The Cornell Language Acquisition Laboratory and Albert R. Mann Library are in the midst of developing an innovative collaboration between a research laboratory and an academic library to plan for the data preservation and discovery needs of the twenty-first century. Digital technology and internet communication now provide the opportunity to revolutionize the research process, through the ability to store, preserve, share, discover, and reanalyze vast amounts of data. While some disciplines, such as genomics or astronomy, have already developed sophisticated information technology infrastructure for these tasks, others are only beginning such work. In many, if not most research fields, it is especially difficult for those uninitiated to discover where data are located, what they describe, and how they may be used. This project has begun to tackle these issues by taking advantage of the library's existing expertise in preservation, archiving, and metadata creation, building on the existing ontology-software tools the library has developed, and introducing a new conceptual framework that divides the tasks of data sharing into discrete levels that may be managed and presented in defferent ways not only for different audiences but respecting political divisions and control issues that will always be present throughout the laboratories and institutions of academia.