> On 11 Mar 2017, at 16:41, ned+ietf@xxxxxxxxxxxxxxxxx wrote: > > > > It would be very helpful to have this text in the RFC. > That would be a regression. On the contrary - what you have now is incomplete if not actually broken, and this completes/fixes it. You have a specification that says (a) JSON can be encoded in UTF-8, UTF-16, or UTF-32 and that (b) That no charset parameter is needed to distinguish the charset and moreover one has "no effect". But you don't explain how why this is the the case, or what compliant implementation do to determine the charset. If you believe clarifying this is problemtic in some way, you need to explain why. > A better change would be to remove the fiction that JSON exists in UTF-16 or > UTF-32 forms. That's of course another way to fix the problem. But it's a far more significant change, and I have to wonder if there is a consensus to do it. Personally, I have no preference as to the approach that's used. But moving forward as-is is really not acceptable. Ned