Metadata Creation before Digitization: Strategies to Unveil Hidden Collections

Metadata Creation before Digitization: Strategies to Unveil Hidden Collections Anton duPlessis, Clinical Assistant Professor (1) duplessis@library.tamu.edu , Lisa Furubotten, Associate Professor (2) cuaco@library.tamu.edu , Felicia Piscitelli, Associate Professor (1) f-piscitelli@library.tamu.edu , and Alma Rivera, Academica del Area de Servicios Tecnicos/Automatizacion (3) alma.rivera@ibero.mx Angel Villalba Roldan, Head of Cataloging (4) angelv@iib.unam.mx listed in alphabetical order (1) Cushing Memorial Libraries and Archives, Texas A&M University Libraries (2) Texas A&M University Libraries (3) Biblioteca Francisco Xavier Clavigero, Universidad Iberoamericana Ciudad de Mexico. (4) Hemeroteca Nacional de Mexico Many libraries are concerned about the invisibility - because inadequately registered - of their valuable research collections of primary source materials including rare books, photographs, and documents. Whether caused by limited staffing or lack of expertise, the unsatisfactory or nonexistent bibliographic description impedes discoverability and access while also precluding consideration for digitization or digitization on demand. Essentially -- without at least rudimentary descriptive metadata -- if you do not know what you have, you cannot evaluate the materials for digitization or easily digitize on a large scale, particularly when institutional preference and workflows favor collections with previously existent descriptive metadata. We discuss a straightforward, inexpensive technique to generate item descriptions which can be expressed in various metadata schemas such as Dublin Core or MODS, while also simultaneously producing traditional MARC21 records for the library catalog. Students, with a user-friendly template, and guided by librarians and curators at Cushing Library, created descriptions for materials in the Mexican Colonial Collection. An interdisciplinary, international team tested different software seeking one which permitted the construction of a tool with a friendly interface permitting non-catalogers to input basic descriptive data, affording an opportunity to reevaluate the prevalent idea that special collections metadata generation requires highly specialized professionals.  Challenges included determining basic element sets for different type of collections, solutions for cross walking the data among different metadata schema, and the problem of adequate file naming for matching records to files upon the subsequent generation of digital versions of collection items.  As well as a tool and methodology that facilitates training and supporting student employees in creating basic data for describing homogeneous special collection materials (with an acceptable error rate), the project produced DC and MARC records that permit scholars to discover, and retrieve items within the Mexican Colonial Collection, and identify those which should be digitized.  Further testing with other collections and libraries will be done in the near future. The funding of this project was granted by the Cataloging Hidden Special Collections and Archives Program from the Council of Library and Information Resources (CLIR).