You gave one example not even from the United States, one of self-initiative using features provided by the device, two where an emergency authorization was reasonable (and presumably was obtained, if it was needed), and one where it isn't very clear but apparently things were done backwards--snoop on people first, then decide afterwards you shouldn't have so you don't investigate.

You're not even talking about the same thing. None of the examples required anything remotely as extensive as Snowden, for example, revealed.

I understand that it is hard to give compelling details of how expansive data-monitoring operations have played critical roles in, say, the capture of terrorists because the data on that isn't public, but I'm not going to assume it works if you're not going to assume that the abuses are there too, all hidden. We have ample evidence from the past that intelligence has been used to avert harm. And ample evidence from the past that intelligence has been abused for political and personal ends. Without very good reason to believe otherwise, we should operate as if both continue to be true.

There's no reason why you can't have effective means of apprehending criminals without significant risk of undetected abuse, save for attitudes that we don't want to subject the process to intense scrutiny, or the opposite attitude that the government never has cause to intervene to save people in critical situations.

Again, the prior events to address are NOT ones in the USSR, but the ones in the U.S. like COINTELPRO.