Re: [MMUSIC] Roman Danyliw's Discuss on draft-ietf-mmusic-sdp-uks-06: (with DISCUSS and COMMENT)

["Martin Thomson" <[e-mail]>]

I've combined the changes with Ben's.  See    https://github.com/martinthomson/sdp-uks/pull/11

On Tue, Aug 6, 2019, at 12:24, Roman Danyliw via Datatracker wrote:
> (1) Section 3.2.  There are a few places where further clarity on error
> handling would be helpful:

Added requirement to use illegal_parameter.

> (2) I support Ben’s DISCUSS on the clarity of Section 3.2. I would add a few
> additional observations:
> 
> ** Per the sentence, “The resulting string is then encoded using UTF-8”,
> shouldn’t this JSON text already be UTF-8 per Section 8.1 of RFC8259 -- what’s
> the new encoding to be done?

A JSON text is a string of tokens, not octets, by my reading.  The UTF-8 encoding is something that happens to JSON text to produce a sequence of octets, so you have to say that.

I realize that isn't clear from RFC 8259, but the risk I'm hedging against is the UTF-8 encoding not happening, not that it might happen twice (which is a nonsensical operation in many contexts, but I will concede still possible in some).  At least I hope that someone will realize that encoding twice is not a sensible option.