EAGER: Revolutionizing Wikipedia’s Relationship with New and Emerging Knowledge

Millions of people view Wikipedia’s articles daily, and many services rely on the quality of Wikipedia to power their business and AI ventures. This project will contribute new and validated, high-fidelity designs for the fundamental mechanisms Wikipedia uses to organize knowledge, with the goal of enriching the types of knowledge it includes and expand the evidentiary standards it employs. Wikipedia's success rests on a radical knowledge production process—anyone can edit the articles—coupled with a strict evidentiary epistemology that mandates "reliable sources" and a "neutral point of view". But along with the success has come systematic problems and controversies over what knowledge is "valid." These processes may have the effect of excluding much valuable knowledge, including (1) knowledge based on different ways of knowing, such as oral histories; (2) less established knowledge, including emerging knowledge about new medical treatments and therapies; and (2) knowledge directed toward different goals, for example personal experiences that contextualize textbook knowledge. These exclusions occur through the decisions of Wikipedia editors: which edits they allow and which they undo and what rules they enforce. Research will fill the theory-practice gap in Wikipedia research via well-evaluated, high-fidelity prototypes that can change the evidentiary practices in the online encyclopedia. This goal will be accomplished through four phases of work: Phase 1: Translate theory and practice into low-fidelity prototypes for new knowledge representations. Phase 2: Validate initial designs with a small panel of Wikipedia experts. Phase 3: Develop the most promising ideas from Phase 2 into high-fidelity, functional prototypes. Phase 4: Evaluate the high-fidelity prototypes through an asynchronous online community for design critique and generation. Expected scientific outcomes of this project include new design prototypes and policies for Wikipedia that expand and enrich its knowledge-production epistemology, and expert and stakeholder insights into the potential feasibility, efficacy, and acceptability of the designs and policies. This multi-stage, iterative design process is applicable to other complex and risky arenas, such as designing interventions for credibility assessment in social platforms.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.