Re: [Json] Unpaired surrogates in JSON strings

[Douglas Crockford <[e-mail]>]

On 6/6/2013 12:16 AM, Joe Hildebrand (jhildebr) wrote:
> On 6/5/13 7:08 PM, "Douglas Crockford" <[e-mail]> wrote:
>
>> The application that does that is JavaScript. Any 16-bit value can be
>> put next to any other 16-bit value and then JSON encoded.
> Agree.  This is a relatively-outdated worldview that we're stuck with, and
> is why I didn't say MUST in my suggested language.
>
>> The meaning of
>> 'character' throughout the RFC is ECMAScript's, which is roughly the
>> same as Unicode's code point.
> I'm not convinced yet.
>
> "\uD834\uDD1E".charCodeAt(0).toString(16);
>
> Yields:
>
> 'd834'
>
> That's not a code point.  That's half a surrogate pair for a code point
> encoded in UTF16.  It's only the same in the BMP.
>
What  then is the standard name for a 16-bit element of text? When 
JavaScript was created, that word was character. What is the word now?