A Generalized Framework for Participatory Transcription: Lessons, Challenges, and Interdisciplinary Impact from the Zooniverse Platform

Text transcription is one of the most common forms of participatory activity undertaken by staff, researchers, and community groups who use collections from galleries, libraries, archives and museums (GLAMs) in different disciplinary contexts including citizen history or humanities and citizen science. This article furthers participatory transcription theory and practice by providing a framework of four common challenges in the creation, design, and execution of participatory transcription projects: 1) variety; 2) units of transcription; 3) single-track transcription or multi-track transcription with aggregation; and 4) bias. For each challenge, we articulate guiding questions based on participatory transcription and citizen science literature, conversations with other practitioners, and our own contributions to several large transcription projects and platforms including Zooniverse, which we present as a case study. Through this case study, we illuminate the affordances and challenges of participatory transcription and data aggregation methods through reference to multiple projects, publications, and hitherto unpublished information. We find that different project or platform design choices interact with the baseline challenges of text as a data type. This framework draws on and can be applied to other transcription projects and platforms beyond Zooniverse, and will support deeper understanding of participatory transcription practice by those creating new projects and those working with existing data. We conclude that, despite the rise of AI and threats to democratic institutions, participatory projects are vital in bringing together GLAMs, researchers, and the public in uniquely powerful ways.