Redirects links

And also closes #2524 again. Fixes definition of operators, and does
some reflow.

Author: JJ

doc/Language/operators.pod6

@@ -1529,33 +1529,43 @@ if C<$a eqv $b>, then C<$a cmp $b> always returns C<Order::Same>.
 
 =head2 infix C<coll>
 
-Defined as:
+Defined asN<These are not yet implemented in the JVM>:
 
     multi sub infix:<coll>(Str:D \a, Str:D \b --> Order:D)
     multi sub infix:<coll>(Cool:D \a, Cool:D \b --> Order:D)
-    multi sub infix:<coll>(Pair:D \a, Pair:D \b)
-    multi sub infix:<coll>(Str:D \a, Str:D \b)
+    multi sub infix:<coll>(Pair:D \a, Pair:D \b --> Order:D)
 
-C<coll> is a sorting operator that takes pairs of C<Str>s, C<Cool>s or C<Pair>s and returns an C<Order> that uses the C<$*COLLATION> order. The default behavior disregards diacritic marks and capitalization, for instance.
+C<coll> is a sorting operator that takes pairs of C<Str>s, C<Cool>s or C<Pair>s
+and returns an C<Order> that uses the C<$*COLLATION> order. The default behavior
+disregards diacritic marks and capitalization, for instance.
 
     say "b" cmp "à";  # OUTPUT: «Less␤»
     say "b" coll "à"; # OUTPUT: «More␤»
 
-In the first case, lexicographic or codepoint order is taken into account. In the second, which uses C<coll>, the diacritic is not considered and sorting happens according to intuitive order.
+In the first case, lexicographic or codepoint order is taken into account. In
+the second, which uses C<coll>, the diacritic is not considered and sorting
+happens according to intuitive order.
 
 =head2 infix C«unicmp»
 
+Defined asN<These are not yet implemented in the JVM>:
+
     multi sub infix:<unicmp>(Str:D \a, Str:D \b --> Order:D)
-    multi sub infix:<unicmp>(Pair:D \a, Pair:D \b)
+    multi sub infix:<unicmp>(Pair:D \a, Pair:D \b --> Order:D)
+    multi sub infix:<coll>(Pair:D \a, Pair:D \b --> Order:D)
 
 Unlike the cmp operator which sorts according to codepoint, C<unicmp> and
-C<coll> sort according to how most users would expect, that is, disregarding aspects of the particular character like capitalization.
+C<coll> sort according to how most users would expect, that is, disregarding
+aspects of the particular character like capitalization.
 
     say 'a' unicmp 'Z'; # Less
     say 'a' coll 'Z';   # Less
     say 'a' cmp 'Z';    # More
 
-The main difference between C<coll> and C<unicmp> is that the behavior of the former can be changed by the  L<C<$*COLLATION>|/language/experimental#index-entry-%24%2ACOLLATION-%24%2ACOLLATION> dynamic variable.
+The main difference between C<coll> and C<unicmp> is that the behavior of the
+former can be changed by the
+L<C<$*COLLATION>|/type/Any#index-entry-%24*COLLATION-%24*COLLATION> dynamic
+variable.
 
 =head2 infix C«leg»
 
@@ -1564,7 +1574,8 @@ The main difference between C<coll> and C<unicmp> is that the behavior of the fo
 
 X<String three-way comparator>. Short for I<less, equal or greater?>.
 
-Coerces both arguments to L<Str|/type/Str> and then does a lexicographic comparison.
+Coerces both arguments to L<Str|/type/Str> and then does a lexicographic
+comparison.
 
     say 'a' leg 'b';       # OUTPUT: «Less␤»
     say 'a' leg 'a';       # OUTPUT: «Same␤»