 |
 |
Stevens,Pat wrote: >As I read this I began to think that we should rethink the philosophy of a single authorized version of the name to which all other variants are linked. Instead we could link multiple versions to a name identifier. > > > Pat, from what I have heard, there are folks working on (or at least thinking about) the idea of an "author identifier". This would solve a number of problems, at least for those current authors that would participate: it would bring together articles written under variant forms of the name; it would give authors and publishers a way to identify authors unambiguously (all of the William Jones', for example). The down side is that there's a huge backlog of authors in our environment, so we'll still have many ambiguous or un-identified authors in our databases. But we know that it's often necessary to work on the future even when you can't resolve past issues. Interestingly, the authoritative form of the name in library authority files IS a unique identifier. No two authoritative names can be the same, which is why birth dates are sometimes added to make the name unique. It was a pre-automation technique that solved the same problem in a way that was appropriate to the catalog of the past. What is different today is that we could retrieve the authority record dynamically or link records based on an identifier other than the name itself; in the past, the authoritative form of the name had to be placed in the actual bibliographic record in order to have that quality of being a unique identifier. If one relies on the LC Name Authorities file, the LCCN on each record is an identifier for the author, so at least for book authors we have a nearly complete file of IDs that could be used. -- ----------------------------------- Karen Coyle / Digital Library Consultant [log in to unmask] http://www.kcoyle.net ph.: 510-540-7596 fx.: 510-848-3913 mo.: 510-435-8234 ------------------------------------