Describes the "CORC at Cornell" project undertaken by a small cross-functional team. Using CORC and Dublin Core (DC) as a framework, the team explored a distributed model for Internet resource description, in which catalogers, selectors, and reference specialists participated in producing DC and MARC metadata for the library's catalog and gateway. After describing the experimental workflow that was developed and tested in the project, the author analyzes project results. Findings indicate that distributed resource description is both feasible and beneficial.