When I try to click on objects which are at the screen edge, it doesn't work when the pointer is directly at the edge. For instance, I can't throw the pointer to the bottom to quickly select a window in the task list because I then hit the unsensitive bottom edge. This breaks Fitt's Law as one has to accurately have to point at items at the edge instead of just moving the pointer towards them. Other window managers like Metacity don't have this problem.
(In reply to comment #0) > When I try to click on objects which are at the screen edge, it doesn't work > when the pointer is directly at the edge. For instance, I can't throw the > pointer to the bottom to quickly select a window in the task list because I > then hit the unsensitive bottom edge. This breaks Fitt's Law as one has to > accurately have to point at items at the edge instead of just moving the > pointer towards them. Other window managers like Metacity don't have this > problem. > I've found that the only areas that are non-responsive are the right and left edges of the screen. The corners of the screen are responsive. This only affects the cube setup. The wall setup has responsiveness on all edges and corners.
Huh, I thought this kind of behavior was just due to corrupt settings or something. Resetting all compiz settings allowed me to fix this issue when it came up sometimes, but that's not a solution.
For reference, here is the Compiz forum thread I opened about the issue, which seems to have some good information on it's cause: http://forum.compiz-fusion.org/showthread.php?t=6885 And here is the corresponding Ubuntu bug, which also seems to have useful comments: https://bugs.launchpad.net/ubuntu/+source/compiz/+bug/103306
(c&p'd from my compiz forum post) This also hits auto-hide panels on the edges of the screen, or other windows that appear only when mouse is on the edge. Curiously, these programs don't seem to be affected by brightside, which is able to map events to edge motions, but still allows auto-hide panels to be opened. Maybe it's worth looking at how brightside detects pointers-on-edges?
For the record, there has been a fix made by Ubuntu developer Michael Vogt for the wall plugin. The patch is named 02_fix_edges.patch and is present in the compiz-fusion-plugins-main Ubuntu package (see http://archive.ubuntu.com/ubuntu/pool/main/c/compiz-fusion-plugins-main/). As I understand from the thread on Launchpar (https://bugs.launchpad.net/ubuntu/+source/compiz/+bug/103306) Ubuntu developers are unwilling to prepare an analogous patch for the cube plugin since there are already too many patches to maintain which are not integrated to the official Compiz source. I'll attach the patch here, and I hope it gets integrated into upstream Compiz sources so that the Ubuntu team won't have to maintain them indefinitely.
Created attachment 16334 [details] [review] The Ubuntu patch for the wall plugin This patch only fixed the wall plugin. The cube plugin could be fixed as well if this patch got integrated into official Compiz sources... Citing the relevant Ubuntu bug: "Jeremy Nickurak wrote on 2008-03-02: (permalink) I hate to beat a dead horse, but again, since the behavior of the bug is identical with the wall and cube plugins, and both are included in the package, how does the bug not apply to both plugins, aside from the fact that one plugin is enabled by default and the other is not? Or would it be more appropriate to file a seperate bug for this behavior under each effected plugin? --- Travis Watkins wrote on 2008-03-02: (permalink) To be honest I don't much care about what happens with the cube plugin. As is we're carrying a patch against wall that doesn't forward port well and that we'll likely have to carry forever, I'd rather not have another one for cube. For cube you can just disable all the edge flip options (pointer, dnd, move) and that'll fix the problem too."
Does this problem still appear in Compiz 0.9.x? I don't use Compiz anymore and can't test it, unfortunately, but others might be interested.
Mass closure: This bug has been untouched for more than six years, and is not obviously still valid. Please reopen this bug or file a new report if you continue to experience issues with current releases.
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.