Making an edition of a text in many versions

One of the most difficult tasks in the scholarly world is to make an edition from a text in many versions. It is also an area where digital methods show immense possibilities, as computers offer ways of managing large quantities of structured data, of exploring the data, publishing the results, and working with other scholars to do all these things.

This two-week workshop will teach how to use the Textual Communities environment to carry out every stage of preparing a scholarly edition of a text in many versions. Students will learn how to gather images of many versions of texts together, to transcribe their text into TEI/XML, how to collate the versions and prepare an apparatus, how to use phylogenetic, variant database and other tools to analyze the agreements and disagreements between the versions to create stemmata and other visualizations, and how to publish the edition online. Models of the kind of edition which users will be able to make can be seen at (for example, Dante’s Commedia at; part of the Canterbury Tales at

The course will be taught in two one-week blocks:

  • Week 1: Data collection and preparation: gathering images, transcribing, starting collation
  • Week 2: Creating an apparatus, analysis and publication: analysis of the apparatus by phylogenetics and VBase; creation of stemma and variant maps; textual commentary and publication.

The two weeks are distinct and could be taken separately. However a full mastery of all the aspects of making an edition of a text in many versions will require both weeks.

Week one: Data collection and preparation. Nine classes

  1. Introduction. What are: texts, documents, works and how Textual Communities treats them. The architecture of Textual Communities.
  2. Analysis of the text to create a base TEI/XML text structure for each witness to be edited, with image references and base TEI Header
  3. Validation and upload of first witness files to textual communities
  4. Gathering and online presentation of witness images (from manuscripts, printed books, etc) from multiple sources. Upload of single images one at a time; or upload of multiple images by IIIF manifest, or from a zipped folder of images; referencing of other online images, linked to witness
  5. Linkage of text and image files. Commencing transcription of text files using TEI/XML: base text structure
  6. Transcription: treatment of abbreviations, scribal revisions, marginalia, multiple columns, other phenomena
  7. Recruitment and management of a large team of project partners in multiple roles (transcribers, approvers, co-editors, collators etc)
  8. Introduction to collation, using the Collation Editor built into Textual Communities
  9. Advanced use of the Collation Editor: regularization, variant setting

Week two: Creating an apparatus, analysis and publication. Nine classes

  1. Export of completed apparatus created by the Collation Editor in multiple forms,
  2. Introducing NEXUS files and Phylogenetics. Creating hypotheses of textual relations (“stemmata”) using PAUP (“Phylogenetic analysis using parsimony”)
  3. Importing apparatus files into VBase. Introducing VBase searches.
  4. Creating and using variant maps.
  5. Using stemmata / variant maps / VBase towards writing a textual commentary
  6. Introducing the Textual Communities API and JSON
  7. Using HTML to create the base edition
  8. Elaborating the edition: adding images, transcripts, collations through the Textual Communities API
  9. Completing the edition. Adding commentary, introductions, etc.


  1. No knowledge of TEI or XML or Javascript / HTML or existDB or XSLT or any other specialized technologies needed. A knowledge of basic TEI/XML, and of javascript / HTML is an advantage but not necessary
  2. Access to existing digital edition materials would be useful. These could be online, or in the form of files on your computer.
  3. Participants should bring their own laptops, with whatever existing images and transcripts they have.