TS-0009 - Character-set for http binding & json
TS-0004 defines in various places UTF-8 as the character set to be used for non-binary serializations and when using strings/text-based values. In http the request sender should specify the character encoding with the "content-type ... ;charset=..." header field. This is good practize. Also, the default for http/1.1 is a different one (ASCII?).
My suggestion would be to make the "charset=utf-8" parameter optional for http requests and responses (in TS-0009), and define a default fallback to UTF-8.
Further questions:
- The MQTT binding is character-set agnostic, how is it with coap?
- Inside an XML serialization one can specify a different character encoding. What happens if this differs from the default or specified encoding?