Of course you are correct, but the tools available to other governments still concern me greatly. Are we actually presently capable of making sure nobody snoops on our data? If so then I agree it is a better solution. But is it not better to use an imperfect plan which is implementable than wait for a perfect plan (if such a thing exists for cyber-security) to become implementable when a threat is present?

Depends how paranoid you want to be, and to what degree of certainty (after all it's impossible to prove a negative).

But in general, backdoors for "the good guys" are just more surface area for "the bad guys", and developers of products which take security seriously lock themselves out too.

Is this about installing backdoors? Or is it just about ensuring the network doesn't run on known-compromised hardware/software?