"Orphan Works As Grist for the Data Mill" by Matthew Sag | Berkeley Tech. L.J.

Matthew Sag, Orphan Works As Grist for the Data Mill, 27 Berkeley Tech. L.J. (2012). "ABSTRACT: The phenomenon of library digitization in general, and the digitization of so-called “orphan works” in particular, raises many important copyright law questions. However, as this Article explains, correctly understood, there is no orphan works problem for certain kinds of library digitization. The distinction between expressive and non-expressive works is already well recognized in copyright law as the gatekeeper to copyright protection.... The non-expressive use of copyrighted works has tremendous potential social value by making search engines possible, and by providing an important data source for research in computational linguistics, automated translation, and natural language processing. Furthermore, the macro-analysis of text is being increasingly used in fields such as the study of literature itself. So long as digitization is confined to data processing applications that do not result in infringing expressive or consumptive uses of individual works, there is no orphan works problem because the exclusive rights of the copyright owner are limited to the expressive elements of their works and the expressive uses of their works."