Re: [Json] Proposal: the minimal edit

[John Cowan <[e-mail]>]

Paul Hoffman scripsit:

> I'm not sure what you mean by "exclude". The sentence Tim was referring
> to is about inclusion:
> 
>   - ECMAScript implementations can generate and consume code points
>     in JSON strings that are not Unicode characters.

There is no point in putting this item into a list of differences
unless it is not true of RFC-JSON.  If this language is to be added, it
follows that RFC-JSON excludes certain code points that are not Unicode
characters.  But exactly which code points?  As worded, presumably
all of them: the surrogate code points, the non-character code points
(a term of art in Unicode; it does not mean "code points that are not
characters", but refers to a specific list of 66 code points), and the
unassigned (as of some particular Unicode version) code points.

Tim's proposed revision replaced the last five words with "that can never
be Unicode characters", thus excluding from RFC-JSON only the surrogate
code points and the non-character code points, but not the unassigned
code points.  As it turns out, this is indeed what he meant to say.

My preference would be to exclude only the surrogates, not the
non-characters, which is yet a third position.  My suggested wording
would be simply "that are surrogates".

(Strictly speaking, code points are not characters; code points
*represent* characters.  This editorial point is orthogonal to the current
discussion.)

-- 
He made the Legislature meet at one-horse       John Cowan
tank-towns out in the alfalfa belt, so that     [e-mail]
hardly nobody could get there and most of       http://www.ccil.org/~cowan
the leaders would stay home and let him go      --H.L. Mencken's
to work and do things as he pleased.              Declaration of Independence