Scribe: Turning Text into Structured Information through the Power of the Crowd

FAIN: HK-50128-13

New York Public Library (New York, NY 10016-0109)
Benjamin Vershbow (Project Director: January 2013 to May 2016)

Further development of Scribe, an open-source, extensible software platform for crowdsourced transcription of cultural heritage collections, including tools for transcription management, quality control, and data sharing.

The New York Public Library seeks to partner with the renowned citizen science team at Zooniverse to build Scribe: an open source transcription engine geared toward flexible, structured data extraction from a wide range of humanities documents. The final deliverable would be a developer-ready, open source transcription engine with an adjustable data schema, enabling scholars and curators to easily decide on a wide range of interactions with a document, keeping technical hurdles to a minimum. Based on an initial prototype built by Zooniverse, the tool would be developed collaboratively over two years through the release of several new humanities projects from the respective teams. At the end of the project, all code would be open sourced alongside a website publishing best practices and live demos of the various out-of-the-box modes of the engine.

Associated Products