because something is difficult is not a reason in and of itself to not do it.

Users don’t like having the rug pulled out from under them; we use computers primarily with muscle memory so it’s jarring to us when our old familiar workflows are broken.

When things change too much, it scares users away. Why would I re-learn my workflow with a new version of Gnome when I could re-learn it on a more stable platform?

#50 Xan on 07.27.12 at 13:59

I don’t think targeting laptops and workstations is all that useful. Icaza talked about this in his “GNOME” proposal, and I was completely unconvinced.

Windows and Mac won the desktop market, it’s done, they’re moving on, Windows with Windows XP, Apple with secret OSX R&D.

Nobody would buy a GNOME PC; it’s a relatively unknown brand, and it doesn’t have an Office app or Photoshop app or Diablo.

It would be a waste of our time. I don’t know where GNOME should go, but chasing the industry coat-tails of the already-saturated desktop market is misguided and doomed to fail. Just look at OS/2, BeOS, KDE, Amiga, CES, etc, etc. A ton of efforts from people better staffed and well-known than us, and they still didn’t care.

If we really wanted to play in the desktop (we shouldn’t), our best bet would be to just write apps for Windows/Mac, instead of writing a new desktop OS.

#51 Felipe Contreras on 07.27.12 at 14:06

I predicted this was going to happen; decrease in market-share which would result in less contributions, and I explained exactly why. GNOME developers ignored me.

The solution is simple; listen to your users. I tried to push for this in various ways and ultimately I basically had to do the GNOME user survey by myself. Unfortunately I haven’t had time to analyze the data, but judging by the unfolding of events it should not be any surprise to anyone.

But you mentioned that the 10×10 goal was reached, but it obviously wasn’t; GNOME global market-share is nowhere near 10%, not even 1%. This was a tremendous failure, and instead of taking a hard introspective view at the project, you ignored this and acted as if everything was OK.

This hints as to the problem with the projects; wishful thinking, dogmatic views, and circle-jerking. I doubt think is going to happen, as I’m sure most GNOMErs deny there’s even a problem.

#52 Mike on 07.27.12 at 14:08

I agree with your blog post, but I have to strongly disagree on the last point:

“The claimed target users for GNOME are leaving desktop computers behind for types of devices GNOME doesn’t work on.” *I* am leaving GNOME behind, because developers think exactly that and I can’t properly use GNOME on my desktop anymore. It’s just to complicated to get shit done with it since 3.0, compared to other _desktop_ environments and earlier versions of GNOME.

And I guess it has something to do with “GNOME has no goals.”

What happened to the “desktop environment” GNOME? It has transformed into a “tablet environment” with big buttons and bad usability everywhere. Saying that the target audience of GNOME is running away for tablets seems to me like a distorted perception. Just because people buy tablets doesn’t mean they don’t use their computers anymore.

GNOME should rethink its goals, perhaps add a clear mission statement. Does it want to be a desktop environment? Than stop doing all that finger-touchy-looky-stuff. Does it want to be a tablet/toy environment? Well, then call it that and do it right. You can’t have both, though, and this is because of your second point: “GNOME is understaffed.”

#53 Mat on 07.27.12 at 14:12

I think a GUI should have two primary functions, symbolic abstraction and application service- both are subjective to the user. As there exist difference learning types, a efficient work flow depending on identification of user interactions with a combination of inherent comprehension and cultural precondition.

Conclusion: A generally usable GUI shoud support at least alternative waysof interaction for different requirement without hiding. One would not want to educate people on a specific behavior for establish GUI´s !

Thats the reason why KDE and other projects are prefered (with exaptions of course). There offer the ability of specific adaption an easy to understand way – it´s so simple.

#54 alex on 07.27.12 at 14:31

I think gnome has a good chance to establish if finally C for high-level projects is ditched and if the things get cleaner and smaller this way.
For example evolution: nice program except the fact that the data saving model they use(d) is horrible. Using gconf for saving addressbooks and tasklists is just insane.
An other example: python:
why isn’t python2 already ditched in favour of python3? (ok, I know: lack of man power and priority) But Python isn’t so small you can have dozen versions of it. Sadly I have no time in the moment for porting.

#55 ocrete on 07.27.12 at 14:37

Maybe the reason third party applications don’t care about GTK 3 is that you broke compatibility, but without introducing any compelling features..Maybe once you have completed the clutter/GTK merge there will be more compelling reasons to use it, but that’s gonna be GTK4. If I were an application developer, I might as well skip GTK3.

#56 Dan on 07.27.12 at 14:47

This post on Reddit does a fantastic job of addressing many of the points made in this article:

#57 adicahya on 07.27.12 at 14:53

I think all the open source desktop environment have the same problem (both Gnome and KDE).
They are all build with classic desktop point of view, in term of concept, UI element,”mouse” interaction and connectivity.

I think now the game is changing. PC is no longer what it use to be, its connected, social and be part of some ecosystem (services, cloud, application store, linked to devices etc).

This whole thing change the consumers behavior. They want or getting used to that ecosystem thing. This thing make open source desktop less and less attractive.

IMHO..all the open source desktop community should make something called “open source ecosystem” (free is no longer enough, facebook is free right?).

In this 4th ecosystem (beside Apple, Microsoft and Google), you either build some sort of application store, social interaction platform (diaspora?), cloud storage (ownCloud?), browser etc. Your advantage maybe the transparency, but ordinary people might don’t care about that either.

Then, build a software around that. Build a new modern computing (not desktop) environment that can be used across new PC form factor (translate: touch interface).

If we go further, think about design your own hardware (hey, KDE did that). Design and build your own hardware to match your software.
Make it an open source tablet or ultrabook :)

In the other hand, i think Ubuntu has taken this step, so does Mozilla with FirefoxOS…

I know, is a tough world today…but like in nature, all must adapt to survive.

#58 pinky0x51 on 07.27.12 at 14:55

It was quite an interesting article and also comments to read. Many people say that nobody wants GNOME3 so I need to raise my voice: I really like GNOME3!

For me GNOME3 clearly shows a vision. In my view GNOME3 is the only free Desktop which can compete against proprietary solutions. It looks beautiful. Everything fits together, visually and functionally (IM, GNOME contacts, online accounts, the new discussed “setup wizard”,…).

At some point the designer might be to keen about touch and mobile devices. But this isn’t something which could be adjusted. Still I have to admit that at the moment I prefer XFCE for tasks where I need a lot of windows (e.g. software development because XFCE makes it easier to distribute terminals, etc on different workspaces and access them). But as a general every day desktop GNOME3 is really great. And I think that’s what most people are looking for.

What really concerns me is the number of developers. But to be honest that’s not really surprising to me. I’m also a software developer and I really understand the comment from romu. The last few years my main programming languages/frameworks where: python, C++, Qt, PHP.
From time to time I look into GNOME/Gtk+ apps because I would really like to help GNOME and also XFCE. But what I see doesn’t look like fun to me. Writing graphical interfaces in C just looks like a lot of pain. There is no real documentation what I should use, plain Gtk+? Clutter? C? JavaScript? Which IDE should I use? I really appreciate the work of the Gtk+ developers, the XFCE developers, what the GNOME3 developers have achieved with GNOME3 and all the developers of all those great Gtk+ programs out there. GNOME3 (and XFCE) are great user platforms with a slightly different audience. Many Gtk+ apps are the best programs in their field. But I’m really not surprised that it is hard to get new developers for the development platform.

#59 Stan Steel on 07.27.12 at 15:00

I agree with all the above comments that suggest Gnome should have stayed focused on its core competency. The desktop. You don’t have to compete on every platform. I would make the claim that an overwhelming percentage of people who use Gnome on Linux use it as part of a workstation and they use this workstation to solve problems; not for entertainment.

#60 orclev on 07.27.12 at 15:09

Personally, I liked Gnome2 and used it as my primary window manager for years. When Gnome3 came out it was a complete disaster (as others have commented), although it’s my understanding that it has since at least had most of the bugs fixed (personally I think the UI is still a complete disaster, but that’s more of a preference). However, in the interim I’ve switched to using a tiling window manager (XMonad to be specific) and really have no interest in going back to a more traditional window manager. If I did I’d probably switch back to something like XFCE or Blackbox, because they do everything I need out of a window manager (that is, manage the windows) with a relatively light footprint. Gnome3 had some interesting ideas I think, but in practice it didn’t work out very well.

#61 Billy on 07.27.12 at 15:12

Gnome has the opportunity to be the elegant, simplified free desktop. Unity is trying too hard to be OS X.

#62 pibarnas on 07.27.12 at 15:21

Have you tried to use gnome3 with a 20 inch or above screen? And the trial to be a semantic desktop? Type to say what I want? Is that really the advance? Well, terminals have been done it for 60 years. I think the egocentric way to dictate the most useful and intuitive UI went wrong. I think it was too complex to make things like gnome2 (what cinnamon appears to be proving it was wrong too) or it was a selfish step to wipe out all old configs, but they underestimate the most important person in that scene: the user. Because the dev first is an user in my opinion. I leave gnome behind to use lxde (with a solid openbox wm, what metacity or other will never be) or xfce, which seems to catch all gnome fans in the process, or maybe mate in the future, it it shows itself viable. It’s a pity.

#63 Jason on 07.27.12 at 15:25

The worst part of all of this is that everybody had seen it coming, and here it is.

I hope this is the beginning of the end for Gnome, Gnome was the de-facto Linux DE and it carried with it the responsibility of such, alienating the users was unforgivable.

#64 mark on 07.27.12 at 15:33

Please do not dare to complain ever.

You Gnome developers ignored the users FIRST.

Of course you developers now chase after the money, after having ripped about the users.

I will remember the names of all those traitors who ignored users.

#65 Dagmar d'Surreal on 07.27.12 at 15:50

Dropline isn’t exactly “my” website, but it’s being mentioned to provide some context of where I’m coming from. For over a decade I have been using and have packagized GNOME more times than I can easily count–so I’m no fresh daisy here. Unity was a mistake. A big one. A mistake that was a slap in the face to many users.

I’m astonished you can’t seem figure out that a huge part of the problem was the change to GNOME3 and Unity. Don’t try to blame Linus’ “bashing” in some passive-aggressive way. His comments were pretty accurate.

Basically, the decision to essentially force everyone to use Unity and GNOME Shell was a huge error, counter to very many of the things GNOME has done in the past. Sure, if you had a touchscreen there was some extra utility there, but if you (like 99.9% of the existing userbase) didn’t have one, most of Unity’s new functionality didn’t make sense and more importantly, annoyed the hell out of you. This wasn’t a matter of “users fear change”–the thing just pursued minimalism and looking pretty single-mindedly enough to ignore that useability was getting tossed by the roadside.

Compounding this error was that it didn’t appear that anyone was even trying to keep the GNOME2 stuff in a usable state in the poster-child distros which are frankly how the majority of the userbase are getting the software–because even they too were “sold” on getting people to use Unity. Linux people might not be *entirely* about choice, but they sure don’t like it when you take it away from them as was done with this. The message was clear and we saw it over and over–if you didn’t like Unity, that was just tough and it was also frequently implied that you were in some way merely incapable of learning a new interface.

So, when you alienate a whole bunch of your users, who happen to also be your developers, you really shouldn’t be surprised when everyone (developers included) starts running for the hills.

#66 Martin on 07.27.12 at 16:07

I have to admit that I switched my private desktops from GNOME to XFCE (Debian testing), because I got depressed by the dark background and no option to configure another colour :~( It sounds so trivial, but such is life. I still use GNOME2 on my office desktop (Debian stable) and like it better than XFCE.

Anyway, all applications I use use GTK+ (Thunderbird, Firefox, Quod Libet, Gimp, dia, inkscape, wireshark, even Emacs), and both GNOME and XFCE (and LXDE and Cinnamon) use this great library. So maybe all the users of GTK+ should be involved more into working on it?

[WORDPRESS HASHCASH] The poster sent us ‘0 which is not a hashcash value.

#67 anonim on 07.27.12 at 16:10

I think there’s space for gnome to compete in other form factors, thinking that there’s no room for an alternative is just giving up before even trying. Nokia’s N9 has proven that there’s a lot of room for competition and lot’s of people willing to use something other than apple/google.

Another battlefield gnome should work is providing the tools for contributors to join more easily, I think here gnome should learn from others and provide a more integrated enviroment. IOS/OSX, Android, Qt all have very good tools/docs to get you started and have something working in no time, gnome should catch up.

my 2 cents.

#68 Robert Swain on 07.27.12 at 16:30

Hussam Al-Tayeb: I think people would likely be using Ubuntu professionally and therefore would be forced to use Unity. However, I don’t think Unity is anything like being a better visual environment than Gnome Shell.

I use shell during my professional working hours and I’m very happy with it, even since the first release. I find Unity clunky for my work-flow. I find that Gnome Shell does most of what I want a desktop environment to do while I’m working – not really be noticeable at all.

Gnome 3 and Gnome Shell have been a step in a different direction from Gnome 2 and panel, sure, but it’s a step in a direction that I like and a step in a direction that I find much more usable, much more attractive when I do pay attention to it and it gets out of the way.

If I had a feeling that Gnome doesn’t have goals that lead me to state that that part of the article was an interesting point, I think that comes from me being quite satisfied with Gnome 3 and Gnome Shell and where does the desktop itself go from there without doing unnecessary things?

Canonical’s WebApps looks like another way of launching a web page in Firefox to me, but I personally think that’s moving in the right kind of direction. I think web applications are the future – whether local using standardised web APIs to interact with local hardware, or remote.

Embracing that we (can) use the web for basically everything above the platform and that writing applications using standardised web APIs is a valid and smart option (very easily portable applications!) is probably the direction the desktop will take.

I expect devices we use professionally in the software development world will persist to be more desktop-like, but perhaps something suited to our needs can evolve in a Firefox OS-ish environment. Time will tell I expect.

#69 Ryan Fugger on 07.27.12 at 16:48

As a developer, I’ve really appreciated how Unity and Gnome Shell have removed a lot of noise from my screen and put it in a meta-place. It’s forced me to use the keyboard shortcuts I should have been using all along, except that so many options were on my screen all the time I was continually drawn to use the mouse. I don’t think that they represent a dumbing down of the interface, and actually think they’re better suited to power users.

I can do without all the social media/chat integration though.

#70 Alexandre on 07.27.12 at 16:51

For god’s sake, if you are a GTK+ developer, why on earth are you talking about purely GTK+ apps in terms of GNOME?

And no, both GIMP and Inkscape actually can use GTK+3. For GIMP it started during a GTK+ hackfest (should sound familiar to you, no?) over a year ago. Inkscape has optional support for GTK+3 since June.

A bit of reasearch never harms :)

#71 Sam Thursfield on 07.27.12 at 16:53

To be clear, nobody in GNOME has a particular desire to kill GNOME 2. Nobody really has a desire to maintain it either, but contributors who would like to spend their time modernising the code in the 2-24 branches of the various components and continuing the 2.x series it would, provided the job was done well, be welcomed. I hope the MATE project achieve just that.

#72 bart9h on 07.27.12 at 17:33


Thanks for what you have done in the past.

Fortunately, MATE saved the world.

#73 John Smith on 07.27.12 at 17:51

As for me I switched from Gnome to XFCE. Why? Because Gnome 3 seems to be mostly focused on tablets and smartphones but in no way it’s suitable for my desktop. Overgrown window button? Just one? Inability to set display DPI? No obvious way to shut down my PC into the mode I want? Gnome guys are kiddng. It’s really loses to Gnome2 and even to lightweight gtk2-based XFCE. No, really, Gnome3 environment for retards by default. So let’s retards to use it if they want.

And the the UBER EPIC FAIL is: Gnome3 devs put so many efforts in tablet/smartphone UIs but … where can buy tablet runnng Gnome UI? Don’t you mind making one? It’s uber-stupid. And KDE guys seems to be much smarter in this regard. They both understand that desktop PC is not a tablet so they have 2 kinds of UI. And they’re about to create KDE tablet after all.

#74 Stine Kristiansen on 07.27.12 at 17:53

I used to love Gnome. Tried KDE and others but Gnome was perfect. Then Gnome3 came and removed many of the things that I loved. I could cry but why did Gnome3 remove the feature of vertical panels at the same time the PCs started to get lower screens? These things combined really break my ability to work as efficient as I used to.

[WORDPRESS HASHCASH] The poster sent us ‘0 which is not a hashcash value.

#75 particleSwarm on 07.27.12 at 17:59

I’m now using MATE. MATE is essentially GNOME 2. I think the GNOME team should embrace MATE, drop GNOME 3 and continue from there.

#76 After two weeks of using GNOME 3, I officially hate it « Felipe Contreras on 07.27.12 at 18:05

[…] Staring into the abyss […]

[WORDPRESS HASHCASH] The comment’s server IP ( doesn’t match the comment’s URL host IP ( and so is spam.

#77 icy on 07.27.12 at 18:08

Gnome is really beautiful, i love the new feel how to use it. But beeing forced to use this “new way of using a desktop” sucks a lot..

#78 khisbeos on 07.27.12 at 18:14

I hope this will open the discussion about the future of GNOME. Xan made a very valid (and kinda funny) point. I’m not a UX expert but I think the same UI can’t work on desktops and on tablets.
Anyways, I want to say good job, GNOME developers; we can love or hate GNOME3 but it’s a very complex and time consuming work, and it’s very well done.

#79 LorgnocOspYe on 07.27.12 at 18:31

People who are satisfied usualy don’t come and comment how satisfied they are. The most vocal are those who are not satisfied. That is why you see more complains than praises about GNOME 3; it doesn’t mean nobody likes it.

And, because I see many commenters claimiing nobody likes GNOME 3, which is not true at all, I feel the need to leave the kind of comment you usually don’t see.

GNOME 2 looked like a Windows and Mac OS copycat and sucked!

GNOME 3 is the best DE avaiable for Linux and is amazing!

If it was not for GNOME 3 revolution, GNOME would stay stagnant and irrelevant; and even more irrelevant as the desktop computers become less relevant than new form factors.

But now, GNOME is inovative and promissing. When you see machines on the market shipping GNOME OS, competing with Windows 8 machines, I will tell you: I told you, but you ignored.


#80 Один из разработчиков GNOME обобщил проблемы в развитии проекта | — Всероссийский портал о UNIX-системах on 07.27.12 at 18:35

[…] Отте (Benjamin Otte), участвующий в разработке GNOME, опубликовал заметку с критикой ситуации, сложившейся в проекте […]

#81 Mario on 07.27.12 at 18:36

QT out of the game. I doubt that. QT is already working in Wayland, and years ago an agreement was made that if the holder of the commercial QT went down, QT opensource will stay free for development. A toolkit that can run on the 4 major systems (Windows, Mac, Linux, BSD), and also being ported to some of the smart phones and is out of the game…. please.

#82 Greg on 07.27.12 at 18:42

“People who are satisfied usualy don’t come and comment how satisfied they are.”

Except for things like this.

“The most vocal are those who are not satisfied. That is why you see more complains than praises about GNOME 3; it doesn’t mean nobody likes it.”

Some people just go away rather than complain.

There are maybe three things that give me hope for Unity as something I can someday recommend to someone. The rest . . . no. Not unless you don’t want a GUI.

#83 davek on 07.27.12 at 18:53

GNOME3 tried to be a new way to do things, when it should have been a new way to create things. We don’t want or care about a new workflow process… it is up to the individual user or distro to do that. You customize your desktop to how you work – having someone else try to shove a paradigm down your throat is offensive.

Think what GNOME2 did. GNOME3 should have taken that and abstracted away all the rigid/hardcoded features. We want customization. Let me repeat that:


If you like a slideout menu in the top left, great. Let users put it in a popup bottom right if they want, or one in every corner, wherever. The goal of GNOME4, or wherever you go to from here, should be abstraction and customization. Nothing should be set in stone. Apple menus at the top? Make that an option with configurations. Then you let your users/adopters/developers create the desktop/workflow paradigms as plugins/themes. Problem solved.

[WORDPRESS HASHCASH] The poster sent us ‘0 which is not a hashcash value.

#84 Vitali on 07.27.12 at 18:57

Great thanks for your job, friend! Gnome3 – is my favorit DE!
Very sad, but i can’t help to you to build this beautiful DE.
Ehh =(

#85 Cyber_k on 07.27.12 at 19:10

Basic requirement of a desktop is to be able to put stuff where you want it, grouped as you want it.

Imagine if your physical desk forced everything into predetermined piles, forcing someone else’s sense of order on you, with most of it a layer down below the surface.

#86 Denys Rtveliashvili on 07.27.12 at 19:20

> Nobody can say if GNOME 3 is better or worse than GNOME 2.

Completely wrong. People can say and do say it all the time. You just don’t want to listen.

> There is no recognized metric anywhere.

Yes there is. It is public opinion. People are screaming as Gnome 2 was buried alive and Gnome 3 was raised instead.

> GNOME describes itself as a “community that makes great software”, which is as nondescript as you can get for software development.

Wrong again. Such description tells a lot. Because it is not a “community that makes useful software which is a pleasure to use”. You can measure if something is useful or not by whether it helps people or not. You can tell if something is a pleasure to use too, although it is far more subjective. But how do you define “great”? I do not know how, but it must be something bizzare, as “great” software where you have to guess that pressing “Alt” while keeping system menu open shall _on_the_fly_ change some “suspend” to “shutdown” and things like that is a weird and unnatural thing.

> GNOME is losing market- and mindshare.

Of course it does. What else did you expect?

1. You have killed almost all customizations you could do in Gnome 2 and created a system which is pretty dumb even with all extensions on.

2. You cannot set up shortcuts properly. Gnome 3 STEALS “windows” key. You cannot assign and use shortcuts which involve it unless you launch gconf-editor and tweak some arcane settings to disable that ghastly behaviour.

3. I have already mentioned that issue with the system menu where some self-proclaimed expert in ergonomics has hidden the “shutdown …” item and that menu changes itself when you press “Alt”. Do doors in your home appear when you sing “Row, row, row your boat”? No? Then why menu is doing that? Ah, I know why. Because nobody is doing it so as Gnome 3 does it must be great, an innovation.

4. Gnome 3 is do bad by default that you have to install a dozen of extensions just to make the experience to be bearable.

5. Yes, I know about Cinnamon and Gnome Fallback Mode. To be clear, they do not provide a Gnome 2 – like environment. Gnome 2 is not about location of panels or color schema. It is about being able to change things, to tune it for yourself and make the desktop environment to be an continuation of your body. Good environment is the environment you do not notice. As for Gnome 3, everyone notices it all the time and often searches online for solutions of usability problems, in vain.

> core developers are leaving GNOME development.

No surprise. People are leaving project which is hated or just ignored by Linux community.

> GNOME is understaffed.

Yes, which is a consequence of previous issue.

> claimed target users for GNOME

And who are these people?

I used to be one, not anymore.

Thank you guys for all of your efforts.
I was using Linux as my primary (and often only) OS for about 15 years. But as Gnome 2, the only half-suitable environment has been abandoned and dropped, time to change horses one more time. This time the next desktop environment would be Mac OS X. At least those, who design it, understand that usability is far more important that some ephemeral greatness.

P.S. Such a disgrace.. After all these years Linux still does not have even a descent switcher for keyboard layout. Switching between more than three layouts is an unreal pain. Although groups in KDE’s switcher make it be a touch easier.

#87 MorayJ on 07.27.12 at 19:30

I can’t comment on the blog posters move, other than to express gratitude for the work he’s done and I think his public statement will certainly help debate amongst the Gnome folk.

But this has turned into one of those candles for all the shell haters to flutter around, so I’d like to join a few others in stating that I think shell is great and getting better.

Countless times I’ve read people claiming to be developers and complaining that the shell doesn’t cater for them – and then complaining they can’t possibly deal with pressing the alt key to shut down the computer. They clearly don’t develop in emacs that has a few key bindings. Honestly RTFM.

And I went back to Gnome 2 for a moment when I installed Debian. It was like being trapped on railway lines. Everything methodical but longwinded.

I know everyone is saying shell is designed for touchscreen, but it seems more designed with the idea of moving away from the computer as filing cabinet paradigm. And too right. Times are changing and Gnome had to change.

There is MATE – that is the point of opensource, surely? But really, Gnome 2 is not the future.

How shell develops should be interesting. The configurability of extensions should really put paid to people’s fears. Shell could, ultimately, underpin a lot of very different looking desktops.

But really just to say, I love it and I wish the best for it. I agree with a previous poster that highlighting the short supply of developers may steel a few more people to dare to step up to the plate.

#88 Dagmar d'Surreal on 07.27.12 at 19:46

There’s plenty of developers–they’re not in short supply. What’s in short supply is people who would care to continue dealing with Unity.

To the people saying “but I like it fine” keep in mind that just because you like it doesn’t mean other people do–and all those other people have been jumping ship in droves.

#89 Benjamin Otte, sobre GNOME: el proyecto ya no tiene una meta | ADCLIC Blog en español on 07.27.12 at 20:03

[…] información | Swfblag This entry was posted in General and tagged Gnome by RDaniel. Bookmark the […]

[WORDPRESS HASHCASH] The comment’s server IP ( doesn’t match the comment’s URL host IP ( and so is spam.

#90 Anonymous on 07.27.12 at 20:49

As a linux home user, I find that gnome 3 feels a little clumsy and difficult to use, so perhaps the first should goal could be to remove rough edges. Here are some ideas

Reduce the number of random freezes: It would be nice to have automatic error reporting ( like kernel oops) and perhaps gnome-shell could be made more error tolerant.

Provide extensions, that provide all functionality from gnome 2: You are fairly close, but I need a simpler way to handle multiple desktops.

It could be nice to have more usability tests. Perhaps you could inteview people that have been using gnome 3 in a coorperate environment for several months. They probably have some small scale suggestions for how to improve the gnome shell.

#91 2cents on 07.27.12 at 21:02

Sadly, GNOME 3 (and Unity) killed the linux desktop for me.

#92 Van de Bugger on 07.27.12 at 21:08

Someone said “nobody likes GNOME3″…

I use GNOME 3 at home, and I LIKE it. There are few features which I miss (like sensors applet), but I can live without that.

I want to use GNOME 3 in office, but I have 3 monitors in office, and GNOME 3 Shell can’t stand with it and crash, so I have to use GNOME 3 + OpenBox. :-(

#93 Benjamin Otte, sobre GNOME: «el proyecto ya no tiene una meta» : – Tecnología – Gadgets – Culto Geek on 07.27.12 at 21:11

[…] Más información | Swfblag […]

#94 Jamie on 07.27.12 at 21:16

I find it hard to take most gnome 3/shell complaints seriously. One minute it’s awful because it takes too long to enable the right extensions, the next minute desktop environments should be all about customisation because people love spending time picking options. One minute it’s too much like OS X, the next minute people are leaving it in disgust to go use OS X. It gets in your way, it’s not in your way enough. And so on and so on for ever and ever and ever and ever. People love to whinge.

Then there’s the “going to and ticking a button is too much hassle, I’m switching distro and forking my own desktop environment”.

#95 aris on 07.27.12 at 21:30

To some of the comment authors:

It’s funny. People work and release this or that window manager. “OMG! It’s a {Windows,MacOS,BeOS,OS/2,…} copy!” The first time they try something bold, inovating in some aspects, like I read a ton of comments asking for in the past, people get upset. Stop whining. If you want to help, point precise use cases where it got worse so the developers know what to improve or redesign. Better yet, join the project commenting on the features when they’re proposed.

And above all, open source is not about others working to please YOU. Pick something else if you’re not willing to help improving things.

#96 Chris Duffy on 07.27.12 at 21:40

I’m not a true developer, “influencer,” or probably your target audience, but I wanted to come here and say that in spite of all this internal bad news, I think GNOME 3 is really beautiful and the best Linux desktop environment yet. (I’ve been using it at home and on my personal computers since 2006)

It’s the first thing I install on a new build followed by emacs (which I’m still learning) and chrome. Probably in the minority, but wanted to let you know the product and hard work is truly appreciated. If there’s something people like me can do, let us know.

#97 Mark H Weaver on 07.27.12 at 21:46

I’m sorry to hear of such troubles within the GNOME community. You may be encouraged to know that after many years of failing to persuade my Apple-loving friends to use free software, GNOME 3 is what finally convinced one of them to switch. My friend, who has been a loyal Mac user since 1984, tells me that large numbers of Mac users are very unhappy about the direction Apple is taking, and many of them are switching to GNU/Linux or at least considering it. For example, he tells me that this unhappiness with Apple is common among readers of MacInTouch, including the person who runs that site. Such users are looking for beauty and simplicity, and GNOME 3 is well positioned to benefit from this. My friend tried Unity for a week or so, but ultimately decided that he preferred GNOME 3.

Please do not lose hope! The complaints about GNOME 3 can be addressed through the extensions system over time. On a more personal note, I’ve been using X11 since 1988 and GNOME since the 1.x days, and although I avoided GNOME 3 for several years because of all the complaints I heard, when I finally tried it, I loved it, and it is now my primary desktop! I can launch and/or switch to applications very efficiently using the keyboard, so I’m not bothered by the less efficient method of moving the mouse to the upper left corner. I also love the new notification system, the instant-messaging integration, and the calendar integration in the top panel clock. For those who prefer other interfaces such as an applications menu, there already exist extensions to add those things. Nowadays, the Gnome Shell can be easily moulded to almost any interface you prefer, so I don’t understand why people are so dead set against it.

#98 foo on 07.27.12 at 21:48

To all those whining about GNOME 3, wake up and smell the roses. GNOME 3 looked like an utter disaster to me, but after having used for a while after Debian switched to it, I find it no worse than GNOME 2. That said, I don’t like some of the directions it goes in, for eg emphasising “apps” over “people”, “places”, “objects” and so on.

[WORDPRESS HASHCASH] The poster sent us ‘0 which is not a hashcash value.

#99 Anon on 07.27.12 at 21:52

I love how a lot of people just mindlessly bash Unity and talk about how it looses marketshare, yet Ubuntu is actually growing and Unity is garnering popularity.

I’d like to point out that the ‘that if you like it doesn’t mean other people like it’ point works both ways! You’re the minority – a lot of people actually do like it. And it does enable new users to switch and actually gives Ubuntu it’s own look. I’m not saying Gnome-shell is bad or Cinnamon is bad, but that you are not deciding for everyone! If Canonical stack with Unity all through 12.10 so far and they actually have OEMs making machines with Ubuntu, I think they did get something right.

Now, I know that you like bashing something, but give it a try – if you still don’t like it, that’s fine – but keep it to yourself! I do not see people going around shouting about how doughnuts with holes are better than the ones without or how they don’t like the new Peugeot 207 because of it’s look and how therefore no one should like it. So please, keep it to yourself. Unless you have valid criticism, which is most welcome, do not just repeat Unity is bad or Gnome-shell is bad. There are good things and bad things about them.

#100 Dirk on 07.27.12 at 21:53

Amusing that GNOME, Unity and Windows 8 apply the exact same mistake while OSX, XFCE and KDE got it. Innovation not requires throwing away everything and do different. Keep the concept and tweak. Evolution not Revolution.

Former GNOME 2 lover and OSX and KDE user now.

#101 Matt Street on 07.27.12 at 22:05

For many of us gnome stopped meeting our needs, and seemed to not care when we brought that up.

I’m all for other people using it, but there isn’t much point in me supporting it or fighting to use it.

#102 Justin on 07.27.12 at 22:30

GNOME does have a goal for 2012, which is to make it the year of accessibility, which any idiot with a browser can see by going to

#103 finger on 07.27.12 at 22:59

NO. GNOME 3 is absolute fail. Metrics? Usability! GNOME 2 was almost perfect. It was transparent. I don’t remember thinking about GNOME 2, when I was using it. It was just there. I configured it as I want.

GNOME 3 is horrible. A little tiny example. I can’t move clock from the center of top panel. This is madness and total idiocy all-in-one.

I don’t know what caused this. If you mention absence of goals than I’m sure – this is it. But I can’t imagine how you can make such a horrible thing as a GNOME 3. Goal was “working Free desktop environment”, wasn’t it? Good! Then tell me, pretty please, how come you just trashed all working and great design decisions and made this horrible software?

#104 Dan on 07.27.12 at 23:29

Gnome 3 would not be in this shape if the UI wasn’t horrible. The only UIs I’ve seen that are worse are Unity and KDE4.

[WORDPRESS HASHCASH] The poster sent us ‘0 which is not a hashcash value.

#105 Ahleron on 07.27.12 at 23:30

As a former Gnome user, I found Gnome 3 to be a total train wreck. Simple example, why the hell is the freaking day and time in the center of the top bar? I don’t want that centered and staring at me all day. I’d rather have some representation of what I’m working on there. Better yet…do what I used to be able to do under Gnome 2, and let me, the user decide what goes there. I had been looking forward to Gnome 3, but the lack of user and configuration options is antithetical to Linux and open source desktop usage. Gnome 3 turned me into a KDE 4 user.

#106 neelwebs on 07.27.12 at 23:51

>Previous supporters of GNOME are scaling back their involvement or have already dropped GNOME completely.
The problem is, nobody likes GNOME 3. People loved GNOME 2, but when GNOME 3 came out, GNOME started to die a slow, but painful death, as people went to other UIs like Unity, KDE, and XFCE. Right now I am using a GNOME 2 fork called MATE ( on Debian Testing. Non-Linux systems like FreeBSD and Solaris contuined to maintain GNOME 2 on their own.

At least we aren’t using Windows 8, where you have to accept whatever M$ gives to us. Sadly, Microsoft wants to force us, Linux users to use Windows 8 with the so-called “secure boot” which makes Windows attractive to the small fraction of Linux users.

>The claimed target users for GNOME are leaving desktop computers behind for types of devices GNOME doesn’t work on.

This is the bad part. Most people are moving to devices which are closed like the iPad. If someone really wants to use Linux in 2020, then that person would somehow have to handle jailbreaking/rooting their device, just to use Linux or other FOSS OSes.

#107 Oliver on 07.28.12 at 00:05

Sure GNOME have some problems but I think that you are overreacting a bit and see things from a negative point of view.

a) Even though some notably Linux distributions
do not use the GNOME shell + mutter they are based on and use GNOME technology + GNOME programs so in practice they are still GNOME, for me GNOME is much more then a a shell and a window manger.

b) The company’s you are mentioning have had different kinds of problems and are focusing on stuff where think they can earn money, and that is not the Linux desktop. On the other hand it seams that consulting company’s like Igaila, CodeThink and Collabora are increasing their investments in GNOME technology.

#108 argus on 07.28.12 at 00:18

GNOME lol, I stared until 4.

#109 Trae "OctobrX" McCombs on 07.28.12 at 00:28

Up until a year ago, I used Linux exclusively for my desktop Operating system since 1996. I’ve gone through KDE 0.x and GNOME 0.x [ WindowMaker, AfterStep, E, etc… ] and although I used KDE for 2 years around 2001,2002, I’ve mostly stuck with GNOME and have always preferred it.

I was given a Mac at work, and it’s done most of what I have needed for the past year. I have missed my Linux Desktop. Recently, I’ve wanted to try and get back into Ubuntu.

Unity has always been scary to me. It has done things I didn’t like, and never felt “Native” or “Easy to use”. The folks doing their usability studies must be on crack. I liken Unity to Windows 8 and the attempts are re-inventing the wheel and getting lost along the way.

There are many things I would like to say about the fall of the Linux Desktop, and how we’ve lost our way, but I doubt anyone cares, or would it even fit in this comment.

I stopped participating in the community a long time ago, I accept part of the responsibility for it’s demise. Perhaps, somewhere along the way; good folks will get involved and make it not suck.

Till then, I’m lost on Linux looking for a Desktop Enviornment that doesn’t suck.

Cinnamon? No thanks. Missing tons of features that Gnome2 had. Perhaps that’s what people are going towards, the “No Configuration” window manager. Let’s just give folks what THEY (we think they) need and nothing more.

#110 Josh on 07.28.12 at 00:33

If you think you can’t say whether GNOME 3 is better or worse than GNOME 2, then I will say it for you: it’s worse. Much worse.

I don’t know how much criticism GNOME has to get before you reverse course, but you are reaping what you sow. You make a bad product, people hate it, there is much criticism, you double down on that direction, people stop using GNOME, developers leave project.

How is this not getting through to you guys?

#111 Paulo on 07.28.12 at 00:38

Certainly, lack of direction might feel like lack of leadership.
A sense of direction and unity of direction is an integral part of any successfull enterprise. Goals are just that.
Not surprising that people are jumping ship, since it “dead in the water”.

#112 Mike Fisher on 07.28.12 at 00:41

As a professional sys admin, the Gnome 3 shell is pretty much useless to me.

As others have already said, a good desktop environment is one that doesn’t get in your way. Gnome shell not only gets in my way, it actively tries to prevent me from going around it. The lack of basic Ui customization ability and absence of documentation are both major road blocks. The ability to modify the environment with plug-ins is a nice idea, but not a substitute for basic, expected functionality. (e.g. power off)

As others have said, the idea that a single interface can fit desktops and tablets is fantasy. The use cases are too different. I *don’t want* the same interface on my desktop and my smart phone any more than I want my laptop and my TV to have the same “HD” 1600×900 screen format. (anyone remember 1600×1200?)

Admittedly, I still use Gnome 3. Locked in fall back mode. With Nautilus disabled, and the default window manager replaced by fvwm. Basically, I use the panel and app integration. Period. The same set up functions equally well on a single screen laptop or a three screen workstation. It does what I want and stays out of my way.

Maybe that should be the mission statement for Gnome 4: Make great software that does what users want and stays out of their way.

#113 Jeff on 07.28.12 at 00:56

The soon coming future name of the game is HTML5 and WebGL.

This is the environment that will not only scale across most devices, it will also allow for more connected applications (between users, data, developers, etc).

IMHO what desktop developers need to do is transition their application domain specific knowledge to a web environment, eg. by making using web components that build up all sort of applications. These developers could join Mozilla and other groups to provide a free and flexible UX for the future.

All is going to the web, mark my words!

#114 williambraski on 07.28.12 at 00:58

Idea for direction: I’ve installed some extensions that let me turn the environment into something I want. It got me thinking…

Focus on extensions. Let people download what they want to turn the desktop into their perfect environment. If people want traditional task bars and all that noise, extensions.

Give distro maintainers an easy way to roll their own environment using extensions that fit their goals.

Turn Gnome into what Linux itself is: a sandbox. A platform that can be modified and tweaked infinitely, via extensions. Something that anyone can take and turn into the DE that works exactly how they want. From the newb to the seasoned distro maintainers.

For the most part, I like Gnome-Shell. My biggest gripe is that, let’s face you, the developers were arrogant. I think the attitudes of those working on the project has done just as much, if not more, to drive users away than the system itself.

#115 alex on 07.28.12 at 01:28

why the hell people complain about gnome extensions?
They are much more flexible than the applets.
The only issue was that some really useful extensions needed to be coded.

Imagine: you can create an applet like before but also nearly every aspect of gnome-shell.
And the thing with the necessary extensions will become untrue next release.
Reason will triumph and you can shutdown the PC the way everyone wants to.

(there is one fatal issue remaining:
why has empathy no support for OTR and Pidgin encryption nor plugins but is standard)

#116 JiyinYiyong on 07.28.12 at 01:30

After installed several extensions, gnome-shell seems really nice to me. It’s modern and much more convinient than GNOME2. But as a Linux fan who’s learning web development, I’m so energetic about implement my own design on my own desktop. But to a notice, it’s really confused to learn about GNOME3 while there’s not much docs about it.
Despite that, I want a desktop, with which I could use APIs to tweak the behaviors how each window and each workspace response to my action and my comands. It won’t be fun if I have to learn C before I can programe a desktop. There’re beautiful languages like CoffeeScript and great tools like Looking Glass, why I must have to know C(even Python is kind of boring to me).
In my opinion, desktop should be more like a platform than a product. Apps like Office and Empthy don’t need to be pre-installed to make it clumsy. To be extreme, I think a desktop should simpely be a basic workspace with consoles, then I need to choose modules to install to have docks, browsers, workspaces on my desktop, or I can write my own version. Things like Unity might be a great product, but not a free desktop. What I want it a free desktop I can programe components and behaviors in my design,

#117 Ron on 07.28.12 at 03:13

I work for a company building commercial touch screen based devices running on Linux. With GTK, it was as if we were being actively discouraged from building anything on it. The API documentation is horribly out of date and incomplete. There are hardly enough working examples to refer to for GTK 3. The GTK portal itself comes across as a poor attempt to throw together bits and pieces about GTK.

Our biggest problem is that GTK is written in C not C++, which in 2012 doesn’t make much sense for a toolkit specifically dealing with graphic and other objects.

We finally abandoned GTK, switched to QT and finally moved to android last year. That’s been working out great so far. We have much shorter release cycles, code is much easier to manage, and bugs are much easier to fix.

#118 Zac on 07.28.12 at 03:20

The problem stems from Gnome in practice is controlled by Red Hat. Red Hat is doing well in the corporate sphere and they are satisfied with that and has no interest in getting Linux out into the average consumers hands, therefore there is no goals or little innovation. I appluad Red Hat for very well contributing but they had the capability to get Linux out there to the masses, but they no desire to. But want more, and it can be more.

#119 anotherone on 07.28.12 at 03:28

If you want listen someone that loved gnome and some variations,
I’m yeat a great fun of gnome (2.3). I guess I can help. Not coding.
I’m very very unhappy because of these news … if water in eyes ( really want crying, i have only two ways so … use just console or openbox + tint2, because of the rest is a bull …. i don’t want just a soft in my desktop, i want the good ones, coded for the MEN, that know what is code, to mee these people are the hackers that i admire. Like all of you that worked in gnome 2) ….

I guess if you talk with the … You will understand me. I guess that is the people that loves gnome like me.

Gnome == The Best Desktop with language support (LANG==^^,) I have used ….

To me nevermind what Linus said. I give a congratulations for the linux. But me or i use a console, or openbox + tint2 or gnome … Or i will never more working with linux.

#120 Ian on 07.28.12 at 03:55

I abandoned Windows 2000 because the “designers” got their fingers into things and produced the mess that was Windows XP.

For years, Gnome2 was the #1 WIMP desktop environment available at any price.

These days, I actually prefer the Windows 7 machine my employers give me.

I don’t own a mobile phone, because I don’t like them. I don’t own a tablet, because I don’t like them. They’re crippled devices with inefficient input devices and small screens. They’re junk.

I’ve been sitting at desks using computers for work and play since I was a child… but between KDE4, Gnome3, Unity and now Metro, sitting at a desk using a computer has become unpleasant… I might actually just give up on computers entirely and switch to gardening.

These bastards turned my wonderful tool into a fucking toy. I hope they rot in hell.

#121 steevithak on 07.28.12 at 03:59

> Nobody can say if GNOME 3 is better or worse than GNOME 2.

Well, let me help you out with this. GNOME 3 is worse than GNOME 2. It’s much, much worse. It’s almost unusable for doing actual work. I know a lot people who use GNU/Linux and not one of them likes GNOME 3. Most have already abandoned it for something else. I keep hanging on, hoping maybe if things get bad enough, the developers will actually start listening to users and push things back in a more sensible direction.

I posted in my blog a while back a lengthy review of some of the horrible misfeatures of GNOME 3 and got quite a few emails expressing similar feelings. I read in the link above that Linus shares some of my complaints too.

#122 Eh on 07.28.12 at 04:04

If GNOME is really a Red Hat project, why doesn’t Red Hat management fix it?

That is, fire all the idiots currently in charge of GNOME, and get some smart people (possibly newly hired) to maintain and lead it.

Maybe they aren’t aware of the situation?
Can anyone inform the relevant managers at RedHat of the need to get rid of those cretins?

#123 Leif on 07.28.12 at 04:16

It’s a bit ironic. We’re in a period where the Windows desktop is losing relevance quicker than ever yet we’re debating the relevance of the GNOME desktop? Now’s GNOME’s opportunity to gain some ground. Smartphones and the “browser-age” is totally eroding the need for the Windows desktop. Give a user access to an excellent web browser and your “desktop” is mostly done. The rest of the chore of the desktop is to stay out the way and just-work.

GNOME 3 is situated quite well for this post-Windows era I think. GNOME Online Accounts, GTK Webkit2, gjs, GNOME extensions, etc.

Honestly, I don’t care much about a lack of “native” GNOME apps. GNOME should concern themselves with empowering shell extension authors (maybe provide web services hooks if they don’t exist already). And beef up GNOME Web App integration.

Full disclosure; I’m a Java enterprise server side + web app developer who happily uses GNOME 3 daily for running GNOME Terminal, Netbeans and Chrome.

#124 Michael on 07.28.12 at 04:39

Well, I realy like simple user interfaces. I’ve always liked the GNOME approach of getting rid of the superfluous crap that achieves nothing. I want a user interface that I don’t even notice exists. If I need to customise then it’s a huge fail. I’m mostly happy with GNOME shell; it lets me do what a window manager should, it has a much better way of switching windows than Alt-Tab, and the workspace management is a breeze. Every time GNOME ventures to push the usability bar higher there is always resistance, like from GNOME 1 to GNOME 2. What makes GNOME special is that they adopt the “Simple as possible. No simpler” mantra. On top that they’ve made it easy to extend for those who have a real need to do so. KDE on the other hand just continues to replay the same mistakes for the sake of familiarity and keeping existing users happy. I would be very sad if GNOME declines because I prefer their approach.

#125 Andrew on 07.28.12 at 05:18

Meanwhile, Transmission 2.61 moves to GTK+ 3.

I used GNOME 3 for a few weeks and it was pretty good. It took a few weeks to get used to, though. I won’t touch KDE because of poor usability decisions (especially Plasma… and modifying panels).

At the moment I’m still on GNOME 2, but I will eventually upgrade. I’m convinced that, in terms of usability, GNOME 3 is still competitive.

I’m not sure how GNOME could improve, but GNOME 3 did appear to rub a lot of people the wrong way. If too many people dislike GNOME 3 then changes should be made to it.

Anyway, just my opinion.

#126 Aron on 07.28.12 at 05:27

People aren’t leaving for tablets. At least not the ones that would ever have used GNOME in the first place.

We have tablets *in addition to* our desktops. We want our desktops to work like desktops (i.e. input- and multitask-centric) and our tablets to work like tablets (output and singletask-centric).

I used Linux for a long time. Since the very first “distros” began to emerge in the early-mid ’90s. The one-after-another succession of a crap KDE release and an unusable GNOME release was enough for me.

As an iPad and iPhone owner, OS X was the natural next step. If either KDE 3 or GNOME 2 had remained on their development trajectories and evolved, rather than devolved, I’d still be a Linux user today.

Instead, after 25 years of being a “fringe” desktop user (Unices and Linux exclusively, with no Windows or Mac OS in the house), in the last couple of years I’ve ended up with an OS X desktop and Windows 7 in a Parallels VM. I have a CD case full of commercial Linux software, an external drive full of Linux downloads and code, burned Linux distro DVDs and CDs and even full Slackware 2 and Slackware 3 floppy sets still in the closet—and no Linux installed anywhere.

And with the end of Linux on the desktop, old-school UNIX on the desktop lost. New school UNIX on the desktop (OS X) won.

And that’s that. Too late to turn back now, GNOME.

On to the future.

#127 Andrew Clunis on 07.28.12 at 06:21

I’d like to suggest that perhaps GNOME ought to consider power users and developers as a greater target of their UX, since such folks make up a very large component (dare I say majority? sadly, I suppose we lack data) of the user base. We use and value Free Software because it empowers us and necessarily tends to be very technically accessible. The GNOME Project has been focusing on the consumer market, a strategy which one can argue has only been marginally effective.

Of course, software development isn’t free of charge (as in beer). Until such time as the market of actual desktop users (power or otherwise) has a means to more directly economically incentivize GNOME development, there won’t be that much talent available to focus on the platform. There’s already a lot of economic incentive (and thus, capital availability) to get work done on the Linux kernel, for instance.

I spoke with some Red Hat folks recently, and the Enterprise Desktop sadly remains low-priority for them. They keep GNOME alive on the backburner, it seems (which I do muchly appreciate).

Anyway, I hope these remarks are helpful. I use and value GNOME very much, and see a lot of potential.

I must add: Shell extensions are a stroke of brilliance.

#128 Wil Hepworth on 07.28.12 at 06:34

If Gnome were a business in the true sense, reality would have sunk in a long time ago and Gnome 3 would have been retired or overhauled. The fact that Cinnamon exists should be a clue to Gnome that all is not well. Instead, Gnome have put their fingers in their ears and are chanting “La, la, la, can’t hear you”. It should have gone the way of Microsoft Bob, instead it is Bing, an eternal also-ran.

#129 Herbert Eisenbeiss on 07.28.12 at 07:05

Oh on the contrary, of course GNOME had goals, mainly: alienating themselves from their loyal user base and building a so called “desktop”, that only works good on desktops with GNOME 3.0, not listening anymore about what the people wanted or not, and giving us that piece of crap to work with it on a daily base. Yikes!

Face it: there is no GNOME based tablet or something like that coming soon, so what was the point in making GNOME shell into something that only works well on a tablet, if at all?

The second important thing is: you – the developers – took away our freedom of choice, by turning off important options we used in former times to tinker with or buried those options somewhere deep in gconfd, so that finetuning GNOME nowadays became something even more rubbish than tinkering with the registry of Windows. It feels almost like the same?

And then you wonder why nobody cares about plain GNOME anymore, why there are so many forks instead (MATE, Cinnamon) and Canonical favors Unity instead?

That’s your fault, the fault of the developers. You simply produced an unholy mass of “user environment”, that feels like crap and nobody really wants to work with, because instead of enabling the users to make things easier, to get their job done, it often just stands in their way.

So it’s no wonder that companies, one by one, are dropping their support for GNOME and user count is dropping in favor for XFCE, Cinnamon, MATE and so on.

My simple advise: if you really want to play big games again with GNOME and let it become a big player, you should endorse Cinnamon Desktop as the new default UI for GNOME 3.X and just forget about the rest.

#130 JosH on 07.28.12 at 07:47

For me, there is one main reason behind this all: The arrogant attitude of the developers. Not listening to your users is a guaranteed way to death. “We like it this way, so you have to do it this way” is never working out well. Unless you are Apple and “this way” turns out to be a very good one.

#131 Fred on 07.28.12 at 08:47

Staring into the abyss, really? Come on, Gnome 3 is the only usable DE right now. Gnome 2 has always been buggy (gnome-panel anyone?) and has no hardware acceleration. KDE and Unity are also slow and buggy. Gnome 3 at least is fast and clean and does not get in your way. Gtk3 is also the best widget toolkit available on Linux, at least give GNOME some credit for that.

Why do people have so high expectations on DEs? It’s only about switching windows and starting programs, duh. GNOME3 does that just fine (and a lot more).

#132 Marc Stürmer on 07.28.12 at 09:06

Well, interesting post but you still do not get the real point which is sitting so plain visible right before your eyes: distribution makers want to make money and profits, so they simply care about what their customers want and not really at all about what developers of desktop environments want.

If GNOME 3.x would have been really well accepted by the community and users, then there would be not that drastic decline in distributions shipping it and companies supporting it.

Let’s tell you one plain, hard and simple fact: most users don’t care at all about a certain desktop environment, if it is free software or not, they don’t love it, they just use what’s in their opinion sucking less at the moment to get the job done. All software sucks more or less, so they choose the piece of software being available at the moment which sucks less compared to the rest.

It is that simple; if you turn their tool in a new version of it into something that sucks considerably more than the previous one, you are just alienating your user base and making them go away, considering and taking other choices.

That’s just what happened with GNOME 3.X, and it seems that this decline still continues, quite and simple because most developers of GNOME fail to understand what the users really do care about and what not!

Even the comments here are full about it, what users want and what not, so if you really want to get better support again by companies, distribution makers and get GNOME better staffed again, FOR HEAVEN’S SAKE WAKE UP AND START LISTENING TO US AGAIN!!!

#133 Anomymous. on 07.28.12 at 09:17

For future reference. This blog post was also slashdotted

#134 Anonymous on 07.28.12 at 09:36

I really liked GNOME 2 but after giving GNOME 3 a try, I found that it just doesn’t fit my needs. And I guess I’m not the only one, if all the variants are some sort of metric.

But all the work done for GNOME 2 is not lost. Fortunately it was forked and renamed MATE and now I’m happily using it.

[WORDPRESS HASHCASH] The poster sent us ‘0 which is not a hashcash value.

#135 GNOME y el abismo del desarrollo Open Source | Incognitosis on 07.28.12 at 10:00

[…] la barrera: uno de los desarrolladores de GNOME ha escrito un inquietante artículo llamado “Staring into the abyss” (“Contemplando el abismo”) en el que deja claro los graves problemas por los que […]

#136 Vulcan on 07.28.12 at 10:24

The “Linus’s bashing” link misses a colon after “https”

#137 Alex on 07.28.12 at 11:00

I personally found Gnome 3 a great improvement over Gnome 2 (which I used for a few years), and I don’t understand what people think when they say the opposite. Gnome 2 was a good desktop environment, but it required quite a bit of tweaking to be a productive work environment. When I switched to Gnome 3, I felt uncomfortable for a period of time, but later on I noticed that I now work much better with the same applications. In fact I even prefer Gnome 3 to Unity. So from my point of view the direction of change in Gnome is right.

#138 David Gerard on 07.28.12 at 11:12

You guys said “If you don’t like GNOME 3, don’t use it.”

So we took you at your word.

The CADT model remains predominant in GNOME: throwing everything away and writing something new is always much more fun (and better for the resume) than just fixing the remaining bugs in something that basically works.

I didn’t leave GNOME, it left me.

#139 AFellowContributor on 07.28.12 at 11:36

I blame GNOME 3 and only GNOME 3. Such a rapid, total change in the approach to the familiar interface we came to love forced people who used GNOME (including myself) to switch to alternatives like MATE or Cinnamon to preserve even the basic functionality of our desktop. People liked GNOME 2, they want an improved GNOME 2, not the crippled, unattractive GNOME 3 that forces us to use the fallback mode to get even the basic functionality.
IMO Gnome should return to it’s 2.x roots and do what it always did best: deliver a consistent, attractive interface that people actually ENJOY. Work on that, improve that, add features on that but for gnome’s sake ***drop that awful 3.x version*** and get back to what people who like gnome actually want.

#140 craig on 07.28.12 at 11:52

@Juan re: #26: those “3 pillars” you mention are precisely the things that are driving users away in droves. If that’s the plan for revitalising gnome, it doesn’t seem likely to succeed.

[WORDPRESS HASHCASH] The poster sent us ‘0 which is not a hashcash value.

#141 Axel FILMORE on 07.28.12 at 12:16

KDE tried to imitate Windows, they have failed, GNOME tried to imitate Apple, they have failed.

Google creates something new, Google Chrome, Android, they succeed.

It seems that the Cathedral works better than the Bazaar, it seems that dictatorship and anarchy are not a good way to develop softwares.

GNOME guys should concentrate on GTK+ and Vala instead of distroying their desktop a little bit more on each release.

#142 gavinw on 07.28.12 at 13:13

Gnome3 is great.

#143 Anon on 07.28.12 at 13:14

I use gnome. I’m not a gnome dev; but I have banged out a few junk programs for personal / work use.

I think If it were easier to [learn to] develop for gnome, you’d get more gnome developers. Grow them organically, instead of trying to convert them from MSVS or something else.

Also, I’ve always been envious of the fancier widgets in Qt.

PS You should fix the Linus’s bashing link.

#144 GNOME:凝视着深渊_互联网资讯最新报道_野火集 on 07.28.12 at 13:14

[…] GTK+开发者Benjamin Otte在博客上谈论了GNOME项目的停滞和衰退。Otte说,核心开发者一个个离开,GNOME现在人手短缺,GTK只有一位全职开发者(也就是Otte本人),GNOME各个子项目的 全职开发者加起来可能不到20人,其中有10人是 Red Hat的雇员,GNOME几乎成了一个Red Hat资助项目。 […]

[WORDPRESS HASHCASH] The comment’s server IP ( doesn’t match the comment’s URL host IP ( and so is spam.

#145 FunkyM on 07.28.12 at 13:16

Thanks for the facts from a top contributor of GNOME.

I hope a solution is found to finally get back to a nice GNOME eco-system and make all the forks die again.

As a developer it’s not nice to see when your contributions are thrown away for something half-baked while ignoring some good arguments from experts, too.

One might argue that some developers are also currently holding back with any contributions since one knows that your extension, theme or app most likely might break in the next GNOME3 release due to the steady refactoring of several parts of the platform.

Some developers simply don’t like to code into the trashcan.

Talking about goals, GNOME2 had a very comprehensive HIG that set clear goals and rules for any part of the system and kept it consistent.

GNOME3 clearly lacks any good HIG and thus the “OS” lacks goals and rules (no, some mockups and loose descriptions on a wiki are not the same).

It feels like an experiment that is changing constantly towards an unknown/undefined direction and being inconsistent in many areas and missing overall architecture.

I hope that some people will:
– Accept facts, critizism and arguments
– Face the situation
– Analyze the facts openly and thus involve the community into discussions
– Set goals for improvement for the whole platform
– Start completing the goals while keeping the community involved

A “next file chooser improvement” or “best document history tracker ever” doesn’t really help out of this situation right now, or does it?

#146 Alessandro Lattao on 07.28.12 at 13:19

I’m a happy GNU/Linux user since 1997 and now I’m a happy GNOME 3 user.
I love GNOME 3 because it’s fast, simple and elegant.
Perfect for my needs.
Keep up the good work, solve problems and ignore whiners :)

#147 qqqq on 07.28.12 at 13:20

Comment #100 wrote:

“Innovation not requires throwing away everything and do different. Keep the concept and tweak. Evolution not Revolution.”

I totally agree! Look at the Linux Kernel!

Gnome 2 was so good, I was happy with it, even wanted to contribute… Now Gnome 3 is just unusable. I switched to XFCE but I *still* miss Gnome 2. KDE is too bugged, Unity is almost as bad as gnome-shell, E17 is never released…

One of the problems I see is that you’ve walked too much on the gnome-shell path that now you won’t ever want to admit it’s a failure and go back to something reasonable because it will feel like a huge waste of time.

If I suddenly became a billionaire I’d probably hire a team of developers and make them create a good desktop environment. I really don’t think it will be too hard to steal most of the user base from the current DEs we have…

#148 peak on 07.28.12 at 13:28

I like Gnome 3. It is just what I want.

#149 Uigingigh on 07.28.12 at 14:07

End user, I use Ubuntu with Gnome 3, which while confusing at the start, I prefer to the large icons of Unity. I read that TB was being wound down in part because there is only so much that you can do with an e-mail client. I really like Evo and would say that the same is possibly true. I don’t know how innovative you can get with a technology like e-mail which is mature.

[WORDPRESS HASHCASH] The poster sent us ‘0 which is not a hashcash value.

#150 Younis on 07.28.12 at 14:22

Dear Sir,
First I’d like to that I don’t know anything about Desktop Development or many stuff you’re talking about.I’m a Web Designer/Developer when I saw Gnome 3 I were very very impressed by it looks how it presented a new way to handle multi-windows and the one that had me that removing the minimize button.Gnome 3 is the best UI I have ever seen till now it make the desktop UI more like Touch UI in a way it didn’t mess the functionally of the mouse and it made the user experience very very uniform not like the Mac and the iPad or not like Microsoft Windows and ‘The Surface’ both of these big well controlled, staffed, funded companies failed to introduce a new unified way to use a PC’s in their many form factor may Apple will do it in the future but Gnome did it now.I not just a fan boy I talking about ‘Think Different’ Gnome did it and the ‘Problems With Gnome 3′ it’s not with Gnome Project ! let’s face the face no one like Linux (OEM I meant) Linux have many distro and I tried very hard to differ one from anther and you can ? what made you choose this one or that one “Fedora,Ubuntu,Mint,Debian,PC linux,Cent OS,Gentoo …. it’s a long list” no one though hey let us make a good apps that actually work and look beautiful ,no every single one kept bragging about has code is more efficient Sucre ….. Ubuntu came with ‘Untiy’ which is the way looks like a mutant child from OSX and Windows 7 and the kick for me they say no to Gnome 3 !!! all I want to say the problem is within our-self as a Linux community Linux companies we all fail to present a useable software to every day use but I still hope one day desktop have Gnome 3 and Phone And Pad, but it just me :)

#151 Cheesed Off on 07.28.12 at 14:50

Here’s a few suggestions,

1. Sack the Gnome team and replace them with sensible people.

2. Regenerate Gnome 2.x and possibly keep 3.x for those that want a dumb user experience.

3. Listen to your users and stop this useless mantra about you know whats best for them.

And btw, the only metric you need to measure a success of any project is the actual uptake of the project when it is rolled out – its really that simple.

#152 brad Yan on 07.28.12 at 14:53

I think GNOME 3 is too complicated.I spent too much time to turn off my PC,I must find the turn off button.Simplicity is very very important.UI is just the UI.The core of linux distribution is linux kernel,not the UI.Let user spent less time in UI is good for user.

#153 ongoto on 07.28.12 at 15:05

I can agree with every comment so far except for one thing; “Gnome 2 is dead” ( because of an announcement ). That’s silly. Gnome 2 didn’t suddenly stop working. It booted the day after the announcement and continues to do so for me. People should insist on having Gnome 2, and should keep using and supporting distros that still offer it and insist on more doing likewise.

Gnome 3 should be the one that get’s the funeral if there is to be one. They should also be the ones who have to rename all the libraries and software they clobber, not the Mate people.

My 2 cents.

#154 Random User on 07.28.12 at 15:21

It seems to me that the issues are not being clearly identified. There is a difference between the GNOME “project” failing and the GNOME “product” failing.

If the goal of the “project” is just to give the developers something fun to code, then the goals of the project are just the personal choices of a few people and GNOME 3 can be considered a success. However, it would be only natural for those people to have differing views on what would be fun next. Therefore the project should just die off as the developers find other projects that interest them.

If on the other hand the goal of the “project” is to make a desirable “product”, then the developers should listen carefully to the users. Users have been saying for years what they don’t like about GNOME and the dumbed down direction that it has been taking. Of course it is loosing market share. When the developers don’t care what the users think, what else could possible happen.

I had always assumed that developers didn’t listen because their goal was not to make a good “product”, but just to have fun, which is totally and completely fine. As a developer, that is their right. If the developers goals and the users needs happen to overlap for a while, great. When they stop overlapping, each can go their own way.

So, to avoid the abyss, GNOME developers first must decide what the purpose of the “project” is: to just have fun making something that very few want to use, or to produce a good GNOME “product” that many want to use. If the developers choose fun only, they should not be surprised by a loss in market share. If they want to make a product they have to start thinking clearly and listen to users, starting with Linus.

#155 xugang on 07.28.12 at 16:42

I don’t care about gnome3 ,it’s not friendly,it’s just link a tablet environment,i suggest the gnome should drop gnome3,and return to gonme2,if not, it will die like nokia

#156 Matz on 07.28.12 at 17:48

I have the feeling that it became kind of cool to dislike Gnome 3, so I’d like to share some thoughts on this.

If a DE like Gnome, KDE, Xfce or [you name it] wants to be competitive, it:
1. has to offer features the user is used to from other DEs
I think Gnome 3 is at eyelevel with all other major DEs. Ironically in some points it isn’t compared to Gnome 2, but nothing that can’t be changed or added with an extension.

2. has to be innovative
Gnome 3 definately is innovative: the possibility to add funtionality to your desktop via extensions is great! Like firefox, you get a good core functionality and can extend it the way you like.

3. has to have sexy look and feel
If you like it or not: you have to make concession to the zeitgeist. OS X is sexy, Windows 7 is a little bit sexy (probably the most sexy Windows can ever get if you look at Windows 8), and KDE 4 is sexy. Gnome 2 maybe was sexy when it was released but definately isn’t anymore nowadays, so Gnome had to change. Otherwise you wouldn’t get any new user because it looks old or doesn’t look modern (it’s dumb but that’s how it is)

Ofcourse the user experience is different to Gnome 2 as it was to Gnome 1. Keep in mind, that Gnome is intended to be a DE not only for developers but also for endusers. Hence the goal to keep it simple and the inevitable arrogance to decide, which decisions to leave to the user and which not. E.g. the panel in Gnome 2 was completly configurable in Gnome 3 it is not (without extensions). Why? Because most endusers just don’t configure it. Heck, most wouldn’t even get the idea to do so! So you don’t want the clock in the center? Write an extension to move it somewhere else. (And btw: Ever tried in Windows to move the clock to the center? Not possible, is it? Think about it.)

I think Gnome 3 is necessarily different from Gnome 2 in order to survive. So please give it a chance to evolve to a mature DE as Gnome 2 did.

#157 anonymous on 07.28.12 at 18:02

GNOME 3 is something between XFCE and Unity, that’s why users leaving GNOME. I choose XFCE (Xubuntu) because it’s speed and fast response on user action’s is exactly what I wanted GNOME 2 to be.

#158 another random on 07.28.12 at 18:20

This is how I’d steer gnome: right into andriod.

the two desktops are merging, because neither is dogmatic about the core. so push it more, get a gnome app marketplace out there, and make it work on linux desktops and android, and figure out how to get the platforms to use the best of bread under the hood.

* I’d look at ways to bring gnome apps to andriod,
* make gnome-shell the best way to do wayland
* ensure systemd, pulseaudio, clutter, cairo?, dbus, wayland are working on becoming the way google wants to do andriod.
* touch
* marketplace

#159 For Linux desktop to survive on 07.28.12 at 19:22

IMHO, for a desktop to survive:

– It has to offer the convenience for common UI tasks, and is familiar to current users. Gnome 3 failed this.

– It has to have something innovative. Eg, it has to grow, otherwise it is dead. Every DE put this as top priority, but misses other points here.

– It has to have something attractive, eg. some eye candy. XFCE is kind of primitive in this aspect.

– It should be elegant, not messy or too fancy (eg. not too much eye candy). KDE fails on this.

– It has to be extensible, the easier the better. So people can add what they think indispensable. Not sure which DE achieves this.

– It better be backed by some organization _long term_. Unity is ugly, hard to use, but it is backed by Ubuntu and will persist and get chance to improve.

#160 z on 07.28.12 at 19:57

You know you’ll never grow your developer community by making a desktop that is hostile to any advanced users whatsoever, be they developers or simply power users.

Make a desktop for idiots and only idiots will use it. People who dev open source software on their free time do it to scratch their own itch. Working for Gnome won’t scratch my itch because it is fundamentally hostile to anything I’d do with it.
Linus Torvalds said it best in the past. Any change he’s going to make in gnome’s source code is going to be rejected unless it’s just a bugfix.

The last line in your blogpost also shows that you’re infected with that mental virus.
“The claimed target users for GNOME are leaving desktop computers behind for types of devices GNOME doesn’t work on.”

Wake up. You’re not going to revitalize a project like Gnome by giving the finger to 90% of current linux users and developers. Target the kind of idiot who doesn’t know how to work with a computer and can only use an iPad and you’ll **never** get the developer community you need to sustain your project.

#161 francesco44 on 07.28.12 at 20:36

Hello Folks

As many people participating to that discussion….I loved Gnome 2. It was fast, simple to understand, stable… I switched to XFCE more precisely Xubuntu. Not half as good as Gnome 2 but so much better than Unity.

For me the discussion is perverted by the confusion of two levels of questions:

1-The convenience for the user;
2-The underlying structure of the user interface (language, style of programming)

Gnome 2 was perfectly convenient for the users. I am not qualified for judging if C++ is better than C. I do not know, either, if the underlying structure of Gnome 2 was adequate for maintenance or evolution

But I think these question should be clearly treated separately. The first one is a question of man-machine interface which can be treated with trial and errors with real peoples.

The second on is a question of programming evolution or dynamics and ease of maintenance.

In France we use to say people “talks with their feet” that is they leave the concert hall if the music or the show is bad. Many people have left Ubuntu because of Unity. Many people have also rejoined XFCE. Aside from the technical underlying question (C++ or C, GTK or…) the verdict of the user base seems quite clear.

Best regards to all the developpers of Gnome 2. I hope they will put the torch in the hands of the future generation of open-source enthusiasts.

#162 Greg on 07.28.12 at 20:51

I’m curious. I see lots of people equally dissatisfied with Ubuntu Unity, but why?

Things I know: The launcher buttons don’t work as expected; they work like Windows 7, for the most part. (Also like the MacOS X Dock? I don’t use Mac enough, so I don’t know.) Window switching is broken by default because Alt+Tab switches between apps. There’s some inherited awfulness from the recent vandalism of Nautilus. There’s a single menu bar at the top, except for when it doesn’t work. The menu bar is only visible on mouse-over. There are no applets allowed in any of the standard, uh, furniture? No gnome-panel, in other words.

That seems like a long list, so maybe this sounds like a joke: Otherwise, I see little difference from the usual desktop. So what’s the big deal? I see the single menu bar, but not the mouse-over behavior, as a good thing. GNOME 3 seems to me to have the same problems, or very similar ones, with a few more (vanishing workspaces, for example) without the benefit of the single menu bar. The vandalism of Nautilus is what leaves me screaming. I wish they’d left it alone and forked or written something new, but I wish they’d done that way back when they abandoned spatial mode.

I do like out of GNOME 3 the idea of a scriptable framework, but I get the impression it doesn’t even get close to the level of Apple’s Automator. Wasn’t GNOME originally supposed to use guile? I know that was abandoned long ago. Still, I think there’s a good idea there that isn’t working out right, just as there was with CORBA, just as there was with Eazel Nautilus, just as there was with the removal of options during GNOME 2.x.

I can’t fail to mention that the dual menu system of GNOME 3 is just an awful idea. I’m not going to back that up; it would take more time than I’m prepared to give.

#163 Linux Torvald's Cat on 07.28.12 at 22:05

A lot of linux is abandonware, just look at a lot of the gnome2 themes and theme engines. Linux is a mess. If i’m running gtk3, gtk2, and KDE4 I have to manage THREE THEMES. Get real. I’ve been using linux since 93, and switched 3 years ago to OSX. I run virtual box with Mint Mate for those few programs (Gimp, Inkscape) I need. I do not need linux for anything else. Apple’s iWork ($79) suite is light-years ahead of LibreOffice or KOffice in capabilities. I like Linux for server stuff of course, but not on the desktop. Its rediculous that the Linux community is split between KDE and GNOME anyway, why does the Linux community waste its efforts on duplicate projects, thats EPIC FAIL. How many years, decades will go by until Linux stops the madness you don’t have much resources why would anyone split them into two desktops. Makes no sense. Meanwhile I think Win7 and Mac OSX have won the desktop. I think its hilarious that the CEO of RedHat uses an iPad, and Linux Torvalds uses a MacBook Air (with linux of course). Linux is great on the server, but on the desktop I got tired of the wars between factions…. I want to get real work done and do real things not “tinker” or particpate in endless wars.

#164 noobermin on 07.29.12 at 00:30

May be I’m a retard, but i like GNOME 3. There needs to be some polishing, perhaps, but I’ve switched to it mainly.

It would be nice to have some real statistics. I mean, just look at politics: you’d think half of USA would support Ron Paul because his supports are so loud and talk frequently but he barely garnered any votes in the primaries. Just because some people (and some famous people like Linus) complain doesn’t mean everyone agrees.

#165 Patrick on 07.29.12 at 00:45

How can I help?

#166 Ian H on 07.29.12 at 02:45

I think gnome needs to concentrate on what it does best – building flexible and powerful tools which can be used to build desktop environments – rather than on the actual design and promotion of a specific desktop environment.

That is the way things are heading anyway.

[WORDPRESS HASHCASH] The poster sent us ‘0 which is not a hashcash value.

#167 Dan Mashal on 07.29.12 at 05:15

I plan to include MATE in Fedora 18.

#168 Techview-Podcast-12-29(Folge165) « Techview-Podcast on 07.29.12 at 06:10

[…] Gnome steht vor dem Abgrund !? […]

[WORDPRESS HASHCASH] The comment’s server IP ( doesn’t match the comment’s URL host IP ( and so is spam.

#169 GNOME 已墜入地獄深淵! | 自由軟體技術交流網 on 07.29.12 at 06:28

[…] 的開發者 Benjamin Otte 日前在其 Blog 上撰文指出:GNOME 實際上已墜入地獄深淵。文中 Otte 列出幾項 […]

#170 Gnomeless on 07.29.12 at 06:47

“Some ideas and redesign are very good, but the no compromise rule goes way too far.”

This is probably the logical conclusion of the Gnome philosophy (attitudes) and ultimately why its dying by attrition.

Gnome devs have been notoriously refractory to outside input , so sure that they are right and that the userbase and everyone else is wrong. The result is a clique who exclude others, get all the control (and all the work load).

As they grow and move on there are no one on the periphery that are contributors that may get more involved. They’ve long ago got tired of being rebuffed and having suggestions and code rejected out of hand. They gone elsewhere.

No compromise attitude is why Gnome is failing its userbase and fails to attract contributors.

Hubris ?

#171 Matěj Cepl on 07.29.12 at 07:41


#172 Amir on 07.29.12 at 08:04

every one agrees that Gnome 3 could be better. I think google’s android is a good example of bringing new things to a GUI, google released android 3.x version for only tablets not the phones, and android 2.x was generally for devices with smaller display. then google built android 4 family and used experiences from 2 previous versions and android once again became one. Gnome can use this strategy, I mean gnome shell is good for tablets and gnome 2.x for desktops. if Gnome project could continue both projects, at the right time, two of them could be one. I think this way is better than fork everything.

#173 tom on 07.29.12 at 08:17

Gnome people could join forces with Elementary Os people. They have potential and solid goals. Cheers!

#174 The PC Samurai on 07.29.12 at 08:38

Note bashing GNOME but I think the Elementary OS is a good example of having Goals that can give value to the users.

#175 David Kastrup on 07.29.12 at 11:10

One advantage commercial projects have over volunteer ones is that there is a finishing line, when resources will get cut and moved elsewhere.

“I have a great idea how we can inconvenience every existing user and start from scratch in a manner where we’ll be half as good as we are now, in about half the time we took for getting where we are now” requires fooling more people where it hurts. “Let’s waste a few dozen manyears and see whether it leads somewhere else” is a harder sell in commercial settings.

A good software manager will recognize when to drag off programmers kicking and screaming from a project which they are damaging with further improvements.

That’s challenging to get right in commercial settings. It is close to impossible in volunteer settings.

#176 bill w on 07.29.12 at 11:38

I used gnome for years, and I’m grateful for what the gnome project has accomplished. However, I tried using Gnome3 and I find it frustrating and less productive. Something as simple as switching windows is no longer intuitive. I used Gnome3 for several weeks to insure that my dislike wasn’t just a knee jerk resistance to change. It looks like Gnome3 is trying to emulate Windoze7 or Mac I don’t know either well enough to know which, but I didn’t find it to be an improvement. I wish I could suggest a better direction. Gnome tools are great.

Good Luck

#177 dave F. on 07.29.12 at 12:19

Here’s my advice for the Linux desktop: burn it to the ground and start all over. Try to avoid the embarrassment of creating ten thousand half-assed desktop environments this time and concentrate on developing one that makes sense and works.

#178 ben on 07.29.12 at 13:07

Gnome3 can do many thing, just slow, very slow.
It’s the most slow DE of KDE4, UNITY base GTK3, XFCE(Of course).
Are the Gnome3 team think every guy use I7+GT6XX so wrote the code so tragedy?

You can run fedora in virtualbox at 1 core.
KDE4 fast, and Gnome3 just slow.

#179 frodowiz on 07.29.12 at 13:08

if it were my problem i would set up a webpage titled “what do users want” and listen to what they say. ubuntu is very good at this and routinely determine targets and send out statistics, poll results and usability results. your customers are whoever uses the product- not the art department.

i have a lot of respect for you devs anywhere on any project but i am a consumer too. i have the irs, my bosses, the neighborhood gossips, the criminals and crappy tv all wanting something from me. my one retreat is opening my laptop and doing my thing. i am going to use the interface that most suits me. i am going to go where i can be heard- or at least feel i am being heard.

this is what gnome needs to focus on. making me feel like i am being heard.

#180 Alejandro Nova on 07.29.12 at 14:22

I’ve used GNOME 3 with a touchscreen convertible, and it was a dream, but simple things took too long and multitasking was difficult even with a touchscreen. Plasma Active (the KDE touchscreen interface) is much more logical, but was being bogged down by Nepomuk on the time of testing, so I had to drop it. Now I use KDE 4.9 in that computer, and I’m quite happy.

And a lot of things have happened in between.

1. Akonadi stabilized. From the unmitigated mess that was in the ~KDE 4.9 era, Akonadi turned into a fast backend for PIM storage. KDE PIM 4.9 is a dream, even with experimental backends enabled.

2. Nepomuk improved quite a lot. 4.9 has seen its Nepomuk performance go UP, and its system load go DOWN (in both cases dramatically) again. Now it’s quite usable, and functionally better than the Microsoft “Libraries” thing (yes, I’m commenting from Windows 8 RP now… I love testing unfinished things ;)). If you try it, it will be slow, for you, only ONCE (while it’s indexing things, and that indexing can be throttled to minimize CPU load)

3. Plasma bugs, even with multimonitor, are a thing of the past.

So, that takes me to this. If GNOME has so few developers, why don’t try to work together with KDE? If Evolution-Data-Server is so bloated, why don’t replace it with Akonadi (Akonadi, BTW, doesn’t require kdelibs, and I don’t know if it even requires Qt) I don’t even have to speak about Nepomuk; if you ask the Zeitgeist development team, they are already integrating Nepomuk, and a Zeitgeist + Nepomuk experience is supported, and even encouraged.

The “Akonadi and Nepomuk are buggy and slow” mantras and excuses are valid only while Akonadi and Nepomuk are buggy and slow, and that’s not the case anymore with lots of typical use cases (including using KDE PIM). KDE 4.10 is going to debut with “social feeds” delivered right on your desktop, mail feeds, and a deeper integration of Nepomuk and Akonadi.

#181 NIH on 07.29.12 at 15:03

They won’t use any KDE-developed library as they didn’t use libappindicator, unity or whatever already-ready-to-use solution available

#182 thaodan on 07.29.12 at 16:16

I also suggest to work together with KDE and use already working frameworks so that only difference is the frontend, this woud make development faster and it woud be easyer for application developers to buid applications for Linux cause they now frameworks work on KDE and Gnome.

#183 thaodan on 07.29.12 at 16:18

I want to add that Unity implements many things that are already present and suggest to only use them if there’s no counter part that is already working and used.

#184 afanen01 on 07.29.12 at 18:11

Of what use is another Gnome 3 lover’s voice? Even if this gnome 3 lover uses gnome on a dual-screen setup with two 22 inch monitors?

My frustrations with Gnome are few, and are similar to the kinds of frustrations I’ve always had with Linux: almost every time I launch vmware/virtualbox, i find that i have a new kernel and i need to recompile something. I cannot for the life of me understand why i get a new kernel every week.

#185 Richard Neill on 07.29.12 at 21:36

I’ve never understood why software developers (in virtually every big project) don’t seem to internalise the following rules:

* If you change something, make sure the user can change it back.

* Don’t deploy a new version until it has substantially all of the features of the previous version available and working.

KDE4 made this catastrophic error, which they are just about recovering from now (though I happened to use an old PC running KDE 3.3 last week and was surprised at how much prettier and more responsive it was than 4.8). Gnome failed to learn, and did it with Gnome3. MS are about to do it with Win8.

Imho, the real problem is that KDE/Gnome try to be too much. The Unix philosophy is to have a collection of small tools. What we should do is make a clearer split between:

1. The Window manager: WM, panel, panel applets, configuration tools for X (kbd/mouse/scrrensaver), wallpaper, desktop icons, menus, theming.

2. A collection of interchanageable apps (nautilus, gnome-office, file-manager, etc)

3. The “social desktop” stuff.

#186 Leslie Satenstein on 07.30.12 at 02:23

Gnome 3 is a new paradigm. It is a different way to present a graphical user interface. G3 is new, and with each update, features that need to be added, will arrive.

If the Gnome people had the staff, they would have created Gnome3 under a new name, and would have retained Gnome2. But there is a limit to resources who need salaries and to numbers of volunteers. Ergo, they froze Gnome2, and concentrated on G3.

That means that I have set aside Fedora 17 as my serious system, and have concentrated on Debian and Centos. Both have Gnome2 interfaces.

I very much enjoyed G2 with Compiz, and the wobbly windows. (We all need to show wobbly windows to a W7 user), or show the desktops on a cube.

If I load Ubuntu, it is essentially to bring the software up to date with the latest patches. Ditto for Fedora 17.
On small netbooks, Unity or Gnome2 take up too much of the screen width, particulary with the favorites bar.

(I use 2 laptops, and 3 physical desktops and multi-boot each)
I develop software for both 32 and 64 bit Intel systems, and I need true machines to do my testing.
My installed testing distributions include Debian (32 & 64bit), Fedora (32 and 64bit), Ubuntu 12.4 (32 and 64bit), Linux Mint and Scientific Linux (32 and 64bit) and Windows (home and professional versions 64bit).

I use Gnome2, G3, and KDE. I have not stayed with xfce, because of eye appeal.

#187 Tristan on 07.30.12 at 05:12


‘if it were my problem i would set up a webpage titled “what do users want” and listen to what they say.’

‘If I had asked people what they wanted, they would have said “a faster horse”‘ — Henry Ford

Or a more up-to-date quote from Ars Technica’s John Siracusa review of Mountain Lion, in response to Mac fans’ very similar complaints about Apple’s relentless iOS-ification of their beloved OS X:

‘The fact is, we are not the center of the market, and haven’t been for a long time. Three decades ago, the personal computer industry was built on the backs of technology enthusiasts. Every product, every ad was created to please us. No longer. Technology must now work for everyone, not just “computing enthusiasts.”‘

#188 Eh on 07.30.12 at 07:31

> A good software manager will recognize when to drag off
> programmers kicking and screaming from a project which they are > damaging with further improvements.

Most of the core GNOME 3 developers are employed by Red Hat.

Hence, Red Hat management could do exactly that (fire them and hire someone else), and it would probably be by far the biggest contribution to desktop Linux they can make right now.

#189 Enrico on 07.30.12 at 09:26

I think that Gnome 3 UX is the best that exists at all. Allan Day, Matthias Classen, Jimmac they make the best possible work in the field of UX.

Integration of all functions such as Document, Music, Video, Contacts, Chat, Email, Virtual Machines (Boxes) with uniform Gnome 3 UX, is in this area excellent ! It is really beautifully and logically designed.

It is clear to me why they decided a few years ago much more to touch interface. Touch devices today are the only branch, that has an increasing tendency, Desktop PC and Notebooks fall in sales.
I personally use ThinkPad X220 Tablet

90% of PC users are common users, these users such as managers, doctors, lawyers, writers, architects, designers, retirees, these people are not stupid, but they want to use computers to write documents, print and scan documents, share and provide documents, create slides, email and chat with others, photo view and search, while relaxing video and music experience. These users want to use the computer only for work, they want to set anything (most of them can not even change the wallpaper on Windows or even burn CDs)

The rest 10% of users are administrators and programmers, etc (and they know how to set everything).

I think it is better to make a simple but beautiful and intuitive environment for average 90 users, than an environment where you can change everything through different adjustment for 10 users.

The programming is also fine, Gnome Shell is most writtern in JS with CSS theming etc. So technologically it does not lag behind Qml (which incidentally came even later). And we have Vala.

Cinamon, Mate, Unity, so ok, they using it on their own way , but they still use Gnome technology.

Gnome is default and prefered by Debian, and most of distros is Debian based distributions.


I wish that the Gnome UX team continue to work on these great ideas and even more with other Gnome core applications.
I wish It will be more documentation to beginner available to start work on Gnome
I wish to have Vala perfect IDE for quick and easy programming.
I wish that all pages have been modified according to a new look (live.gnome)

In these areas I want Gnome to help in my spare time.

#190 RomGnome on 07.30.12 at 10:06

the only thing I can say is that I’m a daily user of Gnome 3 since it came to Fedora (I had to leave Ubuntu after the one of the early version of Unity bricked my computer, I haven’t tried it since then, it’s probably much better now).
After a few hours to get used to it I started to like it. and it’s getting better after each release. Sure there are stuffs to improve but Extensions do a good work at correcting the problems.
I hope that developpers will continue to work on Gnome 3, it seems to be a very worthwhile project !

#191 Tiao on 07.30.12 at 11:05

Using Gnome3 on a E350 processor, makes my load average goes to constantly at 0.81, even with all social and indexing features off.

Bring back Gnome2.

#192 PizzaPill on 07.30.12 at 13:13

Hi Benjamin,

I can only support your observations, I recently started a Poll about Linux Desktop Usage and KDE is far ahead.

The Poll can be found here:

It wasn’t quite the result I expected tbh.

#193 apo on 07.30.12 at 13:30

*crouch* I like gnome3. With a panel (from extension) productivity on a desktop is like gnome 2 but the interface is more polished.

Working on a 13″ screen is fun with gnome 3. Cannot see why most of you are raging.

#194 klapauzius on 07.30.12 at 14:11

The GNOME community are fooling themselves and the few users they have left.

Every time GNOME3 is discussed sooner or later they’ll tell you how “clean”, “modern”, “customizable” blablabla GNOME3 is supposed to be compared to the “hopelessly outdated” GNOME2.

So i asked a GNOME dev once on a mailing list: If GNOME3 is so “clean”, “modern” and “customizable”, why didn’t you simply ship it with a set of extensions to make it look and behave exactly like GNOME2 with the new “clean”, “modern” and “customizable” stuff underneath ?

Answer: It’s not possible. We’d have to rewrite the extension system for that.

That’s when i switched to XFCE.

#195 Will B on 07.30.12 at 16:25

Sorry if I’m repeating something already said, but I truly believe the negative changes with GNOME came when GNOME 3 was released. Sure, GNOME 2 may not have been ‘exciting’ and ‘fresh’, but it served a purpose…to make the user productive. GNOME 3 may be great for mobile devices, but, let’s face it, not everyone is using a mobile device to develop software, run a business or service customers.

Personally, I prefer the GTK toolkit over others, and I really hope that GTK doesn’t die. I hope that the GNOME developers (how few there are) get together, and address the deficiencies in GNOME, and that there will instead be a ‘GNOME Desktop Edition’ and ‘GNOME Mobile Edition’ instead of forcing one type of GUI on every user.

For now, Xfce 4.10 is doing the job quite nicely, and I truly hope they do not go the same road that GNOME has.

#196 Ikem on 07.30.12 at 16:44

Ikey Doherty is working on SolusOS.

For that he’s patching Gnome 3.x.

His goal is to make it work and look like Gnome 2.x.

Maybe he could need some help.

#197 Is GNOME In A Shell? on 07.30.12 at 17:37

[…] is con­cern that GNOME might wash out with the tide. As GNOME users can clearly see from this blog post, at least one devel­oper is show­ing some deep-felt con­cern over recent devel­op­ments. Core […]

#198 pirast on 07.30.12 at 18:49

having considered the fact that GNOME is understaffed, it would be probably best to concentrate efforts.
that would mean:
– drop gtk+ and glib (with only 1 full-time developer working on it) in favour of qt which is superior and has more active developers right now (the question is for how long taking the situation of Nokia in account)
– develop a new, better GNOME based on a new, improved, brilliant GNOME HIG which is easier to use than KDE (KDE offers too much functionality)
– reuse as much code as possible (kdelibs,…..)

#199 Geoff on 07.30.12 at 19:00

Put me down for loving gnome-shell.

I think it’s a wonderfully inspired and philosophically well-considered new shell environment. I use it daily and I’m extremely productive with it, so I have to conclude that people that declare it be somehow objectively “unusable” are off their rockers. If it’s just not your thing, that’s okay.

There are still problems, like any young shell, but they’re fixable.

#200 KORT: Piraterijboete gaat niet naar muzikanten | Tech-nieuws on 07.30.12 at 20:18

[…] Benjamin Otte beschrijft de langzame dood van Gnome in een blogpost. Volgens de programmeur is Gnome richtingloos sinds de originele doelstelling om een open en gratis […]

#201 Is GNOME “Staring into the abyss?” on 07.30.12 at 20:40

[…] Benjamin Otte, a leading GNOME developer thinks GNOME, once a popular Linux/Unix desktop but now more often used as a foundation for other desktop interfaces, is “staring into the abyss.” […]

#202 Staring Into The Abyss: Some Thoughts | jonobacon@home on 07.30.12 at 21:30

[…] week Benjamin Otte shared some thoughts about GNOME that were pretty stark. It gathered some steam and hit Slashdot and this all happened the week […]

#203 Nemo on 07.30.12 at 22:00

Gnome3 was an epic fail since day zero. The main flaw is the idea of forcing people to use their computer in ONE single way that is what Gnome3 designers want, for no other reason than they can. And since personal computers aren’t smartphones either tablets, the whole thing is a nonsense.
Like Linus said, after day zero the other issue was nobody at Gnome wanted to hear that the project was an epic fail.

#204 Leon on 07.30.12 at 22:07

As an end user, the biggest issue was simply the challenge of using Gnome 3 with proprietary drivers. There was an apparent mindset to make Gnome 3 use 3D acceleration and use the opensource drivers, and that this decision would somehow influence rapid evolution of 3D support in the opensource drivers and was even written >>roughly<< in those words on GNOME's website. While this is a noble long term goal, it was executed in a way that immediately sacrificed the user experience to to the point they had no choice but to use another DE. People generally used the proprietary driver when they need 3D to work now instead of some future date long after the device is considered obsolete. Also, the interface of the shell is almost as annoying as Unity :P

#205 V.M. on 07.30.12 at 22:14

Tiao: “Using Gnome3 on a E350 processor, makes my load average goes to constantly at 0.81″
There is something wrong with your installation, I use the same CPU (AMD E350) and the average load caused by Gnome Shell is about 5% (Fedora 17).

The GS is really nice, the transition phase is always difficult, the dust will settle down and things will be better again. Please keep up the good work.

#206 Grigio on 07.30.12 at 22:55

Seriously don’t You see that everything is moving towards the web, Javascript/coffeescript, webapp, ..?

GNOME 3 failed, it broke with GNOME 2 but I still don’t see the benefits, a new FREE or proprietary to succed should beat the current desktop/mobile OSes with:

– Web technologies (HTML5/CSS3/JS/WebGL) as 1st class citizens, i mean a very powerful accelerated webview/browser to make native webapps ala PhoneGap
– A “native” simple and cool: text-editor like Sublime Text 2 (sorry Gedit abandonware), a “native” video-editor, X/Gtk window emulation for legacy apps.

Chrome OS and Ubuntu OS are in different ways toward this path. Other projects are very far or proprietary OSes

#207 Stanislav Maslovski on 07.31.12 at 06:24

I think the best solution is just let it (i.e., GNOME) die. It has been in a frankenstein state for far too long, IMO. Admit a defeat and let the other environments take its place.

#208 Paul on 07.31.12 at 06:41

¿A suggestion?

– Discard Gnome Shell. Almost nobody likes it.
– Try to save the few good things of Gnome 3 (for example GTK3) and transform Gnome 3 Fallbak in the official Gnome 3 (Everybody loves Gnome 2).
– Manage to make that Compiz come to run in Gnome 3. Let’s Admit It: The Golden Age of Gnome was Gnome 2.X + Compiz.

#209 Claudia Huurkman on 07.31.12 at 08:37

I know I’m going to write a lot of things that could be wrong because I have looked at Gnome only from the outside as a user. Also my english is terrible and I’m not attacking you, but the project itself.
But you know… you guys created a lot of problems to me with your choices. Time for me is money. I was going to give money to your foundation yearly, but I don’t believe in projects like yours anymore. I invested my time using your software. I trusted you. So now I just want to vomit what I think.

I don’t understand why GNOME guys wanted to quit with their users base. Because tablets and smartphones became popular? Please give me a break.
If Linux was not able to conquer the PC market when there was Windows Vista or now that Windows 8 is coming out and people hate it…why they should get better results on tablets?
At least with a PC you can install a distro. With a tablet do you think users will be willing to hack it just to use Gnome? Or do you think you can compete with Google or Apple? Hardware manufactures will care to pre-install a OS made by a small foundation?

People don’t use Linux because there aren’t enough good programs. There are a lot of programs that are not professional or are semi-abandoned.

That is because most of developers for the desktop are students and when they go out of college they get a job have no time to continue. Or because they want to try new things and since they do it in their free time they feel no obligation to users.

While for servers application there is an interest by the industry so they finance the programmers. On the desktop there is no way to make money and pay the open source developers. Maybe just for some programs used in the office and backed by companies like Redhat. But how do you make money with Gimp? (I know gimp is not part of Gnome but it is just an example). The only programs that work are the ones that are baked by money. Mozilla Foundation found the way to survive. Blender foundation found the way. Gimp did not and indeed sucks and have a few developers. If you use it now, version 2.8, still looks like a software from the 90’s. And Gnome foundation can take money just from companies, not from users, because most of the users don’t understand your project. Now a lot of companies that invested in open source are not there anymore: SUN, Novell, Nokia is dying and I don’t know…Google is giving something? I heard now you want to involve much more the people for giving donations…well too late, you piss off users then ask for money? Very bad idea.

I remember at that Guadec where you started to talk about Gnome3. You were all using OSX and you were convinced that the problem was that Gnome2 was ugly and looked like old and that the future were the netbooks, tablets and smartphones. You, Canonical and Microsoft in the latest years seemed that just wanted to follow Apple’s trend.

Beauty was not the problem for the users. It was a problem for you guys that are excited only by new shiny things.

Some of you are very good programmers and very smart, but you really can’t empathize with users or understand what people want.

I remember De Icaza showing the Compiz cube everywhere. It took years for him to understand that the cube’s metaphore for workspaces was bullshit.
And he is a very smart guy…
How can guys like you pretend to design the future of Gnome and to conquer the world?

People don’t want to use Gnome or Linux in general because there are NO GOOD PROGRAMS for the Desktop. The Gnome2 interface was FINE well not perfect but at least USABLE, the PROGRAMS were not GOOD at ALL.

Emule had one of the worst interfaces ever seen, but reached hundreds of millions of users because it had a feature people wanted: to download everything.
If you offer programs that are very useful even my butcher will use Linux.

And it’s not with stupid things like Tracker, Beagle etc… things that track my ass and show my boyfriend’s picture when I search a document that has nothing to do with him that you will make people change their mind. Every time I install a recent distro I have to turn off everything that invades my privacy.

Also the most funny thing is that the BEST open source programs for the desktop are ALSO available for EVERY other OS: Windows and OSX
But the best Windows programs are not available for Linux.
How can you compete? How can you reach the 10% or 20% or whatever? Are you nuts?

There are not enough good programs for Linux and the GUIs to configure hardware are bad. Like the last one I’ve seen to configure the graphic tablet…please is that a good job? Something to show around? Every time with every distro I have to get nuts to make it work. It seems you guys hate people using graphic programs. But the funny thing is that the best tools on the desktop are graphic programs: Blender, Mypaint, Inkscape, Krita, Hugin, Shotwell…but you seem to hate them and their users.

What is the meaning to have Epiphany when there are already much better browsers for Linux? But there are not in many other fields!
What is the meaning of GNOME itself?

Try to explain to a user this: “well you know…there is Linux…oh wait Linux is the kernel…what is a kernel? Too hard too explain….well the OS should be called GNU/Linux…yes beautiful name I agree… and you need to install a distro…what is a distribution?…well…a distribution just take care to package everything…oh what is packaging?…OK OK but let’s talk about the Desktop you can have: Gnome, KDE, XFCE. What are they? Well… I don’t know really… I think they are desktop enviroments and… What is a Desktop Enviroment? mmmm… No no you can use KDE programs on Gnome and viceversa and… Ok let’s install Ubuntu…wait Unity sucks! Let’s try Gnome3! oh wait it sucks even deeper. I see you’re confused… ok let’s buy a Mac.”

And last note: What was the meaning to hire that CEO from HP? I remember on Gnome Planet people saying: “Oh guys you don’t get it but she is doing great things for the future of GNOME!!!”
And on her blog she was writing those long posts about buying books for everything, like “my son didn’t eat the salad and now I search on Amazon the solution”. She is the classic CEO that takes decision only about things solved by others. But as Einstein said intelligence is not about knowledge, but about imagination.
Wow I see how many companies are now baking Gnome. Really a good Job!
The only company investing in the Linux Desktop is Canonical that is managed by that Steve-Jobs-Wannabe and is not giving anything back. Probably because you guys are so complicate to collaborate with. Linus Torvalds didn’t know how to submit a patch. Canonical didn’t know how to propose features. Or they’re all dumb or you’re open like a secret lodge.

And about Shuttleworth…a guy that thinks HUD is a great thing. Like some of you guys blogged on Planet Gnome years ago Ubiquity for Firefox was great. Did people care about Ubiquity: NOT AT ALL
Have you ever worked 5 minutes with average users? They HATE the keyboard.
Mark made a lot of money, but with something that has nothing to do with the Linux Desktop. He has no taste and when he speaks he uses so many words for not saying anything. Mark’s Keynote: bla bla bla Really good job guys! bla bla bla Ubuntu is about quality bla bla bla
He said that Ubuntu now is better than OSX? Oh nevermind I know you have nothing to do with him. But it’s several months that I moved from a state of shock to another one where I’m pissed off about the state of the Linux Desktop.
How could you guys did what you did? Linux Hater’s blog was so right.

#210 Is GNOME “Staring into the abyss?” – Mr Pod Blog on 07.31.12 at 10:22

[…] Benjamin Otte, a heading GNOME developer thinks GNOME, once a renouned Linux/Unix desktop yet now some-more mostly used as a substructure for other desktop interfaces, is “staring into a abyss.” […]

#211 pirast on 07.31.12 at 10:39

honestly, just looked at GNOME again (on a normal computer). Using cinnamon it just feels right and has a lot of potential (especially for users who do not want to use metro and want a free desktop os). So much better than unity and everything else.
I do not see why having read the “bright future plans” GNOME wants to get mobile centric. Best would be to offer a normal desktop, looking like cinnamon, and a touch optimized desktop (GNOME 3). Could all be based on GNOME shell if it could be extended in that way.

But: There are some decisions that I do not understand – why are you changing scroll bars? Just to make it look like Mac OS X? On a desktop, it is not an improvement (they are to small, hard to catch).

I agree that GNOME needs a plan, but a good one (and not a GNOME OS plan).

#212 klapauzius on 07.31.12 at 12:08

And here’s a GNOME developer’s take on this article and the consequences for GNOME:

“And as for taking the desktop to the next level I think the main challenge for GNOME is not inside the GNOME 3 desktop, but instead it is a question of if the open source community can come together around pushing things like the Linux OpenGL support forward and make sure we have a strong foundation for supporting games and improving visual effects in the desktop”



more visual effects

He left out Facebook and twitter integration – probably just forgot it.


But the OSS world will just move on. That’s the beauty of it.

#213 Da li GNOME projekt zuri u bezdan? | Linux Za Sve on 07.31.12 at 12:36

[…] tri dana je Benjamin Otte, jedan od vodećih GNOME developera, objavio zanimljiv članak na svom blogu. U njemu izražava svoju zabrinutost budućnošću GNOME projekta i navodi ključne probleme u […]

#214 Job on 08.01.12 at 01:20

1st, Windows is a messy to setup and complex to users. Mostly does not know how to setup many things that they need. But, Windows is everywhere than users feel that is better. Gnome3 try to convince that they Want common users, but they don’t hear their users. Mostly spread of Linux come from technical people, that do these for friends and in the work. Like I do. My users, are happy with Linux, many found Gnome2 more easy and logical than Windows, and its true. But, Gnome3 people, does hear their community, thinking that they does not know users. Its wrong.

#215 MB on 08.01.12 at 04:02

As a VERY long time Gnome and Linux user, I have mixed opinions about this.
When I moved to Gnome, it was because it blew the competition away. CDE, fvwm, these things were terrible. Gnome was very very promising, and up until the Ximian era Gnome was the best there was. The Ximian change was *terrible*, but eventually Gnome got beyond it, and Gnome 2 became a decent basic environment. Gnome 3 just reflects another big shift – and once again, it’s terrible. I expect Gnome will get beyond it, as well… of course, I won’t be using it, unless they port it to tablets. Ubuntu seems to be one of the only distros poised to leap into this space – and they’re going to do it with Unity.

#216 romu on 08.01.12 at 07:41

As I said here above, I’m a big fan of Gnome 3, and I’m happy to read I’m not alone.

I’m surprised to read messages about Gnome 3 as a less productive DE as Gnome 2. Unfortunately, such messages come without any arguments to justify this assessment.

To me, that’s the opposite, I find Gnome 3 really more productive than Gnome 2, here are some examples:
– ability to launch applications with the keyboard only, in Gnome 2, to do the same, you must install something like Gnome-Do
– all is smoothly integrated with the “Super” key magic, dock, virtual desktop, applications menus

For this second point, this is where, in my opinion, Gnome 3 is superior to Unity. In Unity you have to remember one shortcut to drive virtual desktop, the super key to launch applications where there is only one key on Gnome 3.

And I don’t mention compiz which appears as a technical dead end if you read some linux web sites. Finally, I also find that Unity is too resources consuming compared to Gnome 3.

But, I also read and agree with some arguments against Gnome 3, especially the ones about the web browser:
– The browser war is over, Firefox and Chrome won, why spend time and money on Epiphany?
– I don’t work only with Gnome, but also with Windows 7, Epiphany doesn’t run on Windows and can’t offer the Firefox’s sync feature.

Instead, I would propose to get rid off Epiphany and focus the efforts on making an excellent Gnome plugin for Firefox, this one exists, it’s called Adwaita, and maybe it deserves some efforts to make Firefox the official Gnome web browser.

Why Firefox and not Chrome? First Mozilla the fully commited to free software, I can’t say the same for Google, then Firefox is more memory efficient than Chrome.

2 cents more.

#217 Glyn on 08.01.12 at 10:10

Whilst I appreciate the value of highlighting these issues, I think it is also necessary to look outside the box somewhat for other probable causes, certainly regarding contributor numbers. We rely totally on volunteers to create our site content and despite record numbers of folks being unemployed with internet access, or working from home, our recruitment levels are dropping all the time. Maybe folks are just too busy trying to stay one step ahead of life in general. :) Back to another point in the topic, having looked at the likes of Xubuntu/Voyager, I can’t really make an argument for Gnome at all.

#218 pirast on 08.01.12 at 15:22

the point that you have to press “ALT” to be able to shutdown your system in GNOME 3 clearly shows that developers do not understand what users want.
Microsoft made the same mistake with Vista (Suspend was default action), but corrected it with 7. GNOME just sticks to a wrong decision.

#219 Brian Fagioli on 08.01.12 at 15:22

Maybe I am in the minority as a huge Gnome fan. I have been on and off as a Linux user. I dont like KDE or Unity. Gnome 3 on Fedora is what really made me a huuuuge gnome fan. I find it to be a very productive UI. I add no extensions to it and LOVE it. Without gnome I guess I would move to xfce but I wouldnt be happy about it. Keep Gnome 3 alive!

#220 S05E12 – The Portland Sketch Book – MP3 | Ubuntu Podcast on 08.01.12 at 20:36

[…] GNOME loses its way […]

#221 S05E12 – The Portland Sketch Book – OGG | Ubuntu Podcast on 08.01.12 at 20:36

[…] GNOME loses its way […]

#222 S05E12 – The Portland Sketch Book | Ubuntu Podcast on 08.01.12 at 20:39

[…] GNOME loses its way […]

#223 Jason Bourne on 08.02.12 at 00:21

I tried GNOME Shell for 2 seasons. About 2 months the first season, and 3 weeks on the second season. I definitely hate it, because I can’t get work done.

I went distro hopping and moved to Arch, CentOS and Debian. Unfortunately, Arch dropped its installation framework; CentOS has software from the last century and Debian can’t seem to understand that I have a new ATI GPU…

So I’m gonna stuck with Ubuntu 12.04 and Unity. Unity is not my cup of tea, but I’m trying to make it grow on me. It feels more comfortable than GNOME Shell. I think GNOME 3 needs to dig the old roots, admit the fail and redo it. But I know it won’t be possible – with those issues risen up, who will be interested? It’s up to Red Hat now decides what to do – Fedora has gone down in popularity quite a lot – 4th on Distrowatch and keeping falling. It will be up to Red Hat.

And I gotta admit that Ambiance is some hell of a theme, I never get tired of it.

#224 Hank on 08.02.12 at 02:03

You guys failed. You didn’t care about the people who helped make gnome what it is when a shiny new project was proposed.

Who are you making software for?

#225 Tarnylo on 08.02.12 at 06:28

@Benjamin Otte:
What kind of help do you need on GTK+ ?

#226 klapauzius on 08.02.12 at 07:57

I think the GNOME team misunderstood the nature of OSS.

If you screw up and deliver a failed project in the corporate world (like Win 95 or OS X Lion) you have a userbase that you can force your failed project on for whatever reason (psychology of prior investment, inability to move on, whatever).

In the OSS world, this is different. People will look at your project, realize it’s a failure and just move on.

So in the end GNOME3 shows that the OSS model works. Failed projects go where they should ( > /dev/null), true innovations persist.

That’s the beauty of OSS. Choice.

#227 klapauzius on 08.02.12 at 14:21

Here we go again.

The next version of Nautilus won’t have F3-Split View any more.

It also won’t have compact view any more.

Why ?

Just because.

That’ll lose them the last 20 users or so they’ve left.

Ah well. R.I.P. GNOME


#228 Los Dos Costados de GNOME on 08.02.12 at 14:39

[…] […]

[WORDPRESS HASHCASH] The comment’s server IP ( doesn’t match the comment’s URL host IP ( and so is spam.

#229 JO on 08.03.12 at 10:57

@Mike Fisher: It would be sweet if you could put a step by step on how to get that desktop up and running on a blog or the Ubuntu forum or something.

#230 GUADEC 2012: Discussions! « andré klapper's blog. on 08.03.12 at 11:26

[…] people discussed the concerns expressed by Benjamin in a recent blog post. I know that Benjamin deeply cares about GNOME but I am not sure with some of the conclusions […]

#231 La encuesta de Agosto: ¿Que ambiente de usuario dominará Gnome en un futuro? | La mirada del replicante on 08.03.12 at 19:36

[…] las últimas semanas se han producido opiniones de todo tipo, desde desarrolladores que critican el proyecto debido a una falta de claridad en los objetivos a conseguir (“mirando al abismo“), a […]

[WORDPRESS HASHCASH] The comment’s server IP ( doesn’t match the comment’s URL host IP ( and so is spam.

#232 GNOME, a la deriva on 08.03.12 at 23:11

[…] y Rhythmbox, por extensión uno de los desarrolladores más importantes dentro de GNOME. En Mirando hacia el abismo Otte trata de manera concisa y explícita los grandes problemas de fondo que tiene el […]

#233 Kevin Matthews on 08.04.12 at 04:33

I am still wondering why Gnome 3 was rolled into production half-baked after KDE’s early release proved ill-advised. If nothing else Gnome 2 should have been offered with some support and security bug fixes for a set period of time. Offering this support would have provided a fallback for conservative users and eased the user-base into the transition.

#234 BrewSpring on 08.04.12 at 10:39

Now I’m try Ubuntu Unity and Windows

#235 staring into the abyss — Swfblag | txwikinger-open-source | on 08.04.12 at 23:08

[…]   […]

[WORDPRESS HASHCASH] The comment’s server IP ( doesn’t match the comment’s URL host IP ( and so is spam.

#236 sebastian on 08.05.12 at 12:41

I really hope that the Gnome organization can get more manpower and resources. For me as a user, the gnome shell is almost “done”. It’s working fine for me and looks absolutely beautifully. Time to fix bugs and document it for the next 2 years. Leave gnome 2 behind in MATE, this was really hard but there’s not much lost if gnome doesn’t have much more developers than MATE anyway? Also for the users, switching your install to MATE takes less time than arguing online about what a bad decision it has been to leave gnome 2 behind.

I absolutely love Gnome as a brand too, the name is cool, the logo was always cool, choice of fonts and graphics is very beautiful but not too shiny, exactly my cup of tea. The UI of Gnome 3 is beautiful and it interfaces well with Ubuntu fonts etc. and runs perfectly on a small netbook.

IMHO the shell will work well on tablets, but forget the stupid phones. Maybe there could be a GnomeOS based on debian, my dream would be a decent hardware like the mintBox (, with a decent performance and price and based on the Gnome brand. Maybe wait for Steam to arrive on Linux and have Steam integrated cleanly in the shell.

In my opinion Gnome should do frequent user surveys (it’s cheap) and connect more with its users and build a even better brand. The wiki unfortunately looks the same as 5 years ago and it is not helpful enough for developers to really get into, even if you wrote code for years. Make the code more accessible to devs so they can write third-party tweak tools for customizing the desktop when Gnome obviosuly don’t want to do that themselves.

A survey function or user interaction is kinda missing in the desktop itself. There’s no “Report Bugs” menu item or “Get Support”or “Ask Gnome” (like or something. After the install you are kind of left alone with it.

#237 Die Woche: Der Desktop zersplittert | on 08.05.12 at 12:45

[…] der langjährige GTK+- und Gnome-Entwickler Benjamin Otte kürzlich in seinem Blog-Eintrag “Staring Into the Abyss” anriss. Das hört sich nach einem Gnome-spezifischen Problem an – letztlich hat es aber […]

#238 Leslie Satenstein on 08.05.12 at 19:48

When Gnome3 came on the scene about 1 year ago, I like others went into shock. We had no warning and we felt ignored. If you read all the messages since the beginning, you had a choice. Learn the new Gnome3, or stay behind with Gnome2.
Over time I continued with G3, and continued with G2 under an older Linux distribution.
However, slowly, G3 had some new improvements, provided as tweaks. I implemented two of them. a) A tweak to include power off as an addition to suspend and b) a tweak with G3 to provide a menu tree.
Here is what I find as a shortcoming.

I would like to be able to add to the favorites bar, a directory or two of my choice. One click to show the directory on the screen.

I would like to relocate the application name data entry on the top right to a better location. Why should I click on the upper left, and then have to move to extreme right? The icons and their functions could be on the left below the Windows title.

I have a wide monitor, why can I not treat it as two monitors? Why do I need two physical monitors to get what I need (documentation on one monitor, my source code on the other).

Why can I not reposition the favorites bar to any side?

Why can I not change the size of the icon display inside the favorites bar.

How can we interface with the Gnome developers in a constructive manner to get enhancements to G3?

G3 is great for wide screen monitors. When using a 4×3 ratio monitor, it is horrid. The favorites bar could be floating and that would help.

Combine the use of the favorites bar with the G2 presentation of multiple desktops. Each desktop, as opposed to being dynamic and requiring the sliding of the mouse to the extreme right, in order to select the one you want next is an invitation to Carpal Tunnel syndrome for your click finger and your wrist muscles.

So, what we should be doing is not griping, but proposing new enhancements.

Mate is coming and perhaps it will satisfy most of the previous commentators (bloggers). I want to be objective. If there is a new interface, learn it well and then take a decision about what you want to use for long term.


#239 Staring into the Axis’ Abyss: the Railgun map | Nekohayo ! on 08.06.12 at 00:02

[…] sites. News sites don’t care about positive things, it is much more interesting to talk about impending doom. Don’t get me wrong, it is quite healthy and necessary to talk about these things, but it is […]

[WORDPRESS HASHCASH] The comment’s server IP ( doesn’t match the comment’s URL host IP ( and so is spam.

#240 GuiMaster on 08.06.12 at 08:21


I have one simple solution to this entire problem:

When something doesn’t work, the best solution IS NOT trying something new. Never! The best solution is ALWAYS to go back to what worked in the past.

There is a great Desktop Environment included with Ubuntu 12.04 called:

“Gnome Classic” or “Gnome Fallback”

Please, to those who contribute to this project: Get to work on giving users a “Gnome Classic” environment that gives all the functionality of Gnome 2x while having all the benefits of Gnome 3x. Make that a primary focus and you’ll start getting users back.

Admit that Gnome 3 isn’t working out the way you wanted and work on improving Gnome Classic. Pride goes before a fall, humility will save this sinking ship!

#241 mikhas on 08.06.12 at 10:15

I find it funny, in a sad way, how this article got so much media coverage, yet we find blog posts on PGO that try to explain us that nothing is really wrong. Not too surprising, those other blog posts get completely ignored (perhaps because people don’t agree with them?).

#242 Hitler goes mad about Gnome3! | woGue on 08.06.12 at 18:50

[…] and now we got your attention lets do the news! There was another respond to Otte’s post about the future of GTK this time from Jean-François. Will we list every single response to this […]

#243 LiNutZ on 08.06.12 at 21:18

“Nobody can say if GNOME 3 is better or worse than GNOME 2.”

Ugh, I can. And so can many others whom use Linux desktops exclusively. Gnome3 is a failure. We don’t want mobile on our boxes. We want “Usability”, ON THE DESKTOP.

> “Distros are dropping GNOME for other environments instead of working with GNOME.”

Do you blame them? Who got up one morning and decided to change our desktop to the hell that is hand held devices?

> “Previous supporters of GNOME are scaling back their involvement or have already dropped GNOME completely.”

Do you blame them? Who wants to support a project that had, and has, the intention of removing the desktop for all intense and purposes and replace it with the hell that is a hand held device? Just because the new fad seems to be down_with_the_desktop, up with replacing it with portable hand held hell devices? Microsoft, Unity, eh… No one gets it yet? How unfortunate. For most of us that GET_IT, we know the PC is not on the way out. And won’t be for a_long_time.

> Most important desktop applications have not made the switch to GNOME 3. From talking to them, it’s not a priority for most of them. The claimed target users for GNOME are leaving desktop computers behind for types of devices GNOME doesn’t work on.

Did I mention that someone got up one day, decided the desktop was on the way out, and that we needed horrible hand held hell software running in our face? Absurd goals by many, Ubuntu Unity, Microsoft, etc…?

Some will never get it.

#244 suvi on 08.07.12 at 06:29

I love gnome!

#245 Jason Simanek on 08.07.12 at 15:41

I think it’s sad that so many apparently advanced Linux users are giving so little opportunity for Gnome to boldly try something different. Creating a new desktop environment is complex and unfortunately Linux users are apparently very impatient.

Did we learn nothing from KDE 4’s initial introduction? I don’t care for KDE 3 or 4, but I would never be so egocentric as to argue that they are wrong for not meeting my particular needs. They apparently meet some peoples’ needs very well and the project appears to be skillfully executed. But I don’t care to use it.

Though I was initially ambivalent, Gnome 3.4 is pretty great once I got over it not being Gnome 2. I love the big click targets, built-in quick launcher via keyboard (replaced Gnome Do), emphasis on keeping system panels out of the way of the application I’m actively using (no docky or bottom panel) and as soon as Ubuntu migrates to the Gnome 3 keyboard shortcut manager I’ll be able to adjust my keyboard shortcuts just like I used to in Gnome 2.

People always respond negatively to change. But if we don’t try new things the desktop will never be any better than it currently is. We should be celebrating Gnome for trying something new, not deriding it. For once a Linux desktop environment is not merely imitating existing examples and all the vocal Linux community wants to do is complain about how their needs are not immediately being met.

#246 mankou on 08.08.12 at 17:30

Debian switched to Xfce for its default desktop, because Gnome “doesn’t fit into the media anymore”. Yeah, ok…

BTW, I think Debian’s decision shows sanity.

#247 Dan on 08.08.12 at 17:54

And now Debian is going to default to XFCE. Gnome3 is truly dying. I’d like to thank the Shell devs for destroying my favorite DE.

[WORDPRESS HASHCASH] The poster sent us ‘0 which is not a hashcash value.

#248 Leslie Satenstein on 08.09.12 at 03:10

To the sour grapes complainers.

Gnome3 (G3) has many advantages under the covers that were not possible with G2. One of those is the ability to use the Video card’s Graphical processor to do 3D rendering. This was not possible with G2. G3 will work better with HTML5 to provide fast rendering.

A second feature of G3 is a better architecture. It will allow you to tweak G3 to accommodate your UI needs.

To accommodate you negative guys, lets make a comparison…
Unity (the other GUI interface) is supported financially by a billionaire. Get it done and get it done now, is the way, and hang the cost.

G3 is created and supported by a small group of paid Gnome developers (mainly RedHat) and the open source community. RH does not have the deep pockets that allow more contract developers to be engaged. Ergo G3’s development is slower, and well tested.

To demonstrate the richness of G3, you recently have for G3, a release of both Cinnamon and Mate, both UIs give you the previous G2 functionality.

In fact, I am using a G3 with two tweaks. One gives me the “suspend, power-off, hibernate” addition to the shutdown menu, and the other gives me a full menu as a directory tree. This latter output is quite good.

Gnome3 is open source, will always be open source. I cannot say the same for Unity — Ubuntu’s GUI interface.

I predict that eventually UBUNTU will be delivered as two versions:– the scaled down freebee, and the pay for the extras, where the extras have to cover more sophisticated software.

To subsidize Ubuntu, you have the Ubuntu store. There is no store to subsidize Gnome.

In the open source world, there is nothing stopping someone from porting Unity to replace Gnome3 for your favorite distribution (Debian, Centos, Scientific Linux, SUSE, Fedora, RedHat and others).

GUI interfaces are rarely supported by rolling maintenance, so you have to wait for the next release. There is a good reason why rolling releases are not supported for GUI software. I challenge you to respond with the answer why.

The next Unity version 12.4 is very attractive, and if it is rock solid, look to it replacing Gnome3 where Gnome3 is currently installed. But also, look to Gnome3 to be updated as well.
Gnome3 have a plan and timetable for enhancements.
I use Fedora with Gnome3 and I do wish it had more functionality. To deliver a solid reasonable bug-free interface, some things were left out, but these will be added as part of new tested deliverables. If you want a better G3, join the project. They can use your help.

I have a small netbook with 16×9 ratio. With G3, the favorite bar is out of the way until requested. With Unity, it is always there (persistent). When I run Firefox with G3, I have full width of the screen, with Unity, I have to scroll left or right.

You must have patience. KDE started with its new interface about 4 years ago. It has taken this long to have a great alternative choice. You also have xfce and more.

#249 mankou on 08.09.12 at 07:08

@ Leslie

What you see as a richness factor (MATE, Cinnamon, etc.) is what I see as an epic fail. Gnome Shell fails to satisfy a great portion of Gnome’s (prior) user-base, so when there is considerable _demand_ for forks, forks happen. If the people behind those forks believed that this is a temporary issue (like KDE), they wouldn’t undertake this huge task.

So, I’d definitely agree with you about the ‘be patient’ thing, but the _main_ problem is not about dealing with a young, immature desktop. The problem is with the project’s mentality of “let’s get rid of the next feature”. After the taskbar, minimize, shutdown etc. a few days ago Nautilus had a couple of existing features stripped off. They’re also looking to get rid of the ‘categories’ in the apps overview. In 4 years how many features will be left to work with?

What I’m saying is, yeah, it’s not that hard to get this extra extension and get “shutdown” back, or just hold the alt-key. The problem is that the developers/designers think that this is perfectly ok. Would you trust a car manufacturer that gives you this car with no keys, saying “you can use the cool looking wires under the steering wheel to start the engine…” ?

#250 Eh on 08.09.12 at 09:03

So Debian dropped GNOME3 too.

Of course, the moment of truth will be the RHEL7 release, and we’ll see whether RedHat will have the “courage” to ship GNOME3 to enterprise, and, if so, what their reaction is going to be.

#251 Greg on 08.09.12 at 13:19

“Gnome3 is open source, will always be open source. I cannot say the same for Unity — Ubuntu’s GUI interface.”

FUD much?

#252 Séb on 08.09.12 at 15:21

I’ve used Gnome 2 for years. It’s a good and simple desktop environment, maybe too old-fashioned nowadays.

Gnome 3 ? Total disaster ! Although it’s somehow pretty, usability is a nightmare.

I like the KISS (Keep It Simple & Stupid) philosophy, but Gnome Shell is joke that makes no sense at all. A simple start menu is far better than this crap.

What’s the point destroying desktop experience (100% of Gnome users) ?

Now, I’m on KDE 4 and I will switch to OS X soon… Isn’t ironic ?

#253 Ziggy on 08.10.12 at 07:09

The Gnome team made their product redundant by changing too much in Gnome 3. Funny how developers don’t realize that their users like to stick to usability concepts they have for years instead of happily accepting their new “improvements”. More efforst should have been spent to sort out annoying bugs and functions such as the sodding thumbnail housekeeping which rendered Gnome 2 desktops unusable for a couple of minutes after login.
Gnome will survive as long as Red Hat uses it in RHEL, but its importance to users is gone.

[WORDPRESS HASHCASH] The poster sent us ‘0 which is not a hashcash value.

#254 yann on 08.10.12 at 12:51

Debian also leave the boat… but some will still sing
“All in all it’s just another brick ‘out of’ the wall.”

Maybe… but the reject is obvious, especially for a desktop environment that never messed with user experience before Gnome 3.

Ubuntu doesn’t seems to meet great success with the other “PlaySkool” interface already out there on PCs too.

So, is there a good news?

Yes: Microsoft is heading towards another debacle with Windows 8… same mistake here!

#255 boylott on 08.10.12 at 15:00

What the !#@k is behind those Gnome-shell, Unity ……?
What they are looking for?

Since unity/Shell first release, I returned to windows

Sade to say but Linux DE are becoming a piece of chit
Where is the main goal, I need my PC to work and work only

#256 thaodan on 08.10.12 at 16:00

@boylott: May try KDE?^^

#257 saco on 08.10.12 at 18:52

I’m not sure what “the claimed target users for GNOME are leaving desktops for smartphones and tablets” means. I don’t know anyone who has completely given up their laptop/desktop for a tablet, or anyone who has only a tablet and no laptop/desktop. As for smartphones, except for people in the 3rd world, nobody does their main computer work on a smartphone.

Who are the claimed target users of GNOME? Programmers? People who don’t want to pay money to Microsoft (although this is kind of nuts since a laptop costs the same with or without Windows)? 3rd world people?

#258 Jason Simanek on 08.10.12 at 19:59

@saco I use Linux and Gnome because I prefer it to OSX and Windows. It’s a subjective and honest preference.

#259 hdaz on 08.10.12 at 22:55

“RedHat will have the “courage” to ship GNOME3 to enterprise”

I would of thought the majority of RHEL enterprise users don’t use Gnome or a desktop at all.

#260 Claudia Huurkman on 08.11.12 at 08:19

>I’m not sure what “the claimed target users for GNOME are leaving desktops for smartphones and tablets” means.

I guess their goal was to target people that use twitter and facebook, browse the Internet, send pictures with instagram and play Angry Birds.

Gnome should change their tagline in: “We want to make great software to dick around”

#261 Amos Batto on 08.11.12 at 16:32

Inspite of all the negative comments, I actually think that GNOME 3 shell was necessary and kudos to the GNOME development team for having the vision to see it. We all know that mobile devices are the future of computing and we will all need a decent desktop environment for our tablets, netbooks, TVs, cell phones, etc.

We needed an alternative to Android which was compatible with all our existing software (LibreOffice, Gimp, FireFox, etc). Android would never give the ability to use the software that we already know and love. After Maemo and Meego crashed, there really was no viable alternative, so GNOME stepped forward and gave us a lovely interface which will work for the next generation of mobile devices. I am very grateful and applaud all the hard work.

The problem is that an interface for mobile devices is not necessarily the best interface for the PC desktop. Trying to force everyone to use an interface designed for mobile devices was a major mistake. Some people like a mobile interface on the PC, but many do not.

GNOME just needs to recognize that most traditional PC desktop users will never be happy with with the GNOME 3 shell and stop trying to convince them. If GNOME wants to be BOTH a PC desktop environment AND mobile desktop environment, then it needs to have two separate interfaces.

Now that the GNOME 3 shell is usable and starting to stabilize, I think that GNOME should focus its development energy on adding functionality to its “classic” mode. Add all the cool things that were available in GNOME 2 to GNOME 3 classic, including the right click menus. This will woo back all the disenchanted users and hopefully attract new developers.

I know that this plan to support 2 interfaces will involve lots of work, but it is the only way that GNOME can regain its position as the preeminent Linux desktop environment and attract help. Most developers only want to work on something that they use, and most developers have stopped using GNOME 3, because they don’t find the shell to be very productive as a PC desktop environment.

Let me just add that all desktop environments are facing the same problem of how to address mobile devices. Currently we see Windows 8 facing the same problems and we see an avalanche of criticism for its Metro interface. PC desktop users hate Metro in just the same way that GNOME 2 users hate GNOME 3 shell. The company which addressed the interface problem the best is Apple which gave its users two separate interfaces for PC and mobile. Now that Apple is trying to add mobile elements to OSX it is also facing criticism from its users. The lesson is that GNOME is not unique.

The only desktop environments which haven’t faced criticism are the ones which have refused to even try and address mobile devices, as KDE, XFCE and LXDE are doing. In the long run, GNOME is better positioned than any other desktop environment, because it can potentially be used on any type of device.

#262 whatever – GUADEC wrap-up on 08.11.12 at 23:25

[…] the financial crisis, companies taking other paths that do not cross with GNOME’s one, some staring at abyss, or other factors, but, for me, GUADEC started a bit nostalgic “the good old times where […]

[WORDPRESS HASHCASH] The comment’s server IP ( doesn’t match the comment’s URL host IP ( and so is spam.

#263 GuiMaster on 08.12.12 at 05:13

Amos Batto said it well. Gnome has to support “Classic” as well as “Shell.” Both need equal support, and if they give the time and effort they’ll begin to recover the users that they’ve lost recently.

#264 mankou on 08.12.12 at 18:22

@ Amos Batto
I agree with everything you said, except:

“The only desktop environments which haven’t faced criticism are the ones which have refused to even try and address mobile devices, as KDE, XFCE and LXDE are doing.”

Have you heard of something called ‘plasma active’ ? How about ‘vivaldi tablet’? If not, google them. You’d be surprised.

#265 Eric Fitton on 08.12.12 at 20:17

@thaodan KDE 4 is why I switched back to windows. Gnome 3 keeps me there. I miss the desktops of four years ago. I can’t believe that my experience w/ Linux is worse in 2012 than in 2008 when I deleted my Windows partition.

#266 Peter Wallis on 08.13.12 at 01:04

I have used Ubuntu Linux since version 7.04. I am not a techo or linux afficionado and just want a system that works every time I switch the box on and for most part Linux has not disappointed me. The interface was easy to use and understand wnd it was easy to find the things I am looking for.
I liked what happened in Gnome’s development until you guys bastardized a good desktop with this abortion of a release. Its a desktop, not a mobile phone!
So now I use Linux Mint which has at least in part made the desktop halfway user friendly.
Its almost as if you guys are on the payroll of the evil empire. Why would you want to turn a perfectly good and working project into something that makes life difficult? All that does is make new users run back to Windoze, but maybe that is the whole idea.

#267 Peter Wallis on 08.13.12 at 07:15

As an addendum to my previous post:
I strikes me as ironic how when we get something wrong we NEVER look at whatour involvement might be. We blame the economy, the weather, the moronic customers, the inflexibility of users to adopt the brave new world.
The reality is, you guys stuffed up. You had a gerat idea that didn’t translate into user acceptance. The computer history is littered with applications that were fantastic and then made themselves obsolete by pursuing some “cool” improvement that nobody wanted or needed.
You guys have a great product and some very loyal followers. But goodwill will only carry you so far and then you join the dinosaurs.
My old boss in Germany used to say “It takes 200 years to build a business name and 10 bad transactions to loose the lot”
Maybe you should take the complaints of your userbase to heart and turn the ship about, before you hit the rocks.

#268 From the 200 extensions to the new 200 Gnome Apps! | woGue on 08.14.12 at 01:28

[…] every single thing but his code!) will blame Gnome3; when a Gnome Developer will have a “bad moment” to make a huge story out of it, and let them collect every single user complaining about […]

#269 Pavel on 08.17.12 at 08:50

The main problem of GNOME project is that developers don’t care about users anymore, they care about writing “great programs”. This point of view is completely wrong and devastating. Do they care about changing the classic desktop environment paradigm will force users to leave? No, they don’t. Do they care about software functionality that users really WANT to see in GNOME? No. Do they care about developers, artists and designers who can’t anymore work with GNOME because of constant API changing? No, they don’t. They care about great programs that nobody wants to use.

Distros are leaving GNOME, yes. Because distros care about users they have. Because distros collaborate between upstream projects. But GNOME is too great for that type of thoughts, eh? Ubuntu? F_ck Ubuntu and their patches. Our programs is just too great. F_ck collaboration, f_ck users, dedicated to noone, thanks to no one, art is over.

“The claimed target users for GNOME” is just nonsense. There are no target users for GNOME, there are target users for Linux desktop. But, again, GNOME is too great to be associated with Linux. We have too much ego to be just Desktop Environment, we want to be an OS.

#270 george on 08.17.12 at 17:47

Gnome 3 is a great platform. I’ve used a Macbook for the last year, and I still miss my thinkpad running Fedora 16 and Gnome 3.

What Gnome 3 needs to succeed is ask user and developer groups what they want.

The introduction of Gnome 3 was very badly managed from a marketing point of view. They basically refused all criticism.
I still miss the possibility to see an icon constantly on the screen to tell me I have Skype notifications. For me it was a huge regression, and Gnome did not listen to the many users missing those kind of things.

It’s so sad to see this, because Gnome 3 is so much better than Unity, Windows, and to some extent Mac Os X!

#271 Pavel on 08.18.12 at 16:01

>Gnome 3 is so much better than Unity

Unity is built on top of GNOME 3 :)

#272 Rory on 08.19.12 at 02:47

Just a little user fish, but I left GNOME when Ubuntu went to Unity/Gnome3.
For one simple reason: lack of control/options/customisability compared to Gnome2 – that was all removed.

#273 Elhana on 08.19.12 at 12:15

Gnome developers are simply retards which got fired from the real software houses.

It took them 10 years to add a fucking line wrap button on their notepad toolbar (and it is not done yet)! No wonder they are in such a dung pit now.

#274 SuspiciousLizard on 08.19.12 at 19:50

The biggest problem with Gnome 3 is not that it exists; I’m perfectly fine with it existing, and it does have a niche following who greatly appreciate it. Gnome 2 was essentially a finished product, so I understand why developers got bored with minor tuning and wanted to create something new and exciting. There’s nothing wrong with that, in and of itself.

The real problem with Gnome 3 is that its user interface is an fundamentally different project from Gnome 2, and yet the only difference in the overall project’s name was an increase in version number. gnome-shell targets an entirely different type of user from gnome-panel and demands an entirely different workflow, and it’s also nearly impossible for actual end users to customize it to suit their needs…unlike the eminently drag-and-droppable gnome-panel. Gnome 3 is a new avant-garde desktop environment, which inherently precludes it from being accepted as a natural successor to Gnome 2.

By giving an experimental desktop like Gnome 3 the same name as Gnome 2 with an updated version number, the Gnome developers forcibly pushed Gnome 2 out of the respositories with naming conflicts. They didn’t just add a new choice of desktop environment; they did everything in their power to eradicate a choice that a lot of people relied upon for their workspace!

This is a huge deal, because it sent Gnome 2 users scrambling to find a desktop environment that actually suits their needs. Sure, people can fork Gnome 2, and some have, with Mate. That project is now picking up nicely, and I think it has a lot more life in store for it than cynics thought at the outset. Cinnamon (previously MGSE and based on Gnome 3 but now a separate environment) is also being developed, so it will be interesting to see where those take us. Both are vastly superior to Gnome 3’s fallback mode, which has adopted little of Gnome 2’s functionality and none of its customizability.

It’s great that these projects now exist, but they’re STILL not in most package repositories after all this time, and they face a lot of resistance from the, “Derp, Gnome 2 is dead, and change is progress, so move on with your lives” crowd. This is a selfish, arrogant, and controlling attitude, and it undermines the inclusive pluralism that has always been a trademark of the FLOSS desktop scene.

A lot of people see Unity and Gnome 3’s application-orientation and reliance on search a great new paradigm, and I’m happy for them, but it simply doesn’t suit everyone’s workflow. I for instance am a great deal more task and data-oriented, and I work better with the desktop metaphor and the filesystem’s implicit “method of loci” than I do with a focus on applications and searching. This extends to the positioning of minimized windows and other areas as well. Unity and Gnome 3 fight me all the way by insisting their layouts are better for me. They aren’t.

To give an analogy, consider the usefulness of important/export functionality between email clients and web browsers: They work great when you’re just switching applications, but the abstraction breaks when you’re trying to back up a bare minimum of personal data for a fresh installation (without the rest of the cruft in ~/)…to do that, you actually have to know where your personal data is stored in the filesystem and separate it from the stuff you don’t want to keep. Similarly, consider the usefulness of sharing your data across platforms to keep your desktop/laptop/tablet/phone/etc. synchronized: It works, but the filesystem specifics are so abstracted away that you have no real understanding of or control over what exactly is being shared and how.

Gnome 3 and Unity give me the same feeling of being kept “sheltered” and separated from my actual data storage: The additional layer of abstraction provided by the applications/search focus mildly obfuscates the information that keeps me organized and sane. I can still access the filesystem from both of them of course, but the elimination of the desktop metaphor puts the filesystem “off to the side” instead of keeping it an integral part of computer use, which is what I want: For instance, I rely heavily on the desktop metaphor and use the desktop as a temporary working area where I can keep my files messy before organizing them and moving them to the rest of my tidy filesystem.

There are still a lot of other desktop environments like KDE, XFCE, and even LXDE that better preserve the desktop metaphor that many of us depend upon, but if Gnome 2 users really preferred them, we would have been using them already! Thanks to Gnome 3, we’re forced to choose “second best,” risk technical difficulties with third-party repositories to use something like Mate, or wait for it to finally get into the distro repositories over the objections of the naysayers…and aside from Linux Mint, it’s taking a while.

The bottom line here is that people should be able to keep using what they like if change doesn’t suit them, or if they have differing opinions about the meaning of “progress.” Would anyone seriously suggest that any and all hand saws are now obsolete in the face of circular saws, and use their influence to argue against stores carrying hand saws? Would anyone seriously suggest eliminating chess now that Electronic Talking Battleship exists? A desktop does not have to be “cutting edge” or constantly changing to remain relevant to users. For that matter, if it’s an essentially finished product like Gnome 2, it doesn’t even need to be actively developed aside from bug fixes, at least until some hypothetical point in the future where user applications are heavily GTK3 based (even then, there’s nothing keeping GTK2 and GTK3 libraries from coexisting). It will still work exactly as its users expect. Not everyone wants or needs change for its own sake, and we’d like to decide for ourselves when a new desktop looks more appealing and useful instead of being strong-armed into it by packaging politics.

#275 SuspiciousLizard on 08.19.12 at 20:17

Actually, I should add that the Mate developers are planning on progressively updating their codebase to use GTK3 anyway. (After all, a lot of Gnome 3’s technical infrastructure is vastly cleaner and superior to Gnome 2’s.) Once that move is complete for gnome-panel and similar components, I imagine they might even drop most of the forked Gnome 2 code and start sharing most of their codebase with Gnome 3. Depending on how Cinnamon turns out, it might end up becoming the better alternative too. Regardless, I think the point of Mate right now is that it’s currently the only way old Gnome 2 users can continue with our preferred interface uninterrupted…or it would be, if it were in enough repositories.

#276 Siv on 08.20.12 at 09:33

Standing in the wings looking in as a Windows user who is now faced with the new Windows 8 Metro UI, I would say when Windows 8 ships in October, the Linux community will have one of their biggest opportunities to grab serious productivity users from Microsoft.

Any sane person knows that touch on mobile devices and mouse on desktop PCs cannot realistically be combined, they are to my mind mutually exclusive. I know Microsoft are doing it to unify their interface across all platforms and save having to maintain multiple code bases into the bargain.

I think Windows 8 will do well with the casual home user who really use their PC to play games and surf the web.

Business users however will steer well clear of it sticking with Windows 7 until it has to be prised from their cold dead hands.

SO if the Linux community can provide a desktop operating system with a familiar desktop user interface that looks similar to Windows 7/XP but offers new levels of performance on older hardware (pretty much what Gnome 2.x did) but providing improvements in terms of multitasking, memory management and media playback you might be surprised how many new productivity users come on board.

I would have thought you could take Gnome 3, listen to the end users that want a mobile/tablet type touch centric O/S and turn it into “Gnome Touch Edition 1.0″ and take the Gnome 2.x codebase improve the multitasking, memory management and media playback goodness and call that “Gnome Desktop Edition 3.0″.

If you built a really nice equivalent of XP Mode (I know there are things like Wine and Virtual Box already, but you need to make it as slick as XP Mode) so that Businesses could run legacy XP era applications seamlessly you would be onto a winner.

I think LibreOffice is looking like a very usable alternative to MS Office for most business users who don’t need half of what is in MS Office. So combined with a nice friendly Gnome 2.x interface there would be an alternative.

I think Windows 9 may look a lot different to Windows 8 when the backlash from business becomes clear, so you probably need to move fast to capture the user base.


#277 david on 08.20.12 at 10:54

Well, I have to say that it is about time that you noticed. I noticed all of these problems with GNOME years back.

When you folks came out with GNOME-3, I looked at it then said to myself: I don’t remember any real deep documentation explaining the justification behind why the new design was better. No metrics. No UX studies. Nothing.

Just some “oh, we think this is the way to go” hand-wavey stuff published in articles with no specifics on anything. Just cheerleaders… and then people posting their disgust.

I gave both GNOME-3 and Unity a try.

And both were failures.

After trying them myself, I am convinced even to this day that no proper study was ever done. Somebody or so group within your org who was very charismatic convinced the whole lot of your to go on a fools errand.

And this is where it landed you all.

Ubuntu users defected to Mint to keep using GNOME-2.

Then when Mint and others tried to go GNOME-3, they either got backlash or defectors, too.

My advice: use real statistics and studies on UX when making UX decisions.

This blunder was so grand, I couldn’t decide if you all did it to yourselves or if you’d been infiltrated by people hellbent on derailing your project.

I’ve seen both types in my lifetime.

And for GNOME, I can’t tell which one your afflicted with.

#278 Le projet Gnome : est-ce le début de la fin ? « SAM7BLOG on 08.21.12 at 20:28

[…] effet, il y a deux semaines j’avais lu sur le blog d’un des développeurs de gnome, un billet qui a beaucoup circulé et fait parler de lui, pour résumer Otte fait savoir que le projet Gnome manque cruellement de développeurs, […]

#279 xyroth on 08.23.12 at 12:10

I have been using Linux since we had to install it using floppy disks, and a lot of the comments here sound very familiar.

You have a long developed, mostly mature project who spot a real problem (the base libraries are incompatible with multi-touch) and decide to do something about it.

Then they make classic mistake no1, and go away and specify and implement it mostly in secret, in a way that was mostly incompatible with the old libraries.

Then they make classic mistake no2, and have the replacement libraries be both incompatible, and required to have the same name.

This is guaranteed to break lots of stuff, and could trivially have been avoided by using a prefix (ie MT-libname) and gradually migrating the old api to call the new libraries (as wine does with their direct x support).

They then make classic mistake no3, and have each significant release break a new subsystem, and then moan that programmers are not rushing to move to the new ever-changing api, without acknowledging that this means that they need to port the old api (not the old code) to use the new libraries.

Then when users complain about the breakages, they say “but you haven’t given it a chance”, and generally don’t listen when users say that they did, but it is not compatible with their work flow.

There is no such thing as the one true way, people do different things, and need different ways of doing them.

If the kernel can radically rewrite it’s internals without needing to break it’s high level api’s, why can’t a desktop.