GUADEC in retrospection
15. July 2008
GUADEC
GUADEC itself was great, nice people, awesome location and fantastic weather. Thanks to everyone making this possible. I have a little thing to critisise though: IMHO the schedule had some issues! Some really interesting talks were scheduled in the Opening Hours or in the After Hours which not really many people attended. On the core days, some important talks happened at the same time and I don’t think it’s a good idea to schedule anything against the Lightning talks. Some other people I talked to had a similiar opinion but that may of course still just be a small group. Anyway, I would love if some experienced GNOME people could check the schedule for such conflicts next year.
Party
- Opening Cocktails: missed due to great view from appartment…
- Boat Party: Maybe the best GNOME party ever, I doubt that everybody can remember the whole evening. Talked to lots of people though sometimes conversation was rather off-topic. Thanks to the people at Collabora for organizing the event!
- Closing Party: Well, either I was to tired from the day before or it simply wasn’t that good.
GNOME 3.0 (aka 2.30)
The last days I thought a lot about what our goal for 3.0 should be and how we could probably achieve them. First, I think the decision to break API/ABI is good in general as it will remove maintaining tasks for old and deprecated modules from some very important people. Doing this together with Gtk+ 3.0 also sound very reasonable.
Schedule: To achieve more than just the usual small steps between our releases and instead have the time for some bigger things I would propose to leave out GNOME 2.28. That does not mean there will be no GNOME 2.28 release but that there will be no new features but of course string additions, documentation additions and bug-fixes. That would leave developers a reasonable timeframe to get some bigger changes in and port everything to 3.0 in time.
Gtk+ 3.0: Most of the things that Imendio proposes sound reasonable but we should really try to get CSS-like theming because it will attract much more designers to create themes.
Goals: Define clearly now what our goals for GNOME 3.0 are and how we want to achieve them. Maybe the best thing is to have a brainstorm place (on the wiki) and somebody who organizes those ideas in a way that the community/release team can make decisions based on clear concepts. Some of the things that are important to me:
- Better file handling: Frederico had some very nice ideas on that and I think he is right with most of the things he proposed. Of course it would be cool to have a versioned file-system but we haven’t yet and of course some UI ideas might not be perfect. Who cares? Let’s implement what we can do now immediately and enhance it when we have more technologies available. This is something that does not really have to wait for 3.0
- Rethink the panel: Our current panel is not really a modern user interface and as Frederico also said, you have to click too often. And it wastes too much screen-space. I don’t have the perfect solution yet but having those fading sidetabs might be an idea but I am sure there are some other interesting things to think about. I would really encourage people to create mock-ups for a future desktop design.
- More power to the WindowManager: I don’t really like the idea of having an entry in the titlebar but it’s really time to use the available space for something more useful than the window title. Another thing I would love to have is a “Minimize to Icon” button.
- Better default theme: Clearlooks is nice but if you compare it to other desktop environments it looks really conservative. I don’t like the idea of being conservative…
15. July 2008 at 17:30
Which fading sidetabs are you refering to?
15. July 2008 at 18:49
@Vax: Frederico did a keynote a GUADEC presented some UI and file organizing ideas. Seems it’s not online somewhere but he presented some tabs that fade in when you move the mouse to the left side of the screen.
15. July 2008 at 19:47
Fading in just because you’re at the side will lead into it also fading in sometimes when really not desired. Also, this creation by Rasterman is interesting http://www.rasterman.com/files/eem.avi because I really don’t believe that having constantly a panel at the screen is good…
15. July 2008 at 19:59
We proposed not doing a big step. A lapse of a year means that a lot of QA has to be done, this as the development time more than doubles. I do not see any benefit for doing this. We like incremental changes.
If you’re working on something big within a module, branch it.
15. July 2008 at 20:14
I like Clearlooks ;-), but maybe progress bar animations should be enabled by default (I manually enable them from gtkrc file and seem to work ok)
Thanks
17. July 2008 at 1:05
Clearlooks’ aesthetic choices informed my decision to use GNOME over KDE… sad but true. I like it. I don’t agree with the “assume the user’s an idiot” philosophy that I’ve bumped into occasionally.
My big idea for GNOME 3 is making a GNOME desktop that does what you want to! It really shouldn’t be too hard to fall back to sensible defaults or make GUI applications for advanced users’ configuration needs.
23. July 2008 at 0:06
Hey,
I did the schedule, so let me explain:
> Some really interesting talks were scheduled in the Opening
> Hours or in the After Hours which not really many people attended.
If you look at the schedule, there is no slot left in Core days. The alternative was rejecting the talks that didn’t fit in Core. How would have that been better?
> On the core days, some important talks happened at the same time
Guess what? No matter how you lay the talks out, there exists at least one people who finds it suboptimal. In fact, for myself, there was more than one talk I wanted to attend for any given timeslice during the Core days. It’s just impossible to avoid that in any way other than totally giving up on parallel talks. Unfortunately that means rejecting most of the proposals.
I think the schedule was in fact quite decent. You may notice that on the first two Core days, there’s three tracks going on: “Embedded Devices”, “Multimedia”, and “GNOME in the Cloud”. And for each of those tracks, no two talks are scheduled in parallel. The last Core day has “Debugging and Tracing” and a “Infrastructure” tracks going on.
> I don’t think it’s a good idea to schedule anything against the Lightning talks.
Sure, I don’t like it either. But that just meant 4 fewer slots for the Core days.
> Anyway, I would love if some experienced GNOME people could
> check the schedule for such conflicts next year.
Many such people did, and many talks where moved around before we arrived at the current one. You should trying scheduling one day and see how impossible it is to make everyone happy.
Anyway, for next year, I believe we are ready to go to four Core days, and that gives us a bit more flexibility…
Cheers,
behdad