XGL - your opinions

After recently interviewing Dave Mitchell, I've been interested in what everyone thinks about XGL - both in it's current state and where it's set to go.

For those who havn't come across XGL before (perhaps you don't even use X at all - perhaps you do everything in the console!)

WikiPedia entry for XGL, http://en.wikipedia.org/wiki/Xgl (2006)

Basically it drastically extends the ability of X by allowing it to use your graphics card interface rather than the CPU to render images, has has been done previously. As a result it's a lot faster, has more capability and uses less CPU time.

This has been around *way* before Vista's Aero system. Not to mention that as it's open source many many developers can hack away at it and make new extensions and other bits and pieces for it.

Right now it's not terribly user friendly. The window wobble is cute, but can get annoying. The rotating desktops on a cube is also an interesting thing - and it does show off how they can render live video on a 3D object from within X - it's not going to make my workflow any faster.

I can think of a bunch of things that'd be nice. 3D icons for one - that is, not 2D images of 3D objects, but actual 3D objects that give user feedback through point of view and lighting. Adaptive lighting conditions (so, during the day the desktop is lit as if by daylight, and at night it's lit as if by, say, a desk lamp. Real shadows for icons and text. Being able to put windows on angled "stands" to place them out of the way temporarily.

Do you think it's a good idea? Do you think that people will find better things to do with it than wobble windows and spinning cubes?

 

The 2D desktop will be around for a very long time, but the 3D rendering of XGL leads to some very interesting possibilities.

Think of the possibilities for image rendering by game engines. True 3D images instead of the shading and lighting tricks used now. True 3D scientific and engineering modeling applications.

Whatever the application, it won't take software engineers long to figure out it max potential.

 

Right now, it is in its infancy. Love to see what happens later.

Right now, its just oohs and aaahs. Nothing really practical.

Though, what SPD states, is very interesting.

 

Sounds like just what Linux needs to be competitive.