Re: [Json] Allow any JSON value at the top level

[Jacob Davies <[e-mail]>]

On Mon, Jun 10, 2013 at 1:17 PM, John Cowan <[e-mail]> wrote:
> Jacob Davies scripsit:
>> The original specification has an asymmetry in what values may be sent
>> at the top level motivated by an encoding-detection scheme that is now
>> obsolete.
>
> How do we know what the motivation is?  I think the self-delimiting nature
> is more important.

The encoding-detection is the only motivation mentioned in the RFC,
and then only implicitly: "Since the first two characters of a JSON
text will always be ASCII characters [RFC0020], it is possible to
determine whether an octet stream is UTF-8, UTF-16 (BE or LE), or UTF
-32 (BE or LE) by looking at the pattern of nulls in the first four
octets."

Of course there can be other notable benefits or drawbacks regardless
of the stated motivation. Self-delimiting is a benefit, it appears.
The fact that a sub-part of a valid JSON text is not also a valid JSON
text is a drawback in an otherwise symmetric data model.

> In any case, even if numbers, strings, booleans, and null were allowed at
> the top level, all application/json entity bodies would still begin with
> a non-null ASCII character, which is sufficient for encoding detection.

Good to know.