Shell Yes!

Back home after a typically exceptional week at GUADEC.  It is always really inspiring to be surrounded by so many people who are passionate about GNOME.  This year was particularly special for me because we are really starting to see the next wave of heroes becoming more involved with the project – and making an impact.

The overall tone of the conference was amazingly positive and productive, welcoming and fun.  Which seems to be a bit of a contrast to the debates and discussions that have occurred online in reaction to some of the talks.  It is at once sad and funny.  Well, tragic really.  Deciding who is right isn’t interesting if the question isn’t the right one to ask in the first place.

We are dividing and conquering ourselves – lost before we begin.  Are we so hungry and desperate that we must devour each other in order to survive?  I hope not.  These tribal distribution boundaries are guarded jealously and they fight over the smallest (one) percentage of the market and mindshare.

I think it is time we reunite.

If we want to change the game, think big, and demonstrate that we can truly be relevant – we need to work together. If we want to change our approach from mere assembly to something that we design and construct with consideration in a unified and coherent way – then we need to start at the source. We need to start with GNOME.

What lies underneath is mostly just implementation detail. What matters is what we expose to the user and the developer. I propose that we take notes from Android, WebOS, Meego, and others and consider Linux an implementation detail and start to define the OS as we see fit.

The question is not can we, but will we?

Shell YES!

Please see the slides from my talk for a bit more on this topic.  Read the notes for each slide for some idea of what I was saying since I wasn’t reading off the slides.  Hopefully the video will be available soon because there were some fantastic questions asked in the discussion period that followed the slides.

This entry was posted in Uncategorized and tagged , , , . Bookmark the permalink.

14 Responses to Shell Yes!

  1. questioner says:

    why shell can’t be based on already existing code (e.g. unity) since it has proven to be really nice and working? does gnome really need to reinvent the already-existing whell only because the project didn’t start “from the inside”?
    if i was some-gnome-developer i’d prefer sticking with already existing and proved-working code instead of starting on my own, that’s why opensource exists, right?

    note: i’m not affiliated with ubuntu, canonical, kde, redhat, suse, whatever. just wondering watching the situation from blogs, etc.

  2. Is there a chance to get the slides as PDF instead of Flash?

  3. mccann says:

    There should be a link to download as .odp on the slideshare page.

  4. M. says:

    @questioner You should ask that to Canonical. GNOME isn’t “reinvent the already-existing”, in fact they’ve been talking about the Shell for a long time. Unity was announced in May, I suppose.

  5. questioner says:

    since when talking/gimping/inkscaping does count (if this is the case, i’m not sure..)?

    i thought code was the thing that really matters.. and no, i’m not ironic.

    note: i’ve seen a similar discussion in gnome-shell-ml. i didn’t subscribe/intend to do it. i just find this whole situation..funny. imo hackers should discuss code, not politics. how come there’s no gnome hacker that represents this position? i find this somehow strange. in the blogo-social sphere i see only chitchats/rants against canonical..

  6. drago01 says:


    Take a look at … the shell code has been available for a long time before unity was even announced.

    It was/isn’t just talk but its been developed in the open for a while.

    So your question is kind of backwards ;)

  7. questioner: The blog post you’re looking for is (scroll down to Relationship to Gnome Shell).
    Is it a problem that a lot of design in free software are developed in secret? Yes!
    We need to fix that together.

    BTW, applauds for the GNOME Shell for being 100% transparent in their design process, we need more of that!

  8. Jon: thanks for a great talk at Guadec, I became really inspired!

  9. questioner says:

    @drago01: gnome-shell could have been out in the open for a while, but since it’s really far from mature (and it has been already delayed to next year!) while the other “vision” is already on real life devices (read “vision” as: unity, indicators, etc.)… what’s really stopping everyone?

    @andreas: i think that part of that paragraph is not true anymore. gnome-shell vision now really resembles unity with a few visual tweaks (and i think i also read somewhere now gnome-shell wants to address netbooks and such devices..).

    @all: my posts are meant to be read in sense of “optimizing manpower”. Do gnome really need to put efforts in coding something that’s slightly different from something else that’s already available? Unless there are critical issues in the code, and i guess this is not the case, i think this is really a waste of time.
    Is it a problem that a lot of design in free software are developed in secret? Maybe (i think this deserves another discussion), but nothing stops gnome developers from partecipating in/forking Unity as some other people are already doing now. The code is available (and it’s not hidden somewhere in the dark, it’s wild in the open), download, (edit?), compile, run. gnome-shell can be now, if “gnome” wants.

  10. bkor says:

    questioner: gnome-shell is developed in the open, the other one is not (atm maybe, but no guarantee for how long). Your asking the wrong developers.

  11. Pingback: GNOME Shell GUADEC wrap-up « fishsoup

  12. Simon says:

    @questioner – ultimately, the answer to your question is that Gnome Shell predates Unity by several years, and therefore can hardly be accused of avoiding stuff done outside of Gnome.

  13. nicu says:

    I think you have it backwards: GNOME is an implementation detail. We don’t use desktops only to stare at how awesome the Desktop Environment is, we use them to get the work done and for that we need fast and direct access to the applications, with a desktop to stand in our way as little as possible.

  14. Giovanni Campagna says:


    Except that GNOME is not just Shell.
    GNOME is Nautilus (and Midnight Commander, and Terminal with good old bash), Eog (and F-Spot), Rythmbox (and Totem, or Banshee), Brasero, gEdit, Calculator, File Roller, Office (Abiword, gNumeric, ease…), Evolution, Epiphany, Empathy, XChat, Anjuta, as well as the miriad of small configuration utilities.

    Yes, GNOME is pervasive in the everyday activities. It is true that an user should not know “it is Nautilus from GNOME” more than “it is the window with my files”, but we should all work towards consistent applications and behaviour across every distro, platform or hardware system.

    The user doesn’t care if it is Linux, FreeBSD, NetBSD or GNU Hurd. It doesn’t care if the app uses HAL or udev, PolicyKit or gksu, ConsoleKit or POSIX job control, but it will care if notifications have actions or not, if configuration is under System -> Preferences or ->System preferences, if window buttons are left or right.
    This is GNOME OS: a consistent experience everywhere.

    (of course, this is how I see it, I may have missed the point completely :D )

Comments are closed.