In this paper, we explore a range of issues yet to be addressed in the large-scale use of application profiles. While considerable attention has been paid to human-readable application profiles, there is a growing need for machine-readable application profiles that can support quality control mechanisms including, but not limited to, data validation. We examine these issues in the context of the evolving Semantic Web and the DCMI commitment to RDF and the challenges presented. We frame the discussion in terms of select functions to be served by application profiles and our notion of data profiles. While much remains to be done to address these issues, positive movement toward solutions is dependent on the appropriate framing of those issues in terms of the needs of large-scale applications such as metadata aggregators.