> IMO, if we really wanted prefix:local to mean > concat(uriOf(prefix),local)), then that convention should have been > enforced for element names as well as for references to them. For better > or worse, I think the cat left the bag when Namespaces 1.0 was published. What about concat(uriOf(prefix), "/", local) ? Couldn't that be treated as a more modern interpretation of QNames (including element generic identifiers)? It's hard to imagine what software this would break. But I'm sure others are more imaginative than I am. That said: I would much prefer that the value spaces of CURIEs and URIs be distinct. If we could go back in time we could add an extra colon to QNames (which would make them more "namespace-like" in a C++/Perl sense) and that would be solved. There is nothing technical that would prevent a migration to double-colon syntax but the weight of legacy would take a long time to overcome! Paul Prescod
Received on Thursday, 29 June 2006 05:56:00 UTC