A TEXT CREATION PARTNERSHIP
Companion

WELCOME

This site takes up where the University of Michigan Library's WordPress site (now discontinued) left off. It is designed to preserve and make available facts about the history and policies of the TCP project that may be of continued value or interest. Much of the information here was generated during the production of the texts; much by engagement with users; more may well be added as users demand it.

A few matters present in the WordPress site have been omitted: links to projects that use the TCP texts have been dropped (there are too many, and maintaining a set of links is more trouble than it is worth); so have most of the links to presentations designed to promote participation in the partnership. Links to YouTube tutorials on using the TCP interface have been dropped as obsolete; links to matters concerning partnership agreements and arising during vendor negotiations have been dropped as of only historical interest. Anyone looking for that material need only ask: it has all been backed up offline. Though created with the approval and assistance of the University of Michigan Library, this is an unofficial site, independently maintained by editors of the Text Creation Partnership 'present at the beginning' in 1999 and engaged continuously with the project ever since. Questions should continue to be addressed to tcp-info@umich.edu.
At present, the following matters are included:

  1. This page (index and introduction)
  2. FAQ: The mandatory list of Frequently Asked Questions.
  3. Capture and encoding decisions (production documents)
  4. TCP staff and contributors (to give credit where it is due)
  5. About the TCP and the TCP projects
  6. Why not OCR?
  7. Why (and what flavor of) markup?
  8. The original TCP partners gratefully listed
  9. Citing the TCP texts

INTRODUCTION

The Text Creation Partnership was conceived in 1999 as a joint enterprise between the University of Michigan Library, Bodleian Libraries at the University of Oxford, ProQuest, and the Council on Library and Information Resources. ProQuest provided millions of page images from its EEBO product for transcription, as well as catalog records and other metadata. Gale/Cengage later contributed its ECCO images, and Readex/Newsbank its early American ("Evans") images to ECCO-TCP and Evans-TCP respectively. The governing principle of the project saw it as fundamentally library-based and library-informed -- as an innovative way for libraries around the world to:

As of today (March 2026), the project has produced approximately 77,000 accurate, searchable, full-text transcriptions of early printed books, which were previously only available as unsearchable page images.  The titles transcribed represent roughly 90% of surviving early (pre-1700) printed English titles, as well as a healthy smattering of French, and Latin, and nearly complete coverage of early printed Welsh and Scottish Gaelic. Approximately 5,000 early (pre-1800) American imprints are also included, again mostly in English, along with a sampler of about 3,000 18th-century British imprints. Most of these full-text transcripts can be found, searched, and read in the following digital collections:  About 4,000 EEBO titles, 600 ECCO titles, and a handful of Evans titles have yet to go online, but are available as bulk downloads, as are all of the online files, in a variety of formats.