3: E.g. If you include a LoRa sensor, the LoRa Manager will name it like this;
Device name: <Manufactorer> - <DevEUI>
Function names: <DevEUI> - function name (from manufactorer documentation).
If you add a Z-wave device the Zwave Manager will name it like:
Device name: <Node id>. Sensor type
Function names: <Node id>. Function type (from Zwave standard)
If you add a boat via SignalK there will not be any device but the funktions are called the same thing as the path in SignalK
If you add a Nagios installation to it every host will be a device and all services will be functions. With the same name as in the Nagios installation.
So there is no common rule that applies to all subsystems. But you can always change the name of the functions and devices as you see fit.
You have to create your own functions and devices only when the subsystem (or integration) doesn't do it for you. So when we are playing with Node-RED like in the Basic course, there are no subsystem involved. Then we have to do everything ourselves and that is quite easy too.
I hope that explained it for you and keep up the good work! 😃