I think I have done it with the whole gestures and touch in coding, which results in the screen shot below. The squares are for illustration...not final.
- When screen is touched (square in the middle) you can move to the other squares which results in a defined action for that square.
- If you stay in the white area nothing happens. Also because it's very irritating when moving one pixel to the left and immediately one pixel down two actions are performed. See it as a safe zone.
- The light square is a single action, e.g. one down
- The middle dark square is a repetitive action (slow), e.g. down
- The dark square is a repetitive action (fast), e.g. down
- When moved to the white area the actions are stopped
On top of all this the following is also working:
- Swipe to any direction is movement to that direction
- When touch and immediately lifted is Ok/Select
- Double tap = action
- Two finger single tap = action
I have made everything configurable.
I think when you see the squares you can navigate more precise than when not seeing them. At least I am experiencing it like that. Of course they have to be made a little bit better and not so obvious.