Often debouncing circuits are done like this :
In this case, when switch is closed, C1 discharges in its internal series resistance, so discharge is very quick. If there is some rebounds and we suppose C1 internal resistance << R1, capacitor discharge time << capacitor charge time. Consequently, during rebounds when closing switch, response time will be very quick. I assume this configuration is useful when we a quick falling edge detection is required.
But, is this configuration safe for C1 lifetime?