Value space of ID datatype

Part 2 says "The value space of ID is the set of all strings that match
the NCName production in [Namespaces in XML] and have been used in an
XML document."

The "and have been used in an XML document" makes no sense to me.  The
value space is surely the set of all strings that match the NCName
production.  In addition there is the implicit uniqueness constraintthat
you can't have the same NCName occurring on more than one distinct
element.  (XML also has the constraint that you can't have a single
element with multiple IDs, but I don't see any need for XML Schemas to
enforce this.)

The "and have been used in an XML document" is perhaps an attempt to
capture the uniqueness constraint in the specification of the value
space, but I don't think that can work.

James