Text Encoding Initiative

The TEI was founded in 1987 at a conference organized by the Association for Computers and the Humanities, with initial funding from the U.S. National Endowment for the Humanities and the European Union. Its goal was to create a common encoding scheme for literary and linguistic texts, addressing the fragmentation of early digital projects. The first full version of the Guidelines (P1) appeared in 1990, followed by P2 in 1994, which became widely adopted. Subsequent revisions (P3, P4, P5) have kept pace with evolving web standards, with P5 (released in 2007) being the current stable version, maintained through a continuous release process.

The TEI Guidelines define a modular schema that can be customized for specific projects. The core module includes elements for text structure (e.g., <div>, <p>), while additional modules cover features like verse, drama, manuscript description, and linguistic annotation. The TEI uses a customization mechanism called ODD (One Document Does it all), which allows users to specify which elements and attributes they need, generating a tailored schema in Relax NG or W3C XML Schema. This flexibility has made TEI applicable to a wide range of text types, from ancient inscriptions to modern novels.

TEI is the de facto standard for digital scholarly editions, used by major projects such as the Women Writers Project, the Perseus Digital Library, and the Oxford Text Archive. It underpins national corpora like the British National Corpus and the Text Creation Partnership. Beyond academia, TEI has been adopted by libraries and archives for digitization, enabling interoperability and long-term access. The TEI Consortium, a nonprofit membership organization, oversees the standard's governance, with an annual conference and a vibrant community of scholars and developers.¹

Beyond its mainstream use, TEI has niche applications: it is used for encoding endangered languages, enabling documentation and revitalization efforts. The TEI also includes modules for performance texts, such as theater scripts and musical scores, with specific elements for stage directions and notation. A lesser-known feature is the <teiCorpus> element, which allows multiple texts to be combined into a single corpus file. The TEI's history includes early debates about the role of SGML, its predecessor, and the shift to XML in the late 1990s. The TEI also supports the encoding of critical apparatus, allowing for complex textual variants, a feature that has been crucial for genetic criticism.²

ODD

One Document Does it all: a TEI specification format that defines the schema and documentation in a single source.

SGML

Standard Generalized Markup Language, the predecessor to XML, on which early TEI was based.

Critical apparatus

The set of notes and variants that accompany a scholarly edition, which TEI can encode.