Remove the idea of negative codepoints

Author: lizmat

S15-unicode.pod

@@ -93,12 +93,11 @@ With NFG, strings start by being run through the normal NFC process, compressing
 any given character sequences into precomposed characters.
 
 Any graphemes remaining without precomposed characters, such as ậ or नि, are
-given their own negative numbers to refer to them, at least 32 bits in
-length. This is done to avoid clashing with any potential future changes to
-Unicode.
-
-The mapping between negative numbers and graphemes in this form is not
-guaranteed constant, even between strings in the same process.
+given their own internal designation to refer to them, at least 32 bits in
+length, in such a way that they avoid clashing with any potential future
+changes to Unicode.  The mapping between these internal designations and
+graphemes in this form is not guaranteed constant, even between strings in
+the same process.
 
 The Perl 6 C<Str> type, and more generally the C<Stringy> role, deals
 exclusively in NFG form.
@@ -152,12 +151,9 @@ Unicode. C<Str> deals exclusively in the NFG form of Unicode strings.
     ord(Str $string)
     ords(Str $string)
 
-These give you the numeric values of the characters in a string. C<ord> only
-works on the first character, while C<ords> works on every character.
-
-Some the values returned may be negative numbers, and are useless outside that
-specific string. You must convert to one of the codepoint-based types for a
-to-Standard list of numbers.
+These give you the numeric values of the B<base character> of graphemes in a
+string. C<ord> only works on the first graphemes, while C<ords> works on every
+grapheme.
 
 =head2 Length Methods
 
@@ -371,12 +367,11 @@ An error may be issued if the given category name is not valid.
     Stringy.ords() --> Array[Int]
 
 The C<&ord> function (and corresponding C<Stringy.ord> method) return the
-codepoint number of the first codepoint of the string. The C<&ords> function and
-method returns an C<Array> of codepoint numbers for every codepoint in the
-string.
+codepoint number of the base character of the first grapheme of the string.
+The C<&ords> function and method returns an C<Array> of codepoint numbers
+of the base character for every grapheme in the string.
 
-This works on any type that does the C<Stringy> role. Note that using this on
-type C<Str> may return invalid negative numbers as "codepoints".
+This works on any type that does the C<Stringy> role.
 
 =head2 Character Representation