Exciting GSoC Projects For 2011

[phoronix]

Phoronix: Exciting GSoC Projects For 2011

Besides the exciting Mesa / Wayland / X projects accepted as part of this year's Google Summer of Code program, there's a number of other interesting projects for other open-source projects. Some of the other accepted projects include modularizing KDE's KWin compositing window manager, USB 3.0 support for Haiku, GIMP's GEGL library supporting OpenCL, a compositing window manager for Fluxbox, and bringing NetworkManager to FreeBSD...

http://www.phoronix.com/vr.php?view=OTM3OA

[deanjo]

SaX3 for openSUSE, a new version of SaX for xorg.conf settings generation.

One of the top voted must have features.

https://features.opensuse.org/308357

Despite all the efforts that have gone into making xorg.conf obsolete people have found that a once powerful tool is sorely missed in the latest openSUSE as "good enough autodetection" isn't good enough for a lot of end users. BTW, "for openSUSE" isn't necessarily true. SaX3 is aimed to make it easy for other distros to adopt (ala no ycp).

http://michal.hrusecky.net/2011/02/g...1-ideas-sax-3/

One feature that is also a little bit important is to make it distribution independent to make it easy for other distributions to use it, improve it and contribute back, so we will make a world better place for all of us

[Naib]

forgot the

Bring STEAM to linux

[kraftman]

One of the top voted must have features.

https://features.opensuse.org/308357

Despite all the efforts that have gone into making xorg.conf obsolete people have found that a once powerful tool is sorely missed in the latest openSUSE as "good enough autodetection" isn't good enough for a lot of end users.

It's true. Maybe my mouse will start finally working in the next OpenSuse edition.

[pingufunkybeat]

AFAICT, the effort to remove xorg.conf is not meant to burn it with fire and never use it, but rather to let you have a sensible session without one, and have a 3-line xorg.conf instead of cryptic 15400 line one.

I, for one, am glad that the modelines nonsense is not needed to get a mouse cursor anymore, and that the mouse will just work for most people without needing to choose between an imps/2 driver and who knows what else. xorg.conf was in serious need of a facelift, and I'm glad it got one.

[XorEaxEax]

Definately alot of exciting projects in this years SOC, when I read 'Super optimizations for LLVM' it got my hopes up for PGO (profile-guided-optimization) finally being available for llvm but I guess it's perhaps too big a project for a SOC. Still, 'superoptimization' sounds very interesting as does the work on Polly (polyhedral optimization). GCC optimization for cpu/memory usage is obviously a great project, same goes for the warnings plugin since this is an area where GCC could use improvement.

Since Haiku is my favourite OS I'm happy to see it 8 projects approved this year, most excited about the language bindings to the API, hopefully this will help people get more native projects going.

And as usual, hoping for a great turnover on projects in applications like Inkscape, Blender, Gimp, SDL, FFMpeg, since I use them frequently.

[droidhacker]

Hmm, the hardware accelerated h264 for android + vlc for android would be pretty cool
... vlc uses ffmpeg, so accelerated vlc for android. Means a REAL video player that can handle anything you can throw at it. Android currently suffers (like every other mobile OS) from SEVERE media container limitations, even when the content is decodable by hardware. The options are either to remux it into a compatible container (which is probably compatible with either the video *or* the audio and not both...), or to do it in software (poor performance, high battery consumption).

[Djhg2000]

I fail to see the excitement in

- Intel IA64 architecture user-level emulation support.

What will this really mean for home users?
Isn't IA64 a server architecture and very incompatible with AMD64?