The iSchool Drexel University

This Note reports on efforts to build a generalizable OAI-PMH workflow to retrieve metadata sets from unrelated digital libraries. This effort is part of a wider effort to build a database to aggregate metadata from different digital libraries, which can then be used as the basis for content analysis and data mining experiments with the metadata records. A pilot metadata harvest from three digital libraries using OAI-PMH encountered a number of issues, arising from idiosyncratic legacy characteristics of each of the three metadata sets. In rthe end, the harvests had to be manually tailored to each library. OAI-PMH proved to be a useful approach, but only after communication with each digital library had identified important characteristics of each metadata set, including many legacy characteristics, which had to be acounted for in the harvest