voicexml | NodeJS VoiceXML Interpreter | Runtime Evironment library

To my understanding, the functionality of a element is to define prompt audio to be played, followed by the system waiting for a user utterance. My impression comes from the VoiceXML 2.0 specification for prompts:

Conceptually, prompts are instantaneously queued for play, so interpretation proceeds until the user needs to provide an input. At this point, the prompts are played, and the system waits for user input.

Additionally, I understand the difference between a and a to be that:

1. a is a control item and is "not for gathering input", and
2. a is an input item whose value is "gather[ed] from the user"

Lastly, I know that a can be a direct child of both a and .

1. I am pretty sure that a collects input when inside a , but what is supposed to happen when it is inside a , since blocks are control items?

2. In the case that a in a does not collect input and only plays audio, why wouldn't you instead write the text for a prompt as content inside a , omitting the ?