...to partially destroy the micro and/or external chips when experimenting, resulting in very very strange observations later. Just take fresh parts and try it again.

Take care that you never write a "0" to P2.0 as long as your 74HCT165 is connected to the micro! The reset level of P2.0 is "1" helping you to keep the 74HCT165 intact after power-up. But remember that there can be a brief period of time between supply voltage rise-up and reseting the port pins, where the port pins' status can be undefined. This happens usually with micros not being able to provide an asynchronous reset. I'm not familiar with your micro, but this behaviour can result in momentary shorts and damage.

Sometimes it's helpful to insert a small resistance at input of micro to prevent damage from such shorts. In combination with AT89S52 I often take 330R.

Kai