Re: [Json] REMINDER - WGLC Ends 2013-10-11

On Fri, Oct 11, 2013 at 2:06 PM, Jim Schaad <ietf@augustcellars.com> wrote:

>
> 1.  It seems odd in Section 1 that a value is defined for object, but it is
> not for array.  It might be better to move the definition of value in the
> paragraph of what JSON can represent so it does not appear to be special
> for
> object.
>

Don’t get it, which para is that?  I’m looking at instances of “value” in
Section 1 and not seeing it.

>
> 2.  In section 1.2, My reading of the following text ||"Hello world!",
> "42",
> and "true"|| is that there are three strings here, but I think it was
> supposed to represent a string, an number and a Boolean.  I am not sure how
> to make this clearer, but killing the quotes on the last two items might be
> reasonable.
>

Yep, done.

>
> 3.  It seems a bit odd to title section 2 "Grammar" when it does not have
> the entire grammar.  It also seems odd that the introduction paragraph
> talks
> about things which are not in this section, but are in the following
> sections.
>

Yeah,  the title should be “Tokens” or some such, but it’s survived all
these years without causing breakage, I think we can live with it.

>
> 4.  The following is technically a change in the grammar, however is there
> a
> reason why an integer cannot be 00, but 1E00 is legal?  Do we want to
> change
> the definition of the exponent so that it matches that of integer?
>

Eeeek, please no.

> 5.  There has been mention that one reason for using ABNF is that it can be
> machine consumable, does it make any sense to create an appendix which has
> the full grammar in it for simplicity of reference?
>

Evidence suggests that people have no trouble constructing JSON parsers, so
probably not a huge ROI here.

>
> _______________________________________________
> json mailing list
> json@ietf.org
> https://www.ietf.org/mailman/listinfo/json
>