Scribe is a highly configurable, open source framework for setting up community transcription projects around handwritten or OCR-resistant texts. Scribe is particularly geared toward digital humanities, library, and citizen science projects seeking to extract highly structured, normalizable data from a set of digitized materials (e.g. historical manuscripts, account ledgers, catalog cards, or maritime logbooks).

Scribe is not a crowdsourcing project in a box, but it establishes the foundation for a developer to configure and launch a project far more easily than if starting from scratch. Informed by previous crowdsourcing efforts by NYPL Labs and Zooniverse, Scribe handles many of the technical operations common across transcription projects and offers highly configurable task flows and schemas, allowing project administrators to focus on the specifics of their materials and audience.

One of Scribe’s more distinctive affordances is a division of labor model intended to lower barriers to participation and to ensure higher quality results. Scribe breaks the crowd’s work into three discrete task flows: Marking (identifying document structure), Transcription (data entry), and Verification (quality control). Accuracy is further ensured by automated transcription analysis that helps determine consensus among contributors.

A good deal of thought and work has gone into Scribe, but it is most definitely a first draft. We hope you’ll work with us to improve it. Or fork it and take it in other directions.  Read more...