In article <[EMAIL PROTECTED]>, matthiasm <[EMAIL PROTECTED]> wrote:
> Well, he used no mouse but a teblet with a pen and absolute > positioning. As a result, my wonderful new widget was completely > unusable for him. And since he deleted the older version of my > software, he could not use my app at all (and nearly missed his > deadline). I'm intrigued by that. How did you reposition the pointer in order to break the gesture? The two common cases (besides these "fancy" ones) where I used warping was for linear scrollers (like the unbounded roller) and visual panning. If the object being manipulated was a widget, users preferred the mouse to be repositioned on one end of the widget itself. For example, if the object was a roller, and you would like to decrease a value, you would drag down, and the mouse would be warped up again at the start of the wheel continuosly. Under pan, however, most preferred to have the pointer warp at screenful boundaries (even for small views). Also, when releasing the button, the mouse would be backported to its starting position (in viewport terms). Having the pointer visible helped them "grasp" the motion even when they were lost in empty space. Ahh, those little things. > I stopped assuming that everyone uses a three button mouse with a > scrollwheel. I actually bought a tiny tablet and started checking my > apps for usability with a pen... . I like tablets for drawing, but not much else. I'm using a trackball most of the times, to reduce shoulder stress, but, to be perfectly honest, the mouse is still the winner. Higher precision, steady positioning, ability to click without inducing excessive motion (almost impossible to do with a pen), straight lines. Of course, having to use a trackpad makes you rethink about the necessity of precise positioning indeed, but that's not something I would use in a CAD :) _______________________________________________ fltk mailing list [email protected] http://lists.easysw.com/mailman/listinfo/fltk
