Linked data is the language of the web, and is a profoundly important opportunity for libraries to expand their reach. This talk will highlight how University of Alberta and University of Pennsylvania are leveraging key community partnerships (Share-VDE, LD4, PCC, Wikidata) and institutional strategic objectives to drive implementation choices, presenting generalizable approaches to successful linked data strategies in libraries. A general linked data vision takes a holistic approach to library implementation by supporting overall library strategic goals and objectives. Linked data objectives are not a separate or compartmentalized goal, but rather the overarching long term effects of linked data support library operations. Indeed, as library linked open data have proliferated on the web, linking external datasets with classic sources of library information have led to increased exposure of library resources, collections, and expertise. This talk will contextualize linked data from the perspective of current systems; extensions that can be made from where any library is at; and from any level of resource constraint. It is often possible to include linked data in more traditional representations, or to make connections between linked data and more familiar formats. This presentation will introduce examples of these mixed-format or “hybrid” linked data environments as they are expected to be the most common way in which linked data is used in production in the next few years. For the foreseeable future, linked data strategy is based on experimentation, innovation, and collaboration. This will encompass work in a hybrid environment that will include MARC records, sometimes enhanced with linked data identifiers, as well as more “native” linked data formats.