Chris and I arrived at this understanding of what the rule should be: If a reference to a peer map cannot be resolved and there is a reference to a key name in the peer map's scope, the system SHOULD report the key reference as unresolvable but can indicate that, because the scope a peer map, it is possible the key could be resolved once the peer map is available. In this processing context, if there is an otherwise lower-precedence local-scope key definition with the same qualified name then that key definition should be used. That is, in the case where the root map cannot be resolved, the processor MUST treat the key space associated with the peer map's scope as being empty and normal key processing rules apply when resolving key references. Cheers, Eliot ————— Eliot Kimber, Owner Contrext, LLC
http://contrext.com