On Wed, 05 Aug 2009 12:08:54 +0200 Simon Thum <[email protected]> wrote:
> Harald Braumann wrote: > > If it's not possible yet, are there plans to implement such > > functionality in the X server? Like sending an event whenever the > > pointer stops? > Well, X is in pretty much the same situation here; it just knows when > the mouse moves, or when it did so last time (plus something else > happens). > > There is a distinct advantage the server has, namely, it knows more > detail. So it could do a better job than a client/wm. But still that's > far from perfect, which I guess is why reading tea leaves has been > left to the client. I agree completely, but the question is: does the client have all the tea leaves to read from? Here's where my knowledge of the technical details is not deep enough. From what I understand, the WM would have to grab the mouse (probably bad) and then receive all motion events (maybe a problem over slow links). > A possible alternative could be to synthesize inferred proximity from > the driver and make $WM use that. This however doesn't exactly solve > the problem, is hard to do (especially above the driver), so I guess > devs will hate it. Cheers, harry
signature.asc
Description: PGP signature
_______________________________________________ xorg mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/xorg
