Re: [Json] I-D Action: draft-ietf-jsonbis-rfc7159bis-04.txt

[Julian Reschke <[e-mail]>]

On 2017-07-21 03:42, Peter Saint-Andre - Filament wrote:
> I have three comments about this new sentence:
> 
>     When transmitting over a network protocol, or as a payload of a
>     network protocol intended to be interpreted as part of a protocol,
>     JSON text MUST be encoded in UTF-8 (Section 3 of [UNICODE]).
> 
> 1. Is the term "network protocol" sufficiently clear? It might be
> interpreted as a communications protocol at the network layer (e.g.,
> IP), not as an application-layer protocol above the network layer -
> which is where most or all usage of JSON occurs.

No, it's not totally clear, and that's indeed a problem when combined 
with a MUST.

> 2. What are we trying to say with "as a payload of a network protocol
> intended to be interpreted as part of a protocol"?
> 
> 3. Why do we reference Section 3 of the Unicode Standard for the
> definition of UTF-8, not RFC 3629? (Among other things, there's no
> guarantee that Section 3 of http://www.unicode.org/versions/latest/ will
> continue to talk about UTF-8.)

Right.

> I suggest:
> 
>     When an entity transmits JSON text over a network, e.g. as the
>     payload of an application protocol, it MUST encode that text using
>     UTF-8 [RFC3629].

This gets us back to the suggestion to apply this rule to the use of the 
application/json media type (and types derived from it).

Best regards, Julian