Roy Tennant, OCLC
MARC was created for printing cards. Online displays usually just a reformatting. Uncontrolled text strings with inconsistent punctuation,
identification, linkage, quality problems.
Hamlet problem: lots of editions displayed in a list.
same names, such as John Rock.
Name entities: works, persons. Then relationships between them.
canonical identifiers, such as author authorities in OCLC.
OCLC has done works, doing persons next in linked data.
Google papers on knowledge vault. Write programs to extract values (knowledge values)
Entity.JS where you can see some of OCLC’s assertions.
Fiction Finder: solving the Hamlet problem.
VIAF identifiers in Wikipedia: solving the Wang / Li / Zhang problem (most common Chinese names).
Linking works and translations: solving the language problem.
Cleaning up and normalizing data, such as all the different data in 245 subfield h. Solving the quality problem.
Webinar upcoming about what catalogers can do now to prepare for linked data.
Library linked data in the cloud by Godby et al.
You don’t need to do anything about linked data now. There is a Bibframe mailing list.
Updated to add links.