Re: Linked Open Data prototype at the German National Library

From: Haffner, Alexander <A.Haffner_at_nyob>
Date: Mon, 19 Apr 2010 10:17:27 +0200
To: NGC4LIB_at_LISTSERV.ND.EDU
Karen,

We did consider the RDA preferred and variant name for a person and we
discussed the usage quite intensively.

Before I clarify our solution some information first...
Our current cataloguing system offers the following subfields for name
representations: 
*	the applied cataloguing rule
*	a personal name (i.e. "Hildegardis" in Hildegardis Bingensis)
*	a subfield for one or more forenames (i.e. "Johan Wolfgang" in
Johan Wolfgang von Goethe)
*	the surname (i.e. "Goethe" in Johan Wolfgang von Goethe)
*	a name prefix (i.e. "von" in Johan Wolfgang von Goethe) 
*	and a qualifier (i.e. "Bingensis" in Hildegardis Bingensis)

For the prototypical RDF representation we made the decision to
represent each name in two ways:
1. a one-to-one reflection of our current data set state - we specified
a entity called "NameOfPerson" (similar to FRAD) which works as domain
for the application of gnd-properties to represent the single parts of a
name (see the name parts above; this is the way Adrian explained
already). We didn't apply URIs to those entities to keep consistent to
our current system. I know, this is suboptimal...
2. but we decided as well to offer an additional literal representation
for each name (like designated in RDA Element Sets). And of course we
would prefer the usage of rda:preferredNameForThePerson and
rda:variantNameForThePerson. We raised the discussion about and some
colleagues have concerns "pressing" our (RAK-WB) names into RDA-names
without knowing the authorized form of RDA-names (we faced issues like:
is an academic title part of a RDA-name, at what place do we have to put
the surname prefix, how do we have to deal with names like Hildegardis
Bingensis). Finally we used as well the gnd:preferredNameForThePerson
and gnd:variantNameForThePerson to specify the literal (DNB-specific)
representation. 

Once we have answers on all our open questions (hopefully at the latest
after the RDA toolkit publication) we intend to adjust our algorithms to
match the RDA-name constraints. 

Best, Alexander


--
Alexander Haffner
Deutsche Nationalbibliothek 
Informationstechnik
Adickesallee 1
D-60322 Frankfurt am Main
Telefon: +49-69-1525-1766
Telefax: +49-69-1525-1799
mailto:a.haffner_at_d-nb.de
http://www.d-nb.de
Received on Mon Apr 19 2010 - 04:19:10 EDT