Re: [Json] Differences between RFC 4627 or the current ECMAScript specification

[Tim Bray <[e-mail]>]

I’m with Rob on this one. There is de facto a stable ECMA spec that has
language defining JSON, the definition is different, and people who go and
try to do some of the things it allows will likely have interoperability
issues, so I’d put this in.  Yeah, they might do another spec or specs but
that wouldn't invalidate our reference to the existing one.


On Thu, Sep 26, 2013 at 11:36 AM, Paul Hoffman <[e-mail]>wrote:

> On Sep 26, 2013, at 10:31 AM, R S <[e-mail]> wrote:
>
> > Charter:
> > >
> > > All differences between RFC 4627 or the current ECMAScript
> specification > will be documented in the new RFC.
> >
> > The ECMAScript specification allows primitives at the root level,
> specifies exactly how to interpret numbers, and can handle " bit sequences
> which cannot encode Unicode characters" just fine.
>
> <no hat>
>
> Based on what we have learned in the last six months, it might be better
> for this RFC *not* to do what the charter says.
>
> - TC39 is actively revising ECMAScript and it is not clear whether the
> -bis draft of their version will be out first.
>
> - Some of what ECMAscript says about JSON is intertwingled with the
> definition of ECMAscript, such as "exactly how to interpret numbers"
>
> I'm no longer sure that a long-lasting RFC interpreting parts of another
> SDO's spec is a good idea.
>
> --Paul Hoffman
> _______________________________________________
> json mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/json
>