Post GUADEC notes (aka GNOME in revolution mode)

This GUADEC brought a lot of momentum to GNOME: fresh new ideas and nice discussions and plans about our future platform and desktop. For the first time in the last 2 years, I have the strong feeling that GNOME is about to switch to a “revolutions mode”. This means that several small and big revolutionary things will start to bubble inside the project from now on. I’m really excited about it! (Obviously, I’m not saying that nothing interesting happened before this GUADEC. It’s just that now I see more cohesion around some goals inside community.)

The Online Desktop

The Online Desktop is something that, if done right, can bring a big and positive impact on the GNOME user experience. What I like about this idea:

  • It takes advantage of the whole world of cool web services out there
  • It makes the GNOME Desktop much more fun (I remember Vincent commented about it 2 years ago… He is totally right!)
  • Consequently, it brings GNOME much closer to the user who uses the computer to strengthen his/her social and emotional ties

I don’t think we should work on an “online mode” (with a special UI) for GNOME Desktop though. IMO, the way to go is to turn GNOME into a “web-aware” desktop environment. some examples:

  • If you’re viewing some photos or watching a video, it should be easy and seamless to upload them to your favorite photo/video sharing web service
  • It should be natural to keep track of what your friends are doing on web (what Mugshot does)
  • When you import some photos from your camera, it should be possible to upload them straight away to the web

Bringing web-awareness to GNOME involves creating a basic platform (mostly a set of D-Bus services and GTK+ widgets) that could be shared by any desktop component who wants to be web-aware. This way those bits can be used on GNOME Mobile as well.

We should not limit ourselves on the data sharing on web services because, in essence, what we want (as users) is to share media with other people. We should consider having an integrated communication framework as part of the Online Desktop effort. Some examples:

  • If you’re viewing some photos or watching a video, it should be easy and seamless to share them with one or two friends (via messenger, e-mail, local network share, etc). If your friends are online you can simply start chatting with them after you shared something with them. Something like Web Swarm (in Mugshot) does for links but for any shared media
  • There should be some central place to start communicating with someone. it doesn’t matter if it’s via e-mail, messenger or whatever else
  • It should be easy to make collaborative work on the same content (documents and media). OLPC is doing a great job on this subject

We already have a lot of code available for covering some of those use cases. Some of them: Conduit for easy syncing with web services, Empathy and Mission Control for a Telepathy-based integrated communication in the desktop, Mugshot (server and client) stack for web services activities tracking, Soylent for “people browsing”, and so on. There will be this big effort to add the “online” bits to applications and other desktop components.

“Bling!” on GNOME

There were some nice talks about adding more “Bling!” to GNOME. It was nice to see things like Clutter, Lowfat, GTK+ 3.0 (oops, 4.0!) discussions, compositing, etc. There two things I’d like comment on:

  • We have this challenge of making GNOME sexier. Ok, cool! But let’s keep this in mind though: we should not make GNOME unusable on a computer without 3D acceleration. At the same time that we want to impress the user with cool effects (which should improve the usability btw), we need keep in mind that there are lots of people using low-power/old computers with GNOME on social projects out there (Brazil has many of those).
  • In order to bring “Bling!” to GNOME in a consistent way, those bits should be in the platform level with components that could be easily reused among applications. Also, we would need to have clear guidelines in HIG. Otherwise we’ll end up in a “Bling!” hell!

GNOME Platform

It was nice to see that important changes are about to come in our platform. Most of them I’ve already commented before: GVFS, GTK+ 3.0, New applet API, GSettings, Mallard and others. Next year will be really nice for GNOME Platform!


That’s GNOME: a bunch of extremely talented and generous people working on the common goal of creating free, easy-to-use, and innovative technologies to everyone!

Published by


Lucas Rocha is just a brazilian guy who loves hacking and music. He lives in the frozen lands of Finland with his lovely wife Carol. He works for Nokia in the development of Hildon and Maemo. In his free time, he's a happy GNOME contributor. He has a mustache, a beard and big smile in his face.

10 thoughts on “Post GUADEC notes (aka GNOME in revolution mode)”

  1. Hey Lucas! :)

    Could you update the Soylent link to this one?:

    “Project Soylent” was the original (somewhat different) idea from a few years ago, and didn’t really go anywhere. So anything related to it is really out-of-date.


  2. The “bling!” factor never impressed me much. Not too long ago, Gnome used Enlightenment as the default window manager. We moved to Sawfish/Sawmill and then to Metacity, paring down the window manager into something “serious” and simple.

    I was never too impressed with Enlightenment. It’s fun, sure, lots of glowing and shiny buttons (esp. e17) but everything is just so *heavy* and *fuzzy*. I prefer simple and snappy. Lately, I’ve been very impressed by XFCE, it’s slowly growing more robust while remaining quick and straightforward.

    For example, I have never understood why music players need to have an interface entirely unlike anything else on the system (eg. Winamp/xmms). At least now with the current generation ~ Muine, Rhythmbox, Banshee, etc. ~ things look normal and more *usable*.

    I definitely agree that if we’re going to start incorporating Clutter, Lowfat, Compiz/Beryl, Pyro, etc. we are going to need a HIG which addresses these interfaces. They need to be standardized and accessible if they are going to be an official part of Gnome. Just doing things because Apple did them isn’t good enough, they need to have a purpose (demoware is not a good reason). Are ideas like iTunes’ coverflow useful? Personally, I prefer just scrolling through a list of albums (Muine) rather than through a 3D flipbook (iTunes), the pictures are good but is there any purpose to making it 3D?

    Another concern is *reduced* functionality with increased “bling”. I put my desktops in a 3×3 square so that every desktop has 4 adjacent desktops. Every desktop is home to a separate application (terminal, GEdit/vim, Firefox, Nautilus, Muine, etc) so I can easily switch between them. With Compiz/Beryl, switching between desktops is not only slower but can only be done linearly (as far as I’m aware) so that every desktop only has 2 adjacent desktops. It looks cool but reduces my ability to actually work. 3D file managers (“It’s a Unix system, I know this!”) never caught on because file management (along with most other things) is a bear in 3D.

    What is the purpose of these projects? We are putting them out there, saying that if the architecture exists, the projects will come. No offense to Rasterman et al. but there’s a reason Enlightenment isn’t taken seriously and why it’s not popular. Shiny effects aren’t what bring people to the desktop, Vista proved that. “The wow starts now!” wasn’t effective once people realized that things simply didn’t work and were awkward, ugly and confusing when they did.

    Project Ridley (“libgnome must die!”) intrigues me more than the new “bling” libraries. I also see Mono as bringing a more pluggable, rapid development, integrated architecture to Gnome if done right (we have to watch the resource consumption and speed on this a *lot*, along with Python now that both are officially used by Gnome). I’m not saying that we need to be XFCE but, as the previous commenter said, “Let’s not lose our heads here.”

  3. Please do lose your heads. 6 months later anyways you notice that almost nothing useful for users was implemented. I’m all for the idea but experience has taught to be cynical towards these initiatives.

    My own pet peeve is the “desktop searches”.. It’s the DESKTOP part that sucks, many of us have a significant part of everything data already in the networks.. Like Google Docs. And now finding it has become hard, and it requires distinct tools to do several searches to be able to locate anything. Progress, yippiee.

    Or how about before these very far out (although cool) scenarios fixing some very basic problem: complete lack of usable collaborating platform. (IM, multi-point stuff, voip at the same, no need to WORRY about it being connected like GAIM which poos on its own feet every time, no worrying about actually running the application or knowing there IS such application.. The integration level and comfort is presently plain rubbish.) Where on earth is my sane and nice telepathy+frontends+connectors installed by default on every desktop and them integrated for that pervasive goodness huh? Pathetic.

  4. Great post. I’m excited to see people within Gnome thinking in this direction. I’m also glad to see it tempered with some reality checks, particularly the bits about making sure it doesn’t get so heavy it crushes older hardware. I feel that is extremely important. The whole thing reminds of a quote or proverb that I’ve seen around, “A Plan without Action is a DayDream Action without a Plan is a Nightmare.” It sounds like you are trying to get that balance right, and I hope you succeed.

    Gnome’s KISS philosophy has served it well, if you ask me. Despite the controversy it caused I think the simplification that was done was the right thing to do as it now puts Gnome in a good position to add these sorts of revolutionary features in a sane way. I’m not sure that was forethought or just good luck, but here we are.

    However, I do also have to agree with some of Erik’s points, in principle at least. While this visionary stuff is awesome and exciting, there are a still a number of more fundamental things that need attention within Gnome. The printing system comes to mind. My applications work totally inconsistently in how they interact with printers. Evince behaves one way, gedit another. And this is just within the gnome-shipped apps. Once you start adding Firefox and OO.o to the mix the inconsistency gets overwhelming. I consider myself a pretty die-hard Linux/Gnome fan, and even I find it frustrating. I can’t imagine what someone who was “just a user” would think.

    Additionally, there are a number of features that users have come to expect that Gnome doesn’t offer out of the box. The first one to come to mind here is the ability to automatically rotate amongst a group of desktop images. Sure, one could install “drapes” (or whatever it’s called…) to achieve that, but that shouldn’t be necessary. I think that if some of these comfort features don’t start showing up in the core of Gnome again, we will lose a large number of potential users. Adding them along with these other web-integration features seems to make a lot of sense.

  5. I’m sad, Lucas, because you’re supposed to know enough French to blog in French now. And you didn’t do it. I’m sad.

    But I still love you :-)

  6. @Quentin: Some of us really want to push for paperless office… I was once in a job position where I wanted to propose that a 30 000 people organization would stop nearly entirely stop printing stuff on paper. There were some law based obligations to print transactions on paper for archiving, and I would have added a good deal for print shops for getting prints for brochures etc over night. Ahh. General change resistance, it’s always fun. But it would have fit the organization and been good in overall. It does NOT hurt to fix the Gnome printing though as some people might be still wanting to print :)

    @John: I would myself too add a lot there. Like, default email client to point to Gmail. Open office on menus to change to Google docs. Also, universally working default media player selections are missing. Or, voip/conferencing application (are there sip urls to be handled?).

Leave a Reply

Your email address will not be published. Required fields are marked *

Attribution-NonCommercial 3.0
This work is licensed under a Attribution-NonCommercial 3.0.