EEBO-TCP Gap Resolution

This website combines two different tools into a single workflow that supports finding incompletely or incorrectly transcribed words in EEBO-TCP texts, fixing the defective word(s) and logging the correction in a manner that allows for its review and automatic integration into the source text. At the moment the texts offered for correction are Early Modern plays from the EEBO-TCP corpus, but the workflow can be used for any TCP texts.

The tools consist of LibraryFinder, developed at Washington University, and AnnoLex, developed at Northwestern University. LibraryFinder lets you find which plays are held in US Libraries, as well as the Bodleian and the British Library. AnnoLex helps you locate the defects in a given play, lets you enter corrections, and logs their ‘who’, ‘what’, ‘when’ and ‘where’.

AnnoLex provides access to the digital scan from which the text was transcribed, but many defects cannot be satisfactorily cured without consulting an original copy in one of the libraries that hold it.

Find It

You can start the process by finding a play in AnnoLex. If the digital image is inadequate a single click will take you to LibraryFinder’s data about where originals can be found.

Alternately, you can use LibraryFinder to learn about original copies in a library near you. Once you have chosen a play, a single click takes you to the page in AnnoLex that holds the defects in need of correction.

Fix It

You are the person who fixes a defect. You do so by employing the edit routines of AnnoLex. What you enter there does not overwrite the text but creates a distinct record in a separate table

Log It

AnnoLex keeps track of who does what, when, and where. Your contributions are reviewed before being integrated into the source texts. The cumulative list of corrections become a permanent part of the edited text.


All of the bibliographical and most of the curatorial work was done by undergraduates at Northwestern University and Washington University, with light supervision by Joseph Loewenstein and Martin Mueller. Kate Needham and Lydia Zoells at Washington University compiled the data for LibraryFinder. Nayoon Ahn, Hannah Bredar, Madeline Burg, Nicole Sheriko, and Melina Yeh at Northwestern University fixed some 30,000 defects in about 500 plays.

Stephen Pentecost designed LibraryFinder, Craig Berry designed AnnoLex. The data structure for the curatorial workflow depends heavily on the linguistic annotation of the texts with MorphAdorner, designed by Phil Burns.