That is pretty sweet. I can't wait to get a very usable implementation of that.
2 things, though. First of all, should those windows in the video be in the reversed order? You would think that when you move your head side to side that the window in the front would move most. Maybe its just the way that the video looks, but something seems fishy to me.
Second, when using the head tracker, how does the mouse function? If the slightest motion of the head moves your windows around, how is it possible to reliably click on your 3D windows with your 2D cursor? Also how is the z order determined? Is it just the least recently to most recently activated window? If so, wouldn't the windows do some serious jumping as you switch the focus of the windows? I suppose this could be countered with some smooth animation of sorts, but in my opinion, a 3D computing environment should just have all windows in the layering order you put them in and changing the focus leaves then where they are. That way the windows are where you want them. I suppose that will largely be a personal preference thing, but I suspect more people would agree.
Like I said, I'm really excited to see where this goes and how soon I will be able to implement it.
The front most window is on the screen, so it would not move at all relative to the screen itself. This is actually what allows interaction to be possible, as the top most window doesn't move. Z order is determined by Z order, period, there is no determination. I've tried animation, but failed miserably and decided to leave it for after I get a Wiimote myself and have a use for it. You can move the windows manually, but due to technical limitations (for which fixes have been in the works for many many months), you are not able to correctly click on windows that aren't flat on the screen (there is absolutely no way to transform the clicking into 3d space at this time, we're working on it, the X server guys are working on it, the functionality will come)
To tell you the truth, this video is extremely buggy and doesn't even use Lee's trig calculations. I just recently added them myself, so as soon as SmSpillaz gets the updates we should be able to get a new video out. I just hope my Y values are correct (I may have inverted them
)
You've actually brought up a great point and I'm going to try to get the Z order changing animation working this weekend, now that I have plenty of time to do it.
