I didn't really understand your reply, except it seems to have been misinterpreted my post (but as I put on the disclaimer, I did had difficulty to express my opinion).
What I wanted to say is that if we apply the kerchkoff principle
The portofolio of procedures, and the "how to execute" theses procedures have no reason to be secret (for example slicing the pie, how to kick effectively a door, and so on).
The operationnal security being essentially on the following secret : which procedure(s) you will use for which situation/mission/....
And concerning the "we speak of computer science <=> we want to do 10¹⁴ operations but wen can't do them IRL", I really don't understand.
Kerchkoff is on the information & security theory, not from cryptanalysis.
I only wanted to bring a new angle for the topic (how much of a procedure could be public), that's all.
A little pic I think we will all agree on concerning cryptanalysis, security and IRL
And to continue on the irrelevant part I introduce :
signal (and it's derivate what's app, appolo, etc...) is cryptographically sound ... .
If you want to protect the content
of the messages, that should be good. But it doesn't prevent a specific third partie (the enterprise giving you the service) to know who speaks with who.
This last metadata could be of great interest for some intelligence groups
It is enough to protect against a standard eavesdropper.
Concerning the "tactical" radio , that's an hardware reserved for military and special operations police groups in my knowledge.
As far as I know it, the simple "Men In Blue" doesn't have access to them.