As per the last TAG f2f meeting, we owe the Web Audio WG some more detailed feedback about their APIs.

Specific areas of concern that were discussed in the F2F were:

• Doubling up of normative surface area.
• Lack of ctors
• No explanation of <audio>, only wrapping
• No de-sugaring of most block types into Script Block code
• TAG didn't understand why Script nodes have such a buffering/stuttering penalty...is this only in naive implementations?