June 2, 2010

Got Ovid? Classical knowledge base will assist in citing ancient Greek and Latin texts

Scholars looking for multiple sources and translations from among 1,000 years of Greek and Latin texts will have a powerful new tool in their research arsenal with a database being developed at Cornell.

The Classical Works Knowledge Base (CWKB) -- a relational database and specialized link resolver software -- will facilitate linking from citations of ancient texts to the online versions of those texts. The database will ultimately cover all Latin and Greek authors from Homer to Bede, from approximately the eighth century B.C. to the mid-eighth century A.D.

The Andrew W. Mellon Foundation recently granted $215,000 to the American Philological Association (APA) to implement the project, spearheaded by principal investigator Eric Rebillard, professor of Classics and history, in collaboration with Cornell librarians David Ruddy and Adam Chandler. The APA project also received a Mellon planning grant in 2008.

"I got in touch with University Librarian Anne Kenney for consulting with library specialists about the possibility of using the OpenURL framework for linking citations to full texts. She organized a meeting, and after that the project developed in a collaborative way with David Ruddy in E-Publishing and Adam Chandler in Database Management," Rebillard said.

Rebillard, Ruddy and Chandler have developed a working prototype at http://cwkb.org/. Rebillard expects the fully functional version of CWKB to be online in two years.

CWKB works by parsing OpenURL links (commonly used in libraries to help patrons retrieve scholarly articles) once a citation has been clicked on. OpenURL metadata is sent to the link resolver, which "creates several links -- because you can have several versions for the same citation, in the original language and in translation," Rebillard said.

"OpenURL was created about 10 years ago to solve this problem of linking from a citation to the full text," said Chandler, the database management research librarian who programmed the CWKB software. "The current OpenURL method of journal citation isn't quite what we needed, so we designed another metadata format for linking to these canonical works."

The electronic version of the database of classical bibliography L'Année philologique (The Year in Philology) will be the first abstract and index database to propose such links to CWKB. Many other resources are potential users of the new tool.

"For example, the works of the Founding Fathers are full of references to classical texts," Rebillard said. "It would greatly enhance the reading of the Founding Fathers to have links to those texts."

With applications for canonical citations in other fields and types of literature, the project can serve as a model and tool for scholarship in a number of disciplines.

"We've wanted to keep the OpenURL metadata part of our project as widely useful as possible," Ruddy said. "This work can be applied to any discipline that has developed conventions of textual citation which are reasonably independent of specific editions, such as in Biblical or Shakespearean studies."