On 6/28/07, Steven ** <[EMAIL PROTECTED]> wrote:
On 6/28/07, Peter A Trotter <[EMAIL PROTECTED]> wrote: > > > On 28/06/07, Fabien <[EMAIL PROTECTED] > wrote: > > > > On 6/28/07, Robin Paulson <[EMAIL PROTECTED]> wrote: > > > > > > someone on this list had a similar idea for enlarging number keys i > > > think, they would appear or resize depending on context iirc? > > > http://www.apple.com/iphone/usingiphone/keyboard_large.html > > > > > > That's shiny and beautiful, but of limited use: the key becomes bigger > > *after* it's been hit, too late to help the user aiming. Well, actually it's > > still got a practical use: it helps notice when one hits the wrong key (I > > guess no human being could type without looking at the keyboard on an > > iphone). > > > > This immediately made me think of enlarging keys on a predictive basis. > Not sure it is that useful but it was an amusing thought. > > -Pete > Which immediately makes me think of the way Dasher ( http://www.inference.phy.cam.ac.uk/dasher/) works, with the most common next letter(s) being greatly enlarged from its normal size. This could work for a keyboard too, with the predicted next keys getting larger and less likely keys shrinking. It would have to be handled very well though, or else it would just look horrible and be unusable. -Steven
Ok, so I finally got around to watching the original video. They sort of already do what I suggested. They just don't show it to the user. They enlarge the "tap zones" of the most likely next letters and shrink the "tap zones" of the unlikely letters. So, it's kind of transparent. I'm not sure if that's better or worse than actually increasing the size of the letters. What do you all think? -Steven
_______________________________________________ OpenMoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
