The Future of Compiz In Question

[StefanHamminga]

Really? I thought the biggest problem was that the eyecandy was just that, eyecandy and useless for day to day work (with the exception of certain accessibility plug-ins).

If Compiz wants to attract more developers they should consider providing eyecandy that serves a function beyond that of "bling."

I really don't agree here, in my opinion the 'bling' does a lot for the OS, it is the first thing people see and is also the thing 95% of the regular home and office users look at for a great part of the day. Personally I like the things I spend lots of time looking at to be pretty!

A small example: I've my TV hooked up to a HTPC (actually a bunch of parts with really big and silent coolers thrown in the TV furniture) and use it to play movies, play music, browse a bit, that kind of stuff. Now I set this up with the Ubuntu Netbook Remix GUI and it looks absolutely gorgeous and friends (actual non tech ones) who play with it generally really like how it looks and works. The reactions are mostly: "hey, that Linux stuff looks nice, I'd like to have something like that". Much in contrast to the reactions to a basic, dull Gnome desktop. More functionality but no pretty looks. Nobody would have it.

[etnlWings]

Really? I thought the biggest problem was that the eyecandy was just that, eyecandy and useless for day to day work (with the exception of certain accessibility plug-ins).

If Compiz wants to attract more developers they should consider providing eyecandy that serves a function beyond that of "bling."

There is a lot of really useless stuff in compiz (does anyone use, "Paint fire on screen"?) but even some of the eye-candy is useful in it's own way: plugins like expo and cube make using workspaces more accessible by visualising them well; there's various improvements upon alt-tab functionality and then there's the obvious stuff like full-screen zoom and the scale plugin.

Compiz also offers some tweaks which do enhance usability in their own right: resize overlays offer a way around the usually sluggish task of resizing a window and waiting for it's widgets to constantly be redrawn; maxumize speaks for itself. Edit: I'd also argue that window animations are good in the sense that windows don't just pop on and off screen: you can see where they're being minimized to, etc.

Compiz has a lot of fluff/crud but beyond the self-indulgence, compiz makes the *nix desktop more complete.

[yesterday]

This article doesn't say where those developers are going to. If People are leaving Compiz to work on KWIN or a new compositing manager for GNOME, it's fair to say that project isn't in bad shape, but merely evolving. It's not a failure if the tech used in Compiz finds it's way into more integrated solutions (like KWIN), it's a success. If, however, those developers are leaving Compiz because they are sick of the tech, then that is a problem

[Jimmy]

I found this to be interesting and insightful. Why not Compiz?

http://techbase.kde.org/Projects/KWin/4.0-release-notes#FAQ

It's a good read all the way down to the last line,

KWin aims to provide compositing support, focusing on providing useful compositing features and basic visual effects, while keeping its other strengths.

which pings on usefulness bing a key feature. As pointed out by other posts here many experienced Linux desktop users go for useful before bling. Not that bling doesn't have it's place as a marketing tool .

[bridgman]

Nearly every software project claims that their priorities are "useful features not bling", and "stability not haste", unlike *competing* projects

[RealNC]

I fail to understand how people can live without wobbling windows.

[Jimmy]

I fail to understand how people can live without wobbling windows.

I live with it mostly because it wobbles the performance right out of the applications I use the most.

[liam]

I find it amazing so few use the useful Compiz features

I use Compiz(when I can, the latest Nvidia drivers have destroyed compositing-reinstalling older driver helped naught since they won't instal correctly). I RELY on Compiz to get things done in an efficient way.
It allows me to view all open windows on a desktop with a gesture;all desktops with a gesture. Fast, high quality zooming has been really useful when recording tutuorials. Its tiling feature(found in a few other WM, but none that I know of that don't have problems I don't wish to deal with-GNOME user, BTW)is fantastic. The ability to group windows is(while possible by other means)is easier with Compiz, I find.
There is a good bit of fluff, but so what? Some users like it, and as other have said, it draws attention(and I know my gf loves the wobbly windows-and she is most definitely a power user, so such stuff clearly isn't just for noobs).
I believe I read that compiz++ plans on separating the composition from the wm? I believe this would enable one to run the same WM even if the GL requirements weren't met. Besides that, requiring people to code in C for desktop stuff is insanity. More than likely, IMHO, it will bring more problems than it helps, and performance will be hurt since I just don't think the app coders are being careful enough.
The future is, IMO, Clutter, but NOTHING I have either seen, or heard about, comes close to the configurability of compiz.
Regardless, I think IF compiz really does go away, this would be an extremely bad thing for linux. It gave linux something that, at the time, only Mac had, and, most importantly, showed that a linux desktop(or at least a nice looking one) wasn't an oxymoron.
BTW, there is at least one thing I've noticed that can be worked on in Compiz:vectorize things! KDE was ahead of Gnome when it decided to require svg icons. It seems as though icons could be part of what compiz handles.
Best,
Liam

[yesterday]

I'm not sure why you say that coding with C on the Desktop is insanity. I don't want to get into a language war, but C++ allows you to shoot yourself in the foot just as much as C does.

[liam]

Well, very nearly. The biggest thing in C++ favor, IMO, is the most obvious: having a basically modern object oriented design. I would kill myself if I had to deal with only C for gui stuff.
Don't get me wrong, I've nothing against C. There's nothing better if you don't need the overhead of a more modern language, but I find it takes me longer to produce sane code in c. However, I admit that that may not be the case if I focused more on C, regardless, most cs students/coders are more familiar with vm backed/non memory obsessed lanuages like java or python, ime.

Best,
Liam