After the limited results of the batch process, the NCSU project team opted to do manual searches against DBpedia, Freebase, and VIAF, which also contained Library of Congress and ISNI identifiers. Seven staff members in the Acquisitions and Discovery department were quickly trained on how to search linked data sets and extract the appropriate information for producing URIs. The NCSU project team drew upon their cataloging experience and in the training emphasized that the process was very similar to selecting authorized headings as main or added entries in a MARC bibliographic record. One difference that did need to be stressed was that the form of name used in other linked data sets may differ from what was used in the NCSU ONLD but that it was permissible to link to them as long as both data sources were describing the same organization and just using different labels. For example, even though "ABC-CLIO" was selected as the authorized name in the NCSU ONLD and "ABC-Clio Information Services" was the authorized name used in VIAF, the link could be made to that VIAF entry because it was clearly describing the same organization. The NCSU project team's Access database was also designed to assist with the enhancement process, so staff only had to select the unique identification string for each descripLion, such as "140705396" for VIAF and "09c07t" for Freebase, and the database would generate the full URI, "http://viaf.org/viaf/140705396/" and "http://rdf.freebase.com/ns/m/09c07t," for each data set in the XML output. In order to find DBpedia URIs, staff actually searched Wikipedia because of the matching relationship between the Wikpedia URLs (http://en.wikipedia.org/wild/OxfordUniversity}ress) and DBpedia URIs (http://dbpedia.org/resource/Oxford_University_Press).